United Nations Scientific Committee on the Effects of Atomic Radiation

S O U R C E S , E F F E C T S A N D R I S K S O F I O N I Z I N G R A D I AT I O N

UNSCEAR 2016 Report

Report to the General Assembly

SCIENTIFIC ANNEXES A, B, C and D
SOURCES, EFFECTS AND RISKS
OF IONIZING RADIATION
United Nations Scientific Committee on the
Effects of Atomic Radiation

UNSCEAR 2016
Report to the General Assembly,
with Scientific Annexes

UNITED NATIONS
New York, 2017
 NOTE

The report of the Committee without its annexes appears as Official Records of the General
Assembly, Seventy-first Session, Supplement No. 46 and corrigendum (A/71/46 and Corr.1). The
report reproduced here includes the corrections of the corrigendum.

The designations employed and the presentation of material in this publication do not imply
the expression of any opinion whatsoever on the part of the Secretariat of the United Nations
concerning the legal status of any country, territory, city or area, or of its authorities, or

concerning the delimitation of its frontiers or boundaries.

The country names used in this document are, in most cases, those that were in use at the time the
data were collected or the text prepared. In other cases, however, the names have been updated,
where this was possible and appropriate, to reflect political changes.

UNITED NATIONS PUBLICATION

Sales No. E.17.IX.1
ISBN: 978-92-1-142316-7
eISBN: 978-92-1-060002-6

United Nations, January 2017. All rights reserved, worldwide.

This publication has not been formally edited.

Information on uniform resource locators and links to Internet sites contained in the present
publication are provided for the convenience of the reader and are correct at the time of issue.
The United Nations takes no responsibility for the continued accuracy of that information or for
the content of any external website.
 CONTENTS

Page
Report of the United Nations Scientific Committee on the Effects of Atomic Radiation
to the General Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Scientific Annexes
Annex A. Methodology for estimating public exposures due to radioactive discharges. . . . . . . . . . . . . . . . . . . . . . . 19
Annex B. Radiation exposures from electricity generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Annex C. Biological effects of selected internal emittersTritium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Annex D. Biological effects of selected internal emittersUranium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

iii
 Report of the United Nations Scientific Committee on the
Effects of Atomic Radiation to the General Assembly
Contents
Chapter Page

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
II. Deliberations of the United Nations Scientific Committee on the Effects of Atomic
Radiation at its sixty-third session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
A. Completed evaluations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
B. Present programme of work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1. Developments since the Committees 2013 report on the levels and effects of
radiation exposure due to the nuclear accident following the great east-Japan
earthquake and tsunami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Cancer epidemiology of exposures at low dose-rates due to environmental
radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3. Selected evaluations of health effects and risk inference from radiation exposure . . 4
4. Collection of data on radiation exposures, in particular on medical and
occupational exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
5. Outreach activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
C. Long-term strategic directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
D. Future programme of work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
E. Administrative issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
III. Scientific report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
A. Methodology for estimating public exposures due to radioactive discharges . . . . . . . . . . 9
B. Radiation exposures from electricity generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
C. Biological effects of selected internal emitters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Appendices
I. Members of national delegations attending the fifty-seventh to sixty-third sessions of the
United Nations Scientific Committee on the Effects of Atomic Radiation . . . . . . . . . . . . . . . . . 16
II. Scientific staff and consultants cooperating with the United Nations Scientific Committee on
the Effects of Atomic Radiation in the preparation of its scientific report for 2016 . . . . . . . . . 18

v
 REPORT TO THE GENERAL ASSEMBLY 1

Chapter I
Introduction
1. Since the establishment of the United Nations Scientific Committee on the
Effects of Atomic Radiation by the General Assembly in its resolution 913 (X) of
3 December 1955, the mandate of the Committee has been to undertake broad
assessments of the sources of ionizing radiation and its effects on human health and
the environment. 1 In pursuit of its mandate, the Committee thoroughly reviews and
evaluates global and regional exposures to radiation. The Committee also evaluates
evidence of radiation-induced health effects in exposed groups and advances in the
understanding of the biological mechanisms by which radiation-induced effects on
human health or on non-human biota can occur. Those assessments provide the
scientific foundation used, inter alia, by the relevant agencies of the United Nations
system in formulating international standards for the protection of the general
public, workers and patients against ionizing radiation; 2 those standards, in turn, are
linked to important legal and regulatory instruments.
2. Exposure to ionizing radiation arises from naturally occurring sources (such as
radiation from outer space and radon gas emanating from rocks in the Earth) and
from sources with an artificial origin (such as medical diagnostic and therapeutic
procedures; radioactive material resulting from nuclear weapons testing; energy
generation, including by means of nuclear power; unplanned events such as the
nuclear power plant accidents at Chernobyl in 1986 and that following the great
east-Japan earthquake and tsunami of March 2011; and workplaces where there may
be increased exposure to radiation from artificial or naturally occurring sources).

__________________
1 The United Nations Scientific Committee on the Effects of Atomic Radiation was established by
the General Assembly at its tenth session, in 1955. Its terms of reference are set out in
resolution 913 (X). The Committee was originally composed of the following Member States:
Argentina, Australia, Belgium, Brazil, Canada, Czechoslovakia (later succeeded by Slovakia),
Egypt, France, India, Japan, Mexico, Sweden, Union of Soviet Socialist Republics (later
succeeded by the Russian Federation), United Kingdom of Great Britain and Northern Ireland,
and United States of America. The membership of the Committee was subsequently enlarged by
the Assembly in its resolution 3154 C (XXVIII) of 14 December 1973 to include the Federal
Republic of Germany (later succeeded by Germany), Indonesia, Peru, Poland and the Sudan. By
its resolution 41/62 B of 3 December 1986, the Assembly increased the membership of the
Committee to a maximum of 21 members and invited China to become a member. In its
resolution 66/70 of 9 December 2011, the Assembly further enlarged the membership of the
Committee to 27 and invited Belarus, Finland, Pakistan, the Republic of Korea, Spain and
Ukraine to become members.
2 For example, the international basic safety standards for radiation protection and safety of
radiation sources, currently co-sponsored by the European Commission, the Food and
Agriculture Organization of the United Nations, the International Atomic Energy Agency, the
International Labour Organization, the Nuclear Energy Agency of the Organization for
Economic Cooperation and Development, the Pan American Health Organization, the United
Nations Environment Programme and the World Health Organization.
2 UNSCEAR 2016 REPORT

Chapter II
Deliberations of the United Nations Scientific Committee on
the Effects of Atomic Radiation at its sixty-third session
3. The Committee held its sixty-third session in Vienna from 27 June to 1 July
2016. 3 Yoshiharu Yonekura (Japan), Chair; John Hunt (Brazil), Peter Jacob
(Germany) and Hans Vanmarcke (Belgium), Vice-Chairs; and Michael Waligrski
(Poland), Rapporteur, served as officers of the Committee.
4. The Committee took note of General Assembly resolution 70/81 on the effects
of atomic radiation. It recalled that it had expected to report its long-term strategic
directions beyond the period covered by its present strategic plan (2014-2019), so as
to help to inform future deliberations of the Assembly on the Committees
membership.

A. Completed evaluations

5. The Committee discussed four substantive evaluations in detail, adopted the

scientific report based on the findings of those evaluations (see chapter III) and
requested that the scientific annexes be published in the usual manner, subject to the
agreed modifications.
6. The Committee had decided, at its fifty-sixth session, to initiate work on a new
estimation of human exposures to ionizing radiation from electricity generation.
Accordingly it had decided to review and update its previous methodology for
estimating public exposures from discharges published in its 2000 report. The
Committee discussed and approved for publication the scientific annex updating the
methodology and associated electronic workbooks.
7. The Committee recalled that progress on the scientific annex on radiation
exposures from electricity generation had been hampered by, among other things,
gaps in data on occupational exposures and on releases associated with electricity
generated from non-nuclear energy sources. In comparison, there were abundant
data for the nuclear energy industry, although those data remained somewhat
deficient as regards decommissioning and other aspects of the so-called back end of
the nuclear fuel cycle. The evaluation has been completed on the basis of reasonable
and transparent assumptions where precise data were not available. The electronic
workbooks for implementing the methodology had been used in 2015 to complete,
in an internally consistent manner, the assessment of radiation exposures of
populations from various types of electricity generation.
8. At its fifty-sixth session, held from 10 to 18 July 2008, during deliberations on
its future programme of work, the Committee had decided that work should be
undertaken to address radiation doses and the risks and effects from internally
deposited radionuclides. At its fifty-seventh session, held from 16 to 20 August 2010,
__________________
3 The sixty-third session was attended by observers for the International Atomic Energy Agency,
the International Labour Organization, the World Health Organization, the European Union, the
International Agency for Research on Cancer, the International Commission on Radiological
Protection and the International Commission on Radiation Units and Measurements.
 REPORT TO THE GENERAL ASSEMBLY 3

the Committee had further decided to focus on tritium and radioisotopes of uranium.
At the current session, the Committee agreed that the review of the literature was
now complete, that the material had been streamlined and its structure harmonized,
and that final conclusions had been drawn from the material evaluated. The
Committee accordingly approved the evaluations for publication.

B. Present programme of work

1. Developments since the Committees 2013 report on the levels and effects of
radiation exposure due to the nuclear accident following the great east-Japan
earthquake and tsunami
9. The Committee recalled its assessment of the exposures and effects due to the
nuclear accident after the 2011 great east-Japan earthquake and tsunami, as
presented in its report to the sixty-eighth General Assembly in 2013 (A/68/46) and
the supporting detailed scientific annex. 4 It had concluded in that report that, in
general, doses were low and that therefore associated risks were also likely to be
low. Cancer rates were expected to remain stable. Nevertheless, in the report the
Committee had noted a possibility that the risk of thyroid cancer among those
children most exposed to radiation could increase. However, it also noted that the
likelihood of a large number of radiation-induced thyroid cancers in Fukushima
Prefecture such as after the Chernobyl accident could be discounted because
absorbed doses to the thyroid after the Fukushima accident were substantially lower.
It had concluded that no discernible changes in the incidence of birth defects and
hereditary disease were expected, and that the effects on terrestrial and marine
ecosystems would be transient and localized. Cancer rates for workers were
expected to remain stable.
10. Following its assessment, the Committee put in place arrangements for
follow-up activities to enable it to remain abreast of additional relevant information
as it was published. The Committees report of its sixty-second session to the
seventieth General Assembly included the findings from the follow-up activities it
had conducted up to that time.
11. The Committee continued to identify further information that had become
available up to the end of 2015, and systematically appraised relevant new
publications to assess their implications for its 2013 report. A notable publication
was the report by the International Atomic Energy Agency (IAEA) on the accident
at the Fukushima Daiichi nuclear power station. 5 It describes the accident and its
causes, evolution and consequences based on an evaluation of data and information
from a large number of sources available at the time it was written. That report and
a large proportion of the new publications again confirmed the main assumptions
and findings in the Committees 2013 report. None of the publications materially
affected the main findings in the 2013 report or challenged its major assumptions.
Several publications were identified for which further analysis or more conclusive
evidence from additional research was needed. On the basis of the material
reviewed, the Committee saw no need, at the current time, to make any change to its
__________________
4 United Nations publication, Sales No. E.14.IX.1.
5 International Atomic Energy Agency, The Fukushima Daiichi Accident: Report by the Director
General (GC(59)/14), accompanied by technical volumes 1-5.
4 UNSCEAR 2016 REPORT

overarching conclusions. However, several of the research needs identified by the

Committee had yet to be addressed fully by the scientific community.
12. The Committee plans to continue to identify and systematically appraise new
information on the accident, and evaluate the outcomes periodically at its annual
sessions. It also plans to actively engage with those responsible for formulating,
implementing and advising on major research programmes in Japan, in order to
rapidly assimilate emerging issues, and highlight questions needing further research.
At an appropriate time, dependent on the outcomes, the Committee expects to
consider the need to update its 2013 report.
13. The Committee requested the secretariat, subject to available resources, to
publish the findings of its systematic review of new scientific literature as a
non-sales publication in English and also to foster its publication in Japanese.

2. Cancer epidemiology of exposures at low dose-rates due to environmental

radiation
14. The Committee discussed progress on an evaluation of epidemiological studies
of cancer incidence from low-dose-rate exposures due to environmental sources of
radiation. The Committee acknowledged that the scientific review had improved
considerably. It welcomed the development of an appendix on quality criteria for the
Committees reviews of epidemiological studies. The Committee requested that the
scientific review and quality criteria now be brought into accordance with each
other. It requested that the appendix be finalized for publication as an independent
annex because of its wider application, and expected that both the review and
quality criteria could be approved for publication at the sixty-fourth session.

3. Selected evaluations of health effects and risk inference from radiation exposure
15. The Committee considered progress on evaluations of selected health effects
and the inference of risk from exposure to ionizing radiation. Four scenarios were
proposed for evaluation based on agreed criteria and preliminary literature reviews:
leukaemia after exposure at low dose; solid cancer risk after acute and protracted
exposure; thyroid cancer risk after exposure during childhood or adolescence; and
risk of circulatory diseases after acute and protracted exposure. The Committee
expected that the evaluations would be conducted in line with quality criteria for the
Committees reviews of epidemiological studies, and expected to discuss draft
evaluations at the sixty-fourth session.

4. Collection of data on radiation exposures, in particular on medical and

occupational exposures
16. The Committee took note of a progress report by the secretariat on the
collection, analysis and dissemination of data on radiation exposures, in particular
on medical and occupational exposures. The Committee welcomed the fact that the
General Assembly, in its resolution 70/81, had encouraged Member States to
nominate a national contact person to facilitate coordination of the collection
and submission of data on the exposure of the public, of workers and of patients.
Fifty-one Member States had nominated national contact persons by the
sixty-third session of the Committee.
 REPORT TO THE GENERAL ASSEMBLY 5

17. In 2014, the secretariat had launched an online platform for the collection of
data on medical exposures and had invited all Member States to take part in the
Committees Global Survey of Medical Radiation Usage and Exposures. In preparation
for the Global Survey it had fostered close cooperation with IAEA, the
World Health Organization and the International Radiation Protection Association.
Twenty countries had submitted their first data on medical exposure; however, not
all their submissions were complete. Because of the relatively low response rate to
date, and because of delays brought about by changes in the United Nations
administrative and financial platform (Umoja), the cut-off date for data submission
would be extended until May 2017. The Committee requested the secretariat to
prepare a first evaluation of the results for the Committees review at its sixty-fourth
session, including a detailed literature review. It also requested the secretariat to
accelerate the survey on occupational exposures, fostering close cooperation with
the International Labour Organization and other relevant bodies, and to begin
detailed work on defining and collecting data on public exposures to radiation from
natural and artificial sources.

5. Outreach activities
18. The Committee took note of a progress report by the secretariat on outreach
activities. It acknowledged in particular the work done in Japan to disseminate the
Committees 2013 report on the levels and effects of radiation exposure due to the
Fukushima Daiichi accident and the white paper on developments since that report.
It noted that the General Assembly had encouraged the secretariat to continue to
disseminate the findings to the public. The Committee also welcomed the outreach
activities surrounding the sixtieth anniversary of the Committees inception, the
thirtieth anniversary of the Chernobyl accident, and the fifth anniversary of the
nuclear accident in Japan. The updated publication of the United Nations
Environment Programme (UNEP) entitled Radiation: Effects and Sources, which is
intended as a basic scientific guide for the public, was published in English;
publication in other languages is envisaged. The secretariat had also prepared a
memory stick preloaded with all the Committees publications and all the
resolutions relevant to its activities, in all official languages of the United Nations
where available, as a handy reference tool.
19. With regard to the sixtieth anniversary of the Committee, the Mayor and
Governor of the City of Vienna hosted a reception for invited dignitaries, scientists
and diplomats at the Vienna Town Hall to commemorate the anniversary. The
Secretary-General of the United Nations, Ban Ki-moon, sent a video message for
the occasion, in which he said: From assessing the significance of fallout in the
1950s to evaluating the effects of radiation on the human genome today, the
Committee has always taken an independent and impartial approach. This is crucial
on issues that are often highly emotional and political. Other speakers delivered
messages from the heads of their organizations, including the World Health
Organization, IAEA, the Preparatory Commission for the Comprehensive
Nuclear-Test-Ban Treaty Organization and UNEP. The messages commended the
Committee for the expertise and independence shown in its scientific reviews,
lauded its efforts to share its scientific findings with a broader audience and
encouraged it to further enhance those efforts.
6 UNSCEAR 2016 REPORT

C. Long-term strategic directions

20. The Committee considered its long-term strategic directions beyond the period
covered by its present strategic plan (2014-2019). The Committee took note of the
report of the Secretary-General to the General Assembly at its sixty-ninth session on
the impact of the increase in the membership to 27 States, and possible approaches
to further increase procedures (see A/69/350). The Committee also took note of
General Assembly resolution 70/81 on the effects of atomic radiation, in which the
Assembly requested the Secretary-General to provide it at its seventy-second
session with a list of the Member States that had expressed their particular interest
in membership in the Committee between the sixty-sixth and seventy-second
sessions.
21. The Committee envisages to direct its future work mainly at the following
scientific areas:
(a) Improving the evaluation of exposure levels for people in everyday life,
in occupational environments, during medical procedures, and as a result of
accidents;
(b) Improving the understanding of mechanisms of radiation action and
biological reaction at all levels of biological organization, i.e. from the molecular
level to the population level;
(c) Obtaining more definitive evidence relating to health effects, in
particular health effects from low-dose-range and chronic exposure, and sound
estimates of the health implications of exposure of populations to radiation.
22. The Committee also expects that rapidly emerging issues or significant events
may lead to short-term or longer-term reprioritization, and the programme of work
is changed accordingly at each session. As an example, in recent times the
Committee redirected its efforts towards a timely scientific evaluation of the levels
and effects of radiation exposure due to the 2011 nuclear accident in Japan. 4
23. The Committee is of the view that it will be able to continue delivering
authoritative scientific evaluations in the scientific areas outlined above. The
Committee fully supports the Secretary-Generals view that the primary purpose of
any increase in the number of States members should be to enhance the capability of
the Committee to conduct its scientific work. The Committee believes that there is a
limit of about 30 States members with which the Committees secretariat at its
present size could reasonably cope while at the same time supporting the scientific
work of the Committee. Any increase above that number would require further
strengthening of the secretariats human resources (see paragraphs 35 and 40 of
A/69/350).
24. The Committee thus considers that any discussion of membership should focus
on the Committees ability to continue its delivery of authoritative scientific
evaluations as well as on the secretariats ability to support the Committee in doing
so. However, in view of the ever-increasing scientific database it may be necessary
to implement a range of strategies that will support the Committees efforts to serve
the scientific community as well as wider audiences. Such strategies may also allow
for the inclusion of scientists outside the Committees current membership. There
already are examples of such arrangements, and they have proved beneficial to the
 REPORT TO THE GENERAL ASSEMBLY 7

Committees work while causing only minor or negligible additions to the workload
of the secretariat.
25. While recognizing the importance of including all States members in, inter alia,
the implementation of the Committees strategies, future deliberations and the
production of scientific documents, and while giving due regard to available
resources, the Committee could consider including the following in the strategies
referred to in the previous paragraph:
(a) Establishing standing working groups focused on areas such as sources
and exposure, or health and environmental effects;
(b) Inviting, on an ad hoc basis, scientists from other States Members of the
United Nations to participate in evaluations regarding the above areas;
(c) Increasing the Committees efforts to present its evaluations, and
summaries thereof, in a manner that attracts readers without compromising
scientific rigour and integrity;
(d) While maintaining its lead in providing authoritative scientific
evaluations to the General Assembly, liaising closely with other relevant
international bodies to avoid duplication of efforts to the extent possible.
26. Over the coming sessions, the Committee will work towards implementing the
above strategies.

D. Future programme of work

27. The Committee discussed preliminary plans for five projects and two smaller
activities. The five topics for which projects were proposed were: (a) second cancers
after radiotherapy; (b) assessment of the impact on biota of radiation exposure due
to the nuclear industry; (c) biological mechanisms that may influence health effects
from low-dose radiation exposure; (d) effects of exposure to radon in homes and
workplaces; and (e) epidemiological studies of radiation and cancer. Having
considered the current work programme and the capacity of both the Committee and
its secretariat, the Committee decided:
(a) To start projects in 2016 based on topics (c) and (d), and to focus the
project for topic (c) on cancer and hereditary effects;
(b) To start a project in 2017 based on the proposal for topic (e) in a version
further elaborated by the delegation of the United States of America;
(c) To ask the French delegation to elaborate working material for a more
in-depth discussion on the proposal for topic (a) with a view to accepting the
proposal in 2017.
28. The Committee also requested the secretariat to prepare a short paper on the
scientific view of the Committee on the dose and dose rate effectiveness factor, and
another on the evaluation of thyroid cancer data in regions affected by the nuclear
power plant accident at Chernobyl in 1986, with a view to discussion and
acceptance at the sixty-fourth session of the Committee.
8 UNSCEAR 2016 REPORT

E. Administrative issues

29. The Committee recognized that, because of the need to maintain the intensity
of its work particularly its work to develop exposure databases and to improve
the dissemination of its findings to the public, including in official languages of the
United Nations other than English regular pledges to make voluntary
contributions to the general trust fund established by the Executive Director of
UNEP would be pivotal. The Committee suggested that the General Assembly might
encourage Member States to consider making regular pledges of voluntary
contributions to the general trust fund for that purpose, or to make contributions in
kind.
30. The Committee agreed to hold its sixty-fourth session in Vienna from 29 May
to 2 June 2017. It elected new officers to guide the Committee at its sixty-fourth and
sixty-fifth sessions: Hans Vanmarcke (Belgium), Chair; Peter Jacob (Germany),
Patsy Thompson (Canada), Michael Waligrski (Poland), Vice-Chairs; and Gillian
Hirth (Australia), Rapporteur.
 REPORT TO THE GENERAL ASSEMBLY 9

Chapter III
Scientific report
31. Four scientific annexes (published separately) provide the rationale for the
findings set out below.

A. Methodology for estimating public exposures due to radioactive

discharges

32. From time to time, the Committee has undertaken estimations of public
exposures from radioactive discharges to the environment under normal operations,
primarily from facilities in the nuclear fuel cycle. On each occasion, the Committee
reviewed its methodology for estimating exposures in the light of scientific
developments and, where appropriate, revised it. The Committee decided to update
and extend its past evaluations of human exposures to ionizing radiation from
electricity generation. Consequently, the Committee has reviewed and updated its
previous methodology for estimating public exposures from discharges that had
been published in its 2000 report. Because of the need to be applied more flexibly
for different types of electricity generation and in the interest of transparency, the
methodology was updated to provide results in terms of estimated radiation doses
specific to the discharge of each significant radionuclide.
33. The updated methodology can be used to estimate characteristic individual
doses and collective doses resulting from discharges to the atmosphere, rivers and
lakes, and the sea. Characteristic individual doses are doses indicative of those
received by a typical person living in the area around the discharge point.
A collective dose is the product of the mean dose to a specified population from a
particular source, and the number of people in that population, integrated over a
defined period of time. In other words, a collective dose is the dose received by all
members of a particular population combined, over a defined period of time.
However, the calculated doses are metrics to be used only for the comparison of
different sources of exposure, not to estimate implications for health. Moreover, the
methodology applies only to routine discharges that can be assumed to be
continuous; more sophisticated methodologies are needed to assess exposures from
accidental releases.
34. Radioactive discharges can lead to exposures of the public in a number of
ways, and the updated methodology takes the most important of these into account,
namely exposures from radionuclides external to the body, i.e., in the atmosphere
and on the ground, and exposures from radionuclides inside the body following
intakes by inhalation and ingestion. To enable the estimation of exposures for both
nuclear and non-nuclear forms of electricity generation, the methodology was
extended to cover a wide range of radionuclides. The methodology uses models
based on experimental data and other field observations in order to estimate the
transfer of radionuclides through the environment and thus the resulting exposures
of the public. The updated methodology now takes into account an additional route
of exposure not previously considered, namely the ingestion of crops irrigated with
water that contains radionuclides as a consequence of discharges to fresh water.
10 UNSCEAR 2016 REPORT

35. In the past, world average values for population densities and food
consumption were used because those were considered sufficient to estimate global
exposures from nuclear facilities. However, non-nuclear power stations are found
throughout the world, and population densities and food consumption vary much
more around them. Therefore the Committee has decided to include regional factors.
Even so, the regions now considered are still very large, and other approaches
would be necessary to make assessments for individual sites. Exposures are
estimated using a series of mathematical models, for which the Committee has
chosen parameter values that result in realistic exposure estimates. This is in
contrast to a more cautious approach often used for regulatory purposes, whereby
values are selected so as to deliberately overestimate exposures.
36. As before, estimates can still be made of collective doses to populations on a
local, regional and global scale, as appropriate. In addition, the methodology
provides information on collective doses resulting from a years continuous
discharge into the atmosphere to different population groups as a function of their
distance from the discharge point. The estimates of collective doses to the world
population are now available integrated over periods of 100, 500 and 10,000 years.
37. The methodology has been implemented in a series of electronic workbooks to
provide transparency and facilitate use and revision by the Committee in any future
studies. The workbooks contain information on the most important exposure
pathways and radionuclides and are made available for download on the website of
the Committee (www.unscear.org).
38. The Committee is satisfied that the updated methodology, as implemented in
the workbooks, is robust, builds on the strong position of previous versions and is
suitable for estimating exposures of regional and global populations from routine
discharges of radionuclides to various environments.

B. Radiation exposures from electricity generation

39. The worlds mix of electricity-generating technologies changes over time in

response to the landscape of climatic, environmental, resource, political and
economic challenges. Governments and researchers may conduct various
comparative studies that take into account, among other things, the impact of
different technologies on the public, on workers and on the environment. Exposure
to ionizing radiation is only one of the many factors that may be taken into account
as part of such assessments. However, the Committee considers that an update and
extension of its past evaluations of radiation exposures of the public and of workers
from electricity generation could be a useful source of information for such studies.
40. While interest in the exposure of the public and of workers to radiation due to
electricity generation from nuclear power dates back to the earliest use of the
technology, radiation exposures from the use of other electricity-generating
technologies have not been so comprehensively studied. The Committee has
periodically reviewed exposures of both the public and of workers related to
electricity generation from nuclear power, and has also conducted evaluations for
other forms of electricity generation, albeit to a lesser extent. 6 These evaluations
__________________
6 Sources and Effects of Ionizing Radiation 1977 Report to the General Assembly, with Annexes
 REPORT TO THE GENERAL ASSEMBLY 11

have used a variety of methodologies and relied on data from industrial activities
outside the nuclear sector that are not generally monitored or reported in a
systematic manner, which has made meaningful comparisons between the radiation
exposures from the different electricity-generating technologies challenging.
41. Assessing the collective dose from accidents was out of the scope of the
evaluations of radiation exposures of the public and of workers from electricity
generation; however, the Committee has conducted assessments of past accidents in
its 2008 report; of the Chernobyl accident in its reports of 1988, 2000 and 2008; and
of the Fukushima Daiichi nuclear accident in its 2013 report. It is difficult to make
direct comparisons between exposures from accidents and those resulting from
routine discharges. One of the reasons is that the distribution of doses to the public
immediately after an accident is much more localized geographically, whereas the
collective doses from normal operations for electricity generation are more evenly
distributed over regional or global populations. Nevertheless, the collective dose to
the global population from serious accidents, such as those that occurred at the
Chernobyl and Fukushima Daiichi nuclear power stations, were orders of magnitude
larger than the collective doses to the world population from one years normal
operation of significant technologies of electricity generation, as assessed in the
study.
42. As stated above, the Committee has updated its methodology for estimating
public exposures due to radioactive discharges. The methodology is now more
flexible to be able to address a wider range of electricity-generating technologies. In
addition to including an extensive analysis of the available data, the updated
methodology provides the Committee with a sounder basis for comparative studies
than was possible before. In parallel, the Committee also re-evaluated occupational
exposures arising from different electricity-generating technologies, using data
mainly from dosimetry records of worker exposures. These evaluations comprise the
basis for the current comparative study on radiation exposures of both the public
and of workers from electricity generation.
43. The Committee conducted the comparative study by investigating sources of
exposure from electricity-generating technologies based on nuclear power; the
combustion of coal, natural gas, oil and biofuels; and geothermal, wind and solar
power. Two electricity-generating technologies (nuclear power and coal
combustion) were investigated in detail, because a more robust database existed for
them. The Committee evaluated the main sources of radioactive discharges from
their life cycle. For the life cycle of nuclear power the sources of radioactive
discharges included uranium mining, milling and mill tailings, power plant
operation and reprocessing activities. For the life cycle associated with the
combustion of coal those sources were coal mining, the operation of coal-fired
power plants (both modern and older-style), and deposits of coal ash. For the sake
__________________
(United Nations publication, Sales No. E.77.IX.1); Ionizing Radiation: Sources and Biological
Effects 1982 Report to the General Assembly, with Annexes (United Nations publication,
Sales No. E.82.IX.8); Sources, Effects and Risks of Ionizing Radiation 1988 Report to the
General Assembly, with Annexes (United Nations publication, Sales No. E.88.IX.7); Sources and
Effects of Ionizing Radiation 1993 Report to the General Assembly, with Scientific Annexes
(United Nations publication, Sales No. E.94.IX.2); and Sources and Effects of Ionizing
Radiation 2000 Report to the General Assembly, with Scientific Annexes, Volume I: Sources,
(United Nations publication, Sales No. E.00.IX.3).
12 UNSCEAR 2016 REPORT

of simplicity we will refer to these cycles as the nuclear fuel cycle and the coal
cycle, respectively.
44. To compare exposures, the Committee focused on two metrics. The first
consisted of the collective doses to defined population groups resulting from one
years global and regional electricity generation by each technology, integrated over
specific time periods. The second metric consisted of the relevant collective doses
divided by the amount of electricity generated by each technology. The reference
year used for the comparisons was 2010.
45. The Committee estimated that the contribution from the coal cycle was more
than half of the total collective dose to the local and regional public from the
discharges due to a single years global electricity generation. That estimate was
based on the assumption that the discharges originated from modern coal plants. The
nuclear fuel cycle, on the other hand, contributed less than a fifth. The contribution
from the coal cycle comes from discharges of natural radionuclides (primarily radon
and its radioactive progeny) during coal mining, combustion of coal at power plants
and coal ash deposits. Similarly, almost half of the exposures of the global public
from the nuclear fuel cycle result from discharges of natural radionuclides during
uranium mining and milling activities. These values depend on the share of each
technology in total electricity production; in 2010 the coal cycle contributed about
40 per cent, the largest amount. Although radon and its progeny are relatively
important contributors to the collective doses to the public for both the nuclear fuel
cycle and the coal cycle, the associated individual doses are small compared with
doses due to inhalation of radon and its progeny at levels that occur naturally in
homes.
46. The Committee found, however, that the contribution of a given technology to
the exposures of the global public was not simply a function of how much
electricity that technology generated. There were also differences in the collective
doses per unit of electricity generated by each technology to take into account. In
normal operations, the coal cycle gave a higher collective dose per unit of electricity
generated than the nuclear cycle, and a significantly higher dose per unit of
electricity produced than the other technologies evaluated, with the exception of
geothermal power. Based on the limited information available about radon
discharges from geothermal power plants, the collective dose per unit of electricity
generated by geothermal power could be significant. However, because the use of
geothermal technology is not widespread, its contribution to radiation exposures of
the global public is smaller than that from the coal cycle.
47. Previous investigations into electricity generation from nuclear power have
examined the contribution to public exposures made by long-lived radionuclides,
such as carbon-14, which after being discharged circulate globally and continue to
contribute to radiation exposures of the public centuries into the future, albeit at
extremely small individual doses. The contribution of the globally circulating
radionuclides to the collective dose to the global public depends on the length of
time for which the collective dose is integrated. Public exposures due to one years
discharge of these globally circulating radionuclides continue to increase slowly
over time. Over long integration times, such as hundreds of years, these
radionuclides result in a larger collective dose to the global public from the nuclear
fuel cycle than from the coal cycle.
 REPORT TO THE GENERAL ASSEMBLY 13

48. The Committee also assessed occupational exposures. The largest collective
dose to workers per unit of electricity generated resulted from coal mining, because
of exposures to naturally occurring radionuclides. Of all the collective doses
evaluated, both to the public and to workers, the exposure of workers from coal
mining made the largest contribution, although it has fallen over time because of
improving mining conditions. With regard to the construction phase of the
electricity-generating technologies, by far the largest collective dose to workers per
unit of electricity generated was found in the solar power cycle, followed by the
wind power cycle. The reason for this is that these technologies require large
amounts of rare earth metals, and the mining of low-grade ore exposes workers
to natural radionuclides during mining.
49. The total collective dose per unit of electricity generated in the coal cycle
(i.e., the dose to the global public and all exposed workers combined) was larger
than that found in the nuclear fuel cycle. This held true even if long-lived
globally-circulating radionuclides were integrated over 500 years. When
considering the amount of electricity generated in the year 2010 by each technology,
the coal cycle resulted in the largest collective dose to the global public and workers
combined, followed by the nuclear fuel cycle. Of the remaining technologies,
geothermal energy and the combustion of natural gas were the next largest
contributors.
50. Great care should be taken when interpreting and using these results. Their
only function is to provide an insight into the different magnitudes of radiation
exposure resulting from each technology. They are unfit to be used as the only
metric to determine whether one energy generation technology is preferable to
another. As stated earlier, a number of factors determine why countries may select a
certain mix of energy generation technologies. Radiation exposure is only one of
them.
C. Biological effects of selected internal emitters

51. Internal emitters is the commonly used term for radionuclides deposited in
body organs and tissues following their intake, principally by inhalation or
ingestion, but also potentially through wounds or intact skin. Depending on the
radionuclide concerned and the physicochemical form of the intake, internal
emitters vary enormously by type, pattern and duration of their radioactive
emissions and energy deposition within and between organs and tissues.
52. It is important to study exposures to internal emitters directly because the
radiation from some radionuclides is short-ranged and, to varying degrees, densely
ionizing. Moreover, such radionuclides may be distributed unevenly among body
tissues. Consequently, the nature of the dose delivered by some internal emitters
differs markedly from that delivered by radiation penetrating from external sources
such as the atomic bombs detonated in Hiroshima and Nagasaki, Japan. Most of the
evidence of risk from radiation comes from studies of human exposure to
penetrating radiation, while very few direct data are available on the health effects
from internal exposure. Therefore doses to organs from internal emitters have to be
estimated using models, and risk factors are derived principally from studies on
external penetrating radiation. Under those circumstances it is highly desirable to
14 UNSCEAR 2016 REPORT

validate the underlying assumptions by obtaining real observations of populations

exposed internally to radiation from specific radionuclides.
53. In response to initiatives in a number of countries to estimate the appropriate
doses from tritium and radioisotopes of uranium and understand the corresponding
health effects, the Scientific Committee has reviewed the relevant information on
these radionuclides. Two scientific annexes provide the rationale for the
Committees conclusions set out here.
54. Tritium ( 3 H) is a radioactive isotope of hydrogen that decays solely by
low-energy beta-particle emission. It occurs both naturally, mainly as a result of
interaction between cosmic-ray particles and the upper atmosphere, and artificially,
in the operation of nuclear reactors and other industrial installations, as a substance
used in biomedical research and, in the past, as an ingredient used in a variety of
consumer products. In the future tritium is expected to be used on a large scale in
fusion reactors. In the environment and at the workplace tritium is encountered
mainly as tritiated water in liquid or vapour form. An aspect of environmental and
food-chain transfer that warrants further investigation is the accumulation of tritium
in the organic component of foodstuffs, referred to as organically-bound tritium.
55. Uranium is a naturally occurring element and is ubiquitously distributed in the
environment. There are three naturally occurring radioisotopes of uranium: 234 U,
235
U and 238 U. These are present in rocks and soils and hence in the human diet.
They decay mainly by alpha-particle emission and have very long half-lives.
Internal exposures of workers to uranium are mainly the result of mining activities
and its use as a nuclear fuel. In daily life, people are exposed to uranium originating
mainly from drinking water and foodstuffs. Concern has been expressed over
exposures of military personnel and members of the public to depleted uranium
(isotope mixtures containing a low percentage of 235 U) used in munitions, for
example by the General Assembly in its resolution 69/57 on the effects of the use of
armaments and ammunitions containing depleted uranium.
56. Whereas absorbed doses to body organs as a result of exposures to external
sources of radiation are calculated using anatomical models of the human body,
commonly referred to as phantoms, estimating doses from internal emitters
additionally requires biokinetic models that describe the behaviour of radionuclides
following their intake into the body, principally by inhalation or ingestion. Such
models consider the deposition of inhaled particles and vapour in the respiratory
tract and the passage of ingested radionuclides through the alimentary tract. Models
also represent the subsequent distribution of radionuclides to body organs and
tissues from blood, their retention in those sites of deposition, and their excretion.
The reliability of models used to estimate doses from individual elements and their
radioisotopes depends on the quality of available experimental and human data.
57. For tritium, models are available in the form of tritiated water, representing its
distribution throughout body organs and tissues according to their water content.
Less information is available with which to construct adequate models for the
behaviour of various forms of organically bound tritium and other tritiated
compounds, including amino acids, some of which are involved in the synthesis of
DNA and associated proteins. The absorption of uranium depends partly on whether
it is inhaled or ingested and varies substantially according to the physical and
chemical form of the uranium. Uranium absorbed to blood accumulates mainly in
 REPORT TO THE GENERAL ASSEMBLY 15

the skeleton but with some retention also in the kidneys during the rapid urinary
excretion of a large fraction.
58. Different types of radiation vary in their effectiveness in causing cancer and
other health effects. Two broad categories of radiation are photons and charged
particles such as electrons and alpha particles. Some types of charged particles are
generally more effective at causing cancer per unit of absorbed dose than
penetrating photons. The assessment of such differences relies largely on
experimental data about their relative biological effectiveness (RBE), defined as the
ratio of the absorbed dose of a reference radiation to the absorbed dose of a test
radiation required to produce the same biological effect.
59. There is extensive literature on studies of RBE for tritium beta particle
emissions. Values of RBE for a range of biological end points range from about
unity to several-fold higher compared to gamma rays and X-rays. However, the
ability to draw conclusions for carcinogenesis is limited by the very small number
of relevant studies in mammals. Limited information is available that could be used
to estimate RBE values for alpha-particle emissions from isotopes of uranium.
However, RBE values for alpha particles depend on the particles energy, range and
the dense deposition of energy along short tracks, and the values of RBE will be
largely independent of the radionuclide concerned other than when the radionuclide
determines the origin within body tissues of the emission. Typical values of RBE
reported for alpha particles relative to gamma rays or X-rays are around ten for the
end points of liver and lung cancer, with lower values for leukaemia.
60. While the tumorigenic effects of uranium in animals are likely related to
radiological toxicity due to alpha-particle emissions, some effects are clearly related
to the chemical toxicity of uranium species, particularly in the kidneys. Chemical
toxicity is the limiting factor determining currently acceptable levels of uranium in
drinking water.
61. A number of epidemiological studies have been conducted of workers and
members of the public who may have been exposed to tritium. However, none of
these studies have so far been informative in showing an increased frequency of
cancer in the exposed populations that could be attributed to radiation exposure
from tritium. Epidemiological studies of nuclear workers have shown a weak
association between exposures from uranium and rates of lung cancer, but the data
are not sufficiently conclusive to demonstrate a causal relationship.
62. The Committee considered studies on the health effects of depleted uranium
used in munitions for military applications. No clinically significant pathologies
related to exposure from depleted uranium were found in military personnel or
members of the public. This is consistent with expectations, given the low levels of
measured or assessed exposures.
63. The Committee recognizes that continued research and review is needed to
assess the effects of internal exposures. Further work is required to understand the
effects of uneven delivery of doses from internal emitters within tissues and cells
relative to the uniform delivery of doses from external exposure to penetrating
radiation. The complexity of changing exposures and tissue sensitivities during in
utero and early postnatal development should also be a focus for further research.
16 UNSCEAR 2016 REPORT

Appendix I
Members of national delegations attending the
fifty-seventh to sixty-third sessions of the United Nations
Scientific Committee on the Effects of Atomic Radiation
Argentina A. J. Gonzlez (Representative), A. Canoba, P. Carretto,
M. di Giorgio, M. G. Ermacora
Australia C.-M. Larsson (Representative), C. Baggoley, M. Grzechnik,
G. Hirth, P. Johnston, S. B. Solomon, R. Tinker
Belarus A. Stazharau (Representative), J. Kenigsberg (Representative),
A. Nikalayenka, A. Rozhko, V. Ternov, N. Vlasova
Belgium H. Vanmarcke (Representative), S. Baatout, H. Bijwaard,
H. Bosmans, G. Eggermont, H. Engels, F. Jamar, L. Mullenders,
H. Slaper, P. Smeesters, A. Wambersie, P. Willems
Brazil J. G. Hunt (Representative), D. R. Melo (Representative),
M. Nogueira Martins (Representative), D. de Souza Santos,
L. Holanda Sadler Veiga, M. C. Loureno, E. Rochedo
Canada P. Thompson (Representative), N. E. Gentner (Representative),
B. Pieterson (Representative), C. Purvis (Representative),
D. Boreham, K. Bundy, D. B. Chambers, J. Chen, P. Demers,
S. Hamlat, R. Lane, C. Lavoie, E. Waller, D. Whillans
China Pan Z. (Representative), Chen Y., Dong L., Du Y., Gao H.,
Li F., Lin X., Liu J., Liu S., Liu Y., Pan S., Qin Q., Song G.,
Su X., Sun Q., Wang Y., Xuan Y., Yang H., Yang X., Zhang W.,
Zhou P., Zhu M.
Egypt W. M. Badawy (Representative), T. S. El-Din Ahmed Ghazey
(Representative), M.A.M. Gomaa (Representative), T. Morsi
Finland S. Salomaa (Representative), A. Auvinen, R. Bly, E. Salminen
France L. Lebaron-Jacobs (Representative), A. Rannou (Representative),
E. Ansoborlo, J.-M. Bordy, M. Bourguignon, I. Clairand,
I. Dublineau Naud, A. Flry-Hrard, J.-R. Jourdain,
R. Maximilien, F. Mntrier, E. Qumeneur, M. Tirmarche
Germany P. Jacob (Representative), W. Weiss (Representative),
S. Baechler, A. Bttger, A. A. Friedl, K. Gehrcke, T. Jung,
G. Kirchner, J. Kopp, R. Michel, W.-U. Mller, W. Rhm, H. Zeeb
India R. A. Badwe (Representative), S. K. Apte (Representative),
K. S. Pradeepkumar (Representative), K. B. Sainis
(Representative), B. Das, P. C. Kesavan, Y. S. Mayya
Indonesia E. Hiswara (Representative), Z. Alatas (Representative),
S. Widodo (Representative), G. B. Prajogi, G. Witono,
B. Zulkarnaen
 REPORT TO THE GENERAL ASSEMBLY 17

Japan Y. Yonekura (Representative), K. Akahane, M. Akashi, S. Akiba,

T. Aono, N. Ban, M. Chino, H. Fujita, K. Kodama, M. Kowatari,
M. Nakano, O. Niwa, K. Ozasa, S. Saigusa, K. Sakai, G. Suzuki,
M. Takahashi, T. Takahashi, Y. Yamada, H. Yamagishi, H. Yasuda
Mexico J. Aguirre Gmez (Representative)
Pakistan Z. A. Baig (Representative), M. Ali (Representative), R. Ali
Peru A. Lachos Dvila (Representative), L. V. Pinillos Ashton
(Representative), B. M. Garca Gutirrez
Poland M. Waligrski (Representative), L. Dobrzyski, M. Janiak,
M. Kruszewski
Republic of B. S. Lee (Representative), M. Baek, K.-W. Cho
Korea (Representative), K.-H. Do, J.-I. Kim, K. P. Kim, S. H. Kim,
D.-K. Keum, J. K. Lee, J. E. Lee, S. H. Na (Representative),
S. Y. Nam, S. W. Seo
Russian A. Akleyev (Representative), M. Kiselev (Representative),
Federation R. Alexakhin, T. Azizova, S. Geraskin, V. Ivanov,
N. Koshurnikova, A. Koterov, A. Kryshev, I. Kryshev, B. Lobach,
S. Mikheenko, O. Pavlovsky, A. Rachkov, S. Romanov,
A. Samoylov, A. Sazhin, S. Shinkarev
Slovakia L. Auxtov (Representative), E. Bdi (Representative),
M. Zemanov (Representative), M. Chorvth, A. urecov,
V. Jurina, . Kantov, K. Petrov, L. Tomek, I. Zachariov
Spain M. J. Muoz Gonzlez (Representative), D. Cancio, M. T. Macas
Domnguez, J. C. Mora Caadas, B. Robles Atienza, E. Va
Carruana
Sudan N. A. Ahmed (Representative), I. Salih Mohamed Musa
(Representative), E.A.E. Ali (Representative), A. E. Elgaylani
(Representative), M.A.H. Eltayeb (Representative),
I. I. Suliman
Sweden I. Lund (Representative), L. Hubbard (Representative),
L. Moberg (Representative), A. Almn, E. Forssell-Aronsson,
L. Gedda, J. Johansson Barck-Holst, J. Lillhk, A. Wojcik
Ukraine D. Bazyka (Representative)
United Kingdom S. Bouffler (Representative), J. Cooper (Representative),
of Great Britain J. Harrison (Representative), A. Bexon, J. Simmonds,
and Northern R. Wakeford, W. Zhang
Ireland
United States of R. J. Preston (Representative), F. A. Mettler Jr. (Representative),
America A. Ansari, L. R. Anspaugh, J. D. Boice Jr., W. Bolch, H. Grogan,
N. H. Harley, E. V. Holahan Jr., B. A. Napier, D. Pawel,
G. E. Woloschak
18 UNSCEAR 2016 REPORT

Appendix II
Scientific staff and consultants cooperating with the United
Nations Scientific Committee on the Effects of Atomic
Radiation in the preparation of its scientific report for 2016
L. Anspaugh
B. Lauritzen
M. Balonov
I. Dublineau
H. Grogan
L. Hubbard
B. Lambert
C. Robinson
E. Rochedo
R. Shore
J. Simmonds
R. Wakeford

Secretariat of the United Nations Scientific Committee on the Effects of Atomic

Radiation
M. J. Crick
F. Shannoun
 ANNEX A

METHODOLOGY FOR ESTIMATING PUBLIC

EXPOSURES DUE TO RADIOACTIVE
DISCHARGES

19
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 21

Contents

I. INTRODUCTION ................................................................................................................................... 23
II. GENERAL CONSIDERATIONS .......................................................................................................... 24
A. Dosimetric quantities ................................................................................................................ 24
B. Integration times ........................................................................................................................ 29
C. End points considered .............................................................................................................. 30
D. Discharges considered.............................................................................................................. 31
III. DESCRIPTION OF THE METHODOLOGY ...................................................................................... 32
A. Assessment of doses from discharges to atmosphere.................................................. 34
B. Assessment of doses from discharges to a freshwater environment ................................. 48
C. Assessment of doses from discharges to a marine environment ............................. 64
IV. GLOBALLY DISPERSED RADIONUCLIDES ................................................................................... 70
A. Tritium ............................................................................................................................................ 70
B. Carbon-14 ...................................................................................................................................... 71
C. Iodine-129 ..................................................................................................................................... 72
D. Krypton-85 .................................................................................................................................... 72
V. LIMITATIONS OF THE MODELS AND DATA USED ................................................................... 73
A. Discharges to atmosphere ...................................................................................................... 74
B. Discharges to freshwater and marine environments .................................................... 75
C. Global modelling ........................................................................................................................ 76
D. Comparison of the current and previous methodologies ........................................... 77
VI. APPLICATION OF THE METHODOLOGY ...................................................................................... 78
VII. CONCLUSIONS ..................................................................................................................................... 90
VIII. ACKNOWLEDGEMENTS .................................................................................................................... 91
TABLES ........................................................................................................................................................... 97
REFERENCES ............................................................................................................................................... 127

Electronic attachments (http://www.unscear.org/unscear/en/publications/2016.html)

1. Comparison of the Committees earlier and current methodologies
2. Summary of the Committees pre-2010 methodologies and their development
3. Population densities around uranium mines and nuclear power sites
4. Sensitivity studies in support of the methodology development
5. Electronic workbooks that implement the current methodology
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 23

I. INTRODUCTION
1. For many years the Committee has had a methodology for assessing the radiation exposures of the
general public from discharges of radionuclides to the environment, where the term discharge refers to
authorized releases from normal operations. The methodology does not apply for accidental releases.
Every few years, the Committee has updated its estimates of the global radiological impact of the
various types of nuclear installation, using this methodology and the latest available information on the
level and nature of discharges. The most recent versions of the methodology, which in turn were built
on earlier versions, were published in annex A of the UNSCEAR 2000 Report [U6] and annex B of the
UNSCEAR 2008 Report [U10]. Although used successfully for many years, the Committee decided at
its fifty-sixth session in 2008 to review and, where appropriate, update its methodology [U10] as part of
its strategic plan for 20092013, with a view to subsequently updating its assessments of the levels of
radiation exposure from energy production.

2. The global impact of energy production was one of the Committees thematic priorities identified
in its strategic plan for 20142019 (see [U12]). The Committee decided to specifically assess the global
radiological impact of discharges from both nuclear and non-nuclear sources of electrical energy
production, and thus to review its methodology so that it could be used to:

(a) Evaluate the radiation levels worldwide to which people are usually exposed as a consequence
of electricity generation;

(b) Assess the typical variations in exposure worldwide to different sources of radioactive
discharges;

(c) Identify sources of possible concern for public exposure;

(d) Allow users to derive benchmarks for comparison purposes and to derive relationships for
their investigative work;

(e) Analyse temporal trends in the contributions of different sources to overall public exposure.

3. The Committee recognizes that it is impractical to carry out a full site-specific assessment for
each site of electricity generation in the world that discharges radionuclides to the environment.
However, it decided that, where possible, its approach should be able to take into account differences
among regions of the world. It also decided that the methodology should be robust, transparent and
applicable to the different electrical energy sources, and that it should build on the experience gained in
applying the previous methodology. One change agreed was that for ease of application to different
electrical energy sources and to all parts of the nuclear fuel cycle, the methodology should provide
estimates of the doses 1 per unit discharge of key radionuclides to different environments. With
information on the discharges from the various facilities, these factors could then be used to estimate
doses per unit electrical energy produced (see annex B).

1
Where the word dose is used in this annex without qualification, it should be taken to mean effective dose to an individual or
collective effective dose to a population, according to context, unless otherwise stated.
24 UNSCEAR 2016 REPORT

4. The methodology is thus designed to assess individual and collective doses from unit discharge of
each of the key radionuclides to atmosphere, lakes, rivers and seas. The individuals considered are
those living in the area local to the point of discharge with behaviour indicative of most people living in
that area; the doses to these characteristic individuals are referred to here as characteristic individual
doses and are not to be confused with those to the so-called representative person (previously
termed by the International Commission on Radiological Protection, ICRP, as the critical group) used
for radiation protection purposes [I13]. The Committee agreed that, where possible, estimates should be
available for collective doses as a function of distance from the point of discharge and of the levels of
individual dose that make up these collective doses. The Committee also agreed to take account of
some differences among geographical regions, notably on population densities. For ease of application
and transparency, the methodology has been implemented through a series of Excel workbooks, which
should also facilitate any future updates of the Committees assessments.

5. The Committee intends to use the methodology to conduct its worldwide assessments of the
radiological impact of discharges of radionuclides. The methodology aims to provide best estimates of
radiation doses, in contrast to other established methodologies (e.g. [I2]) which aim to ensure that doses
are not underestimated and thus adopt cautious parameter values and assumptions. The Committees
methodology only applies to continuous routine releases to the environment and is not intended to be
comprehensive, covering all situations and other uses (such as for estimating the risks of radiation-
induced health effects).

6. This scientific annex describes the methodology to assess doses from discharges to atmosphere,
freshwater bodies and sea, together with the approach adopted for globally dispersed radionuclides. The
limitations of the models and data used are also discussed. An appendix provides detailed information
on the implementation of the methodology, including the end points considered, the relevant
mathematical equations and model parameters, a specification of the series of Excel workbooks that
implement the methodology, information on how the workbooks can be used to assess exposures from
different types of discharge, and a description of how progeny of radionuclides were treated and quality
assurance conducted. In addition more detailed supporting information and the workbooks themselves
are provided in electronic attachments, which can be downloaded only from the UNSCEAR website
(http://www.unscear.org/unscear/en/publications/2016.html).

II. GENERAL CONSIDERATIONS

A. Dosimetric quantities
7. Although for strict scientific purposes other quantities are appropriate, the Committee has for
many years used two protection quantities, namely effective dose and collective effective dose for its
evaluations of levels and trends of exposures. These have the advantage of simplifying comparison of
doses from different types of radiation and different distributions of dose within the body, and of
averaging over age and sex; moreover, many regulatory authorities keep records in terms of these
quantities. The quantity effective dose is based on another underlying radiation protection quantity, the
equivalent dose, which takes into account the inferred differences in biological effectiveness of
different types of radiation by applying defined radiation weighting factors to the absorbed dose to an
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 25

organ or tissue. Equivalent doses to different organs and tissues are then combined using defined tissue
weighting factors, which take into account the inferred differences in detriment from irradiation of the
particular organ or tissue, averaged over the sex and age distributions of typical populations. This
enables external and internal exposures from a source of radiation to be added to produce an overall
effective dose 2 from that source, which can then be compared with effective doses from other sources.
In this annex, the definition of effective dose used to express both individual and collective doses is
generally that of Publication 60 of the ICRP [I10]. Although ICRP made some modifications to the
weighting factors in the definition of effective doses in its 2007 recommendations [I13], it has not yet
published a full set of dose coefficients for external and internal exposure and so the Committee has
used the previous values, unless otherwise stated.

8. It is important to note that there are differences between the quantities of individual and collective
dose, although they are calculated in similar ways. In this methodology and any related studies, the
collective (effective) dose is always estimated for a defined population over a specified period of time.
The collective dose used here is the product of the mean effective dose to a specified population from a
particular source, and the number of people in that population, integrated over a defined period of time.
In other words, a collective dose is the dose received by all members of a particular population
combined, over a defined period of time. When evaluating collective doses, the population and time
period should always be specified and any underlying assumptionssuch as the population remaining
the same over the time periodshould be acknowledged. The Committee has used the quantity,
collective dose, for many years to compare the radiation exposures of populations from different
sources of ionizing radiation, or following different protection measures. In particular, the quantity
provides a convenient basis for comparing the impacts of radioactive discharges from nuclear and other
sources of electrical energy, and for incorporation by analysts into measures of economic utility for
decision-aiding. However, collective dose should only be used for comparative purposes; it is
inappropriate to use it for quantifying exposure in epidemiological studies or for making risk
projections. Specifically, the Committee has stated that it does not recommend the aggregation of very
low doses over extended time periods to large numbers of individuals to estimate absolute numbers of
radiation-induced health effects within a population exposed to incremental doses at levels equivalent
to or lower than normal natural background levels [U13].

9. The Committee decided to retain its previous approach where collective dose was estimated for
local, regional and global components. The spatial variation in collective dose can be estimated for
discharges to atmosphere because (a) dispersion can be expressed as a function of distance from the
discharge point and (b) data on the distribution of population density are also available as a function of
distance. However, for aquatic discharges, when using a simple generic approach, it is not possible to
determine collective doses as a function of distance. Because people over a wide area may all obtain
their drinking water and fish from the same water body, there is not a simple relationship between
distance from the discharge point and dose. A simplified approach is therefore used based on a
discharge to a single water body so that, for aquatic discharges, less spatial resolution in collective
doses is possible than for discharges to atmosphere.

10. The Committee considered that the assessment of characteristic individual doses is also useful in
addressing the thematic priority in its strategic plan mentioned above. As noted previously, as far as
possible, the dose estimates are based on best estimates of parameter values. For example, for
discharges to atmosphere, the characteristic individual dose estimated is the dose to someone living

2
For both individual and collective doses, the effective doses estimated here are the sum of the effective doses from external
exposure received during the period of interest and the committed effective doses from intakes by inhalation and ingestion during
the same period.
26 UNSCEAR 2016 REPORT

5 km from the discharge point who is assumed to obtain a proportion of their food from this location,
rather than a more cautious assumption of living 1 km or less from the discharge point. For aquatic
discharges, the characteristic individual dose estimated is the dose to someone assumed to be living in
the area near to the discharge location and to obtain part of their foodand in the case of discharges to
freshwater bodies, their drinking waterfrom the receiving water body.

11. In this regard, for estimating characteristic individual doses, the Committee considered it to be
more realistic to assume that only a proportion of their food consumed was locally produced (locally
produced food is that grown by the individual or bought fresh from local markets) rather than an
approach often used for regulatory purposes of assuming all their food consumed was 100% locally
produced. There is little information on how much locally produced food people consume. In France, a
comprehensive investigation was carried out to determine the amount of locally produced food that
people ate [B3]. This showed that rural populations in France consumed more locally produced food
than urban populations, as would be expected, and that there was a significant variation for different
foods. For example, for milk consumption in France as a whole, only 3% was found to be locally
produced but in rural areas, the figure was 30%, on average. For leafy vegetable consumption in France
as a whole, just over 25% was found to be locally produced but in rural areas the figure was 70%. For
cereal consumption in France, the percentage locally produced was very low, at 0.1% both for the
country as a whole and for rural areas. While recognizing that there are likely to be wide variations in
practice across the world, the Committee agreed that a simple assumption of 25% of the food consumed
being locally produced was reasonable for its generic methodology and that this value be used for both
terrestrial and aquatic foods. This value is consistent with the value used in the Committees recent
assessment of doses in Japan following the nuclear accident at the Fukushima-Daiichi nuclear power
station [U12].

12. The methodology employs published dose coefficients to estimate doses from external and
internal exposure. The dose coefficients used for internal exposure are those for adult members of the
public and are the committed effective doses to 70 years of age per unit intake of radionuclide given by
ICRP [I14]. The dose coefficients for external exposure from radionuclides in the air or deposited on
the ground following discharges to air or water bodies were selected from literature sources relevant to
the situation being considered. The main dose coefficients used for internal and external exposure are
summarized in table 1 both for radionuclides that are discharged and for their key progeny, where
appropriate (see also the appendix). It should be noted that for dose coefficients for internal exposure
the contribution of short-lived progeny is included in the dose coefficient assigned to the parent.
However, for the dose coefficients for external exposure the dose coefficients are for the parent alone
and any contribution from short-lived progeny has to be added separately using the values given in
table 1(b). Except for isotopes of argon (e.g. 41Ar), krypton (e.g. 85Kr), xenon (e.g. 133Xe, 135Xe and
138
Xe) and radon (e.g. 222Rn), which are all gases, the discharges are assumed to be particulate with a
default size of 1 m AMAD. The dose coefficients for tritium in the form of tritiated water (HTO) and
of organically bound tritium (OBT) are provided for use with the specific activity model for tritium in
the environment described in chapter III. The chemical form assumed for each element is generally the
default type for public exposure given in ICRP Publication 72 [I12].
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 27

Table 1. Radionuclide-specific dose coefficients for internal exposure from inhalation and
ingestion and for external exposure from the plume and deposition
Values are given to one decimal place only as this is considered sufficient for this purpose

(a) Radionuclides discharged

Radio- Progenya Lung Committed effective dose Effective dose coefficients for
nuclide absorption coefficients for adults to age external exposured
typeb 70 years for internal exposurec
[I12] Inhalation Ingestion Plume Deposition
Dinh Ding Dex,cloud Dex,deposit
(Sv/Bq) (Sv/Bq) (Sv/(Bq s/m3)) (Sv/(Bq /m2))e
[I14] [I14] [E2, P2] [P2]
3
H M 4.5 1011 0 0 0
14
C M 2.0 10 9
5.8 10 10
2.6 10 18
0
35 f
S M 1.4 10 9
7.7 10 10
3.1 10 18
0
41
Arf n/a 0 0 6.2 1014 0
54
Mn M 1.5 109 7.1 1010 3.9 1014 5.2 1014
58
Co M 1.6 109 7.4 1010 4.5 1014 5.0 109
60
Co M 1.0 108 3.4 109 1.2 1013 2.0 107
65
Zn M 1.6 109 3.9 109 2.8 1014 9.0 109
85
Krf n/a 0 0 9.9 1017 0
90
Sr c 90
Y M 3.6 10 8
2.8 10 8
9.8 10 17
0
106
Ru c 106
Rh M 2.8 10 8
7.0 10 9
0 0
129
I F 3.6 108 1.1 107 2.8 1016 3.1 109
131 d
I F 7.4 109 2.2 108 1.7 1014 2.4 1010
133
Xef n/a 0 0 1.2 1015 0
135
Xe f 135
Cs n/a 0 0 1.0 10 14
0
138
Xef 138
Cs n/a 0 0 5.4 1014 0
134
Cs F 6.6 109 1.9 108 7.1 1014 6.2 108
137
Csc 137m
Ba F 4.6 109 1.3 108 9.3 1017 4.6 1013
210
Pb M 1.1 106 6.9 107 4.5 1017 2.5 1010
210
Po M 3.3 106 1.2 106 3.9 1019 1.2 1013
222
Rnf,g 218
Po 214Pb, 214Bi, 214Po n/a 4.8 109 n/a n/a n/a
226
Ra M 3.5 10 6
2.8 10 7
2.8 10 16
3.8 109
230
Th S 1.4 105 2.1 107 1.5 1017 0
232
Th Ra, 228Ac, 228Th, 212Pb
228
S 2.5 105 2.3 107 7.2 1018 8.7 1011
234
U M 3.5 106 4.9 108 6.1 1018 5.5 1011
238
U 234
Th, 234mPa M 2.9 106 4.5 108 2.5 1018 1.9 1011
239
Pu M 5.0 105 2.5 107 3.5 1018 3.5 1011
240
Pu M 5.0 105 2.5 107 3.4 1018 1.5 1011

241
Am M 4.2 105 2.0 107 6.7 1016 7.9 109
28 UNSCEAR 2016 REPORT

(b) Radionuclides considered to have different chemical forms in the environment or as progeny of
others

Radio- Lung absorption typeb Committed effective dose Effective dose coefficients for
nuclide [I12] coefficients for adults to age 70 external exposured
years for internal exposurec

Inhalation Ingestion Plume Deposition

Dinh Ding Dex,cloud Dex,deposit
(Sv/Bq) (Sv/Bq) (Sv/(Bq s/m3)) (Sv/(Bq /m2))e
[I14] [I14] [E2, P2] [P2]

HTO n/a 1.8 1011 1.8 1011 0 0

OBT n/a 4.1 10 11
4.2 10 11
0 0
90
Y 7.9 10 16
0
106
Rh 9.4 1015 5.5 1015
135
Csf F 6.9 1010 2.0 109 9.5 1018 2.4 1011
138
Csf F 2.4 1011 9.2 1011 1.2 1013 3.9 1012
137m
Ba 2.7 1014 1.7 107
212
Pb 5.7 1015 4.7 1012
214
Pb M 1.4 108 1.4 1010 1.1 1014 5.6 1013
210
Bi 2.6 1016 2.2 1011
214
Bi 1.4 108 1.1 1010 7.1 1014 2.5 1012
214
Po n/a 0 0 3.8 1018 1.9 1023
218
Po n/a 0 0 4.2 1019 2.3 1018
228
Ra M 2.6 106 6.9 107 0 0
228
Ac F 2.5 10 8
4.3 10 10
4.0 10 14
1.7 1011
228
Th S 4.0 105 7.2 108 8.1 1017 6.9 1011
234
Th S 7.7 109 3.4 109 2.9 1016 1.3 1011
234m
Pa n/a 0 0 1.2 1015 1.1 1014
a
The progeny considered in the methodology are listed. For further details see the appendix; The symbol means either that
the progeny are not considered or that the progeny are stable isotopes.
b
This refers to the rate of absorption from the lung. The symbols relate to fast (F), medium (M) or slow (S) absorption.
c
For these radionuclides with short-lived progeny, the dose coefficients for internal exposure for the parent include the
contribution from the progeny and coefficients are not given separately for the progeny.
d
In all cases the dose coefficients for external exposure are for the parent only and the contribution from progeny has to be
considered separately.
e
Integrated effective dose to 100 years following a continuous deposition rate of 1 Bq/(m2 s) for one year. The Petoussi-Henss et
al. values are not used directly but are modified to allow for changes in external dose with time [G2].
f
Considered for discharges to atmosphere only.
g
See section III.A.3 for more information on the treatment of radon inhalation.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 29

B. Integration times
13. The methodology is applicable to discharges that can be assumed to be continuous and takes into
account (a) the build-up of long-lived radionuclides in the environment and (b) the continued exposure
to long-lived radionuclides after discharges have ceased. This is done by considering a years discharge
of a radionuclide, its dispersion in the environment and the subsequent exposures of people over many
years; the resulting dose rates are then integrated to various times. Using these integrals, it is also
possible to consider that the discharges may continue for many years from the same site. This is
because the integrated dose to say 100 years from one years discharge is numerically equal to the dose
in the 100th year from a continuous discharge at a constant rate. The Committee agreed that 100 years
was generally a reasonable assumption for the length of continuous discharge for estimating
characteristic individual doses as it covers the lifetime of individual power plants and also the
possibility that additional power plants may be constructed on the same site. Separate considerations
apply for the disposal of waste residues (mill tailings) from uranium and coal mining operations as
discussed below.

14. Collective dose rates can be integrated to infinity (referred to as the collective dose commitment) to
take account of all possible future doses from a discharge. However collective dose rates are generally
integrated to shorter times; values ranging from 80 to 10,000 years have been used in the Committees
past reports, see electronic attachments 1 and 2 (collective doses integrated to various times are sometimes
referred to as truncated collective doses). For most radionuclides, integrating collective dose rates to
100 years gives a significant proportion of the collective dose commitment [S9]. However, some relatively
long-lived, environmentally mobile radionuclides (3H, 14C, 85Kr and 129I) become globally dispersed and
can continue to contribute to the integrated collective doses for many years. For both 3H and 85Kr (half-
lives 12.6 years and 10.7 years, respectively), the integrated collective dose does not increase beyond
100 years. However, for 14C (half-life 5,730 years), the integrated collective dose increases by a factor of
more than five between 1,000 and 10,000 years; however the additional collective dose is then negligible
if the integration is continued to 1,000,000 years. For the very long-lived 129I (half-life 15.7 million years),
the integrated collective dose increases by about a factor of six between 1,000 and 10,000 years and by a
factor of over 150 between 10,000 and 1,000,000 years (based on table 70 of [U5]). However, for
discharges from nuclear sites, the integrated collective dose from 14C rather than that from 129I dominates
the total integrated collective dose and so most of the collective dose will be received in the first
10,000 years [S9, U5]. Also, when comparing options explicit consideration of collective doses at long
times is often not necessary because the collective doses at these times are all similar and moreover
comprise extremely low levels of individual dose [S9, U5].

15. In general, the uncertainties associated with estimating collective doses increase with integration
time, for example, because of possible major changes in environmental conditions and population
dynamics. The integrated collective dose to 500 years from a continuous discharge over one year has
often been used to provide an upper estimate of the highest future annual collective dose rate if the
practice, and hence discharge, were to continue at a constant rate for that number of years [L1].
Five hundred years is used because it is assumed to be the maximum duration of the generation of
electrical energy by nuclear power. An integration time of 10,000 years was used in the Committees
assessments conducted before the year 2000. Ten thousand years is consistent with the duration of the
warm period between glacial periods [L1] and therefore the maximum number of years over which any
reasonable assumptions about the nature of the global environment can be made.

16. When comparing the radiological impact of different sources, it is important not to introduce any
bias when choosing integration times for collective dose. For example, choosing a short time (e.g. of a
few hundred years) for nuclear power generation would mean that a significant portion of the integrated
30 UNSCEAR 2016 REPORT

collective dose is not included. For non-nuclear energy production, notably that using fossil fuels, the
major radionuclides discharged are from the 238U decay chain, particularly 226Ra and its progeny
including 222Rn. These radionuclides do not become dispersed globally and so once discharged to the
environment need to be considered only for relatively short timescales; a few hundred years is sufficient
to allow for the ingrowth of the radiologically significant radionuclide, 210Po.

17. The conventional mining and milling of uranium ore gives rise to various waste residues, which
are referred to as mill tailings. These contain enhanced levels of naturally occurring radionuclides and
the radon is emitted from the mill tailings for many years following disposal because of the long
radioactive half-life of the parent 238U. In assessing the radiological impact of mill tailings, it is
necessary to allow for the number of years over which the emissions will continue. Previously, the
Committee estimated collective doses from mill tailings assuming that the emissions of radon continued
for 10,000 years [U6]. Since the year 2000, improvements have been made to the treatment of mill
tailings and the rehabilitation of areas where mill tailings had been disposed of. As discussed in the
companion report on radiation exposures from electricity generation (see annex B), enhanced levels of
radon emissions may continue for periods of less than 100 years. The Committee therefore agreed that
the best estimate of the period of radon emissions from mill tailings should be reduced to 100 years, but
that the effect of considering periods of 500 years and 10,000 years should also be evaluated. The same
considerations also apply to the disposal of mine spoil or ash from coal used for electricity generation.

C. End points considered

18. Taking account of the considerations discussed in sections A and B above, the Committee has
developed its methodology to estimate the following characteristic individual doses (expressed as
effective doses, in sieverts, Sv) from each radionuclide discharged continuously at a rate of 1 Bq/s:

(a) For discharges to atmosphere, the characteristic individual dose in the 100th year (as noted
above, the dose in the 100th year of continuous discharge is equal to the integrated dose to
100 years from 1 years discharge) at 5 km from the discharge point (annual individual doses are
also calculated at 50, 300, 750 and 1,250 km from the discharge point for use in estimating
collective doses);

(b) For discharges to freshwater bodies, the characteristic individual dose in the 100th year;

(c) For discharges to a marine environment, the characteristic individual dose in the 100th year.

19. In addition, integrated collective doses (man Sv) from each radionuclide discharged continuously
at a rate of 1 Bq/s for a year are calculated as follows:

(a) For discharges to atmosphere, collective doses are integrated to 100 years (this is equivalent to
the collective dose in the 100th year of continuous discharge) in the following distance bands:
0100 km (local), 100500 km, 5001,000 km and 1,0001,500 km. The results for 1001,500 km
are summed to constitute regional components of collective dose;

(b) For discharges to freshwater bodies, the collective dose is integrated to 100 years. It should be
noted that for freshwater bodies, instantaneous mixing in a single water volume is assumed and,
with this assumption, it is not possible to distinguish between local and regional components of
collective dose;
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 31

(c) For discharges to a marine environment, local and regional components of collective dose are
integrated to 100 years;

(d) For globally circulating radionuclides only (3H, 14C, 85Kr and 129I), the global collective dose is
integrated to 100, 500 and 10,000 years.

20. The results can be scaled to obtain the characteristic individual and collective doses for specific
discharges and sources. In chapter III, results are calculated for discharges from different geographical
regions of the world, considering different population distributions (generic, nuclear coastal and nuclear
inland). Results are also calculated for a generic discharge location situated in a region of very low
population density, which may apply to sites of uranium mines and mill tailings.

D. Discharges considered

1. Electrical energy production

21. One of the aims of the methodology is to allow exposures from discharges of radionuclides to
atmosphere and water bodies to be assessed for a range of facilities associated with electricity
generation, covering both nuclear and non-nuclear power production. The methodology could also be
broadly used to assess routine discharges from other sources, butas noted earlierit is not intended
for site-specific or regulatory purposes. The assessment of exposures due to electricity generation
requires adequate and representative discharge data from the various facilities for input into the models
that implement the methodology. For example, discharges from the nuclear facilities involved in power
production and the fuel cycle listed in table 2 could be considered.

Table 2. Types of nuclear facility for which the methodology may be relevant

Life-cycle stage Facility types

Front end Uranium mine with mill
Enrichment and fabrication

Nuclear reactor operation Nuclear power reactors (e.g. pressurized water reactors,
boiling water reactors, fast breeder reactors, gas cooled
reactors, heavy water cooled reactors, and light-water-
cooled, graphite-moderated reactors)
Research reactors

Back end Reprocessing

Decommissioning
Waste management

22. There are also a number of non-nuclear means of electrical energy production that may have
associated discharges of naturally occurring radionuclides because of the materials used in construction
or at some point in their production cycle. The Committee decided that these should also be covered by
the methodology and they are considered in annex B. These could include the use of: (a) fossil fuels
(coal, gas and oil); (b) geothermal sources; and (c) solar power, wind and biomass.
32 UNSCEAR 2016 REPORT

2. Radionuclides discharged
23. The radionuclides covered by the methodology are listed in table 1 and are considered by the
Committee to give rise to most of the doses from radioactive discharges from different energy sources.
This list of radionuclides includes those that have been demonstrated to give rise to the highest
integrated collective doses in previous assessments [C3, U6]. Naturally occurring radionuclides are also
covered by the methodology in order to assess the radiological impact of discharges from mining and
milling of primary materials, fuel production, and non-nuclear forms of electrical energy production [N4].

24. Although the methodology is designed to use data on discharges of individual radionuclides, in
many cases discharge data are provided only for groups of radionuclides as a whole. For example,
discharges may be given for total alpha and then this value will need to be apportioned among the
relevant radionuclides included in the methodology (e.g. 239,240Pu, 241Am). For discharges specified as
total beta/gamma or particulate there may be many radionuclides that could be discharged, with the
most important for nuclear power plants generally being 60Co, 90Sr, 134Cs and 137Cs but with significant
contributions from others such as 54Mn or for United Kingdom gas cooled reactors only 35S. In using the
methodology, it will be necessary to specify a fraction of the discharge for each relevant radionuclide
and also which representative radionuclides to consider as defaults for the sum of discharges from any
radionuclides not included in table 1.

25. The methodology is designed to allow different assumptions to be made about the state of
equilibrium between components of the decay chains of the naturally occurring radionuclides. Account
is taken of the time necessary for ingrowth to occur and differences between the importance of
ingrowth of progeny for different types of discharge and for different exposure pathways. Where
appropriate, the contribution of progeny is taken into account according to the following assumptions:

(a) Where the progeny are short-lived compared to the parent, their contribution is not considered
explicitly but is generally included with the parent, for example, contributions from 90Y and
137m
Ba are included in the dose coefficients for intakes by inhalation and ingestion for 90Sr and
137
Cs, respectively;

(b) Where the progeny are long-lived compared to the parent, their contribution is not considered
further if, over a period of 100 years, there is unlikely to be sufficient ingrowth of the progeny.
Examples where this is the case are 239Pu/235U, 240Pu/236U, and 241Am/237Np;

(c) The key radionuclides in the decay chains of the naturally occurring radionuclides are
considered separately and other progeny are only considered if they make a significant
contribution to exposures or if it is reasonable to assume that they would be discharged in
secular equilibrium with the parent.

Further information for specific radionuclides and exposure pathways and details of how this is applied
in practice are given in the appendix.

III. DESCRIPTION OF THE METHODOLOGY

26. When radionuclides are discharged to the environment there are various ways in which people can
be exposed to radiation. Figure I illustrates the more important exposure pathways.
Figure I. Exposure pathways following discharge of radioactive material to the environment

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 33

34 UNSCEAR 2016 REPORT

A. Assessment of doses from discharges to atmosphere

27. This section describes the methodology for estimating characteristic individual and local and
regional components of collective dose to the public from a discharge of radionuclides to atmosphere.
The appendix gives full details of all of the equations used. The exposure pathways included in the
methodology for discharges to atmosphere are:

(a) Internal exposure from inhalation of radionuclides in the plume;

(b) External exposure (beta and gamma emitters) from radionuclides in the plume;

(c) External exposure from deposited radionuclides;

(d) Internal exposure from ingestion of radionuclides incorporated in food.

28. Other exposure pathways such as internal exposure from inhalation of resuspended radionuclides
deposited on the ground and from inadvertent ingestion of soil are also possible. However a review of
the literature indicated that the doses from these exposure pathways are negligible compared with those
from the exposure pathways listed above for continuous discharges of radionuclides to atmosphere [J5].

29. Characteristic individual doses are calculated for a member of the public assumed to be living at a
distance of 5 km from the point of discharge. Local and regional components of collective dose from
discharges to atmosphere are derived directly by multiplying individual doses calculated at the
midpoints of distance bands or annuli (of 0100, 100500, 5001,000 and 1,0001,500 km) around the
point of discharge by data specific to each geographical region on the number of people within each
annulus. Region-specific or generic values are used for parameters in the dose calculation as discussed
in the following sections.

1. Population distribution
30. The Committees previous approach was based on simplified assumptions regarding population
density around nuclear sites representative of the situation in the 1980s. Given the importance of
population information in determining the overall magnitude of collective doses, and that there have
been significant changes in population patterns throughout the world, the Committee decided to update
the information on population density. For comparison purposes and to examine the effect of different
population densities, four different sets of population data are considered in the methodology. As a
default case for comparison purposes, the calculation of collective dose from discharges to atmosphere
is based on information on population densities for 2010 weighted by the population in each
geographical region [F1]. The population distribution used for the default case is given in table 3 where
the same population density is used for each annulus (the population is the assumed density multiplied
by the area of the annulus). In addition to values by region, a world-average population distribution has
been obtained as an average of the values for each region weighted by the populations in each region.
The regions considered in the methodology are those used by the United Nations Environment
Programme (UNEP), illustrated in figure II below.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 35

Table 3. Default population distributions for different geographical regions

UNEP region Population density, Population within each annulus

weighteda (km2)
0100 km 100500 km 5001,000 km 1,0001,500 km

Africa 7.9 10 1
2.5 10 6
5.9 10 7
1.9 108
3.1 108
Asia and Pacific 2.8 102 8.8 106 2.1 108 6.6 108 1.1 109

Europe 1.3 102 4.0 106 9.7 107 3.0 108 5.0 108
Latin America and Caribbean 1.4 102 4.3 106 1.0 108 3.2 108 5.3 108
North America 3.2 101 1.0 106 2.4 107 7.6 107 1.3 108

West Asia 1.0 102 3.2 106 7.7 107 2.4 108 4.0 108

World averageb 1.6 102 5.0 106 1.2 108 3.8 108 6.3 108
a
The population densities are taken from FAO [F1] and are values for each region weighted as described in this reference.
b
The world-average value is the average of the values for each region weighted by the population in each region.

Figure II. The geographical regions used in the methodology (taken from UNEP [U3])

31. For nuclear power stations, a more specific analysis of population data is possible. In
collaboration with the Metadata and Socio-Economics Unit of UNEP/DEWA/GRID-Geneva, 3 a
geographic information system has been used to express the number of people living within specified
distance bands around each station. Details of the approach and data used are given in electronic
attachment 3; in essence, the numbers of people living within different distances (100, 500, 1,000 and
1,500 km) around the site of each operating nuclear power station have been analysed. These data have
been used to derive simplified averages by UNEP region for the population in each annulus. The
averages have been estimated separately for inland sites and for sites located on the coast, where the
population distribution is likely to be affected by the presence of a large body of water (see table 4).
Again a world-average value has also been calculated using all the available data. Because there were,
at the time of analysis, no significant operating nuclear power stations in West Asia and only a coastal

3
United Nations Environment Programme (UNEP) Division of Early Warning and Assessment (DEWA) Global Resource
Information Database network (GRID) http://www.grid.unep.ch/.
36 UNSCEAR 2016 REPORT

site for Africa, no population densities are given for these in table 4. The results for the default
population distribution in table 3 could be used to consider nuclear sites in these regions if needed.

Table 4. Population distribution around operating nuclear power stations as a function of UNEP
region and location

Based on information provided by UNEP for the year 2008

Population within each annulusb

UNEP regiona
0100 km 100500 km 5001,000 km 1,0001,500 km
COASTAL
Africa 3.9 10 6
1.1 106 8.7 106 3.4 107
Asia and Pacific 8.3 106 8.8 107 2.0 108 3.5 108
Europe 3.3 106 6.4 107 1.3 108 1.5 108

Latin America and Caribbean 3.6 106 6.2 107 4.8 107 4.3 107
North America 4.7 106 4.0 107 7.0 107 6.7 107

West Asia n/a n/a n/a n/a

World averagec 5.6 106 6.6 107 1.4 108 2.0 108

INLAND
Africa n/a n/a n/a n/a
Asia and Pacific 1.9 10 7
2.0 108
4.6 108
5.0 108
Europe 4.6 106 8.9 107 1.8 108 1.8 108
Latin America and Caribbean 4.6 106 1.5 107 3.7 107 3.4 107
North America 3.0 106 3.9 107 7.7 107 8.5 107
West Asia n/a n/a n/a n/a

World average c
4.6 10 6
7.2 107
1.5 108
1.5 108
a
Note that at the time of analysis there were no significant operating nuclear power stations in West Asia or inland in Africa (see
figure II for definition of UNEP regions) and hence population data are not presented for these regions.
b
Data are presented to one decimal place, which is considered sufficient for this purpose.
c
The world-average is the average of the population in each annulus for all nuclear sites for which data are available.

32. The other case considered in the methodology is for uranium mines and mill tailings sites. Most
of these facilities are situated in areas of very low population densities, as illustrated in electronic
attachment 3. Therefore, the methodology also considers a population distribution with a density of
5 km2 to gain insight into the effect on collective doses. Using this low population density out to
1,500 km results in estimated collective doses from atmospheric discharges that are about two orders of
magnitude lower than those estimated using the population distributions in tables 3 and 4. The low
value for population density should therefore only be used where appropriate and where it can be
assumed to apply over the integration period for the collective dose.

33. There are two main approaches to assessing collective doses from ingestion of terrestrial foods.
The first uses information on the distribution of agricultural production to obtain an overall collective
dose from the production but with no indication of the individual doses that make up the collective
dose. The second approach, which the Committee adopted here, assumes that people obtain their food
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 37

from the area where they live. This enables both individual and collective doses to be estimated.
However, as discussed earlier, in many parts of the world it is not realistic to assume that people derive
all their food from local produce. This is particularly so within a small area that has a high population
density, and therefore low agricultural production. The Committee therefore decided that collective
doses from ingestion (where it is assumed that 100% of the food is obtained from the area considered)
should only be calculated for distances of at least 100 km. For the characteristic individual dose, based
on an individual living at a distance of 5 km from the site, 25% of their food is assumed to be locally
produced as discussed earlier (paragraph 11).

2. Dispersion in the atmosphere

34. In previous assessments, the Committee adopted a simple generic atmospheric dispersion model.
It considered the uncertainties associated with this approach in some detail in its 2000 Report [U6].
Given the generic nature of its assessment objectives, the Committee concluded that the long-term
sector-averaged Gaussian model was likely to provide an appropriate level of accuracy. Crawford et al.
[C5] acknowledge that simple Gaussian models continue to be used because they produce results that
often agree fairly well with measured data and because their results are relatively easy to obtain. The
Committee has decided to retain this approach.

35. The variation in activity concentration in air, Ca(x), with downwind distance beyond 1 km is
approximately given by the following equation:

Ca ( x) = D1 Q x n (1)

where D1 is the annual average dilution factor at 1 km (s/m3), Q is the discharge rate (Bq/s), x is the
downwind distance (km) and n is an empirically-determined index. The Committee agreed to retain the
parameter values used in the UNSCEAR 2000 Report, in which a value of 5.3 107 s/m3 was found to
be the best approximation for D1 with a value for n of 1.42 for all radionuclides except noble gases,
tritium and carbon-14 [U6]. The dilution factor, D1, was found to be relatively insensitive to changes in
the values of parameters, except for wind speed and release height. The value of n was, however, found
to be dependent upon deposition velocity and inversion height. For noble gases, which do not deposit, a
value of n equal to 1.2 is retained. This value is also retained for tritium because, although tritium
rapidly exchanges with water in the soil and vegetation, it is quickly re-emitted to the atmosphere.
Carbon-14 also exchanges with carbon in soil and vegetation and partially returns to atmosphere
through plant and soil respiration. A value for n of 1.4 is therefore used for this radionuclide.

36. As part of the development of the current methodology, the activity concentrations estimated
using this simple relationship were compared with the results obtained using a more complex Gaussian
model, for a uniform wind rose (for which activity concentrations were averaged over 12 sectors) that
allowed for depletion of the plume but not radioactive decay. The results were found to be very similar
(less than a factor of two difference).

37. A reduction factor is now included in the methodology to allow for the radioactive decay of short-
lived radionuclides, notably radon, during dispersion over distances of hundreds of kilometres. This
factor is based on a mean wind speed, ua, of 2 m/s [I2].

38. The deposition rate of radionuclides at a specified distance (in Bq/(m2 s) is determined by the
application of a simple effective deposition velocity that allows for both wet and dry deposition on
vegetation, soil and other surfaces. An annual-average deposition rate of 0.002 m/s [U6] is used for all
38 UNSCEAR 2016 REPORT

radionuclides, with the exception of the noble gases, tritium and 14C for which a value of zero is
applied. As noted above, tritium and 14C exchange with the soil and vegetation; however, this is
modelled using a specific activity approach (as discussed below). Because these radionuclides are
returned to atmosphere, the deposition rate is not used and the effective deposition velocity is taken to
be zero, even for 14C which is partly retained on the ground. This is consistent with the use of the
generic approach to the modelling of dispersion applied in the UNSCEAR 2000 Report [U6] and, with
the application of a value of n of 1.42, 1.4 or 1.2 in equation 1. The use of these parameters ensures that
there is a balance between the activity discharged and the activity deposited within a distance of around
2,000 km, thereby ensuring that the relevant exposures are included in the estimation of collective dose.

39. The calculated activity concentrations in air at ground level depend to some extent on the height
from which the discharge takes place and whether the discharge is from a point source or a wide area.
The heights of stacks from which discharges take place (i.e. point sources) vary considerably. The
discharges from uranium mine and mill tailings, on the other hand, take place at ground level over a
wide area. Electronic attachment 4 discusses the variation in estimated activity concentrations in air
with stack height and the differences for point and area sources. The differences in the concentration
profiles with stack height lessen with distance from the stack. For the case illustrated in electronic
attachment 4, where a uniform wind rose is assumed together with Pasquill Category D conditions for
65% of the time and all other categories for the rest of the time, the estimated concentration in air for a
ground-level discharge is about a factor of three higher than that from a 100 m stack at a distance of
5 km, and less than a factor of two higher than from a 30 m stack. Larger and smaller differences would
be seen for the individual Pasquill categories but this comparison applies to the annual average that is
needed for assessing routine continuous discharges. At a downwind distance of 100 km, the difference
in concentrations for a ground-level discharge and one from a 100 m stack is about a factor of two. For
point and area sources, the differences in estimated activity concentrations at these distances are small.
The ratio of the time-integrated activity concentrations in air for an area source to that for a point source
is 0.9 at 5 km and is 1 beyond 10 to 20 km (see electronic attachment 4). Given these findings, the
Committee agreed that, for its generic methodology, it was reasonable to adopt a single stack height of
30 m and to consider all discharges to atmosphere as point sources.

3. Behaviour of radionuclides in a terrestrial environment

40. Radionuclides discharged to atmosphere may be transferred to plants by a number of processes,
primarily direct deposition onto the surface of plants and uptake by their roots from material deposited
on the soil. The most important processes for the transfer of radionuclides through the terrestrial
environment are illustrated in figure III.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 39

Figure III. The important processes for transfer of radionuclides through the terrestrial environment

41. There have been significant developments in the modelling of the transfer of radionuclides
through the terrestrial environment in recent years, partly as a result of studies following the Chernobyl
accident, see for example [C1]. The standard compilations of the concentration factors, linking different
parts of the environment, e.g. soil and plants, and other relevant parameters have been reviewed and
updated as part of international coordinated research activities [C1, I5].

42. For the purposes of this methodology, the Committee used the dynamic food-chain model
FARMLAND [B6], as implemented in PC-CREAM [S8], to derive integrated activity concentrations in
food (fresh weight) per unit deposition rate; these are the activity concentrations integrated to 100 years
for continuous deposition at a rate of 1 Bq/(m2 s) for one year. These estimated concentrations take
account of the various environmental transfer, migration and loss processes illustrated in figure III and
radioactive decay. Values of radionuclide transfer parameters for terrestrial environments were selected
from international compilations of data [I1, I5], supplemented by additional data [S8]. Expert
judgement was used to select values appropriate for the Committees purposes which are not overly
cautious. The values selected are intended to be appropriate for a range of environments and climates.
The FARMLAND model uses a standard approach to estimate the transfer of radionuclides through
terrestrial food chains and has been found to be in good agreement with measured data [S8]. In a
comparison exercise carried out as part of the IAEA EMRAS II (Environmental Modelling for
40 UNSCEAR 2016 REPORT

Radiation Safety) programme [I6], the results obtained from FARMLAND were found to be in good
agreement with those results from other models. The derived activity concentrations applied in the
Committees methodology are summarized in table 5 and full details of the model and parameter values
can be found in Brown and Simmonds [B6] and Smith and Simmonds [S8].

Table 5. Activity concentrations in foodstuffs integrated to 100 years for continuous deposition at
a rate of 1 Bq/(m2 s) for one year [S8]

Integrated activity concentration in food (fresh weight) (Bq/kg per Bq/(m2 s))
Radionuclide
Cereals Vegetables & fruits Milk & dairy products Meat & offal
(Ccereal,unit) (Cveg,unit) (Cmilk,unit) (Cmeat,unit)
35
S 3.7 105 1.2 105 1.4 106 6.0 106
54
Mn 8.6 104 1.3 105 9.0 104 1.9 105
58
Co 5.3 104 9.3 104 3.9 105 7.4 105
60
Co 1.1 105 1.5 105 2.3 106 1.3 107
65
Zn 1.8 105 2.2 105 7.0 105 5.8 106
90
Sr 7.1 105 7.3 105 1.5 106 3.3 105
106
Ru 9.8 103 1.0 105 1.1 103 9.7 105
129
I 6.0 105 2.3 105 3.1 105 2.1 105
131
I 4.2 104 4.1 104 5.8 104 2.5 104
134
Cs 4.9 105 1.4 105 2.5 105 1.2 106
135
Csa 7.0 105 3.3 105 3.0 105 1.6 106
137
Cs 5.9 105 2.2 105 3.0 105 1.5 106
138
Csa 8.5 100 1.7 102 1.3 102 1.3 103
210
Pb 4.1 104 1.4 105 1.2 104 2.4 104
212
Pba 1.6 102 3.2 103 2.1 102 8.5 101
214
Pba 7.0 100 1.4 102 6.6 101 1.1 104
214
Bia 5.4 100 1.0 102 1.4 100 1.7 102
210
Po 4.9 103 9.5 104 8.2 104 1.1 105
226
Ra 2.6 105 3.0 105 1.7 105 3.3 104
228
Raa 8.8 104 1.3 105 7.4 104 1.4 104
228
Tha 5.4 103 1.0 105 1.5 102 1.2 103
230
Th 7.9 103 1.1 105 1.7 102 2.2 103
232
Th 7.9 103 1.1 105 1.7 102 2.2 103
234
Tha 3.4 103 6.6 104 1.1 102 6.6 101
234
U 5.6 104 1.5 105 5.7 104 2.8 105
238
U 5.6 104 1.5 105 5.7 104 2.8 105
239
Pu 1.0 104 1.1 105 1.6 102 8.8 103
240
Pu 1.0 104 1.1 105 1.6 102 8.8 103
241
Am 1.5 104 1.1 105 8.5 101 4.7 103
a
Included as progeny only.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 41

43. The dispersion of 3H and 14C in the environment is more complex than that of other radionuclides
owing to the role of hydrogen and carbon in biological systems. These radionuclides therefore need to
be treated differently from other radionuclides. As previously, the Committee has assumed these
radionuclides reach equilibrium rapidly with their corresponding stable element, and used an approach
based on their specific activity.

44. The specific-activity model for tritium is based on the assumption that tritium behaves as
hydrogen in the environment so that tritium discharged to atmosphere exchanges with the hydrogen in
water in air, soil, plants and animals. The concentration of tritiated water (HTO) and organically bound
tritium (OBT) in plants is determined on the basis of the water content in plants and a partition factor
that takes account of the presence of exchangeable hydrogen in the dry weight of the plant [I5]. Full
details of the approach are given in the appendix. The concentrations of HTO and OBT in animal
products are determined using concentration factors that relate the concentrations in these products to
those in feed, drinking water and inhaled air.

45. The specific activity (or activity concentration) of 14C in elemental or stable carbon is determined
from the discharge by taking account of dispersion in the atmosphere and the concentration of stable
carbon in the atmosphere. The concentration of 14C, expressed in Bq/kg of stable carbon, is assumed to
be the same in plants as in air. The following equation is used to calculate the concentration of 14C in
terrestrial foods, Cf,C-14(x), at distance x from the discharge point:

Sp
C=
f ,14 C
( x ) Cair,14C ( x ) (2)
Sair

where Cair,C-14 (x) is the activity concentration of 14C in air at distance x (Bq/m3), Sp is the concentration
of stable carbon in the crop of interest (grams of carbon per kg fresh weight of crop) and Sair is the
concentration of stable carbon in air (grams of carbon per cubic metre of air).
14
Similarly the concentration of C in animal products, Cf,C-14 at distance x from the discharge point is
given by:

f C Cpasture,14C ( x ) S a
C f ,14 C ( x ) = (3)
Sp

where fc is the fraction of feed containing 14C (assumed to be 1 in the methodology), Cpasture,C-14 (x) is the
concentration of 14C in pasture at distance x from the discharge point (derived as for crops using
equation (2)), Sa is the concentration of stable carbon in the animal product (grams of carbon per
kilogram fresh weight) and Sp is the concentration of stable carbon in the pasture (grams of carbon per
kilogram fresh weight of pasture).

46. The values of the model parameters forming part of the specific-activity model for 3H and 14C are
summarized in table 6 (see the appendix for more information). The symbols used here and in the
workbooks are those used in the IAEA publication describing this model for ease of reference.
42 UNSCEAR 2016 REPORT

Table 6. Parameters and their values associated with the specific-activity model for 3H and 14C
(taken from [I5])

Parameter Symbol Value Unit

Stable carbon concentration in air Sair 0.2 g of C/m3

Stable carbon concentration in cereals (fresh weight) Sp 390 g of C /kg

Stable carbon concentration in vegetables (fresh weight) Sp 30 g of C /kg
Stable carbon concentration in pasture (fresh weight) Sp 100 g of C /kg

Stable carbon concentration in cow meat (fresh weight) Sa 200 g of C /kg

Stable carbon concentration in cow milk (fresh weight) Sa 65 g of C /kg
Fraction of feed containing 14C fc 1

Fractional water content of cereals (fresh weight) WCp 0.12 L/kg

Fractional water content of vegetables (fresh weight) WCp 0.92 L/kg

Fractional water content of pasture (fresh weight) WCp 0.76 L/kg

HTO:H2O water vapour pressures 0.909
Absolute humidity Ha 6 103 L/m3

Relative humidity RH 0.7

Empirical constant CRs-a 0.3
Concentration ratio for HTO intake through milk CRa,HTO 0.87 (Bq/kg fresh weight)
per (Bq/L)
Concentration ratio for HTO intake through meat CRa,HTO 0.66 (Bq/kg fresh weight)
per (Bq/L)
Water equivalent factora, cereals WEQp 0.56 L/kg
Water equivalent factor, vegetables WEQp 0.51 L/kg
Water equivalent factor, pasture WEQp 0.56 L/kg
Partition factor b
Rp 0.54
Concentration ratio for OBT intake, milk CRa,OBT 0.24 (Bq/kg fresh weight) per
(Bq/kg dry weight)
Concentration ratio for OBT intake, meat CRa,OBT 0.4 (Bq/kg fresh weight) per
(Bq/kg dry weight)
a
The water equivalent factor is the mass (kg) of water produced per unit mass (kg) of dry matter combusted.
b
The ratio of the concentration of non-exchangeable organically-bound tritium in combustion water to that in tissues (e.g. leaves).
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 43

4. External and internal exposure

(a) External exposure

47. The Committee considers that its previous methodology [U6] for estimating the dose from
external exposure due to immersion by assuming a semi-infinite cloud of radionuclides is still
appropriate. It acknowledges some limitations with this approach, most notably where activity
concentrations in air are likely to be non-uniform over a distance of a few hundred metres to one
kilometre from the point of discharge. However, given the distances over which the Committees
assessments are made, the application of dose coefficients for external exposure used previously, which
were based on [E2], is unlikely to be a source of significant error. The dose coefficients applied in the
Committees methodology are presented in table 1.

48. The annual characteristic individual dose from immersion in the plume, HE(ex,cloud),i for
radionuclide, i (Sv/a) is estimated at a series of distances, x (m), from the discharge point by assuming
100% occupancy (the fraction of the time that is spent at a particular location) at locations at those
distances (Oann (s/a)) and applying the relevant dose coefficients for external exposure due to immersion
in the plume, Dex, cloud,i from table 1 (Sv per (Bq s/m3)). Account is then taken of the fraction of time
spent outdoors, Oout (dimensionless) and the shielding effect of buildings, Lcloud (dimensionless) as
shown in equation (4) and discussed below.

( x ) Cair,i ( x ) Dex,cloud,i Oann ( Oout + (1 Oout ) Lcloud )

H E (ex,cloud),i= (4)

Where Cair,i (x) is the activity concentration of radionuclide i in air (Bq/m3) at location x. The annual
individual dose from external exposure due to deposited material, HE(ex,deposit),i (x) (Sv/a) at a distance
x (m) from the discharge point is calculated from the time-integrated activity concentration (to
100 years) of the radionuclide i on soil at the location of interest, di ( x) (Bq s/m2), the length of the
discharge, tdischarge (s/a) the relevant dose coefficient for external exposure due to deposition, Dex,deposit,i
(Sv/ Bq s/m2), the fraction of time spent outdoors, Oout (dimensionless) and the dimensionless location
factor that takes account of the shielding effect of buildings, as shown in equation (5) and discussed
below.

) di ( x ) tdischarge Dex,deposit,i ( Oout + (1 Oout ) Ldeposit )

H E ( ex,deposit ),i ( x= (5)

49. The Committees previous methodology applied the dose coefficients for external exposure from
radionuclides deposited on soil provided by Beck [B1]. An alternative approach was used for the
Committees 2013 Report of the levels of radiation exposure due to the nuclear accident at the
Fukushima-Daiichi nuclear power station [U12] based on the model published by Petoussi-Henss et al.
[P2]. The effective dose coefficients calculated using this more recent model [P2] were similar to or
within a factor of two of those used previously. For the sake of simplicity and consistency with its
2013 Report [U12], the Committee decided to apply dose rate coefficients for external exposure, edep
(nSv/h per kBq/m2) from Petoussi-Henss et al. These values are based on calculations using a voxel
phantom and the latest definition of effective dose [I13] (the use of the previous definition of effective
dose in the methodology would make little difference to the calculated effective doses for external
exposure). Where data for particular radionuclides were not available, the data set was supplemented by
factors derived from the United States Federal Guidance Report No. 12 [E2].
44 UNSCEAR 2016 REPORT

50. Petoussi-Henss et al. [P2] modelled the geometry as an infinite mono-energetic plane source
shielded by a soil layer of depth 0.5 g/cm2. Such an assumption gives a good description of the
radiation field after wet deposition on the ground and after dry deposition following the first rainfall
[I15, J1]. With this geometry, the air kerma from a 137Cs (137mBa) source is reduced by a factor of 0.67
compared to a plane source on the groundair interface. The effective dose coefficients were derived
assuming a constant density of air of 1.2 103 g/cm3 and a soil density of 1 g/cm3, which is considered
to be representative of the upper 2 cm of soil [P2].

51. In order to model the reduction of gamma dose rates because of migration into the soil, the
Committee adopted an attenuation function r(t) derived empirically from data on the migration of 137Cs
[G2]:

ln2 ln2
r ( t )= p1 exp t + p2 exp t (6)
T1 T2

where p1, and p2 are dimensionless parameters with values derived empirically, T1 and T2 are the initial
and final times, respectively, following deposition and t is the time after the deposition of interest. The
parameter values in this function were determined from gamma-spectrometric analyses of over 400 soil
samples taken during 19862003 in the areas of Germany (specifically Bavaria), the Russian
Federation, Sweden and Ukraine affected by deposition from the accident at the Chernobyl nuclear
power plant in 1986 (e.g. [G2, J3, L2, S2]). For the purposes of the Committees methodology,
parameter values derived for areas distant from the Chernobyl plant were applied, namely p1 was taken
to be 0.5 and p2 to be 0.5; T1 was assumed to be 1.5 years while T2 was assumed to be 50 years. Thus
the relationship below with T = 100 years was applied to the value of the dose coefficient from
deposition ( edep ) given in [P2]:

t li + ln 2 t
t
li +
ln 2
t

r ( t ) dt edep p1e
T2

( lit )
edep e
==
+ p2 e
T1
edep dt (7)

0 0

Substituting for p1, p2, T1, T2 in the above equations and changing from a decay constant to a half-life,
the effective dose (Sv) from external exposure in year 100, edep, is given by:

ln 2 ln 2
+ T ln 2 ln 2
+ T
edep edep
=
0.5 1 e t1/2 1.5 + 0.5 1 e t1/2 50 (8)
ln 2 ln 2 ln 2 ln 2
+ t + 50
t1/2 1.5 1/2

This can be simplified to

ln 2 ln 2
+ T ln 2 ln 2
+ T
=
edep edep 0.75t1/2 1 e t1/2 1.5
+
25t1/2 1 e t1/2 50
(9)
(t1/2 + 1.5) ln 2 (t1/2 + 50) ln 2

As an example, for caesium-137 and caesium-134 the following values can be calculated:
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 45

edep ( 137
)
Cs+137m Ba = 1.55 107 Sv/(Bq/m2) (10)

edep ( 134
)
Cs = 6.23 108 Sv/(Bq/m2) (11)

52. Despite differences in climate, little variation in indoor occupancies has been found between
countries and there is no evidence of a significant difference due to climate [A1]. Studies show that
indoor occupancies range from around 84 to 91% [G1, O2]. The Committee considered that values of
between 70% (for an outdoor worker) and 90% (for pensioners and indoor workers) were appropriate
for a Japanese population [U12]. In accordance with its previous methodology, the Committee assumed
an indoor occupancy factor of 80% [U6]. In equations (4) and (5) above the outdoor occupancy Oout is
used; this is simply a fractional value and is 1 0.8 = 0.2.

53. People indoors receive some protection from external exposure due to gamma-emitting
radionuclides in the plume as it passes overhead. The reduction in ambient dose equivalent rate depends
on the nature and structure of the building as well as on the energy of the radiation. A number of studies
of the shielding effect of various types of buildings have been published, particularly following the
Chernobyl accident [E2]. For example, the work of Le Grand [L1] and Jacob and Meckbach [J1, J2]
indicated that shielding factors (i.e. the ratio of the ambient dose equivalent rate indoors to that
outdoors) for radiation from airborne radionuclides ranged from 0.01 to 0.7 for single-family houses in
various European countries. For multi-storey buildings the shielding factor varies with storey, with the
lowest values in the basement and first storey and the highest value for the top storey. Le Grand [L1]
estimated that shielding factors appropriate for an airborne plume with photon energy of 0.68 MeV
ranged from 0.01 to 0.1 for the ground floor and from 0.09 to 0.4 for the upper floors of buildings. For
its methodology, the Committee has selected a shielding factor of 0.2 for radionuclides in the plume,
being within the range applicable to single and multi-storey buildings.

54. The UNSCEAR 1982 Report [U4] provided an overview of shielding factors appropriate for
deposited radionuclides, with values ranging from 0.05 in office buildings, 0.2 in masonry homes, and
0.4 in wooden buildings. This information was based on work by Burson and Profio, which showed a
range of shielding factors depending on the nature of the building, with the lowest factors being for
basements of large multi-storey buildings [B8]. The United States Nuclear Regulatory Commission
[N6] and others have suggested a generic shielding factor of 0.5 for regulatory purposes, which it
considers appropriate for photon energies above a few hundred kiloelectronvolts [E2]; for photons of
lower energy, use of such a value may considerably overestimate the dose equivalent [K6]. In their
assessment of the radiation exposure of the population of the United States of America [N3, S1], the
National Council on Radiation Protection and Measurements (NCRP) applied a shielding factor of
0.59 (6%) for terrestrial gamma radiation, which it considered to be appropriate for photon energies of
greater than a few hundred kiloelectronvolts.

55. Location factors are used to express the ratio of external exposures in terms of ambient dose
equivalent rate at a specific location to the ambient dose equivalent at the location for which calculated
or measured ambient dose equivalent was obtained (for example, this might be between the exposures
outdoors in an urban and a rural environment). This is a broader term than shielding factor which just
allows for the reduction of ambient dose equivalent from being indoors. For the purposes of this
methodology, the two terms can be considered to be equivalent, because it is the reduction in external
exposures from being indoors that is relevant. Golikov et al. evaluated the location factors for gamma
radiation in air in rural and urban environments in the Russian Federation five years after the Chernobyl
accident [G2]. In urban areas, they derived location factors for living areas that varied from 0.01 for a
multi-storey house to 0.09 for a wooden house, while they estimated the factor for buildings where
46 UNSCEAR 2016 REPORT

people work to be around 0.02. In rural areas, location factors for living areas ranged from 0.02 (for a
multi-storey house) to 0.13 (for a single-storey wooden building). This work formed the basis for the
time-dependent location factors used in the Committees 2013 Report on the levels and effects of
radiation exposure due to the nuclear accident at the Fukushima-Daiichi nuclear power station [U12].
Three different time-dependent location factors were used for that assessment: (a) for paved external
surfaces, (b) for non-paved external surfaces, and (c) for inside buildings. This time dependence is
important when assessing exposures due to an accident. However, for routine releases it is the dose in
the 100th year of continuous discharge that is of interest and the effect of using a time-dependent
location factor is small compared to differences in location factors for different building types. These
results also show a greater degree of shielding from deposited radionuclides than from radionuclides in
the plume, and this is reflected in the Committees choice of values for its methodology. In accordance
with its previous methodology [U6], the Committee has used a shielding (location) factor of 0.1 to
allow for the reduction in ambient dose equivalent from external exposure due to deposited material
while indoors. This value was chosen as representative of occupancy in a single-storey building and
was considered to be reasonably consistent with the result that would be obtained if an approach based
on time-dependent location factors had been used.

(b) Internal exposure from inhalation

56. As previously, the Committee uses a standard approach to assess the internal exposure from
inhalation of radionuclides in the air following discharges to atmosphere. This approach uses the
estimated activity concentration of a radionuclide in air (section III.A.2 above), an appropriate
breathing rate and the relevant dose coefficient for intake by inhalation (table 1).

57. The Committee continues to use a nominal adult breathing rate of 20 m3/d [U7] in its
methodology to maintain consistency with previous assessments, recognizing differences between this
value and that used by ICRP to derive dose coefficients. As discussed earlier, the Committee considers
that the dose coefficients for inhalation given in ICRP Publication 119 [I14] using the ICRP model of
the respiratory tract [I11] continue to be appropriate (table 1). No account is taken of any reduction in
activity concentrations in air when people are indoors because the Committee considers this to be a
second-order effect that will not have a significant effect on the estimated doses.

58. A different approach is needed for the inhalation of isotopes of radon. In the UNSCEAR 2006
Report, the Committee considered a number of issues related to the assessment of doses from the
inhalation of radon and its short-lived progeny [U9]. It concluded that it should continue to use a value
for the radon dose coefficient of 9 nSv per (Bq h)/m3 for this methodology, with equilibrium factors
(the ratio between the activity of the short-lived radon progeny and the activity that would be at
equilibrium with the radon parent) of 0.6 for radon outdoors and 0.4 for radon indoors (indoor
occupancy being 80% as discussed earlier) to calculate annual effective doses from inhalation of radon
and its short-lived progeny (see the appendix for more details).

(c) Internal exposure from ingestion

59. Members of the public may be exposed through ingestion of food that contains radionuclides
resulting from discharges to atmosphere. As indicated in section III.A.3, radionuclides discharged to
atmosphere may be transferred to human food by a number of routes. The annual effective dose from
the ingestion of food, f, containing radionuclide, i, at distance x (m) from the discharge point in region
r, (HE(ing),f,r,i (x)) (Sv/a) is given by equation (12):
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 47

H E ( ing ), f ,r ,i ( x=
) C f ,i ( x ) Ding,i Flocal I f ,r (12)

where Ding,i (Sv/Bq) is the effective dose coefficient for ingestion of radionuclide i as given in table 1;
Flocal is the fraction of food that is locally produced (dimensionless); If,r is the average consumption rate
of food f in region r (kg or L per year) as discussed below; Cf,i is the activity concentration of
radionuclide i in terrestrial food f at distance x from the discharge point (Bq/kg) given by:

f ,i ( x )
C= C f ,unit,i di ( x ) (13)

where Cf,unit,i is the activity concentrations in foodstuffs integrated to 100 years for continuous deposition
at a rate of 1 Bq/(m2 s) for one year (Bq/kg per Bq/(m2 s)) given in table 5 and di ( x) is the deposition rate
at distance x of radionuclide i (Bq /(m2 s)). The food contamination monitoring and assessment
programme (GEMS/Food) database for 2012 part of the WHO Global Environment Monitoring
System formed the basis for the population-weighted consumption rates used by the Committee to
calculate doses from ingestion for each UNEP region [W2]. For the Committees methodology, the World
Health Organization (WHO) combined the data for specified food groups and reconfigured them for the
UNEP regions and subregions (figure II). The resulting values of If,r for terrestrial foods are presented in
table 7. World-average values have also been derived; these are the averages of the values for each region
weighted by the population densities given in table 3. As discussed in paragraph 11, for the estimation of
characteristic individual doses, it is assumed that only 25% of foods consumed are produced locally (Flocal
is 0.25). However in estimating collective doses, people are assumed to obtain 100% of their food from
the area where they live (Flocal is 1.0), this assumption is considered reasonable because collective doses
are estimated for areas of 30,000 km2 or more.

Table 7. Annual average per caput consumption rates of terrestrial foods by UNEP region

Annual average per caput consumption ratea (kg) for each food type
UNEP region Cereals Vegetables and Milk and dairy Meat and offal
fruit products
Africa 130 220 31 17
Asia and Pacific 140 240 45 30
Europe 110 280 120 65
Latin America and Caribbean 110 210 77 63
North America 88 260 120 100
West Asia 140 180 43 34

World averageb 130 230 65 44

a
Values are rounded to two significant figures in order not to imply great precision.
b
World-average values are the averages of the values for each region weighted by the population densities given in table 3.

60. The Committee considered that the inclusion of all food groups and all 13 regions in the WHO
GEMS/Food [W2] would imply a level of accuracy that was inconsistent with other aspects of the
revised methodology. Although the dietary data in GEMS/Food were based on clusters of countries
with similar dietary habits, this clustering would not have been consistent with other information for
UNEP regions used to calculate doses (e.g. population distribution and energy production). Therefore, a
simplified approach was used based on the UNEP regions shown in figure II; any differences in intakes
and hence doses because of this simplification were deemed commensurate with those due to other
simplifying assumptions made in the methodology.
48 UNSCEAR 2016 REPORT

61. As discussed in section III.A.3, activity concentrations in foods integrated to 100 years for
continuous deposition at a rate of 1 Bq/(m2 s) for one year have been determined using combined food-
and radionuclide-specific transfer parameters. These estimated concentrations take account of a range
of physical processes that do not depend on the chemical element (such as interception by plant
surfaces) and biochemical factors that do depend on the chemical element, such as root uptake and
transfers from animal feed to meat and milk. For tritium and carbon-14, combined transfer parameters
are defined per unit activity concentration in air based on a specific-activity approach. The intake of
radionuclides is then determined using the population-weighted food consumption rates for the different
geographical regions. This is a slightly different approach from that used previously by the Committee,
where simple global average intakes were used, (see electronic attachments 1 and 2); however, the
overall difference in the resulting dose estimates is not significant.

62. A limited sensitivity analysis was carried out to determine the applicability of the generic food-
chain model for different regions of the world and whether the foods and related consumption rates
were appropriate. This analysis is outlined in electronic attachment 4 and the Committee concluded that
it was reasonable to use the food-chain model and the consumption data by geographical region for the
purposes of assessing worldwide exposures.

B. Assessment of doses from discharges to a freshwater environment

63. The extent of dispersion of radionuclides in freshwater bodies varies significantly depending on
characteristics of the water body, particularly the volume of the body into which the discharge occurs
and the water flows. However, the Committee decided that there would be value in developing generic
dose-calculation factors that provide an estimate of the integrated activity concentrations in water from
a discharge of each radionuclide for a year, allowing for dispersion within different types of water
body. An approach for estimating individual doses and for calculating regional components of
collective dose was developed on this basis. (As discussed previously, for discharges to rivers using the
simple, generic approach adopted in the methodology, it is not possible to distinguish between local and
regional components of collective dose).

64. The following exposure pathways are considered for discharges to freshwater bodies (see
figure I):

(a) Internal exposure from ingestion of radionuclides in drinking water;

(b) External exposure (beta and gamma emitters) from radionuclides in sediments deposited on
the riverbank;

(c) Internal exposure from ingestion of radionuclides incorporated into freshwater fish;

(d) Internal exposure from ingestion of radionuclides incorporated into terrestrial foods because of
irrigation.

65. Other exposure pathways such as internal exposure from inadvertent ingestion of water and
sediment, external exposure when swimming or boating, and external exposure from radionuclides
deposited on irrigated land, are also possible. However, the exposure pathways above have been found
to be the largest contributors to individual and collective doses following continuous discharges to
water bodies [J5]. The external exposure pathway, (b) above, is included only in the calculation of
characteristic individual dose. This pathway is unlikely to be a significant contributor to collective
doses from aquatic discharges from nuclear power stations and reprocessing plants, as demonstrated by
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 49

past assessments (e.g. [J6]). Individual doses are estimated from individual consumption rates and
occupancy data, while collective doses from ingestion are calculated using region-specific data on total
fish catches, water abstraction and agricultural production. Thus, the resulting individual and collective
dose estimates are not directly linked in the way that they are for discharges to atmosphere. In effect,
collective doses from discharges to a freshwater body are based on the total usage of water or total
amount of produce from the generic water body, irrespective of the location of people.

66. The Committees methodology initially considered three types of environment for assessing doses
arising from aquatic discharges to freshwater bodies: (a) lakes, (b) small rivers and (c) large rivers.
However, the Committee agreed that the dispersion of aquatic discharges from facilities on the shores
of a large lake can be assessed using the same approach as for large rivers. Given the differences in
characteristics and hence dispersion in different rivers and lakes throughout the world, the Committee
agreed that the distinction between lakes and large rivers was a second-order effect and did not need to
be considered further for its purposes. It should be noted that the Committees methodology does not
apply to closed lake systems, i.e. those without rivers feeding into or out of the lake.

67. Two types of river are defined to allow for the range of discharge conditions that may occur. For
example, inland nuclear facilities (with the exception of mines and mills) are assumed to discharge into
large rivers, and non-nuclear facilities and mines into small rivers. The dimensions and flow rates
associated with a range of rivers into which nuclear and other facilities discharge were reviewed in
order to select appropriate parameter values for the Committees methodology.

1. Dispersion of radionuclides in freshwater bodies

68. The Committees methodology for deriving activity concentrations of radionuclides following
their discharge into freshwater bodies assumes that complete mixing occurs immediately. This is a
simplification because for rivers complete dilution would occur over some tens of kilometres
downstream, the actual distance depending on the width of the river, which is in turn related to the
volumetric flow rate. Nevertheless, significant mixing does occur over relatively short distances. As an
illustration of this, table 8 shows results from using the NCRP screening model, which is designed for
the purpose of demonstrating compliance with environmental standards for discharges of radionuclides
[N2]. The table indicates that the activity concentration in unfiltered water assuming complete mixing
of the radionuclide across the total width of the river essentially equals the actual activity concentration
in unfiltered water in the plume (i.e. the value of the partial mixing correction factor approaches one)
for rivers up to a width of around 100 m at a downstream distance of around 5 km. For wider rivers, the
value of the correction factor would be of the order of one to five at this distance; complete mixing
would, however, occur over a distance of around 100 km for such rivers. Specific information is
available for the TechaIsetTobolIrtyshOb river system in the Russian Federation [N5]. This is a
complex river system and complete mixing does not occur until some distance downstream. (At 27 km
downstream there was not complete mixing with nearly a factor of 5 variation in measured
concentrations at different locations across the river.) However, the abstraction point for drinking water
is likely to be at some distance downstream from the discharge point and fish will be caught over large
sections of the river. Given this and the generic nature of the revised methodology, the distances over
which collective doses are calculated, and other uncertainties associated with the approach, the
Committee considered it unlikely that assuming complete mixing would significantly affect the
collective dose estimate. Although the dose to an individual obtaining their drinking water from the
immediate vicinity of a source would depend on the location of the abstraction point relative to the
point of discharge, because of the generic nature of the approach, assuming complete mixing is also
appropriate for individual dose estimation.
50 UNSCEAR 2016 REPORT

Table 8. Correction factors to complete mixing in a river plume as a function of downstream

distance from the point of discharge for different river widths [N2]

The correction factor for partial mixing represents the ratio of the activity concentration in water in the plume to the activity
concentration in water if, at the same point, complete mixing in the total width of the river were assumed. Complete mixing
would give a value of 1

Distance downstream Partial mixing correction factor at various river widthsa

(km) 10 m 50 m 100m 200 m 400 m
0.1 3.2 6 8 10 12

1 1 2.4 3.2 5 8
5 1 1 1.6 2.6 3.6
10 1 1 1 1.9 3

50 1 1 1 1 1.4
100 1 1 1 1 1
a
Values are rounded to one decimal place.

69. For the purposes of the Committees methodology, the activity concentrations of radionuclides in
river water are therefore derived assuming that the discharged radionuclides are dispersed uniformly
across the river. Under these conditions, the activity concentrations in water depend upon the
volumetric flow rate of the river alone, which, in turn, is primarily determined by the width of the river.
Thus, it is assumed that, for all freshwater pathways, the activity concentration in unfiltered freshwater
is given by:

Cuw,i = Qi / F (14)

where Cuw,i is the activity concentration of radionuclide, i, in unfiltered water at the discharge point
(Bq/m3); Qi is the discharge rate of radionuclide, i, (Bq/s) and F is the volumetric flow rate of the river
at the point of discharge (m3/s). The activity concentrations of radionuclides in filtered water are
determined from the suspended sediment load and the radionuclide-specific partition factors for
freshwater environments. These activity concentrations are then used, for example, to derive activity
concentrations of radionuclides in freshwater fish. To derive activity concentrations of radionuclides in
drinking water, a water treatment factor is applied.

70. For the large river (and as noted earlier, for lakes), a width of 240 m and flow rate of 1,000 m3/s
are assumed for the purposes of the methodology. Table 9 gives some illustrative flow rates and other
dimensions for a range of rivers of different sizes throughout the world. The values assumed for a large
river are consistent with the data for major sections of the Rhne, Loire, Danube and Rhine; these are
rivers into which a number of nuclear installations discharge radionuclides. They are also similar to
other major rivers throughout the world, as shown in table 9, although some rivers have greater flows,
notably the Amazon.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 51

Table 9. Flow rates and dimensions of some typical rivers and those assumed in the methodology

River Country Flow rate Length in Width Depth

(m3/s) specified (m) (m)
country (km)
Amazona Peru, Columbia, Brazil 2.0 105 6.9 103
Danubeb Germany, Austria, Slovakia, 3 103 3 103 5 102 6
Hungary, Croatia, Serbia,
Bulgaria, Romania,
Moldova, Ukraine
Gangesa India, Bangladesh 1.6 104 2.5 103
Kennet (tributary of Thames) United Kingdom 1 101 3.3 101 17 1
Loirec France 7.8 102 5.6 102 2.6 102 3
Mississippia United States of America 1.6 104 3.7 103
Niled,e Egypt 1.8 103 1.5 x 103

Paran f Brazil, Paraguay, Argentina 1.7 104 4.8 103 1.8 104
Pearl Chinag 2.2 103 2.2 103
Rhnec France 1.2 103 4.0 102 2 102 7
Thames h
United Kingdom 5 10 1
2.3 10 2
50 2
Yangtze Chinag
3.2 10 4
6.3 10 3

Yellow Chinag 2.1 103 5.5 103

ASSUMED VALUES IN THE METHODOLOGY

Methodologysmall riveri 1 101 1 102 30 1

Methodologylarge riveri 1 103 5 102 240 4
a
Milliman and Farnsworth (2011) [M2].
b
Maringer (2000) [M1].
c
Smith and Simmonds (2009) [S8].
d
Wahaab and Badawy (2004) [W1].
e
Hamza (2014) [H1].
f
Encyclopaedia Britannica (Parana-River) [E3].
g
Information provided by the Chinese delegation.
h
Hilton et al. (2003) [H4].
i
The dimensions of the small river are similar to those of the River Kennet in the United Kingdom; those of the large river are
similar to those of major sections of the Rhne and Loire in France and other rivers.

71. For the small river, a flow rate of 10 m3/s (width of 30 m and depth of 1 m) is assumed in the
methodology. This is similar to that of the River Kennet in the United Kingdom.

72. The measured suspended sediment loads of the Colorado and Mississippi Rivers have been shown
to vary by three or more orders of magnitude [H2, S10]. However, a value of 5104 t/m3 is considered
to be appropriately representative within this range and is assumed for large rivers; this also agrees with
other data presented for the Colorado River [V1].
52 UNSCEAR 2016 REPORT

73. The suspended sediment loads of small rivers also vary by several orders of magnitude,
depending on flow rate. For low flow rates, values in the range 2 to 5105 t/m3 may be considered to
be typical (see, for example, Hejduk and Banasik [H3]). A value of 2105 t/m3 was adopted for the
purposes of this methodology.

2. Behaviour of radionuclides in a freshwater environment

(a) Transfers in freshwater environments

74. The key element-specific parameters that account for the transfer of radionuclides in an aquatic
environment are concentration factors that relate the concentrations of elements in fish to their
concentrations in water, and partition coefficients that relate the distribution of elements between
sediment and water. Standard compilations of these data [I3, I5] have also been updated since the
Committees assessments were issued in the UNSCEAR 2000 and 2008 Reports [U6, U10, U11].

75. The activity concentration of a radionuclide in an aquatic food is derived by applying an

equilibrium concentration factor, defined as the ratio of the concentration of the radionuclide in the
aquatic food (fresh weight) at equilibrium to its concentration in water (Bq/kg per Bq/L) (note in some
compilations of data the term bioaccumulation factor is used for these values). The most recent
international compilation of concentration factors for freshwater environments [I5] provides transfer
parameter values appropriate for equilibrium conditions. Where the available data permit, ranges of
observed values are presented together with mean values. Those mean values have been used for the
purposes of this methodology supplemented by data from other sources, where necessary; the values
adopted by the Committee for this methodology are given in table 10.

76. The transfer of radionuclides between the water column and suspended and bottom sediments
depends on both the characteristics of the water body and the chemical and physical characteristics of
the radionuclides themselves. The partition coefficient, Kd, relates to the partitioning of radionuclides
between the solid and aqueous phases and is expressed as the ratio of the activity of a radionuclide per
unit dry weight of sediment at equilibrium to the activity of that radionuclide per unit volume of water
(Bq/kg per Bq/L) (this can also be referred to as a distribution coefficient). The most recent
international compilation of Kd values for freshwater environments is presented in [I5], although the
Committee also used values from other sources, where data were not available in [I5]. The data used
are given in table 10 and are intended to represent generic best estimates, recognizing that Kd values
vary depending on the characteristics of the water body.
Table 10. Values of radionuclide-specific parameters used in the model for freshwater environments

Radionuclide Concentration factor for freshwater fish, Activity concentrations in sediment (dry weight) in the Water treatment factor Partition coefficient, Kd
Bfish (L/kg) (fresh weight) 100th year of continuous discharge at 1 Bq/s (Bq/kg) FWT (m3/t)
[B7]a
Small river Large river
[S8] [S8]
3
H (HTO and OBT) 1.0 100 [N2] 0 0 1.0 100 0 [I2]
14
C 5
4.0 10 [I5] 5.0 104
5.0 106
1.0 100
5.0 100 [I2]
35
S 8.0 102 [N2] 2.0 102 1.8 104 5.4 101 2.0 102 [B4]
54
Mn 2.4 102 [I5] 3.1 100 2.0 103 3.6 101 7.9 104 [I5]

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 53

58
Co 7.6 101 [I5] 2.3 100 1.9 103 5.4 101 4.4 104 [I5]
60
Co 7.6 101 [I5] 2.3 100 1.9 103 5.4 101 4.4 104 [I5]
65
Zn 3.4 103 [I5] 4.9 102 4.0 104 5.4 101 5.0 102 [I2]
90
Sr 2.9 100 [I5] 1.2 101 7.5 104 8.1 101 1.2 103 [I5]
106
Ru 5.5 101 [I5] 2.0 100 1.9 103 5.4 101 3.2 104 [I5]
129
I 3.0 101 [I5] 4.0 101 1.4 103 8.1 101 4.4 103 [I5]
131
I 3.0 101 [I5] 4.0 101 1.4 103 8.1 101 4.4 103 [I5]
134
Cs 2.5 103 [I5] 1.8 100 1.9 103 8.1 101 2.9 104 [I5]
137
Cs 2.5 103 [I5] 1.8 100 1.9 103 8.1 101 2.9 104 [I5]
210
Pb 2.5 101 [I5] 8.3 101 1.7 103 5.4 101 1.0 104 [K1]
212
Pb 2.5 101 [I5] 6.3 101 1.5 103 5.4 101 1.0 104 [K1]
214
Pb 2.5 101 [I5] 1.3 103 2.1 104 5.4 101 1.0 104 [K1]
210
Po 3.6 101 [I5] 2.2 100 1.9 103 5.4 101 4.0 104 [S3]
226
Ra 4.0 100 [I5] 6.4 101 1.6 103 5.4 101 7.4 103 [I5]
228
Th 6.0 100 [I5] 4.0 100 2.0 103 3.0 101 1.9 105 [I5]
230
Th 6.0 100 [I5] 4.0 100 2.0 103 3.0 101 1.9 105 [I5]
232
Th 6.0 100 [I5] 4.0 100 2.0 103 3.0 101 1.9 105 [I5]
234
Th 6.0 100 [I5] 3.9 100 2.0 103 3.0 101 1.9 105 [I5]
234
U 9.6 101 [I5] 5.0 103 4.9 105 3.0 101 5.0 101 [I2]
238
U 9.6 101 [I5] 5.0 103 4.9 105 3.0 101 5.0 101 [I2]
 54 UNSCEAR 2016 REPORT
Radionuclide Concentration factor for freshwater fish, Activity concentrations in sediment (dry weight) in the Water treatment factor Partition coefficient, Kd
Bfish (L/kg) (fresh weight) 100th year of continuous discharge at 1 Bq/s (Bq/kg) FWT (m3/t)
[B7]a
Small river Large river
[S8] [S8]
239
Pu 3.0 101 [I2] 4.1 100 2.0 103 2.7 101 2.4 105 [I5]
240
Pu 3.0 101 [I2] 4.1 100 2.0 103 2.7 101 2.4 105 [I5]
241
Am 2.4 102 [I5] 3.5 100 2.0 103 2.7 101 1.2 105 [I5]

a
The water treatment factors used are based on flocculation, coagulation, clarification and rapid sand filtration. Where data were not given for a specific element in Brown et al. [B7], an analogue approach was used.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 55

(b) Transfers from fresh water to terrestrial environments

77. The irrigation of crops with fresh water may lead to the transfer of radionuclides from a
freshwater to a terrestrial environment. This exposure pathway was not included in the Committees
previous methodology but, given the importance of irrigation of crops throughout the world, the
Committee decided that it should be considered. In general terms, the rate at which radionuclides
present in fresh water are deposited on crops and soil is estimated from information on the rate of
irrigation and the length of the crop-growing season. This information is then used to estimate the
resultant activity concentrations of radionuclides in terrestrial crops from their activity concentrations in
fresh water and the derived calculation factors for their activity concentrations in food per unit
deposition rate.

78. There are many different types of irrigation. The most important for the transfer of radionuclides
to crops is sprinkler or spray irrigation because radionuclides in the water will be deposited directly on
the surface of the plants. Other types of irrigation where water is deposited on the surface of the soil
give rise to lower transfers and so were not included in the Committees methodology. Although there
are obvious differences in the behaviour of radionuclides that are deposited on the surface of plants
during sprinkler irrigation and that of radionuclides deposited by wet and dry deposition from the
atmosphere, the Committee considered that the differences in the overall transfers to plants are a
second-order effect in the context of the objectives of the methodology. Therefore, the values of the
transfer factors for deposition from the atmosphere are assumed to apply also to irrigation. These
parameters include the transfers from plant surfaces and from uptake from the soil. The Committee
recognizes that there may be some overestimation in assuming that the rate of uptake from irrigation
water is equivalent to that from deposition from the atmosphere, given that greater losses from plant
surfaces may be anticipated for the higher deposition rates characteristic of irrigation.

79. The Food and Agriculture Organization of the United Nations (FAO) publishes statistics on water
resources in its AQUASTAT database [F4]. The information indicates that 40% of global food
production involves irrigation. Data are available on the overall rates of abstraction of water by source
(whether from surface or groundwater). Information is available on the overall rate of water abstracted
for irrigation, but not on the fraction of irrigation water that is drawn from each source. In the absence
of this information, it is assumed that the fraction of water used for irrigation of the total amount
abstracted is the same for both sources. Using data from the AQUASTAT database, the fraction of
surface water withdrawn for irrigation and the area irrigated per unit volume of water was obtained for
each region as shown in table 11, which also gives world-average values. The appendix describes how
these values are used in the methodology.

80. From this table, it is clear that, with the exception of West Asia, there is only limited abstraction
of water from surface water.

81. The International Commission on Irrigation and Drainage provides data on the area of land under
different types of irrigation for 45 countries [I7]. The Committee has used these data to derive the
fraction of land under sprinkler/spray irrigation for each region and an average for the world (see
table 12). The values in this table are for various times between 1999 and 2012.

82. The fractions given in table 12 are used in the methodology together with the calculated activity
concentrations of radionuclides in water and the abstraction rate for irrigation to derive concentrations
of radionuclides in foods and hence to estimate individual and collective doses. The Committee agreed
that irrigation should only be considered for vegetables and grain and only for large rivers. For small
56 UNSCEAR 2016 REPORT

rivers with low abstraction rates, it is unlikely that there is significant irrigation. Although other crops
are likely to be irrigated, the most important are leafy vegetables and grain [F4]. The irrigation of
pasture is important in some parts of the world (such as parts of the United States) but it is rare in other
major areas with significant cattle production, such as Brazil and Argentina, and so it is not included in
the methodology.

Table 11. The fraction of water for irrigation withdrawn from surface waters and the total area
irrigated per unit volume of surface water available

Parameter
Region Fraction of water for irrigation withdrawn Area irrigated per unit volume of
from surface waters (dimensionless) withdrawn water (m2 a /m3)
Africa 2.7 102 6.0 101
Asia and Pacific 9.4 102 7.9 101

Europe 1.2 102 1.9 100

Latin America and Caribbean 1.0 102 8.2 101

North America 2.4 102 1.3 100

West Asia 5.7 101 4.7 101

World average 3.9 102 8.5 101

Table 12. Fraction of irrigated land that is spray-irrigated (based on information from [I7]

Region Fraction
Africa 0.2
Asia and Pacific 0.1

Europe 0.4
Latin America 0.2
North America 0.5
West Asia 0.3

World average 0.2

(c) Specific activity models for tritium and carbon-14 in fresh water
83. The Committee has adopted a specific-activity approach for assessing the transfer of tritium and
carbon-14 in a freshwater environment. This is the same approach as it used previously and as
described for a terrestrial environment. The assumption that equilibrium conditions exist, which is
implicit in the specific-activity model, is considered to be a good approximation for most aquatic
compartments in the model [E1, K5]. The model adopted is that described in [I5] in which activity
concentrations of tritium in fish are determined on the basis of the HTO concentration in the water
column and the fractional water content of fish. The fractional water content is found to be 0.78 L/kg
for most fish that form part of the human diet [I5]. The concentration of organically bound tritium
(OBT) is also taken into account. The relevant equations and parameter values used in the methodology
are given in the appendix.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 57

84. A specific activity approach is also applied to estimate the total tritium concentration in soil water
by taking account of the activity concentration of tritium in irrigation water, the rate of irrigation and
the effect of mixing with water reaching the soil from precipitation [C1]. The equations and data
applied are presented in the appendix.

85. For discharges to aquatic bodies, the specific activity of dissolved inorganic carbon is assumed to
be in equilibrium within the part of the environment of interest. The methodology includes a simplified
approach based on a dynamic model developed by Sheppard et al. [S4, S5] with some modification to
apply it to irrigation based on the approach outlined in [C1]. The equations and parameter values
applied are given in the appendix.

3. External and internal exposure

(a) External exposure from radionuclides in riverbank sediments

86. Annual individual doses from external exposure during occupancy of riverbanks are calculated
using the time-integrated activity concentrations of radionuclides in sediments, the dose coefficients for
external exposure from surface deposits and the amount of time spent on riverbanks. This exposure
pathway is only considered for the calculation of the characteristic individual dose.

87. For discharges to fresh water bodies, activity concentrations of radionuclides in sediments at a
distance of 5 km downstream from the discharge point are used for the purpose of calculating
characteristic individual doses from external exposure due to radionuclides in riverbank sediments. It is
assumed that there will be continual cycling of radionuclides from the aquatic environment into
riverbank sediments taking account of build-up over the 100 year integration period. Dose coefficients
for external radiation exposure from surface deposits (Sv/s per Bq/m2) from the United States Federal
Guidance Report No. 12 [E2] are used in the Committees revised methodology. However, these dose
coefficients apply to an infinite surface (assuming a surface deposit only); for river banks, a geometry
factor is applied to allow for the finite size of the source of external exposure. Apostoaei et al. [A2]
have discussed appropriate geometry factors (these depend upon the surface area of the sediment on the
river bank, as represented by the width of the sediment on the bank, and upon the radionuclide). They
recommend that factors for 137Cs are applicable in general. The Committee agreed to apply a geometry
factor of 0.2 based on the values given by Apostoaei et al. for riverbank sediments with a width of a
few metres. In addition, the Committee agreed that the methodology should use an annual occupancy
factor of 50 hours for the calculation of the characteristic individual dose.

(b) Internal exposure from ingestion

88. Three sources of internal exposure via ingestion are considered in the methodology: drinking
water; freshwater fish and irrigated terrestrial foods. In estimating the characteristic individual doses, it
is assumed that only 25% of the food is locally produced (see section II, paragraph 11).
58 UNSCEAR 2016 REPORT

Drinking water

89. As described above, the activity concentrations of radionuclides in unfiltered fresh water are
assumed to be a function of the volumetric flow rate of the river, which is, in turn, directly related to the
type of river assumed. In order to derive activity concentrations in drinking water, element-specific
water treatment factors are applied. These are similar to those presented in the UNSCEAR 2000 Report
[U6], updated with information included in the most recent edition of the WHO Drinking Water
Guidelines [W3]. The WHO provides indicators of the performance of a number of common water-
treatment methods for a range of radionuclides.

90. Different treatment techniques are applied across the world and have different effectiveness.
Based on information used for the WHO Drinking Water Guidelines [W3], the Committee assumed
thatfor most countries with high human development indicesflocculation, coagulation, clarification
and rapid sand filtration are used. As there are insufficient data available on the techniques employed in
less developed countries, the Committee decided to apply a single set of water-treatment factors based
on the data for countries with high human development indices. The fractions of each radionuclide
removed by standard water-treatment processes assumed in the methodology are shown in table 10.

91. Previously, the Committee used a value of 500 litres per year (1.4 litres per day) [U6, U7, U9] for
the worldwide average individual rate of ingestion of drinking water. The Committee has reconfirmed
that this value should be retained because it is appropriate for average individual consumption rather
than the somewhat higher values typically used for protection purposes.

Freshwater fish

92. Individual doses from the ingestion of freshwater fish are calculated using the activity
concentrations of radionuclides in fish and the per caput consumption rate; collective doses are
calculated using the activity concentrations and fish-catch data. Information on fish-catch data for each
region in 2007 was obtained from [F3]. The total fish catch for each region was divided by the size of
the population for the same region to derive per caput consumption rates, which were used to estimate
characteristic individual doses. Table 13 shows the resulting annual per caput consumption rates of
freshwater fish for each region and a world-average value.

Table 13. Annual per caput consumption of freshwater fish by region

These values represent the total intakes of freshwater fish; the factor of 25% to account for local consumption of food is
applied subsequently when calculating characteristic individual doses

Region Annual consumption (kg)

Africa 2.7

Asia and Pacific 7.8

Europe 3.4
Latin America and Caribbean 1.6

North America 4.5

West Asia 1.3

World average 5.7

 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 59

93. For the calculation of collective doses, the Committee has derived generalized fish-catch
information for a range of rivers using data from FAO [F4]. Aquaculture (e.g. the intensive farming of
freshwater fish) is excluded from consideration because a large proportion of fish reared in aquaculture
come from lakes and ponds [F4].

94. Typical values for the quantity of freshwater fish caught per unit distance and volume of water for
typical small and large rivers have been derived for use in the Committees methodology for each
region separately and as a world average. From the FAO data, information was derived on the catch of
freshwater fish including and excluding that obtained from aquaculture, plus the annual fish catch for
each river in a country. Using information on the length and volume of the relevant stretch of each
river, annual fish catches per unit length and per unit volume were calculated and the results are shown
in table 14. Based on this information, generic values for small and large rivers were adopted for use in
the methodology. It is therefore assumed that one tonne of freshwater fish is caught annually per
kilometre from small rivers (of length less than 500 km) and that ten tonnes are caught annually per
kilometre from large rivers (length greater than 500 km). Because of the volumes of water assumed for
the two river sizes, which also differ by a factor of 10, the freshwater catch per unit volume is
independent of river size.
 60 UNSCEAR 2016 REPORT
Table 14. Freshwater fish-catch data for a range of rivers

Data taken from FAO [F1] unless otherwise stated

Annual fish catch

Country Country-wide annual fish catch River River Annual fish catch
for the river(s)

Including Excluding Excluding Length in specified Per unit volume, excluding Per unit length, excluding
Volume (m3)
aquaculture (t) aquaculture (t) aquaculture (t) country (km) aquaculture (t/m3) aquaculture (t/km)

Australia 4.0 103 1.2 103 All rivers 6.2 101

China 1.9 107 1.6 106 Pearl 2.8 104 1.4 103 2.3 109 1.0 105 2.0 101
Egypta 9.4 105 2.3 105 Nile 2.3 105 1.5 103 1.5 102

France 4.4 104 2.6 103 Loire 1.0 103 5.6 102 4.3 108 2.0 106 1.9 100

Rhne 1.5 103 4.0 102 6.3 108 2.0 106 3.9 100

All rivers 2.6 103 3.0 10-1

Germany 5.6 104 2.1 104 Danube 7.9 103 5.0 102 2.1 108 4.0 105 1.6 101
Japan 6.7 104 2.5 104 All rivers 1.4 101

Romania 1.6 104 5.7 103 Danube 5.7 103 4.0 102 3.2 109 2.0 106 1.4 101

Sudanb 8.1 104 7.9 104 Nile 5.4 104

Nile tributaries 2.5 104

(incl. Blue and
White Nile)

United Kingdom 1.6 104 2.5 103 Thames 2.3 102 2.3 102 2.5 107 9.0 106 9.8 101

Kennet (tributary 6.0 100 3.3 101 6.1 105 9.0 106 1.7 101
of Thames)

All rivers 2.5 103 8.0 101

United States 2.5 105 1.4 104 All rivers 3.4 101
a
Additional data obtained from Hamza [H1].
b
Data for 2012 from internal report published by the Sudan Ministry of Animal Resources and Fisheries [S11].
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 61

Irrigated foods

95. As discussed in section III.B.2(b), the ingestion of irrigated foods is included in the methodology
for the estimation of both characteristic individual and collective doses. In the calculation of individual
doses from this pathway, it is assumed that cereals and vegetables are irrigated. For the assessment of
collective doses, it is assumed that only cereals are irrigated, because these represent the bulk of
irrigated crops intended for human consumption [F4].

96. As illustrated in table 15 [C1], there are variations in irrigation rate which are likely to depend to
some extent upon factors such as the type of crop and the climate. A single value for a daily irrigation
rate was considered to be appropriate for the purposes of this methodology. The United States NCRP in
its screening methodology uses a value of 5 L/m2 for the volume of irrigation water applied per unit
area for a period of 150 days annually[N2]. The FAO AQUASTAT database [F4] contains data for a
wide range of countries on irrigation water withdrawals and/or requirements, and on the total areas of
irrigated crops harvested. These data could then be used to indicate values of daily irrigation rates in
L/m2 assuming a total irrigation period of 150 days in a year. Table 16 illustrates the range of irrigation
rates thus obtained from the FAO data. The irrigation rate varies from country to country and although
some of the highest values are for arid areas of the world, some countries that are generally arid (e.g.
Peru) do not have particularly high estimated irrigation rates. From the data in tables 15 and 16 a single
value for the daily irrigation rate of 4 L/m2 was adopted for the methodology.

Table 15. Irrigation rates averaged over the growing season for various crops and climates [C1]

Country or area Crop type Daily irrigation rate (L/m2) Reference

Min Average Max

Canada Forage 0 [C6]

Garden vegetables 0 1.8

France Fruit 2 6 [C4]

(Loire Valley)
Garden vegetables 2 6

Grain 0 2.4

Maize 0.8 2.8

India Banana 2.6 [P1]

Gram 1.8

Ground nut 6.3

Improved jowar 2.0

Pigeon pea 1.6

Rice 0.9

Sugar cane 4.1

Turmeric 2.1

Wheat 4.3

Republic of Korea Rice 5 8 [J4]

62 UNSCEAR 2016 REPORT

Country or area Crop type Daily irrigation rate (L/m2) Reference

Min Average Max

United States Alfalfa 4.9 7.1 [U14]

(California)
Barley 2.4 3.2

Fruit orchards 3.9 4.9

Garden vegetables 4.2 5.2

Hay 3.2 5.4

Oats, rye 2.0 2.5

Pasture 3.7 4.4

Grapes 1.7 [B5]

Table 16. Effective irrigation rates obtained from data in the FAO AQUASTAT database [F4]

Country Daily irrigation rate (L/m2)

Algeria 5.2
Argentina 3.5

Australia 3.3
Bangladesh 2.8
Brazil 1.9

Canada 4.1
China 1.8
Egypt 5

France 1.4
Greece 2.8
India 2.8

Indonesia 2.3
Japan 8.6
Kenya 3.2
Peru 3.5
Qatar 7.3

Russian Federation 6.5

Spain 4.5
United States 5.1

97. Collective doses from irrigation are determined by assuming that, on average, the fraction of
water abstracted for irrigation from the type of river of interest, in a given geographical region, is the
same as the fraction of water abstracted for irrigation purposes from all surface waters (rivers, lakes and
so on). The fraction of water abstracted for irrigation and the amount of land irrigated per unit volume
of water, was derived from information in the FAO AQUASTAT database [F4]. This information,
together with the annual yields of cereals [F1] (given in table 17), the integrated activity concentrations
of radionuclides in cereals for unit deposition rate (given in table 5), and the percentage of irrigation
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 63

that is sprinkler irrigation (see table 12 and section III.B.2(b)), is used to calculate collective doses from
irrigation. Table 17 shows the parameter values used in the methodology for the estimation of collective
doses in different regions from the irrigation of cereals; world-average values are also given.

Table 17. Parameter values used in the model for the irrigation of cereals

Region Annual cereal yield Fraction of surface water Area irrigated per unit
(kg/m2)a used for irrigationb volume of water
Firr,surface withdrawnb
Airr,unit (m2 a/m3)
Africa 2.3 101 2.7 102 6.0 101
Asia and Pacific 2.9 101 9.4 102 7.9 101
Europe 3.6 101 1.2 102 1.9 100

Latin America and Caribbean 3.3 101 1.0 102 8.2 101
North America 3.0 101 2.4 102 1.3 100
West Asia 2.9 101 5.7 101 4.7 101

World average 3.0 101 3.9 102 8.5 101

a
From FAO 2010 [F1].
b
From FAO 2014a [F3].

4. Individual and collective doses

98. Annual characteristic individual doses arising from the ingestion of drinking water and freshwater
fish are calculated using the derived activity concentrations of radionuclides in water and fish, the
relevant consumption rates, and the dose coefficients for ingestion (see table 1). Because the activity
concentrations in unfiltered fresh water are assumed to be a function of the river flow rate alone, these
individual dose estimates are directly related to the type of river assumed.

99. Collective doses are calculated using the total abstraction rate of drinking water and the total catch
of fish associated with the freshwater body. The abstraction rate of drinking water is assumed to be a
fraction of that abstracted for municipal use. The abstraction rate was derived from FAO statistics
which indicate that around 110% of domestic water is used for drinking [F1]. Since not all municipal
water will be for domestic use, the lower end of this range is applied in this methodology. The total
amount of water abstracted depends on the size of the water body with smaller amounts removed from
small rivers than from large rivers; the water flow is also inversely related to this size. This means that,
although for small rivers the activity concentrations per unit discharge are higher than for large rivers,
the amount of water abstracted is lower and thereforeusing the Committees methodologythe
collective doses from drinking water are estimated to be essentially independent of river size.

100. The irrigation pathway is considered using generic irrigation rates and transfer factors appropriate
for a terrestrial environment. Characteristic individual doses from this pathway are determined using
region-specific population-weighted consumption rates and the assumption that 25% of the food is
locally produced. Collective doses are determined from information on the region-specific proportion
of water abstracted from surface waters for irrigation purposes, the area irrigated and the annual yield
of cereal products.
64 UNSCEAR 2016 REPORT

101. Collective doses from the consumption of water and freshwater fish are calculated on the basis of
the total abstraction of water and fish-catch data, respectively, and are thus related to the river as a
whole and cannot be divided into local and regional components.

C. Assessment of doses from discharges to a marine environment

102. As indicated above, the Committee decided to develop generic dose calculation factors that take
account of dispersion within different types of water bodies. The approach used for assessing doses
from discharges into a marine environment is based on a simple two-box compartment model, the
characteristics of which are broadly representative of areas where discharges into a coastal environment
occur (see figure IV, which is based on [C2]). The larger compartment (referred to as the regional box)
could also be used to assess the dispersion of radionuclides discharged into the deep sea from oil and
gas platforms, if required.

103. The following exposure pathways are considered for discharges to a marine environment (see
figure I):

(a) External exposure from radionuclides (beta and gamma emitters) in sediments;

(b) Internal exposure from ingestion of radionuclides incorporated into marine foods.

The first pathway is considered for the calculation of characteristic individual doses only; a review of
previous studies assessing collective dose indicated that this pathway provided less than 1% of the
collective dose. Similar to discharges to atmosphere and freshwater environments, other exposure
pathways (e.g. inadvertent ingestion of water or sediments, and exposures during swimming) are
possible; however, the ingestion of marine foods has been found to be the most important exposure
pathway in published assessments of collective doses (for example, see [J5]).

Figure IV. Simple two-box compartment model representing a marine environment

Q is the discharge rate (Bq/a); l,r is the effective rate of transfer between the local and regional compartments (a1), taking
into account exchange between the compartments; r,g is the effective rate of transfer between the regional and global
oceans (a1), taking into account exchange between the compartments; s,l and s,r are the rates of transfer to sediment for
the local and regional compartments, respectively (a1), and li is the radioactive decay constant for radionuclide, i, (a1)
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 65

104. Figure IV illustrates the simplified approach used to model the movements of radionuclides between
marine compartments. The model considers inputs (such as discharges and incoming transfers of
radionuclides) and losses (such as radioactive decay, sedimentation and outgoing transfers of
radionuclides). This allows the activity concentrations in the local and regional compartments to be
determined. The equations and parameter values associated with this approach are presented in the
appendix. The model includes transfer to other marine areas but any contribution to collective doses from
the global oceans is only considered for the long-lived radionuclides (3H, 14C and 129I) in a separate global
circulation model as discussed below. The size of the regional compartment and the related transfer
parameters have been chosen to ensure that the model is a reasonable representation. A comparison with a
more complex multi-compartmental marine model [S7, S8], which has been validated against
measurement data, has shown that the simple model is adequate for the purpose of this methodology.

105. The dimensions and characteristics applied in the methodology are presented in table 18.

Table 18. Characteristics of the marine compartment model

Marine compartment
Characteristic
Local Regional
Volume (m ) 3
1 10 9
1 1015

Effective rate of transfer between compartments (a1):

Local to regional, l,r 2 101
Regional to global, r, g 1

Depth (m) 10 1 000

Suspended sediment load, SSL (t/m3) 2 104 1 107
Sedimentation rate, SR (t m2 a1) 1 104 1 105
Length of coastline (km) 10 1 000

OTHER PARAMETERS

Volume of global oceans, Vg (m3) 1 1018

Volumetric exchange between local and regional compartments, 2 1010
Vl,r (m3/a)
Volumetric exchange between regional compartment and global oceans, 1 1015
Vr,g (m3/a)

106. The transfer of radionuclides between water and sediments is dependent on a combination of
factors that relate to the characteristics of the water body and of individual radionuclides. The key
radionuclide-independent parameters are the suspended sediment load and sedimentation rate, which
are determined by the nature of the coastal or marine compartment. Values have been chosen from the
range of those associated with European waters [S7]. Sedimentation is also dependent upon the
radionuclide-dependent partition factor (Kd) for a coastal or marine environment; these values have
been derived from [I3].

107. The purpose of the local marine compartment is to allow doses to a characteristic individual living
in the area and collective doses to the local population to be calculated. The selected dimensions of the
local marine compartment allow tidal, bathymetric and sedimentary conditions to be broadly
homogenous throughout the compartment and have been selected so that the compartment is
sufficiently large to represent a source of shellfish for the local population. The dimensions of the local
66 UNSCEAR 2016 REPORT

marine compartment are typical of those used in the MARINA II project for European waters [S7]. The
dimensions of the regional marine compartment are significantly larger than those generally adopted for
site-specific modelling: approximately equivalent to the size of the north-east Atlantic. This again
reflects the purpose of the methodology; the regional marine compartment is intended to represent the
source of exposure for the regional component of collective dose, which is assumed to extend for a
distance of around 1,500 km from the point of discharge.

108. The depth of the sea in the regional marine compartment is assumed to be 1,000 m, which is the
depth down to which fish are assumed to be caught.

109. The sensitivity of the calculated activity concentrations of radionuclides in water to changes in the
volume of the local compartment and movements between the local and regional compartments was
investigated. The calculated activity concentrations in water in the local compartment (assuming
constant transfer rates between the local and regional compartments) decrease proportionally as the
volume of the local compartment increases. A water transfer rate of 2040 a1 is typical of those found
in European waters based on the volume exchange rates and water volumes given in [S8]. The
Committee adopted a volume of 1 109 m3 with a water transfer rate of 20 a1 for the local marine
compartment as being reasonably representative of coastal conditions into which nuclear sites are
known to discharge.

110. The values of the sedimentation rate and suspended sediment load adopted for the local
compartment are typical of values for local compartments around the United Kingdom (e.g. the south-
western part of the North Sea) [S7]. The values adopted for the regional compartment correspond to the
north-east Atlantic [S7], considered to be typical of large deep oceans.

1. Radionuclide transfers in a marine environment

111. The key element-specific parameters that take account of the transfer of radionuclides from water
to other parts of a marine environment are (a) concentration factors, which relate concentrations of
elements in fish and shellfish to their concentrations in water, and (b) partition coefficients, which
express the distribution of elements between sediment and water. The most significant international
compilation of concentration factors (also referred to as bioaccumulation factors) and partition
coefficients, Kd, for coastal and deep-sea environments is presented in [I3]. Following a critical review
of the data, these values have been adopted and applied in the present methodology, and are given in
table 19.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 67

Table 19. Radionuclide-specific parameters used in the marine model

Dose coefficient for Partition coefficient (L/kg)

Concentration factor (L/kg) (fresh weight)
external exposure (dry weight sediment)
Radio- from surface
nuclide Fish Crustaceans Molluscs deposita, Ocean Open ocean
Bfish Bcrust Bmolluscs Dex,deposit margin Kd,regional
[I3] [I3] [I3] (Sv/s per Bq/m2) [E2] Kd,local [I3] [I3]
3
H 1 10 0
1 10 0
1 100 0 1 100 1 100
14
C 2 104 2 104 2 104 1.3 1020 1 103 2 103
35
S 1 100 1 100 3 100 1.3 1020 5 101 1 100
54
Mn 1 103 5 103 5 104 7.9 1016 2 106 2 108
58
Co 7 102 7 103 2 104 9.3 1016 3 105 5 107
60
Co 7 102 7 103 2 104 2.3 1015 3 105 5 107
65
Zn 1 103 3 105 8 104 5.4 1016 7 104 2 105
90
Sr+ 90Y 3 100 5 100 1 101 1.6 1018 8 100 2 102
106
Ru + 106Rh 2 100 1 102 5 102 3.5 1016 4 104 1 103
129
I 9 100 3 100 1 101 2.0 1017 7 101 2 102
131
I 9 100 3 100 1 101 3.6 1016 7 101 2 102
134
Cs 1 102 5 101 6 101 1.5 1015 4 103 2 103
137
Cs+ 137mBa 1 102 5 101 6 101 5.8 1016 4 103 2 103
210
Pb 2 102 9 104 5 104 2.1 1018 1 105 1 107
212
Pb 2 102 9 104 5 104 1.4 1016 1 105 1 107
214
Pb 2 102 9 104 5 104 2.4 1016 1 105 1 107
210
Po 2 103 2 104 2 104 8.1 1021 2 107 2 107
226
Ra 1 102 1 102 1 102 6.1 1018 2 103 4 103
228
Ra 1 102 1 102 1 102 0 2 103 4 103
228
Th 6 102 1 103 1 103 2.1 1018 3 106 5 106
230
Th 6 102 1 103 1 103 6.4 1019 3 106 5 106
232
Th 6 102 1 103 1 103 4.6 1019 3 106 5 106
234
Th 6 102 1 103 1 103 7.5 1018 3 106 5 106
234
U 1 100 1 101 3 101 5.9 1019 1 103 5 102
238
U 1 100 1 101 3 101 4.2 1019 1 103 5 102
239
Pu 1 102 2 102 3 103 2.8 1019 1 105 1 105
240
Pu 1 102 2 102 3 103 6.0 1019 1 105 1 105
241
Am 1 102 4 102 1 103 2.3 1017 2 106 2 106
a
These dose coefficients are for the parent only; the contribution from short-lived progeny is added separately (see appendix).
68 UNSCEAR 2016 REPORT

2. External and internal exposure

(a) External exposure from radionuclides in beach sediments

112. Annual characteristic individual doses due to external exposure from occupancy of beaches are
estimated on the basis of the modelled activity concentrations of each radionuclide in sediments, the
relevant dose coefficients for external exposure from surface deposits and the amount of time spent on
the beach.

113. The activity concentration of each radionuclide in sediments is estimated using the activity
concentration in water, the relevant partition coefficient Kd, an assumed average thickness of sediment
of 5 cm and an assumed density of sediment of 1.2 t/m3 [I2]. The dose coefficients for external
exposure from surface deposits are taken from [E2] for each radionuclide. A factor of 0.5 is applied to
the dose coefficients to account for the geometry of the radionuclide distribution on a marine shoreline
[E2]. The estimates of effective dose include contributions from beta and gamma irradiation of the skin
from radionuclides in the sediment. Any additional dose from irradiation of the skin due to direct
contact with sediment is not included in the methodology because it is deemed not a major contributor
to the overall characteristic individual doses.

114. An average individual beach occupancy rate of around four days per year is assumed, based on
data for the Nord-Cotentin area of France [R1]. Because this exposure pathway is unlikely to be a
significant contributor to collective doses from nuclear installations [J6], it is not included in the
estimation of collective dose.

(b) Ingestion of marine foods

115. Annual individual doses from ingestion of radionuclides in marine foods are estimated from the
modelled activity concentration of each radionuclide in fish and shellfish (comprising crustaceans and
molluscs) harvested from the relevant marine compartment, the region-specific or world-average annual
consumption of the food (for adults) and the relevant dose coefficient for ingestion of each
radionuclide. Information on the annual consumption of marine fish, crustaceans and molluscs was
obtained from WHO [W2] and the values used are given in table 20.

Table 20. Per caput annual consumption of marine foods by region

Based on data provided by WHO [W2]

Per caput annual consumption of food (kg)

Region
Fish Crustaceans Molluscs
Africa 6.6 0.1 0a
Asia and Pacific 6.9 1.4 2.4
Europe 13 0.9 1.0
Latin America and Caribbean 5.9 0.6 0.4
North America 8.2 2.8 1.4
West Asia 4.5 0.3 0a

World average 7.5 1.1 1.6

a
Consumption rate is less than 2 102 kg/a and treated as 0.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 69

116. Activity concentrations of radionuclides in aquatic foods are estimated from the product of the
modelled activity concentrations in water and concentration factors that relate the activity
concentrations in sea food to those in water (see table 19). In order to apply this approach, it was
necessary to make some assumptions about the origin of the food and therefore the activity
concentrations in water that are appropriate to adopt. Thus, it was necessary to make an assumption
about the proportions of fish and shellfish that are derived from each compartment in the model.

117. One approach applied to estimate doses to individuals is to assume that a certain proportion of
fish and shellfish is derived from the local and regional marine compartments. The proportions assumed
would depend upon the purpose of the assessment. In a similar manner to that used for terrestrial and
freshwater foods, it is assumed in the Committees methodology that the characteristic individual
obtains 25% of their fish consumption from the local compartment, with the remaining fraction
obtained from the regional compartment. All of the crustaceans and molluscs are assumed to be
obtained from the local compartment.

118. The respective local and regional components of collective dose from marine discharges are
estimated using (a) data on the average crustacean-catch and mollusc-catch per unit length of coastline
appropriate for the local compartment, and (b) data on the average fish-catch per unit volume
appropriate for the regional compartment. Catch data for fish, crustaceans and molluscs were derived
from the FAO FishStatJ software [F2], which comprises the FAO databases on capture and aquaculture
for major fishing areas from 1955 to 2012. The area of sea and length of coastline associated with each
of the major fishing areas were determined using a geographical information system (ArcGIS,
version 10) and then used to estimate the relevant values for the UNEP geographical regions employed
in the methodology; world-average values were also derived. The edible fraction of the catch of fish,
crustaceans and molluscs is assumed to be 0.5, 0.35 and 0.15, respectively [S6]. Table 21 shows the
catch data for estimating collective doses for the different regions.

Table 21. Annual catch of fish, crustaceans and molluscs used for estimating collective doses from
discharges to marine environments [F2]

Annual catch

Region Mass of fish Mass of crustaceans Mass of molluscs

per unit area of sea per unit length of coastline per unit length of coastline
(kg/km2) (kg/km) (kg/km)
Africa 2 102 4 103 5 103
Asia and Pacific 2 102 1 104 5 104
Europe 2 102 1 103 5 103

Latin America and Caribbean 9 101 7 103 1 104

North America 6 101 7 103 7 103
West Asia 2 102 5 103 5 103

World average 1 102 8 103 3 104

70 UNSCEAR 2016 REPORT

(c) Individual and collective doses

119. Annual individual doses arising from the ingestion of marine fish and shellfish are estimated
using the modelled activity concentrations of radionuclides in water and fish, the per caput annual
consumptions given in table 20 and the dose coefficients for ingestion of the relevant radionuclides
given in table 1.

120. To estimate collective doses, region-specific information on the catch of fish and shellfish per unit
area of sea or length of coastline are applied from table 21.

IV. GLOBALLY DISPERSED RADIONUCLIDES

121. The Committee has periodically assessed public exposure to long-lived globally dispersed
radionuclides discharged from nuclear power and reprocessing plants since 1982. The radionuclides of
particular interest are 3H, 14C, 85Kr and 129I. The features of the approaches used and the evolution of the
Committees approach are summarized in electronic attachment 2. The estimated collective doses per unit
discharge integrated to various times after discharge are summarized in table 22 at the end of this section.

A. Tritium
122. The Committees previous approach to estimating the doses from the global circulation of tritium
discharged from nuclear installations [U6] was based on a comparison of models developed by Kelly et
al. [K2], NCRP [N1], Bergman et al. [B2] and Killough and Kocher [K4]. The relevant concentration of
tritium for each compartment of the Committees model was determined from the total amount of
tritium in the compartment divided by the volume of water represented in the compartment. The
concentration in humans was then estimated from the modelled concentration of tritium and the
relevant fractional intake from each compartment of the model.

123. The global collective dose from discharges to the near-surface atmosphere within the 3050
latitude band of the northern hemisphere was estimated by the Committee using the model developed
by Killough and Kocher [K4]. This gave a global collective dose per unit discharge of tritium of
2.3 man Sv/PBq [U6]. The global collective dose arising from discharges to the ocean estimated by
NCRP [N1] and Bergman et al. [B2] was around one tenth of that arising from discharges to
atmosphere. A value of 0.2 man Sv/PBq was therefore adopted in the UNSCEAR 2000 Report [U6] for
the collective dose per unit discharge of tritium to the ocean. The Committee still considers these
estimates to be appropriate for a world population of 10 billion.

124. The approach used by the Committee to estimate the collective dose commitment is summarized
as follows [S8]:

S (t ) = c p I c (t ) f p ,c Rc U p P (15)
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 71

where S(t) is the collective dose commitment truncated at time t (man Sv); Ic(t) is the time-integrated
activity concentration in compartment c at time t ((Bq a)/kg); fp,c is the fraction of the individuals
intake of water due to pathway p; Rc is the total consumption rate by an individual of water from
compartment c (kg/a); Up is the dose per unit intake from pathway p (Sv/Bq); and P is the number of
people in the exposed population. This model was implemented in the PC-CREAM 08 computer
system [S8] and has been used to derive values for the global collective doses from unit discharges of
tritium for the Committees methodology. An important factor in the collective dose estimation is the
assumed number of people in the world population. The estimate of the United Nations for this for
1 December 2014 was 6.4 billion (6.4 109) [U2]. The value for the global population of 10 billion
used in the Committees methodology is equivalent to the United Nations median estimate of the
projected population for 2060 [U1] and is deemed reasonable as a rounded value for collective doses
integrated into the future.

B. Carbon-14
125. Carbon-14 is the largest contributor to the collective dose from global dispersion of long-lived
radionuclides discharged from reprocessing of nuclear fuel and is a significant contributor to that resulting
from operation of nuclear reactors. A 23-compartment model [T1] (see figure VI, annex A of the
UNSCEAR 2000 Report [U6]) was applied by the Committee to estimate the activity per unit mass of
carbon in each environmental compartment over time. Once mixing had been achieved, a specific-activity
approach was used to estimate the collective dose commitments from 14C, assuming that the specific
activity of 14C in the carbon ingested by humans was the same as that in the compartments that directly
related to the intake of food (i.e. ground vegetation for terrestrial foods and surface water compartments
for marine foods). The collective dose commitment per unit discharge to atmosphere, truncated at
10,000 years, was estimated to be 109,000 man Sv/PBq. The model can be used to estimate collective
doses for discharges to any compartment; the collective doses from unit discharge to the surface waters of
oceans were found to be about the same as those from unit discharges to atmosphere, but doses from unit
discharges to deep oceans were around 20% lower [U6]. Killough and Rohwer [K3] found that the
estimates from six models differed by a factor of only 1.5, suggesting a remarkable level of agreement
(although it has not been possible to fully validate any of the models against global measurements). This
has been attributed to the long half-life of 14C relative to its rate of movement in the environment, which
makes calculated dose commitments insensitive to the detailed structure of the models [K3].

126. The Committee has therefore not modified its approach to the estimation of collective dose
commitment from 14C from that used in 2000. The approach to derive the collective dose commitment
is implemented in PC-CREAM 08 and is summarized as follows [S8]:

=
S (t ) I (t ) f
c p
c p ,c Rc U p P (16)

where S(t) is the collective dose commitment truncated at time t (man Sv); Ic(t) is the time-integrated
activity concentration in compartment c at time t (Bq a/kg); fp,c is the fraction of the individuals intake
of carbon from pathway p; Rc is the total intake rate of stable carbon (kg/a); Up is the dose per unit
intake from pathway p (Sv/Bq); and P is the size of the exposed population. As before for tritium, the
world population is assumed to be 10 billion.
72 UNSCEAR 2016 REPORT

C. Iodine-129
127. The Committee used the global circulation compartment model for 129I developed by Titley et al.
[T1] for the assessment it presented in the UNSCEAR 2000 Report (see figure VII, annex A [U6]). The
inventories of stable iodine in the compartments of the model and the transfers between the
compartments were estimated from environmental measurements and the requirement that the total
mass of iodine in the environment was balanced. Iodine intakes by humans from each model
compartment were estimated from the average inhalation and food consumption rates combined with
the concentrations of stable iodine in the atmosphere and foods respectively, or using a specific-activity
approach. Five exposure pathways were considered as follows, with the values of individual intakes of
stable iodine used given in brackets: inhalation (0.29 g/d); deposition from the atmosphere onto crops
followed by ingestion by humans or by dairy or beef cattle subsequently ingested by humans (6.6 g/d);
ingestion of surface water (5.3 g/d); ingestion of marine fish and shellfish (11 g/d); and root uptake
from soils and surface waters followed by ingestion of crops and animal products (200 g/d).

128. The PC-CREAM computer code [S8] also implements the model by Titley et al. The approach
used to estimate the collective dose is similar to that used for tritium and 14C, except that individual
intakes were estimated using transfer factors to relate the activity concentrations in food and air to the
activity concentrations in the various compartments, rather than to the intake of stable iodine. The
Committee still considers this approach to be appropriate.

D. Krypton-85
129. The Committees approach to estimating the global dispersion of 85Kr is described in detail in the
UNSCEAR 1988 Report [U5]. It was based on the approach presented in the UNSCEAR 1982 Report
[U4], where a simple two-compartment model was used for the discharge of 85Kr [C2]. The two
compartments represent the tropospheres of the northern and southern hemispheres. A transfer
coefficient of 0.5 a1 between the compartments was used. The time-integrated activity concentration in
air per unit discharge was 1010 (Bq s)/m3 per Bq [U4]. The dose coefficients to convert from activity
concentration in air to absorbed dose rate in air and absorbed dose rate in skin from the emitted beta
radiation were derived from [C2].

130. The value of the collective effective dose equivalent per unit discharge used in the UNSCEAR
1982 Report [U4] was 0.17 man Sv/PBq (based on a world population in 1982 of 4 billion). This was
scaled to 0.2 man Sv/PBq for an assumed population in 1988 of 4.6 billion [U5]. It was noted that this
collective dose commitment would be delivered during the first 50 years after discharge.

131. No changes have been made to the previous approach to modelling the dispersion of 85Kr. Around
50% of the effective dose arises from gamma irradiation of the whole body, and 50% arises from beta
irradiation of the skin [I8, Z1]. The value of dose rate for unit activity per mass of air is 8 109 Sv/a
per Bq/kg (the value is given per unit mass of air because this expression is required for the global
circulation model and was obtained using a density of air of 1.225 kg/m3), based on an assumption of
immersion in a semi-infinite cloud (with no shielding) [S8]. The collective dose commitment is then
estimated as follows:

S (t )= I (t ) F P (17)
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 73

where S (t) is the collective dose commitment truncated at time t (man Sv); I(t) is the time-integrated
activity concentration in air of the appropriate hemisphere at time t ((Bq a)/kg); F is the dose rate per
unit concentration in air (Sv/a per Bq/kg); and P is the number of people in the exposed population. As
for the other globally circulating radionuclides, the Committee has taken for its methodology values of
collective doses derived by PC-CREAM 08, which implements the model used by the Committee and
uses a world population of 10 billion.

Table 22. Collective dose commitments from globally dispersed radionuclides

Collective dose commitment truncated after a given time

from a radionuclide discharge of 1 Bq/s over a year (man Sv)
Radionuclide
To atmosphere To a marine environment
100 years 500 years 10 000 years 100 years 500 years 10 000 years
3
H 1.0 108 1.0 108 1.0 108 1.1 109 1.1 109 1.1 109
14
C 2.8 104 5.6 104 2.7 103 1.1 104 3.5 104 2.5 103
85
Kr 8.0 109 8.0 109 8.0 109
129
I 2.7 103 3.1 103 7.4 103 3.3 106 6.7 106 1.4 104

V. LIMITATIONS OF THE MODELS AND DATA USED

132. The methodology is generic and intended for use with discharges of radionuclides from nuclear
energy generating technologies, and a number of non-nuclear electrical energy production sources,
throughout the world. It builds on previous work that has provided the Committee with robust results
that have been suitable for its purposes. In developing the methodology further, no intentional bias has
been introduced to either under- or overestimate radiation exposures; the aim has been to be generic and
as realistic as possible. It is difficult to quantify the uncertainties, because of the generic nature of the
methodology. Nevertheless, this section outlines the limitations and uncertainties associated with the
methodology. The Committee notes the following generic limitations:

(a) The input to the models is assumed to be a continuous discharge, and annual average
parameter values are used. The models therefore do not apply for short-duration planned or
accidental releases of radionuclides to the environment;

(b) Although some parameter values are used that are specific to geographical region (such as for
food consumption), most of the models are generic and are not intended for detailed site-specific
dose assessments, risk assessments or demonstrating regulatory compliance;

(c) The assumptions and data used for estimating individual doses are intended to apply to a
characteristic individual living local to the discharge point with typical habits and behaviour;

(d) In order to be realistic rather than cautious in estimating characteristic individual doses, it is
assumed that only 25% of food intake (both terrestrial and aquatic) is locally produced. The
resulting ingestion doses are sensitive to this assumption, being directly proportional to the food
intake, and the overall dose will be similarly sensitive where ingestion is a major component.
However, it should be noted that this assumption is not used in the assessment of collective doses;
74 UNSCEAR 2016 REPORT

(e) The assessed collective doses are directly proportional to the assumed population densities for
releases to atmosphere or to the assumed total amount of drinking water or aquatic foods for
releases to water bodies. The methodology gives results for releases to atmosphere for four
different population distributions with values provided by geographical region for three of them.
The effect of using different distributions can be seen in the results presented in the following
section but are generally only relatively small. However, the use of a very low population density,
suitable for remote sites, leads to collective doses around two orders of magnitude lower than when
the default population densities are used. Therefore, the value for the very low population density
should be used with caution.

(f) The models and data are thought to be the most appropriate for use currently and are also
assumed to apply for representing the future. No account is taken of possible future changes, such
as to population distributions in different regions or the effects of climate change. The uncertainties
in the results of the methodology increase with time; this is particularly the case for global
circulation models of long-lived radionuclides when integrated for 10,000 years or more.

133. Specific limitations and uncertainties relating to the different areas of the methodology are noted
in the following sections.

A. Discharges to atmosphere
134. A standard Gaussian-plume model is used to estimate the dispersion of radionuclides following
discharges to atmosphere. As discussed in section III.A, the Gaussian model can be implemented
generically for assessing exposures from different sources of discharges throughout the world and it
produces results that agree reasonably well with measurements [C5]. However, as implemented in the
methodology, the model makes no allowance for local topographical features (such as hills, buildings or
the site being on the coast) which can strongly influence dispersion, particularly close to the discharge
point. The height of the discharge, which is related to both the physical height of the discharge stack
and any buoyancy of the discharge caused, for example, by heat (referred to as plume rise), also has a
significant effect on the subsequent dispersion of the radionuclides. However, this effect reduces with
distance from the discharge point (see electronic attachment 4), and so the Committee agreed to adopt a
single value for stack height of 30 m for its generic methodology.

135. A simplified approach has also been retained to account for the deposition of radionuclides in the
plume on the ground. An effective deposition velocity of 0.002 m/s is used to represent both dry and
wet deposition for all radionuclides (apart from noble gases, tritium and carbon-14, for which, a value
of zero is used because noble gases do not deposit, and because deposition is not explicitly modelled for
tritium and carbon-14, as they exchange quickly between the atmosphere and ground and a specific
activity approach is used instead and). If, for example, a higher deposition velocity were used, it would
clearly lead to higher deposition close to the discharge point but lower deposition at greater distances
because of depletion of radionuclides in the plume. Thus, overall, for the estimation of collective doses,
the effects tend to counteract each other (except for very short-lived radionuclides). However, the
estimation of characteristic individual doses is more sensitive to the assumed deposition velocity.
Nevertheless, because this methodology is intended for comparative purposes, the Committee
considered that the simplified approach was justified.

136. The approach used to estimate the transfer of radionuclides through the terrestrial environment to
food is again a standard one. Although the model used, FARMLAND [B6], was developed in the
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 75

United Kingdom, it has been validated against measured data [S8] and is in reasonable agreement with
models used elsewhere in the world [I6]. Also, to some extent, parameter values that are considered to
be widely applicable are used, as discussed in section III. The FARMLAND model was also used in a
modified form for the Committees 2013 assessment of the levels and effects of the nuclear accident
following the 2011 great east-Japan earthquake and tsunami [U12]agricultural practices and
parameter values specific to Japan were used. A comparison of the results of the use of two versions of
FARMLANDthe one using parameter values that are considered to be widely applicable and the
other parameter values that are specific to Japanare compared in electronic attachment 4. Although
there are some differences, these are not significant for the purposes of the methodology.

137. The methodology makes use of consumption rates of terrestrial foods that differ by geographical
region (see table 7). This is a compromise between the use of a single diet for the world and more
country-specific consumption rates. From table 7, it is clear that there are differences in the total
amount of food that is eaten annually, which will have a direct influence on both collective and
characteristic individual doses but these are generally less than a factor of two (differences of up to a
factor of 5 are seen for individual food types). There are also differences in the composition of the diet,
and it is less clear what impact this will have on estimated doses. For discharges from nuclear sites, one
of the most important radionuclides in terms of both collective and individual doses is carbon-14 [U8].
A specific-activity model is included in the methodology, so that the intake of carbon-14 is related to
the intake of stable carbon in the diet. As carbon is an essential component of the diet and is found in all
foods, it is likely that the intake of carbon-14 is insensitive to changes in the dietary composition. This
was investigated in a sensitivity study (described in electronic attachment 4) that considered the intakes
of carbon for diets typical of Japan and the United Kingdom. This study showed that the total amount
of carbon ingested was similar and therefore the dose estimates would be similar even though the foods
from which the carbon was derived were different.

138. The estimation of doses due to external exposure from both radionuclides in the plume and those
deposited on the ground takes account of the shielding effects of buildings when people are indoors.
The methodology uses a single indoor occupancy factor and single shielding/location factors for
radionuclides in the plume and those on the ground. The indoor occupancy factor is 0.8 and any
variations in this for much of the world are likely to be less than a factor of two and so the effect would
be relatively minor. The shielding factors, 0.1 and 0.2 for radionuclides deposited on the ground and in
the plume, respectively, are typical of those for standard single-storey buildings. As discussed in
section III.A, there are differences between the shielding factors for different building types (shielding
factors might be a factor of 5 or 6 higher than used here for wooden buildings and a factor of up to
10 times lower for multi-storey buildings). There is also evidence for location factors changing as a
function of time. The empirical models used are based on measurements and there can be good
agreement between the results of such models and personal dosimeter measurements where local
factors are taken into account. The factors used here are considered to be appropriate for use in the
Committees assessments, recognizing the variations that occur globally

B. Discharges to freshwater and marine environments

139. The methodology includes consideration of two generic freshwater environments for aquatic
discharges: (a) a small river and (b) a large river; the results for a large river are also considered to
apply to a large lake. In all cases, instantaneous mixing is assumed within a single body of water, which
is obviously a significant simplification. In reality, lakes and rivers are complex environments and
rivers have many tributaries and sections with different flow rates, volumes and behaviours. Similarly,
76 UNSCEAR 2016 REPORT

the methodology uses a generic approach for discharges to a marine environment with a simple
compartmental model consisting of two compartments representing local and regional marine waters.
Again in reality, there are significant differences between marine environments depending on local and
regional currents, and the nature of the coastline and local environment. Although there have been
validation studies showing good agreement between models and predictions, these are for specific sites
and situations and are not necessarily applicable to the simplified generic approach adopted here.
However, the Committee considers that its generic approach is adequate for the purpose of assessing
worldwide exposures from radioactive discharges and for comparative studies, but notes the limitations
of the approach for other purposes.

140. Similar to terrestrial foods, the values for consumption rates of freshwater and marine fish are
based on data by geographical region and are suitable for estimating the characteristic individual dose.
There are differences in the values of consumption rates for freshwater fish and some marine fish.
Table 13 for freshwater fish shows differences of up to a factor of 6 between regions, and table 20
shows differences of a factor of 10 or more for the consumption of crustaceans and even greater
differences for consumption rates of molluscs. These differences would be directly reflected in the
values of the calculated individual doses.

141. The methodology now considers the transfer of radionuclides from freshwater to terrestrial foods
by means of irrigation. There are many different types of irrigation systems and significant variations
across the world in their usage, the source of water (surface or groundwater), and the type of crops that
are irrigated. The estimation of the transfer of radionuclides to terrestrial foods through irrigation is
therefore particularly uncertain. The simplifying assumptions in the methodology however are only
intended to be suitable for generic assessments and the Committee considers that its adopted approach
is robust enough and suitable for this purpose. Nevertheless, the limitations should be recognized; the
importance of these limitations will depend on the relative importance of the irrigation exposure
pathway compared to exposures from intakes of drinking water and freshwater fish.

142. The estimation of collective dose from discharges of radionuclides to rivers, lakes and the sea is
not based on the sum of individual doses as is the case for discharges to atmosphere. Instead, the
collective dose is based on estimates of the total amount of fish caught in the part of the environment of
interest, and the amount of water that is abstracted for drinking water and for irrigation. As such, the
methodology assumes that the fish, drinking water and irrigated foods are consumed but no account is
taken of who consumes them. The Committee considers that this approach is adequate for estimating
collective doses for its purposes, but notes that it is not possible with its methodology to break down the
collective dose into individual doses or to distinguish between local and regional components of
collective dose for any aquatic discharges.

C. Global modelling
143. The approach used to estimate collective doses from the four most important radionuclides that
can be globally dispersed (3H, 14C, 85Kr and 129I) is one that is well accepted and has been used by the
Committee for a number of years. The globally dispersed radionuclides, particularly 14C, are the biggest
contributors to the overall collective doses due to discharges from the nuclear industry when these
doses are integrated over 500 years or more. One of the factors that influences the estimated collective
doses is the size of the global population that is assumed, because the global collective dose is
proportional to the size of the global population. A rounded value of 10 billion people is included in the
methodology; the uncertainty associated with this value increases with time. The global models are
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 77

based on movements of water, carbon and other natural elements through the environment. Such
movements are likely to be affected by changes in the climate, sea level, ocean currents and
atmospheric conditions. The model for carbon is a specific activity approach based on the global carbon
cycle and the amounts of stable carbon in different parts of the environment. Increases in stable carbon
due to the burning of fossil fuels will influence the transfers in the future. These possible future changes
are not taken into account in the methodology, which leads to increasing uncertainties in estimation of
doses in the longer term. It is also not possible to validate this type of model whose scope covers wide
areas and time frames.

D. Comparison of the current and previous methodologies

144. The methodology described in this annex has been developed from that used previously by the
Committee [U6, U7, U10]. The main changes that have been made are discussed in electronic
attachment 1, with further information given in electronic attachment 2.

145. The main changes that have been introduced are:

(a) The results of the methodology are now given in terms of the dose per unit discharge rather
than being normalized to the amount of electricity generated (dose factors normalized to
electricity generated can be calculated subsequently as required);

(b) The methodology has been extended for application to the estimation of exposures from
discharges of radionuclides from non-nuclear electrical energy production;

(c) The inclusion of factors that depend on geographical region, notably for population
distributions, consumption rates and fish-catch data;

(d) The inclusion of a more detailed approach for modelling the transfer of radionuclides from
fresh water to irrigated crops;

(e) The results are provided for a range of integration times;

(f) Some updates have been made to the various models included in the methodology.

As noted in electronic attachment 1, many parts of the methodology remain unchanged, because the
Committee considered, following review, that they were still appropriate for its purposes.

146. The results of the various workbooks implementing the methodology were compared with data
derived from the previous methodology [U6, U7]. The estimates of the annual collective effective doses
from the following pathways were compared:

(a) External exposure due to deposited radionuclides from discharges to atmosphere;

(b) Inhalation of radionuclides discharged to atmosphere;

(c) Ingestion of radionuclides discharged to atmosphere;

(d) All pathways from radionuclides discharged to both freshwater and marine environments.

In general, the results are within about an order of magnitude, despite the changes that have been
introduced to parts of the methodology (see electronic attachment 1, tables 26).
78 UNSCEAR 2016 REPORT

VI. APPLICATION OF THE METHODOLOGY

147. The Committee decided that its calculations should be transparent and that the methodology
should be able to be relatively easily applied and updated in the future. It agreed that the methodology
be implemented in a series of Excel workbooks. The appendix describes the end points considered, the
workbook design, the mathematical equations to implement the methodology in the workbooks, the
treatment of progeny, the quality assurance carried out to ensure that the workbooks were implemented
correctly, and details of how the workbooks could be used for assessing the exposures from different
source terms.

148. The results of the methodology, as given by the workbooks, are for specific discharges to
different environments. The results are in the form of dose calculation factors for individual
radionuclides; for radionuclides that are in secular equilibrium with their short-lived progeny, the
factors represent the sums of the contributions from the parent and progeny (see appendix). The factors
are for characteristic individual doses (in units of sieverts) in the 100th year of a continuous discharge
of a radionuclide at a rate of 1 Bq/s, and collective doses (in units of mansieverts) integrated to 100,
500 and 10,000 years for a continuous discharge of a radionuclide at a rate of 1 Bq/s for 1 year (see
section II.C). For most radionuclides the collective doses to 100 years only are provided because this
was deemed to be sufficient. However collective doses integrated to other times are given for 3H, 14C,
85
Kr and 129I.

149. The resulting characteristic individual and collective dose calculation factors for unit discharge of
radionuclides to atmosphere, to the different freshwater environments (the results for a large river also
apply to a lake) and to a marine environmentderived using the methodology implemented in the
workbooksare presented in tables 2331. The doses can then be scaled by the actual rates of
discharge of each radionuclide from a particular site and then summed over all relevant radionuclides to
give the overall characteristic individual and collective doses. Results are given for sites located in
different regions of the world, with additional results for coastal and inland nuclear sites and for sites
situated in areas of low population density.

150. For future assessments, where discharges are reported to the Committee for broad groups of
radionuclides (e.g. gross alpha, noble gases, radioiodines or particulates), some assumptions would
need to be made about the proportion of the different radionuclides constituting the group (see
section II.D.2).
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 79

Table 23. Estimated characteristic individual doses in the 100th year from a continuous discharge
of a radionuclide to atmosphere

The characteristic individual dose is to a person living 5 km from the discharge point obtaining 25% of their food locally. The
regions are as shown in figure II

Individual effective dose in the 100th year of a continuous discharge at a rate of 1 Bq/s (Sv)
Radionuclide
Latin North
Africa Asia & Pacific Europe West Asia
America America
3
Ha 4.2 1014 4.4 1014 4.8 1014 4.3 1014 4.7 1014 4.1 1014
14
C 3.3 1012 3.7 1012 3.7 1012 3.4 1012 3.6 1012 3.6 1012
35
S 5.1 1012 7.2 1012 1.4 1011 1.2 1011 1.8 1011 7.6 1012
41
Ar 4.2 1014 4.2 1014 4.2 1014 4.2 1014 4.2 1014 4.2 1014
54
Mn 1.5 1012 1.6 1012 1.9 1012 1.7 1012 2.0 1012 1.5 1012
58
Co 6.4 1012 6.8 1012 7.9 1012 7.4 1012 8.4 1012 6.7 1012
60
Co 2.3 1010 2.5 1010 3.0 1010 2.9 1010 3.5 1010 2.5 1010
65
Zn 3.0 1011 3.9 1011 6.7 1011 6.1 1011 8.8 1011 4.0 1011
85
Kr 8.7 1017 8.7 1017 8.7 1017 8.7 1017 8.7 1017 8.7 1017
90
Srb 2.4 1010 2.8 1010 3.8 1010 2.9 1010 3.7 1010 2.5 1010
106
Rub 1.9 1011 2.1 1011 2.9 1011 2.7 1011 3.5 1011 2.1 1011
129
I 4.3 1010 4.9 1010 5.6 1010 4.6 1010 5.3 1010 4.5 1010
131
I 1.3 1011 1.5 1011 1.8 1011 1.4 1011 1.7 1011 1.3 1011
133
Xeb 1.0 1015 1.0 1015 1.0 1015 1.0 1015 1.0 1015 1.0 1015
135
Xe 8.6 1015 8.6 1015 8.6 1015 8.6 1015 8.6 1015 8.6 1015
138
Xeb 3.2 1014 3.2 1014 3.2 1014 3.2 1014 3.2 1014 3.2 1014
134
Cs 1.2 1010 1.4 1010 1.7 1010 1.5 1010 1.8 1010 1.4 1010
137
Csb 2.1 1010 2.2 1010 2.4 1010 2.3 1010 2.6 1010 2.2 1010
210
Pb 1.1 109 1.2 109 1.3 109 1.1 109 1.2 109 1.0 109
210
Po 2.1 109 2.3 109 2.7 109 2.4 109 2.8 109 2.1 109
222
Rnb 2.7 1012 2.7 1012 2.7 1012 2.7 1012 2.7 1012 2.7 1012
226
Ra 2.2 109 2.3 109 2.4 109 2.2 109 2.3 109 2.1 109
230
Th 5.7 109 5.7 109 5.7 109 5.6 109 5.7 109 5.6 109
232
Thb 2.8 108 2.8 108 2.8 108 2.8 108 2.8 108 2.8 108
234
U 1.4 109 1.5 109 1.5 109 1.5 109 1.5 109 1.4 109
238
Ub 1.2 109 1.2 109 1.2 109 1.2 109 1.2 109 1.2 109
239
Pu 2.0 108 2.0 108 2.0 108 2.0 108 2.0 108 2.0 108
240
Pu 2.0 108 2.0 108 2.0 108 2.0 108 2.0 108 2.0 108
241
Am 1.7 108 1.7 108 1.7 108 1.7 108 1.7 108 1.7 108
a
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
b
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
Table 24. Estimated characteristic individual dose in the 100th year of a continuous discharge of a radionuclide to rivers

80 UNSCEAR 2016 REPORT

Characteristic individual doses are to people living 5 km downstream of the discharge point obtaining 25% of their food locally. The regions are as shown in figure II

Individual effective dose in the 100th year of a continuous discharge at a rate of 1 Bq/s (Sv)
Radionuclide
Africa Asia & Pacific Europe Latin America North America West Asia
Small Large a
Small Large a
Small Large a
Small Large a
Small Large a
Small Largea
3
Hb 3.0 1012 3.0 1014 3.0 1012 3.0 1014 3.0 1012 3.0 1014 3.0 1012 3.0 1014 3.0 1012 3.0 1014 3.0 1012 3.0 1014
14
C 1.6 108 1.6 1010 4.5 108 4.5 1010 2.0 108 2.0 1010 9.3 109 9.4 1011 2.6 108 2.6 1010 7.6 109 7.8 1011
35
S 6.2 1011 6.5 1013 1.4 1010 1.3 1012 7.3 1011 8.0 1013 4.5 1011 4.9 1013 9.0 1011 9.6 1013 4.1 1011 4.7 1013
54
Mn 5.2 109 3.5 1012 5.3 109 3.5 1012 5.2 109 3.5 1012 5.2 109 3.5 1012 5.2 109 3.5 1012 5.2 109 3.5 1012
58
Co 4.7 109 4.0 1012 4.7 109 4.0 1012 4.7 109 4.1 1012 4.7 109 4.0 1012 4.7 109 4.1 1012 4.7 109 4.0 1012
60
Co 1.2 108 1.1 1011 1.2 108 1.1 1011 1.2 108 1.1 1011 1.2 108 1.1 1011 1.2 108 1.1 1011 1.2 108 1.1 1011
65
Zn 1.0 109 9.0 1012 2.7 109 2.2 1011 1.3 109 1.1 1011 6.9 1010 6.1 1012 1.6 109 1.4 1011 5.9 1010 5.3 1012
90
Src 1.2 109 1.9 1011 1.2 109 1.5 1011 1.2 109 2.7 1011 1.2 109 1.9 1011 1.2 109 2.8 1011 1.2 109 2.0 1011
106
Ruc 1.7 109 3.5 1012 1.7 109 3.4 1012 1.7 109 3.7 1012 1.7 109 3.5 1012 1.7 109 3.8 1012 1.6 109 3.5 1012
129
I 4.7 109 6.1 1011 5.1 109 5.3 1011 4.7 109 7.4 1011 4.6 109 5.9 1011 4.8 109 7.6 1011 4.6 109 6.3 1011
131
I 4.0 1010 1.9 1012 4.0 1010 1.9 1012 4.0 1010 2.0 1012 4.0 1010 1.9 1012 4.0 1010 2.0 1012 4.0 1010 1.9 1012
134
Cs 8.6 109 1.8 1011 1.2 108 2.1 1011 9.2 109 2.0 1011 7.8 109 1.7 1011 1.0 108 2.1 1011 7.6 109 1.7 1011
137
Csc 4.1 109 1.1 1011 6.7 109 1.2 1011 4.5 109 1.2 1011 3.5 109 9.9 1012 5.0 109 1.3 1011 3.4 109 1.0 1011
210
Pb 2.0 108 2.2 1010 2.1 108 2.0 1010 2.0 108 2.5 1010 1.9 108 2.1 1010 2.0 108 2.6 1010 1.9 108 2.2 1010
210
Po 3.4 108 3.5 1010 3.7 108 3.4 1010 3.4 108 3.9 1010 3.3 108 3.5 1010 3.5 108 4.0 1010 3.3 108 3.5 1010
226
Ra 7.6 109 1.1 1010 7.8 109 9.0 1011 7.7 109 1.4 1010 7.6 109 1.0 1010 7.7 109 1.4 1010 7.6 109 1.1 1010
230
Th 3.2 109 3.7 1011 3.2 109 3.4 1011 3.2 109 4.4 1011 3.2 109 3.7 1011 3.2 109 4.6 1011 3.2 109 3.7 1011
232
Thc 3.3 108 2.8 1010 3.3 108 2.5 1010 3.3 108 3.2 1010 3.3 108 2.7 1010 3.3 108 3.3 1010 3.3 108 2.8 1010
234
U 7.4 1010 9.6 1012 7.4 1010 8.4 1012 7.4 1010 1.2 1011 7.4 1010 9.5 1012 7.4 1010 1.3 1011 7.4 1010 9.7 1012
238
Uc 7.3 1010 9.4 1012 7.4 1010 8.3 1012 7.3 1010 1.2 1011 7.3 1010 9.3 1012 7.3 1010 1.2 1011 7.3 1010 9.5 1012
239
Pu 3.5 109 4.1 1011 3.6 109 3.7 1011 3.5 109 4.9 1011 3.4 109 4.1 1011 3.5 109 5.1 1011 3.4 109 4.1 1011
240
Pu 3.5 109 4.1 1011 3.6 109 3.7 1011 3.5 109 4.9 1011 3.4 109 4.1 1011 3.5 109 5.1 1011 3.4 109 4.1 1011
241
Am 3.8 109 3.4 1011 5.6 109 3.1 1011 4.1 109 4.1 1011 3.4 109 3.3 1011 4.5 109 4.3 1011 3.3 109 3.3 1011
a
The dose estimates for large rivers are also assumed to apply to lakes.
b
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
c
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
Table 25. Estimated characteristic individual dose in the 100th year of a continuous discharge of a radionuclide to a marine environment
The characteristic individual dose is for people ingesting crustaceans and molluscs from the local marine compartment plus 25% of their marine fish consumption from the local compartment and 75% from the
regional compartment. The regions are as shown in figure II

Individual effective dose in the 100th year of a continuous discharge at a rate of 1 Bq/s (Sv)
Radionuclide
Africa Asia and Pacific Europe Latin America North America West Asia
3
H 5.0 1017 1.6 1016 1.5 1016 7.0 1017 1.7 1016 4.0 1017
14
C 3.2 1011 1.0 1010 9.4 1011 4.5 1011 1.1 1010 2.6 1011
35
S 1.9 1015 1.1 1014 7.6 1015 3.5 1015 9.5 1015 1.5 1015
54
Mn 2.6 109 5.3 109 5.2 109 5.2 109 5.3 109 5.2 109

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 81

58
Co 4.0 1010 8.6 1010 8.3 1010 8.1 1010 8.5 1010 8.0 1010
60
Co 1.2 109 2.6 109 2.5 109 2.4 109 2.6 109 2.3 109
65
Zn 2.8 1010 3.7 109 2.1 109 1.4 109 5.6 109 5.9 1010
90
Sra 2.5 1013 1.6 1012 1.1 1012 5.1 1013 1.5 1012 2.2 1013
106
Rua 2.3 1011 6.0 1011 5.2 1011 4.8 1011 5.6 1011 4.6 1011
129
I 2.7 1012 7.6 1012 7.3 1012 3.3 1012 7.0 1012 1.9 1012
131
I 3.6 1014 8.8 1014 8.5 1014 5.7 1014 8.3 1014 4.7 1014
134
Cs 1.5 1011 3.1 1011 3.2 1011 2.6 1011 3.2 1011 2.4 1011
137
Csa 7.2 1012 1.5 1011 1.6 1011 1.2 1011 1.6 1011 1.0 1011
210
Pba 2.3 108 4.2 107 2.2 107 1.2 107 5.1 107 4.1 108
210
Po 1.0 108 1.4 107 7.5 108 4.0 108 1.5 107 1.3 108
226
Raa 2.4 108 4.2 107 2.2 107 1.3 107 5.1 107 4.1 108
230
Th 3.7 1010 1.6 109 1.3 109 6.3 1010 1.8 109 3.2 1010
232
Tha 6.3 109 1.6 108 1.5 108 1.3 108 1.7 108 1.2 108
234
U 2.5 1013 6.8 1012 3.2 1012 1.5 1012 5.5 1012 3.3 1013
238
Ua 6.4 1012 3.3 1011 2.4 1011 1.2 1011 3.4 1011 5.8 1012
239
Pu 9.1 1011 3.0 109 1.4 109 5.7 1010 1.9 109 8.5 1011
240
Pu 9.1 1011 3.0 109 1.4 109 5.7 1010 1.9 109 8.5 1011
241
Am 1.5 1010 1.2 109 6.8 1010 4.0 1010 1.0 109 2.3 1010
a
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
 82 UNSCEAR 2016 REPORT
Table 26. Local and regional components of collective dose for discharges to atmosphere from sites in different regions of the world

The collective doses are out to 1,500 km based on the population distributions given in table 3 for different regions of the world as shown in figure II. They can be used for discharges from any type of source for
comparative purposes and are the only values appropriate for non-nuclear sites

Radio- Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (man Sv)
nuclide Africa Asia and Pacific Europe Latin America North America West Asia World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional
3 b 8 7 8 7 8 7 8 7 9 8 8 7 8
H 1.4 10 1.2 10 5.6 10 4.7 10 3.0 10 2.5 10 2.6 10 2.2 10 7.3 10 6.0 10 1.8 10 1.5 10 3.2 10 2.7 107
14
C 1.1 106 5.3 106 4.3 106 2.2 105 2.0 106 1.0 105 1.9 106 9.7 106 4.9 107 2.5 106 1.5 106 7.8 106 2.4 106 1.2 105
35
S 1.8 106 8.2 106 9.1 106 4.2 105 8.2 106 3.8 105 7.5 106 3.5 105 2.7 106 1.3 105 3.5 106 1.6 105 6.9 106 3.2 105
41
Ar 6.1 1010 3.2 1015 2.2 109 1.1 1014 9.8 1010 5.2 1015 1.0 109 5.5 1015 2.5 1010 1.3 1015 7.9 1010 4.2 1015 1.2 109 6.5 1015
54
Mn 3.8 107 1.8 106 1.6 106 7.4 106 9.0 107 4.3 106 7.8 107 3.7 106 2.3 107 1.1 106 5.0 107 2.4 106 9.2 107 4.4 106
58
Co 8.9 107 4.1 106 3.6 106 1.7 105 2.4 106 1.1 105 2.2 106 1.0 105 6.7 107 3.1 106 1.3 106 6.0 106 2.3 106 1.1 105
60
Co 3.0 105 1.4 104 1.3 104 6.3 104 9.6 105 4.6 104 9.4 105 4.5 104 3.1 105 1.5 104 5.0 105 2.4 104 9.1 105 4.3 104
65
Zn 8.5 106 4.0 105 4.3 105 2.0 104 3.7 105 1.7 104 3.5 105 1.6 104 1.2 105 5.9 105 1.6 105 7.6 105 3.2 105 1.5 104
85
Kr 1.4 1011 1.1 1010 4.8 1011 4.0 1010 2.2 1011 1.8 1010 2.3 1011 1.9 1010 5.5 1012 4.6 1011 1.8 1011 1.5 1010 2.7 1011 2.3 1010
90
Src 8.7 105 4.2 104 3.6 104 1.7 103 2.3 104 1.1 103 1.8 104 8.6 104 5.5 105 2.6 104 1.2 104 5.5 104 2.2 104 1.0 103
106
Ruc 3.9 106 1.9 105 1.8 105 8.4 105 1.3 105 5.9 105 1.2 105 5.7 105 4.1 106 1.9 105 6.2 106 3.0 105 1.2 105 5.6 105
129
I 1.6 104 7.6 104 6.4 104 3.1 103 3.3 104 1.6 103 2.9 104 1.4 103 7.9 105 3.8 104 2.1 104 1.0 103 3.6 104 1.7 103
131
I 3.9 106 1.4 105 1.6 105 5.6 105 9.3 106 3.2 105 7.6 106 2.7 105 2.2 106 7.9 106 5.1 106 1.8 105 9.3 106 3.2 105
133
Xec 1.6 1010 8.0 1010 5.6 1010 2.9 109 2.6 1010 1.3 109 2.7 1010 1.4 109 6.5 1011 3.3 1010 2.1 1010 1.0 109 3.2 1010 1.6 109
135
Xe 8.3 1010 1.7 1010 3.0 109 5.9 1010 1.4 109 2.7 1010 1.4 109 2.9 1010 3.4 1010 6.8 1011 1.1 109 2.2 1010 1.7 109 3.4 1010
138
Xec 1.9 1012 1.7 1031 6.7 1012 6.2 1031 3.0 1012 2.8 1031 3.2 1012 3.0 1031 7.7 1013 7.1 1032 2.4 1012 2.3 1031 3.8 1012 3.5 1031
134
Cs 2.9 105 1.4 104 1.2 104 5.9 104 7.3 105 3.5 104 6.9 105 3.3 104 2.1 105 9.9 105 4.4 105 2.1 104 7.5 105 3.6 104
137
Csc 3.5 105 1.7 104 1.4 104 6.8 104 7.9 105 3.8 104 7.6 105 3.6 104 2.2 105 1.0 104 5.0 105 2.4 104 8.5 105 4.1 104
210
Pb 3.0 104 1.4 103 1.2 103 5.5 103 6.0 104 2.9 103 5.0 104 2.4 103 1.4 104 6.7 104 3.5 104 1.7 103 6.4 104 3.1 103
210
Po 4.4 104 2.1 103 1.7 103 8.2 103 1.1 103 5.0 103 9.0 104 4.2 103 2.8 104 1.3 103 5.5 104 2.6 103 1.0 103 4.9 103
222
Rnc 4.0 107 1.7 106 1.4 106 6.0 106 6.4 107 2.8 106 6.8 107 2.9 106 1.6 107 6.9 107 5.1 107 2.2 106 8.0 107 3.4 106
226
Ra 4.3 104 2.1 103 1.7 103 8.0 103 8.4 104 4.0 103 7.4 104 3.5 103 2.0 104 9.5 104 5.4 104 2.6 103 9.3 104 4.4 103
 Radio- Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (man Sv)
nuclide Africa Asia and Pacific Europe Latin America North America West Asia World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional
230
Th 5.7 104 2.7 103 2.0 103 9.8 103 9.5 104 4.5 103 9.8 104 4.7 103 2.4 104 1.1 103 7.3 104 3.5 103 1.2 103 5.5 103
232
Thc 2.9 103 1.4 102 1.0 102 5.0 102 4.9 103 2.3 102 5.0 103 2.4 102 1.2 103 5.8 103 3.7 103 1.8 102 5.9 103 2.8 102
234
U 1.5 104 7.3 104 5.6 104 2.7 103 2.7 104 1.3 103 2.7 104 1.3 103 6.9 105 3.3 104 2.0 104 9.5 104 3.2 104 1.5 103
238
Uc 1.3 104 6.2 104 4.8 104 2.3 103 2.3 104 1.1 103 2.3 104 1.1 103 5.9 105 2.8 104 1.7 104 8.1 104 2.7 104 1.3 103
239
Pu 1.9 103 9.2 103 6.8 103 3.3 102 3.1 103 1.5 102 3.3 103 1.6 102 7.9 104 3.8 103 2.5 103 1.2 102 3.9 103 1.9 102
240
Pu 1.9 103 9.2 103 6.8 103 3.3 102 3.1 103 1.5 102 3.3 103 1.6 102 7.9 104 3.8 103 2.5 103 1.2 102 3.9 103 1.9 102

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 83

241
Am 1.6 103 7.7 103 5.7 103 2.7 102 2.6 103 1.3 102 2.8 103 1.3 102 6.6 104 3.2 103 2.1 103 1.0 102 3.3 103 1.6 102
a
The world-average value is based on a world-average population density given in table 3 together with a world-average consumption rate for terrestrial foods (table 7).
b
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
c
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details of the progeny included).

Table 27. Local and regional components of collective dose for discharges to atmosphere for inland nuclear sites

These collective doses are based on the population distributions out to 1,500 km for inland nuclear sites given in table 4 for the regions shown in figure II. They apply to inland nuclear sites only. Because there were
no inland operating nuclear power plants in Africa or West Asia at the time of analysis, results are not given for these regions

Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
Radionuclide
Asia and Pacific Europe Latin America North America World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional
3
H b
1.2 10 7
3.3 10 7
3.4 10 8
1.6 107
2.8 10 8
2.4 108
2.2 10 8
6.7 10 8
2.9 10 8
1.1 107
14
C 9.2 106 1.6 105 2.3 106 6.7 106 2.1 106 1.1 106 1.5 106 2.8 106 2.2 106 5.2 106
35
S 2.0 105 3.1 105 9.4 106 2.5 105 8.1 106 4.0 106 8.1 106 1.5 105 6.3 106 1.4 105
41
Ar 4.6 109 1.1 1014 1.1 109 4.8 1015 1.1 109 8.3 1016 7.4 1010 2.1 1015 1.1 109 3.9 1015
54
Mn 3.3 106 5.4 106 1.0 106 2.9 106 8.4 107 4.3 107 7.0 107 1.3 106 8.4 107 1.9 106
58
Co 7.8 106 1.2 105 2.7 106 7.3 106 2.3 106 1.2 106 2.0 106 3.6 106 2.1 106 4.7 106
60
Co 2.8 104 4.6 104 1.1 104 3.1 104 1.0 104 5.2 105 9.2 105 1.7 104 8.4 105 1.9 104
65
Zn 9.2 105 1.5 104 4.2 105 1.2 104 3.7 105 1.9 105 3.7 105 6.8 105 2.9 105 6.5 105
 84 UNSCEAR 2016 REPORT
Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
Radionuclide
Asia and Pacific Europe Latin America North America World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional
85
Kr 1.0 1010 2.8 1010 2.5 1011 1.2 1010 2.5 1011 2.2 1011 1.7 1011 5.1 1011 2.5 1011 9.4 1011
90
Src 7.7 104 1.3 103 2.6 104 7.2 104 1.9 104 9.9 105 1.7 104 3.0 104 2.0 104 4.5 104
106
Ruc 3.8 105 6.1 105 1.4 105 4.0 105 1.3 105 6.6 106 1.2 105 2.2 105 1.1 105 2.4 105
129
I 1.4 103 2.2 103 3.8 104 1.1 103 3.1 104 1.6 104 2.4 104 4.3 104 3.3 104 7.5 104
131
I 3.4 105 4.3 105 1.1 105 2.3 105 8.2 106 3.2 106 6.7 106 9.6 106 8.5 106 1.5 105
133
Xe 1.2 109 2.2 109 3.0 1010 9.1 1010 2.9 1010 1.7 1010 1.9 1010 4.0 1010 2.9 1010 7.4 1010
135
Xe 6.3 109 5.5 1010 1.6 109 2.5 1010 1.5 109 4.3 1011 1.0 109 1.1 1010 1.5 109 2.0 1010
138
Xec 1.4 1011 5.8 1031 3.5 1012 2.6 1031 3.5 1012 4.5 1032 2.3 1012 1.1 1031 3.5 1012 2.1 1031
134
Cs 2.6 104 4.3 104 8.4 105 2.3 104 7.4 105 3.8 105 6.2 105 1.1 104 6.9 105 1.6 104
137
Csc 3.0 104 4.9 104 9.1 105 2.5 104 8.2 105 4.2 105 6.5 105 1.2 104 7.8 105 1.8 104
210
Pb 2.5 103 4.0 103 6.9 104 1.9 103 5.4 104 2.8 104 4.2 104 7.7 104 5.9 104 1.3 103
210
Po 3.7 103 6.0 103 1.2 103 3.3 103 9.7 104 4.9 104 8.2 104 1.5 103 9.6 104 2.1 103
222
Rnc 3.0 106 4.7 106 7.4 107 2.0 106 7.3 107 3.6 107 4.8 107 8.7 107 7.3 107 1.6 106
226
Ra 3.6 103 5.8 103 9.7 104 2.7 103 7.9 104 4.1 104 5.9 104 1.1 103 8.5 104 1.9 103
230
Th 4.4 103 7.1 103 1.1 103 3.0 103 1.0 103 5.4 104 7.1 104 1.3 103 1.1 103 2.4 103
232
Thc 2.2 102 3.6 102 5.6 103 1.6 102 5.3 103 2.7 103 3.6 103 6.7 103 5.4 103 1.2 102
234
U 1.2 103 1.9 103 3.1 104 8.6 104 2.9 104 1.5 104 2.1 104 3.8 104 3.0 104 6.7 104
238
Uc 1.0 103 1.7 103 2.7 104 7.4 104 2.5 104 1.3 104 1.8 104 3.3 104 2.5 104 5.7 104
239
Pu 1.5 102 2.4 102 3.6 103 1.0 102 3.5 103 1.8 103 2.4 103 4.3 103 3.6 103 8.0 103
240
Pu 1.5 102 2.4 102 3.6 103 1.0 102 3.5 103 1.8 103 2.4 103 4.3 103 3.6 103 8.0 103
241
Am 1.2 102 2.0 102 3.0 103 8.4 103 3.0 103 1.5 103 2.0 103 3.6 103 3.0 103 6.8 103
a
The world-average value is based on a world average population density given in table 4 together with a world-average consumption rate for terrestrial foods (table 7).
b
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
c
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
Table 28. Local and regional components of collective dose for discharges to atmosphere for coastal nuclear sites

These collective doses are based on the population distributions out to 1 500 km for coastal nuclear sites given in table 4 for the regions shown in figure II. They apply to coastal nuclear sites only. Because there
were no coastal operating nuclear power plants in West Asia at the time of analysis, results are not given for this region

Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
Radionuclide
Africa Asia and Pacific Europe Latin America North America World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional
3
H b
2.3 10 8
6.9 10 9
5.3 10 8
1.6 10 7
2.5 10 8
1.2 107
2.2 10 8
6.1 10 8
3.4 10 8
6.3 10 8
3.6 10 8
1.1 107
14
C 1.7 106 2.8 107 4.1 106 7.7 106 1.6 106 4.8 106 1.6 106 3.0 106 2.3 106 2.7 106 2.7 106 5.0 106

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 85

35
S 2.8 106 4.3 107 8.6 106 1.5 105 6.7 106 1.8 105 6.3 106 1.1 105 1.3 105 1.4 105 7.7 106 1.4 105
41
Ar 9.6 1010 6.1 1017 2.0 109 4.8 1015 8.1 1010 3.5 1015 8.7 1010 3.4 1015 1.2 109 2.2 1015 1.4 109 3.6 1015
54
Mn 6.1 107 9.7 108 1.5 106 2.6 106 7.4 107 2.1 106 6.6 107 1.2 106 1.1 106 1.2 106 1.0 106 1.8 106
58
Co 1.4 106 2.2 107 3.4 106 5.9 106 1.9 106 5.3 106 1.8 106 3.2 106 3.1 106 3.4 106 2.6 106 4.5 106
60
Co 4.8 105 7.7 106 1.2 104 2.2 104 7.9 105 2.2 104 7.9 105 1.4 104 1.4 104 1.6 104 1.0 104 1.8 104
65
Zn 1.3 105 2.1 106 4.0 105 7.2 105 3.0 105 8.4 105 2.9 105 5.1 105 5.8 105 6.5 105 3.6 105 6.4 105
85
Kr 2.1 1011 6.5 1012 4.5 1011 1.4 1010 1.8 1011 8.5 1011 2.0 1011 5.5 1011 2.6 1011 4.8 1011 3.1 1011 9.3 1011
90
Src 1.4 104 2.2 105 3.4 104 6.1 104 1.9 104 5.2 104 1.5 104 2.7 104 2.6 104 2.9 104 2.4 104 4.4 104
106
Ruc 6.2 106 9.9 107 1.7 105 3.0 105 1.0 105 2.9 105 1.0 105 1.8 105 1.9 105 2.1 105 1.3 105 2.4 105
129
I 2.5 104 4.0 105 6.1 104 1.1 103 2.7 104 7.7 104 2.4 104 4.3 104 3.7 104 4.2 104 4.0 104 7.3 104
131
I 6.2 106 6.4 107 1.5 105 2.0 105 7.6 106 1.7 105 6.4 106 9.4 106 1.1 105 9.4 106 1.0 105 1.4 105
133
Xec 2.5 1010 3.8 1011 5.3 1010 1.0 109 2.1 1010 6.6 1010 2.3 1010 4.7 1010 3.0 1010 3.8 1010 3.6 1010 7.1 1010
135
Xe 1.3 109 3.2 1012 2.8 109 2.5 1010 1.1 109 1.8 1010 1.2 109 1.7 1010 1.6 109 1.1 1010 1.9 109 1.8 1010
138
Xec 3.0 1012 3.3 1033 6.3 1012 2.6 1031 2.5 1012 1.9 1031 2.7 1012 1.8 1031 3.6 1012 1.2 1031 4.2 1012 1.9 1031
134
Cs 4.6 105 7.4 106 1.2 104 2.1 104 6.0 105 1.7 104 5.8 105 1.0 104 9.7 105 1.1 104 8.4 105 1.5 104
137
Csc 5.5 105 8.9 106 1.3 104 2.4 104 6.5 105 1.8 104 6.4 105 1.1 104 1.0 104 1.2 104 9.5 105 1.7 104
210
Pb 4.7 104 7.6 105 1.1 103 2.0 103 4.9 104 1.4 103 4.2 104 7.5 104 6.6 104 7.4 104 7.2 104 1.3 103
210
Po 6.9 104 1.1 104 1.6 103 2.9 103 8.7 104 2.4 103 7.6 104 1.3 103 1.3 103 1.4 103 1.2 103 2.1 103
222
Rnc 6.2 107 7.5 108 1.3 106 2.2 106 5.3 107 1.4 106 5.7 107 1.1 106 7.6 107 8.5 107 8.9 107 1.5 106
226
Ra 6.8 104 1.1 104 1.6 103 2.8 103 6.9 104 1.9 103 6.2 104 1.1 103 9.3 104 1.0 103 1.0 103 1.9 103
 86 UNSCEAR 2016 REPORT
Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
Radionuclide
Africa Asia and Pacific Europe Latin America North America World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional
230
Th 9.0 104 1.5 104 1.9 103 3.5 103 7.8 104 2.2 103 8.2 104 1.5 103 1.1 103 1.2 103 1.3 103 2.3 103
232
Thc 4.6 103 7.3 104 9.8 103 1.8 102 4.0 103 1.1 102 4.2 103 7.4 103 5.7 103 6.4 103 6.6 103 1.2 102
234
U 2.4 104 3.9 105 5.3 104 9.5 104 2.2 104 6.3 104 2.3 104 4.1 104 3.2 104 3.7 104 3.6 104 6.5 104
238
Uc 2.1 104 3.3 105 4.5 104 8.1 104 1.9 104 5.3 104 2.0 104 3.5 104 2.8 104 3.1 104 3.1 104 5.5 104
239
Pu 3.0 103 4.9 104 6.4 103 1.2 102 2.6 103 7.3 103 2.8 103 4.9 103 3.7 103 4.2 103 4.3 103 7.9 103
240
Pu 3.0 103 4.9 104 6.4 103 1.2 102 2.6 103 7.3 103 2.8 103 4.9 103 3.7 103 4.2 103 4.3 103 7.9 103
241
Am 2.5 103 4.1 104 5.4 103 9.7 103 2.2 103 6.1 103 2.3 103 4.1 103 3.1 103 3.5 103 3.6 103 6.6 103
a
The world-average value is based on a world average population density given in table 3 together with a world-average consumption rate for terrestrial foods (table 7).
b
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
c
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 87

Table 29. Local and regional components of collective dose for discharges to atmosphere for low
population density sites

These collective dose estimates are for locations with a low population density out to 1,500 km and are based on a population
density of 5 km2. They apply to the world as a whole and are intended for use for uranium mines and mill tailings sites

Collective dose integrated to 100 years from a years continuous discharge

Radionuclide
at a rate of 1 Bq/s (manSv)

Local component Regional component

3
Ha 1.0 109 8.4 109
14
C 7.5 108 3.8 107
35
S 2.2 107 1.0 106
41
Ar 3.8 1011 2.0 1016
54
Mn 2.9 108 1.4 107
58
Co 7.3 108 3.4 107
60
Co 2.9 106 1.4 105
65
Zn 1.0 106 4.7 106
85
Kr 8.6 1013 7.2 1012
90
Srb 6.8 106 3.2 105
106
Rub 3.7 107 1.8 106
129
I 1.1 105 5.4 105
131
I 2.9 107 1.0 106
133
Xeb 1.0 1011 5.1 1011
135
Xe 5.3 1011 1.1 1011
138
Xeb 1.2 1013 1.1 1032
134
Cs 2.4 106 1.1 105
137
Csb 2.7 106 1.3 105
210
Pb 2.0 105 9.6 105
210
Po 3.3 105 1.5 104
222
Rnb 2.5 108 1.1 107
226
Ra 2.9 105 1.4 104
230
Th 3.6 105 1.7 104
232
Thb 1.9 104 8.9 104
234
U 1.0 105 4.8 105
238
Ub 8.6 106 4.1 105
239
Pu 1.2 104 5.8 104
240
Pu 1.2 104 5.8 104
241
Am 1.0 104 4.9 104
a
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
b
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
Table 30. Collective dose for discharges to small and large rivers

88 UNSCEAR 2016 REPORT

Values are for small and large rivers situated in the regions shown in figure II. The results for large rivers are also assumed to apply for lakes

Radio- Collective doses integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
nuclide Africa Asia and Pacific Europe Latin America North America West Asia World averagea
Small Large Small Large Small Large Small Large Small Large Small Large Small Large
3
H b
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 10 7
1.9 107
14
C 2.3 103 1.2 103 2.3 103 1.2 103 2.3 103 1.2 103 2.3 103 1.2 103 2.3 103 1.2 103 2.3 103 1.4 103 2.3 103 1.2 103
35
S 7.4 106 4.5 106 7.4 106 4.8 106 7.4 106 4.6 106 7.4 106 4.3 106 7.4 106 5.2 106 7.4 106 1.7 105 7.4 106 4.8 106
54
Mn 1.5 106 9.2 107 1.5 106 9.8 107 1.5 106 9.4 107 1.5 106 8.8 107 1.5 106 1.1 106 1.5 106 3.6 106 1.5 106 9.7 107
58
Co 1.6 106 1.3 106 1.6 106 1.4 106 1.6 106 1.3 106 1.6 106 1.3 106 1.6 106 1.4 106 1.6 106 3.1 106 1.6 106 1.4 106
60
Co 7.2 106 6.4 106 7.2 106 6.8 106 7.2 106 6.5 106 7.2 106 6.1 106 7.2 106 7.2 106 7.2 106 2.2 105 7.2 106 6.7 106
65
Zn 1.4 104 6.1 105 1.4 104 6.1 105 1.4 104 6.1 105 1.4 104 6.0 105 1.4 104 6.2 105 1.4 104 9.2 105 1.4 104 6.1 105
90
Src 7.2 105 1.0 104 7.2 105 1.2 104 7.2 105 1.1 104 7.2 105 8.8 105 7.2 105 1.4 104 7.2 105 9.8 104 7.2 105 1.2 104
106
Ruc 1.4 105 1.2 105 1.4 105 1.2 105 1.4 105 1.2 105 1.4 105 1.2 105 1.4 105 1.2 105 1.4 105 1.5 105 1.4 105 1.2 105
129
I 3.1 104 3.8 104 3.1 104 4.5 104 3.1 104 4.0 104 3.1 104 3.4 104 3.1 104 5.3 104 3.1 104 3.3 103 3.1 104 4.4 104
131
I 5.7 106 5.3 106 5.7 106 5.4 106 5.7 106 5.4 106 5.7 106 5.3 106 5.7 106 5.5 106 5.7 106 9.1 106 5.7 106 5.4 106
134
Cs 3.5 104 7.8 105 3.5 104 8.7 105 3.5 104 8.1 105 3.5 104 7.1 105 3.5 104 9.8 105 3.5 104 4.9 104 3.5 104 8.5 105
137
Csc 2.4 104 5.5 105 2.4 104 6.3 105 2.4 104 5.7 105 2.4 104 5.0 105 2.4 104 7.2 105 2.4 104 3.9 104 2.4 104 6.1 105
210
Pb 1.3 103 1.2 103 1.3 103 1.3 103 1.3 103 1.2 103 1.3 103 1.2 103 1.3 103 1.3 103 1.3 103 2.5 103 1.3 103 1.3 103
210
Po 2.3 103 2.1 103 2.3 103 2.1 103 2.3 103 2.1 103 2.3 103 2.1 103 2.3 103 2.1 103 2.3 103 2.3 103 2.3 103 2.1 103
226
Ra 4.9 104 5.9 104 4.9 104 6.6 104 4.9 104 6.1 104 4.9 104 5.4 104 4.9 104 7.5 104 4.9 104 3.8 103 4.9 104 6.5 104
230
Th 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.0 104 2.7 104 2.0 104 2.0 104
232
Thc 1.5 103 1.6 103 1.5 103 1.6 103 1.5 103 1.6 103 1.5 103 1.5 103 1.5 103 1.7 103 1.5 103 4.3 103 1.5 103 1.6 103
234
U 4.7 105 5.1 105 4.7 105 5.3 105 4.7 105 5.1 105 4.7 105 4.9 105 4.7 105 5.7 105 4.7 105 1.7 104 4.7 105 5.3 105
238
Uc 4.6 105 5.0 105 4.6 105 5.2 105 4.6 105 5.1 105 4.6 105 4.8 105 4.6 105 5.5 105 4.6 105 1.6 104 4.6 105 5.2 105
239
Pu 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 3.3 104 2.3 104 2.2 104
240
Pu 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 2.2 104 2.3 104 3.3 104 2.3 104 2.2 104
241
Am 3.1 104 1.8 104 3.1 104 1.8 104 3.1 104 1.8 104 3.1 104 1.8 104 3.1 104 1.9 104 3.1 104 3.1 104 3.1 104 1.8 104
a
The world-average collective dose is calculated using world-average values for irrigation (tables 11, 12 and 17) (note that the catch data for unit river length is the same for all regions and the world-average).
b
For 3H the calculated doses include a contribution from HTO and OBT as discussed in section III.A.3.
c
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
Table 31. Local and regional components of collective dose for discharges to a marine environment
These results are for discharges to a marine environment for the regions shown in figure II

Radio- Collective dose integrated to 100 years from a years continuous discharge at a rate of 1 Bq/s (manSv)
nuclide Africa Asia and Pacific Europe Latin America North America West Asia World averagea
Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional Local Regional
3 13 14 12 14 13 14 12 14 12 14 12 14 12
H 8.5 10 4.8 10 3.4 10 5.8 10 6.0 10 4.7 10 1.4 10 2.8 10 1.1 10 1.8 10 1.0 10 5.7 10 2.2 10 3.1 1014
14
C 5.5 107 3.3 108 2.2 106 4.0 108 3.9 107 3.2 108 9.2 107 1.9 108 6.9 107 1.2 108 6.5 107 3.9 108 1.4 106 2.1 108
35
S 4.9 1011 5.0 1013 2.7 1010 6.6 1013 3.9 1011 4.9 1013 9.7 1011 3.0 1013 6.3 1011 2.0 1013 5.4 1011 5.9 1013 1.8 1010 3.6 1013
54
Mn 5.1 107 1.5 109 3.9 106 5.2 109 4.5 107 1.4 109 1.2 106 1.8 109 7.2 107 1.1 109 5.1 107 1.7 109 2.6 106 3.3 109
58
Co 2.5 107 3.7 1010 1.6 106 9.9 1010 1.9 107 3.4 1010 5.8 107 3.9 1010 3.8 107 2.5 1010 2.8 107 4.2 1010 1.1 106 6.1 1010

ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 89

60
Co 1.4 106 7.6 109 8.8 106 2.0 108 1.0 106 7.0 109 3.1 106 8.0 109 2.1 106 5.2 109 1.5 106 8.7 109 5.8 106 1.3 108
65
Zn 2.7 105 3.1 108 1.0 104 1.1 107 1.2 105 1.7 108 5.3 105 5.5 108 4.7 105 4.8 108 3.5 105 4.0 108 7.0 105 7.2 108
90
Srb 7.6 109 2.4 1010 4.1 108 3.3 1010 5.6 109 2.3 1010 1.5 108 1.5 1010 1.0 108 9.8 1011 8.6 109 2.8 1010 2.7 108 1.8 1010
106
Rub 5.7 108 8.9 1011 4.1 107 5.1 1010 4.7 108 7.7 1011 1.4 107 1.7 1010 8.4 108 1.1 1010 5.9 108 9.6 1011 2.7 107 3.3 1010
129
I 3.4 108 2.8 109 1.6 107 3.3 109 3.0 108 2.7 109 5.6 108 1.6 109 3.6 108 1.0 109 3.8 108 3.3 109 1.0 107 1.7 109
131
I 2.4 1010 6.1 1013 1.1 109 7.2 1013 2.1 1010 6.0 1013 3.9 1010 3.4 1013 2.5 1010 2.2 1013 2.7 1010 7.1 1013 7.1 1010 3.8 1013
134
Cs 5.9 108 3.9 109 2.1 107 4.6 109 4.6 108 3.9 109 8.7 108 2.2 109 6.3 108 1.4 109 7.0 108 4.6 109 1.4 107 2.4 109
137
Csb 4.1 108 3.6 109 1.5 107 4.1 109 3.2 108 3.5 109 6.0 108 2.0 109 4.4 108 1.3 109 4.9 108 4.2 109 9.4 108 2.2 109
210
Pbb 2.5 103 1.1 105 1.2 102 2.9 105 1.4 103 8.9 106 5.3 103 1.3 105 4.2 103 1.0 105 3.2 103 1.3 105 7.9 103 1.8 105
210
Po 7.6 104 2.6 106 3.8 103 5.0 106 4.6 104 2.4 106 1.6 103 2.2 106 1.2 103 1.6 106 9.2 104 3.1 106 2.5 103 2.9 106
226
Rab 2.6 103 1.2 105 1.2 102 3.2 105 1.4 103 9.7 106 5.3 103 1.4 105 4.3 103 1.1 105 3.2 103 1.5 105 8.0 103 2.0 105
230
Th 8.7 106 3.5 107 3.8 105 4.4 107 5.8 106 3.4 107 1.6 105 2.1 107 1.2 105 1.4 107 1.0 105 4.1 107 2.5 105 2.4 107
232
Thb 3.1 105 7.3 107 1.4 104 1.0 106 1.8 105 7.0 107 5.9 105 4.7 107 4.7 105 3.2 107 3.8 105 8.6 107 9.0 105 5.5 107
234
U 2.9 108 1.9 1010 1.9 107 5.3 1010 2.2 108 1.8 1010 6.7 108 2.1 1010 4.4 108 1.3 1010 3.2 108 2.2 1010 1.3 107 3.3 1010
238
Ub 1.7 107 6.1 109 7.9 107 7.9 109 1.2 107 5.9 109 3.2 107 3.7 109 2.4 107 2.4 109 2.0 107 7.1 109 5.2 107 4.3 109
239
Pu 1.1 105 9.0 108 8.4 105 2.4 107 9.9 106 8.8 108 2.6 105 8.9 108 1.5 105 5.3 108 1.1 105 1.0 107 5.5 105 1.5 107
240
Pu 1.1 105 9.0 108 8.4 105 2.4 107 9.9 106 8.8 108 2.6 105 8.9 108 1.5 105 5.3 108 1.1 105 1.0 107 5.5 105 1.5 107
241
Am 4.4 106 6.3 108 2.7 105 1.1 107 3.3 106 6.0 108 9.7 106 4.8 108 6.5 106 3.1 108 4.9 106 7.3 108 1.8 105 6.6 108

a
The world-average value is calculated using the world-average annual catch of fish, crustacea and molluscs from table 21.
b
For these radionuclides the calculated doses include a contribution from the ingrowth of progeny (see appendix for details).
90 UNSCEAR 2016 REPORT

VII. CONCLUSIONS
151. An important aspect of the Committees work is the periodic assessment of global and regional
radiation exposures to the public from discharges of radionuclides to the environment. Such studies
require a robust, generic methodology that is defensible and takes into account developments in the
field. The methodology used by the Committee for many years has proved to meet its needs for
worldwide exposure assessments but, as previously, before any major evaluation of exposures, the
methodology has been reviewed by the Committee to ensure its continued validity. An important factor
for the current review was the decision by the Committee to update its evaluations of human radiation
exposures from all significant types of electrical energy production. Following the review, the
Committee decided to make some changes to parts of the methodology, while retaining other parts that
it felt were still appropriate for its purposes. Previously, the results had often been expressed in terms of
dose per unit of electricity generated but, for flexibility, it was agreed that the results be presented in
terms of dose per unit of activity discharged.

152. The updated methodology can be used to estimate characteristic individual doses (typical of the
average person living in the area around the source) and collective doses due to discharges to
atmosphere, to freshwater bodies (small and large rivers or a lake) and to the sea. A wide range of
radionuclides are included to enable exposures to be assessed for nuclear and non-nuclear forms of
electrical energy production. The methodology applies to routine, continuous discharges only and it
cannot be used for accidental releases; nor is it suitable for detailed site-specific assessments of doses to
representative persons for regulatory purposes. As previously mentioned, local, regional and global
components of collective dose are estimated, as appropriate. However, it is now possible to gain
additional insight for discharges to atmosphere on collective doses within different distance bands from
the discharge point. Global components of collective dose are now also available for integration times
of 100, 500 and 10,000 years.

153. The methodology now includes factors for population densities and for food consumption rates
that vary with geographical region. The Committee added this detail because non-nuclear power
stations are found throughout the world where population densities and patterns of food consumption
are significantly different from a generic world average. However, the geographical regions adopted are
still very large and the inclusion of such region-specific data does not mean that the methodology is
suitable for site-specific assessments. Another addition to the methodology is the inclusion of doses due
to the irrigation of terrestrial foods with fresh water into which radionuclides had been discharged. The
extent and nature of irrigation is very variable throughout the world but, for the purposes of this
methodology, a simplified generic approach has been adopted for estimating exposures due to
irrigation.

154. This scientific annex describes the methodology, outlines changes from previous versions and
discusses its limitations and particular areas of uncertainty. Where possible, the results from applying
the methodology have been compared with those from other methodologies and published studies.
Although the purpose of the other studies would have influenced the developers choice of models and
parameter values and hence the estimated doses, the comparisons still helped to give confidence that the
Committees methodology is fit for its purposes.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 91

155. The methodology has been implemented in a series of Excel workbooks to ensure that it is
transparent for use by the Committee in any future studies. It should also be relatively easy to update
the methodology using such a system. The workbooks enable the user to obtain information on the
important exposure pathways and radionuclides to gain insight for the Committees work. The
workbooks have been checked and verified by people not involved in their development.

156. The Committee considers that the methodology should also be transparent to the wider community and
therefore, in addition to the information presented here, further detail is provided in electronic attachments
available on the UNSCEAR website (http://www.unscear.org/unscear/en/publications/2016.html). The
Committee is satisfied that the methodology, as implemented in the workbooks, is robust, builds on the
strong position of the previous versions of the methodology and is suitable for assessing worldwide
exposures due to routine discharges of radionuclides to the environment. It can be used to assessin a
consistent wayexposures due to releases from non-nuclear and nuclear sites of electricity generation
including the related fuel-cycle sites (e.g. uranium mining and milling, nuclear fuel reprocessing and nuclear
power stations that are being decommissioned).

VIII. ACKNOWLEDGEMENTS
The Committee wishes to acknowledge with gratitude the experts directly involved in developing the
methodology and the related workbooks, and those who carried out quality control checks, and support
staff in their institutes who assisted them. In particular, the Committee would like to thank: the
consultants, C. Robinson and J. Simmonds; the expert group, who provided valuable oversight and
advice for the work; staff, T. Anderson and K. Jones, from Public Health England in the United
Kingdom, who developed the workbook system and supported development of the methodology; and
H. Grogan, who reviewed and checked the workbook system. The views expressed in the scientific
annex and related material remain those of the Committee and do not necessarily represent the views of
individual experts, the United Nations or its Member States.

Expert group

Co-chairs: E. Waller (Canada)

Members: T. Anderson (United Kingdom), L. Anspaugh (United States), H. Grogan (United States),
G. Hirth (Australia), K. Jones (United Kingdom), L. Hubbard (Sweden), R. Michel (Germany),
J. Simmonds (United Kingdom)

Observer: G. Proehl (IAEA)

The Committee would also like to acknowledge with gratitude: the staff of the Metadata and Socio-
Economics unit of UNEP/DEWA/GRID-GENEVA who analysed the population data around nuclear
power plants; P. Verger and colleagues from WHO for their analyses of food consumption data;
S. Golikov and M. Balonov for reviewing the model for external exposure from deposited material; and
M. Tzivaki for conducting some specialized quality checks of the workbooks and their results.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 93

APPENDIX A. IMPLEMENTATION OF THE METHODOLOGY

A1. The aim of this appendix is to provide more details of how the Committee has implemented the
methodology as a set of Excel workbooks by describing the workbook design (including the end
points considered, the mathematical equations to implement the methodology in the workbooks, and
information on how the workbooks can be used to assess exposures from different types of discharge),
details of how radioactive progeny have been treated, and quality assurance.

I. WORKBOOK DESIGN
A2. The methodology has been implemented as a series of Excel workbooks from which dose
calculation factors (i.e. the calculated individual and collective doses from unit discharge of
radionuclides under a set of defined conditions) can be derived. The series includes a workbook for
each environment into which a discharge can be made (i.e. atmosphere, freshwater bodies, and marine
environments); a further workbook summarizes the doses due to discharges into each of the three
environments, and another provides dose values for globally circulating radionuclides. Table A1
summarizes the various end points that are produced by the workbooks, and how the relevant
populations and pathways are treated.

A3. The workbook for each environment into which a discharge is made comprises a number of
interlinked worksheets. An initial status worksheet describes the purpose of the workbook and provides
key information about its format and contents, version and quality control. Other worksheets contain
values of the model input parameters (radionuclide-specific ones and others such as habit data);
intermediate calculations (e.g. of activity concentrations of radionuclides as a function of distance for
discharges to atmosphere); and output values for the end points (i.e. dose calculation factors).

A4. Discharges to atmosphere. The workbook first calculates activity concentrations of radionuclides
in air at a number of distances from the point of discharge. Deposition rates and activity concentrations
of radionuclides on the ground surface, in soil and in crops, meat and milk are derived for each
distance. These activity concentrations are used to calculate characteristic individual and collective
doses due to inhalation of the plume, and due to external exposure from radionuclides in the plume and
deposited on the ground. Consumption rates of terrestrial food for different geographical regions are
used to calculate characteristic individual and collective doses from consumption of foods. Table A2
lists the equations that are implemented for discharges to an atmospheric environment that give rise to
exposure through terrestrial pathways.

A5. Discharges to a freshwater environment. The workbook calculates activity concentrations of

radionuclides in water for two sizes of river (the results for a large river are also applied for a lake).
Concentration factors are used to derive activity concentrations in freshwater fish, and irrigation rates
are used to calculate activity concentrations in irrigated crops. Activity concentrations of radionuclides
in riverbank sediment are assumed to be the same as those in riverbed sediment. These values are used
to derive characteristic individual and collective doses from consumption of irrigated foods and water
94 UNSCEAR 2016 REPORT

abstracted from the river, consumption of fish caught from the river and individual doses from external
exposure to radioactive material in riverbank sediments. Table A3 lists the equations implemented for
discharges to a freshwater environment.

A6. Discharges to a marine environment. The workbook is based on the assumption that radionuclides
are discharged to a relatively small, local compartment and water in the local compartment is
exchanged with that in a larger, regional compartment. Activity concentrations in water are calculated
for each compartment, and activity concentrations in fish, crustaceans, molluscs and beach sediment are
derived. The workbook calculates individual doses from external exposure to beach sediment, and
individual and collective doses from consumption of shellfish harvested in the local compartment and
consumption of fish caught in both compartments. Table A4 lists the relevant equations for discharges
to a marine environment.

A7. Summary workbook. This is linked to the above workbooks for each environment into which a
discharge is made and, for this reason, must be saved in the same electronic folder as them. It presents
characteristic individual and collective doses for each geographical region from discharges to
atmospheric, freshwater and marine environments. The doses from discharges of radionuclides with
very short-lived progeny (e.g. 137Cs and 137mBa) include the contributions from radioactive progeny. In
other cases (i.e. radionuclides with longer-lived progeny), the contribution from the progeny are
considered separately (see section II below). If the degree of equilibrium between different members of
the radionuclide chain is significantly different from that assumed here, then the doses from the parent
and progeny have to be summed once the activity of radionuclides in the discharge and the degree of
equilibrium are specified.

A8. Globally circulating radionuclides. The final workbook includes calculation factors for collective
doses from the global circulation of the relevant radionuclides for 100, 500 and 10,000 years following
unit discharge. These factors have been derived using the PC-CREAM 08 computer system [S8], which
implements each of the global circulation models adopted by the Committee.

A9. The results can be scaled by the actual rates of discharge of each radionuclide from a particular
site and then summed over all relevant radionuclides to give the overall characteristic individual and
collective doses. Results are produced for sites located in different regions of the world, with additional
results for coastal and inland nuclear sites and sites situated in areas of low population density.

II. TREATMENT OF RADIOACTIVE PROGENY

A10. For most radionuclides the workbooks for atmospheric, freshwater and marine environments
provide the relevant dose factors separately for the parent and any progeny. (This permits the
workbooks to be used when non-equilibrium conditions apply). Tables A5, A6 and A7 contain further
details on how dose contributions from radioactive progeny have been treated for terrestrial, freshwater
and marine pathways respectively, following the general principles outlined in section II.D.2. The
summary workbook contains summed values based on the assumptions detailed in those tables.
However, for radionuclides that are in secular equilibrium with their short-lived progeny, the doses
presented and clearly marked in the workbooks are the sums of the doses from the parent and progeny.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 95

III. QUALITY ASSURANCE

A11. A systematic review of the completed workbooks was carried out for quality assurance purposes,
firstly by the workbook developers and then by independent reviewers. This included a comparison of
the parameter values and variables with source data, and the workbook equations with the agreed
calculation methodology, defined by the lists of equations and data presented below in tables A2-A4.
This ensured that the workbooks had implemented the methodology agreed by the Committee.

A12. As further verification, a workbook was created that compared a sample of the results of the
workbooks with those published in the UNSCEAR 2000 Report [U6]. The results are discussed in
detail in electronic attachment 1.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 97

TABLES
Table A1. Summary of end points, populations and pathways considered

(a) Characteristic individual dose

End point/pathway Key features of methodology Integration considerations

DISCHARGE TO ATMOSPHERE
Characteristic individual dose Total dose from all exposure pathways assuming Time integration: 100 years
location 5 km distant from site of discharge Individual dose in 100th year following
100 years continuous discharge

External irradiation (beta + gamma Activity concentrations of radionuclides in plume

radiation from plume, and gamma derived using approach in [U6]
radiation from deposited radionuclides) Time-integrated activity concentrations in soil from
[S8]
Dose factors (table A3 below)

Inhalation Activity concentrations of radionuclides in plume

derived using approach in [U6]

Ingestion of terrestrial foods Activity concentrations of radionuclides in

terrestrial foods per unit deposition rate from [S8]
Approaches for 14C and 3H based on [I5]

DISCHARGE TO FRESHWATER ENVIRONMENT

Characteristic individual dose Total dose from all exposure pathways assuming Time integration: 100 years
complete instantaneous mixing Individual dose in 100th year following
For two situations: discharge into large and small 100 years continuous discharge
rivers (for nuclear and mines/non-nuclear
discharges, respectively)

External irradiation from occupancy on Time-integrated activity concentrations of

freshwater sediments radionuclides in sediment from [S8] 5 km distant
from discharge point

Drinking water Ingestion rate from [U6]

Fish ingestion Region-specific consumption rates from [F3]

Ingestion of irrigated terrestrial foods Deposition rate related to generic irrigation rate
and growing period appropriate for grain [W2]
Approaches for 14C and 3H derived from [C1]

DISCHARGE TO MARINE ENVIRONMENT

Characteristic individual dose Total dose from exposure pathways calculated Time integration: 100 years
using a two-compartment model Individual dose in 100th year following
Activity concentrations of radionuclides in 100 years continuous discharge
seawater in local marine compartment

External irradiation from occupancy on Time-integrated activity concentrations of

marine sediments radionuclides in sediments within the local
marine box [S8]

Ingestion of marine foods (fish, Ingestion of crustaceans and molluscs from the
crustaceans and molluscs) local compartment (25%) and regional
compartment (75%)
98 UNSCEAR 2016 REPORT

(b) Collective dose

Component Key features of methodology Integration considerations

DISCHARGE TO ATMOSPHERE
Local component Based on individual dose at midpoint and number Time integration: 100 years
of people in annuli around discharge point Distance integration: using activity
derived from population density information concentrations of radionuclides and
from [F1] for nuclear sites or using a population individual doses at centre point of annular
density of 5 km2 for sites in areas of low area, 0100 km
population (see electronic attachment 3)

Regional component Based on individual dose and number of people in Time integration: 100 years, 500 years
annuli around discharge point for nuclear sites Distance integration: using activity
(see electronic attachment 3) or derived from concentrations of radionuclides and
weighted population density information from individual doses at centre point of
[F1] following annular areas: 100500,
5001 000 and 1 0001 500 km
For discharge continuing for 100 years and
integration times of 100 years

Global component Based on collective dose from globally dispersed Time integration: 100 years, 500 years and
radionuclides [S8], which implements models 10 000 years
referred to in [U6]

DISCHARGE TO FRESHWATER ENVIRONMENT

Local and regional components from Based on total collective intakes from abstraction Time integration: 100 years
freshwater pathways (external of drinking water and freshwater fish-catch data
irradiation from exposure on freshwater from specified lengths of small and large rivers
sediment not included) Consumption of irrigated terrestrial foods based
on generalized yield data and areas of land
equipped for irrigation as function of region
(from [F1])

Global component Based on collective dose from globally dispersed

radionuclides [S8], which implements models
referred to in [U6]

DISCHARGE TO MARINE ENVIRONMENT

Regional component Based on total collective intake from seafood- Time integration: 100 years
(External exposure on marine sediment catch data derived from the representative
not included) regional marine compartment

Global component Based on collective dose from globally dispersed

radionuclides [S8], which implements models
referred to in [U6]
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 99

Table A2. Parameter and equation list: Atmospheric discharge and terrestrial environment

Parameter Symbol Value Units Reference

ATMOSPHERIC DISPERSION

Activity concentration in air at a specified distance calculated using:

Cair,i ( x ) = D1 Q x n e i x ua (T1)

Activity concentration of radionuclide, i, in air Cair,i ( x ) Equation (T1) Bq/m3

at distance, x, from discharge point

Dilution factor at 1 km D1 5.3 107 s/m3 [U6]

Discharge rate Q Unit discharge rate = 1 Bq/s

Downwind distance from discharge point x 5, 50, 300, 750, 1 250 km

Empirical index n Noble gases, [U6]

tritium: 1.2
14
C: 1.4
All others: 1.42

Radioactive decay constant of radionuclide, i li Radionuclide-specific s1 [I9]

Geometric mean of wind speed at the height ua 2 103 km/s [I2]

of release representative of one year

Activity concentration in air of 135Cs (progeny of 135Xe) and 138Cs (progeny of 138Xe) calculated using:

=Cair , progeny ( x ) Cair , parent ( x )

parent
progeny parent
1 e (
(
progeny parent )( x u a )
) (T2)

(note this equation is valid for one progeny and does not include a branching fraction)

Activity concentration of progeny in air at Cair,progeny ( x ) Equation (T2) Bq/m3

distance, x, from discharge point

Activity concentration of parent in air at Cair,parent ( x ) Equation (T1) Bq/m3

distance, x, from discharge point

Radioactive decay constant of parent , progeny Radionuclide-specific s1 [I9]

parent/progeny

Downwind distance from discharge point x 5, 50, 300, 750, 1 250 km

Geometric mean of wind speed at the height ua 2 103 km/s [I2]

of release representative of one year

Deposition rate at a specified distance calculated using:

di ( x ) = VT Cair,i ( x ) (T3)

Total daily-average deposition rate at di ( x ) Equation (T3) Bq/(m2 s)

distance, x, of radionuclide, i, from both
wet and dry processes
100 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Total deposition velocity (wet and dry) VT Noble and non-reactive gases: 0 m/s [U6]
Tritium: 0
14
C: 0
All others: 0.002

Activity concentration of radionuclide, i, in Cair,i ( x ) Equations (T1), (T2) Bq/m3

air at distance, x, from discharge point

ENVIRONMENTAL TRANSFER

Activity concentration in terrestrial food calculated using:

C f ,i ( x ) = C f ,unit,i di ( x ) (T4)

Activity concentration of radionuclide, i, in C f ,i ( x ) Equation (T4) Bq/kg or Bq/L

terrestrial food, f, at distance, x, from
discharge point

Activity concentration of radionuclide, i, in C f ,unit,i PC-CREAM output [S8] Bq/kg per Bq/(m2 s)
year 100 in terrestrial food, f, per unit or
deposition rate for continuous deposition Bq/L per Bq/(m2 s)

Total daily average deposition rate at di ( x ) Equation (T3) Bq/(m2 s)

distance, x, of radionuclide, i, from both
wet and dry processes

14
C activity concentration in terrestrial food (cereal, vegetables and pasture) calculated using:
Sp
C=
f ,14 C
( x ) Cair,14C ( x ) (T5)
Sair

Activity concentration of 14C in food, and C f ,14 C ( x ) Equation (T5) Bq/kg [I5]
pasture, f, at distance, x, from discharge
point

Activity concentration of 14C in air at distance, Cair,14C ( x ) Equation (T1) Bq/m3

x, from discharge point

Concentration of stable carbon in the plant Sp Cereal: 3.9 102 g C/kg [I5]
(fresh weight) Vegetables: 30
Pasture: 1 102

Concentration of stable carbon in air Sair 0.2 g C/m3 [I5]

14
C activity concentration in terrestrial food (animal products) calculated using:
f C Cpasture,14C ( x ) S a
C f ,14 C ( x ) = (T6)
Sp

Activity concentration of 14C in food, f, at C f ,14 C ( x ) Equation (T6) Bq/kg [I5]

distance, x, from discharge point

Fraction of animal feed that is contaminated fC 1 [I5]

Activity concentration of 14C in pasture at Cpasture,14C ( x ) Equation (T5) Bq/kg

distance, x, from discharge point
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 101

Parameter Symbol Value Units Reference

Concentration of stable carbon in the animal Sa Milk: 65 g/kg [I5]

product (fresh weight) Beef: 200

Concentration of stable carbon in the plant Sp Pasture: 1 102 g/kg [I5]

(fresh weight)

Activity concentration of HTO in soil water calculated using:

CR s-a Cair,HTO ( x )
Csw,HTO ( x ) = (T7)
Ha

Activity concentration of HTO in soil water at Csw,HTO ( x ) Equation (T7) Bq/L [I5]
distance, x, from discharge point

Empirical constant CR s-a 0.3 [I5]

Activity concentration of HTO in air at Cair,HTO ( x ) Equation (T1) Bq/m3

distance, x, from discharge point

Absolute humidity Ha 6 103 L/m3 [I5]

Calculated using:

C f ,HTO ( x=
C ( x)
) WC p RH air,HTO + (1 RH ) Csw,HTO ( x ) (T8)
H a

Activity concentration of HTO in food, f, at C f ,HTO ( x ) Equation (T8) Bq/kg [I5]

distance, x, from discharge point (fresh
weight)

Fractional water content of the plant (fresh WC p Cereal: 0.12 L/kg [I5]
weight) Vegetables: 0.92
Pasture: 0.76

Relative humidity RH 0.7 [I4]

Activity concentration of HTO in air at Cair,HTO ( x ) Equation (T1) Bq/m3

distance, x, from discharge point

Absolute humidity Ha 6 103 L/m3 [I2]

Activity concentration of HTO in soil water at Csw,HTO ( x ) Equation (T7) Bq/L [I5]
distance, x, from discharge point

Ratio of HTO vapour pressure to H2O vapour 0.909 [I5]

pressure

Weighted concentration of HTO in drinking water and feed calculated using:

Cpasture,HTO ( x )
CR f ,HTO ( x ) = 0.5 Csw,HTO ( x ) + 0.5 (T9)
WC p

Weighted concentration of HTO in drinking CR f ,HTO ( x ) Equation (T9) Bq/L

water and of water in feed (assumes that
50% intake is drinking water and 50% is
pasture)
102 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration of HTO in soil water at Csw,HTO ( x ) Equation (T7) Bq/L [I5]
distance, x, from discharge point

Activity concentration of HTO in pasture at Cpasture,HTO ( x ) Equation (T8) Bq/kg [I5]

distance, x, from discharge point (fresh
weight)

Fractional water content of the plant WCp 0.76 L/kg [I5]

HTO activity concentration in terrestrial food (animal products) calculated using:

=
C f ,HTO ( x ) CR a ,HTO CR f ,HTO ( x ) (T10)

Activity concentration of HTO in food, f, at C f ,HTO ( x ) Equation (T10) Bq/kg [I5]

distance, x, from discharge point

Concentration ratio: concentration of HTO in CR a ,HTO Milk: 0.87 L/kg [I5]

product (fresh weight) divided by Beef: 0.66
concentration of HTO in water intake

Weighted concentration of HTO in drinking CR f ,HTO ( x ) Equation (T9) Bq/L

water and of water in feed

OBT activity concentration in terrestrial food (cereal, vegetables and pasture) calculated using:
WEQ p R p C f ,HTO ( x )
(1 WC p )
C f ,OBT ( x ) =
WC p
(T11)

Activity concentration of OBT in food, f, at C f ,OBT ( x ) Equation (T11) Bq/kg [I5]

distance, x, from discharge point

Fractional water content of the plant (fresh WC p Cereal: 0.12 L/kg [I5]
weight) Vegetables: 0.92
Pasture: 0.76

Water equivalent factor WEQ p Cereal: 0.56 L/kg [I5]

Vegetables: 0.51
Pasture: 0.56

Partition factor Rp 0.54 [I5]

Activity concentration of HTO in food, f, at C f ,HTO ( x ) Equation (T8) Bq/kg [I5]

distance, x, from discharge point

OBT activity concentration in terrestrial food (animal products)

=
C f ,OBT ( x ) CR a ,OBT Cpasture,OBT ( x ) (T12)

Activity concentration of OBT in food, f, at C f ,OBT ( x ) Equation (T12) Bq/kg [I5]

distance, x, from discharge point

Concentration ratio: concentration of OBT in CR a ,OBT Milk: 0.24 kg/L [I5]

milk (fresh weight) divided by Beef: 0.40
concentration of OBT in intake (dry weight)

Mean concentration of OBT in feed Cpasture,OBT ( x ) Equation (T11) Bq/kg

 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 103

Parameter Symbol Value Units Reference

INDIVIDUAL DOSE CALCULATIONS

Effective dose from inhalation at a specified distance calculated using:

I inh
H E ( inh ),i ( x=
) Cair,i ( x ) Dinh,i Oann (T13)
86 400

Effective dose from inhalation of H E (inh ),i ( x ) Equation (T13) Sv

radionuclide, i, at distance, x, from
discharge point

Activity concentration of radionuclide, i, in air Cair,i ( x ) Equation (T1) Bq/m3

at distance, x, from discharge point

Dose coefficient for inhalation of Dinh,i Dependent on radionuclide Sv/Bq [I14]

radionuclide, i, (see equation (T14) for
calculation of dose coefficient for 222Rn plus
short-lived progeny)

Adult inhalation rate I inh 20 m3/d [U6]

Conversion factor 86 400 s/d

Annual occupancy at location Oann 3.15 107 (rounded) s

Dose coefficient for inhalation of radon and its short-lived progeny calculated using:

Dinh,222Rn= 24
(
DC 222 Rn EF222 Rn,in O222 Rn,in + EF222 Rn,out O222 Rn,out ) (T14)
I inh

Dose coefficient for inhalation of 222Rn plus Dinh,222Rn Equation (T14) Sv/Bq [U6]
short-lived progeny

Conversion factor 24 h/d

Dose conversion factor for radon DC 222 Rn 9 109 Sv per (Bq h/m3) [U6]

Indoor equilibrium factor for radon EF222 Rn,in 0.4 [U6]

Indoor occupancy factor O222 Rn,in 0.8 [U6]

Outdoor equilibrium factor for radon EF222 Rn,out 0.6 [U6]

Outdoor occupancy factor O222 Rn,out 0.2 [U6]

Adult inhalation rate I inh 20 m3/d [U6]

Dose from external exposure from immersion in the plume at a specified distance calculated using:

( x ) Cair,i ( x ) Dex,cloud,i Oann ( Oout + (1 Oout ) Lcloud )

H E (ex,cloud),i= (T15)

Effective gamma dose from external H E (ex,cloud),i ( x ) Equation (T15) Sv

exposure to radionuclide, i, at distance, x,
from discharge point
104 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration of radionuclide, i, in air C air,i ( x ) Equation (T1) Bq/m3

at distance, x, from discharge point

External dose coefficient for immersion in the Dex,cloud,i Dependent on radionuclide Sv per (Bq s/m3) [E2, P2]
plume

Annual occupancy at location Oann 3.15 107 (rounded) s

Fraction of occupancy that is outdoors Oout 0.2 [U4]

Building shielding factor in plume (location Lcloud 0.2

factor)

Dose from external exposure from deposits calculated using:

) di ( x ) tdischarge Dex,deposit,i ( Oout + (1 Oout ) Ldeposit )

H E ( ex,deposit ),i ( x= (T16)

Annual effective dose from the deposited H E ( ex,deposit ),i ( x ) Equation (T16) Sv
radionuclide, i, at distance, x, from
discharge point

Total daily average deposition rate at di ( x ) Equation (T3) Bq s/m2

distance, x, of radionuclide, i, from both
wet and dry processes

Duration of deposition tdischarge 3.15 107 (rounded) s 1 year

Dose coefficient for external exposure from Dex,deposit,i Dependent on radionuclide Sv per (Bq/m2) [E2, P2]
deposited material, integrated to 100 years
(time dependence taken into account using
equation given by [G2])

Fraction of occupancy that is outdoors Oout 0.2 [U4]

Building shielding factor from deposited Ldeposit 0.1

material (location factor)

Effective dose from ingestion calculated using:

H E (ing ), f ,r ,i ( x=
) C f ,i ( x ) Ding,i Flocal I f ,r (T17)

Annual effective dose from the ingestion of H E (ing ), f ,r ,i ( x ) Equation (T17) Sv

food, f, containing radionuclide, i, at
distance, x, from discharge point in region r

Activity concentration of radionuclide, i, in C f ,i ( x ) Equations (T4), (T5), (T6), (T8), Bq/kg or Bq/L
terrestrial food, f, at distance, x, from (T10), (T11), (T12)
discharge point

Dose coefficient for ingestion Ding,i Dependent on radionuclide Sv/Bq [I14]

Fraction of food that is locally produced Flocal Individual dose: 0.25

Collective dose: 1.0

Amount of food, f, ingested per caput in a I f ,r Dependent on region and kg or L [W2]

year in region, r food
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 105

Parameter Symbol Value Units Reference

Total effective dose from ingestion calculated using:

H E ( ing ),r ,i ( x ) = H E (ing ), f ,r ,i ( x ) (T18)
f

Total annual effective dose in region, r, from H E (ing ),r ,i ( x ) Equation (T18) Sv
ingestion of radionuclide, i, in food at
distance, x, from discharge point

Annual effective dose from ingestion of food, H E (ing ), f ,r ,i ( x ) Equation (T17) Sv

f, containing radionuclide, i, at distance, x,
from discharge point in region, r

Total individual effective dose calculated using:

H E ( atmos ),r ,i ( x ) =
H E (inh ),i ( x ) + H E (ex,cloud),i ( x ) + H E ( ex,deposit ),i ( x ) + H E (ing ),r ,i ( x ) (T19)

Total effective dose in region, r, from H E ( atmos ),r ,i ( x ) Equation (T19) Sv

radionuclide, i, at distance, x, from
discharge point

Effective dose from inhalation of H E (inh ),i ( x ) Equation (T13) Sv

radionuclide, i, at distance, x, from
discharge point

Effective dose from external exposure to H E (ex,cloud),i ( x ) Equation (T15) Sv

radionuclide, i, at distance, x, from
discharge point

Effective dose from the deposited H E ( ex,deposit ),i ( x ) Equation (T16) Sv

radionuclide, i, at distance, x, from
discharge point

Total effective dose in region, r, from H E (ing ),r ,i ( x ) Equation (T18) Sv

ingestion of radionuclide, i, in food at
distance, x, from discharge point

COLLECTIVE DOSE CALCULATIONS

Collective effective dose calculated using:

,i ( k
S E ( atmos ),r= x , xk +1 ) N r ( xk , xk +1 ) H E ( atmos ),r ,i ( x ) (T20)

Collective effective dose in region, r, in the S E ( atmos ),r ,i ( xk , xk +1 ) Equation (T20) man Sv
annulus limited by xk and xk+1

Total population in region, r, in the annulus N r ( xk , xk +1 ) Dependent on region see electronic

limited by xk and xk+1 attachment 3
and [F1]

Annular distances (the outer distance for ( x1 , x2 , x3 , x4 , x5 ) {0,100,500,1 000,1 500} km

each annuli)

Total effective dose from radionuclide, i, at H E ( atmos ),r ,i ( x ) Equation (T19) Sv

distance, x, from discharge point where x is
the midpoint between xk and xk+1
106 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Local collective effective dose calculated using:

2
S E ( atmos ),r ,i ( local ) = S E (atmos),r ,i ( xk , xk +1 ) (T21)
k =1

Local (0 km to 100 km) collective effective S E ( atmos ),r ,i ( local ) Equation (T21) man Sv
dose in region, r, from radionuclide, i

Total annual collective effective dose in S E (atmos),r ,i ( xk , xk +1 ) Equation (T20) man Sv

region, r, from radionuclide, i, in the
annulus limited by xk and xk+1

Regional component of collective effective dose

5
S E ( atmos ),r ,i ( regional ) = S E (atmos),r ,i ( xk , xk +1 ) (T22)
k =3

Regional component (100 km to 1500 km) of S E ( atmos ),r ,i ( regiona Equation (T21) man Sv
collective effective dose in region, r, from
radionuclide, i

Total annual collective effective dose in S E (atmos),r ,i ( xk , xk +1 ) Equation (T20) man Sv

region, r, from radionuclide, i, in the
annulus limited by xk and xk+1

GLOBAL DISPERSION

Collective doses from tritium, 14C, 85Kr and 129I were calculated by PC-CREAM (incorporates global dispersion models referenced in [U6])
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 107

Table A3. Parameter and equation list: Freshwater environment

Parameter Symbol Value Units Reference

FRESHWATER DISPERSION

Activity concentration in unfiltered water calculated using:

Q
Cuw , s = (F1)
Friver, s

Activity concentration in unfiltered water Cuw ,s Equation (F1) Bq/m3

Annual discharge rate of radionuclide, i Q 1 Bq/s

Volumetric flow rate of river of size, s Friver,s Large: 1 000 m3/s

Small: 10

Activity concentration in filtered water calculated using:

Cuw , s
Cfw , s ,i = (F2)
1 + K d ,i s

Activity concentration of radionuclide, i, in filtered water Cfw , s ,i Equation (F2) Bq/m3

from river of size, s

Activity concentration in unfiltered water Cuw ,s Equation (F1) Bq/m3

Sedimentwater distribution factor K d ,i Dependent on element m3/t [I2, I5]

Suspended sediment load in river of size, s s Large: 5 104 t/m3 [O1]

Small: 2 10 5

ENVIRONMENTAL TRANSFERS

Activity concentrations in freshwater fish calculated using:

Cfw , s ,i Bfish,i
Cfish, s ,i = (F3)
1 000

Activity concentration of radionuclide, i, in freshwater fish Cfish, s ,i Equation (F3) Bq/kg

Activity concentration of radionuclide, i, in filtered water Cfw , s ,i Equation (F2) Bq/m3

from river of size, s

Concentration factor for freshwater fish (fresh weight) Bfish,i Dependent on food and Bq/kg per [I5]
element Bq/L

Conversion factor 1 000 L/m3

Activity concentration of HTO in fish calculated using:

WC f Cfw , s ,HTO
Cfish, s ,HTO = (F4)
1 000

Activity concentration of HTO in freshwater fish Cfish, s ,HTO Equation (F4) Bq/kg
108 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Fractional water content of fish (fresh weight) WC f 0.78 L/kg [I5]

Activity concentration of HTO in filtered water from river of Cfw , s ,HTO Equation (F2) Bq/m3
size, s

Conversion factor 1 000 L/m3

Activity concentration of OBT in fish calculated using:

Cfish, s ,OBT =
(1 WC ) WEQ
f f R f Cfw , s ,HTO
(F5)
1 000

Activity concentration of OBT in freshwater fish Cfish, s ,OBT Equation (F5) Bq/kg

Fractional water content of fish (fresh weight) WC f 0.78 L/kg [I5]

Water equivalent factor for fish WEQ f 0.65 [I5]

Partition factor for fish (geometric mean, fresh weight) Rf 6.6 104 kg/kg [I5]

Activity concentration of HTO in filtered water from river of Cfw , s ,HTO Equation (F2) Bq/m3
size, s

Conversion factor 1 000 L/m3

Activity concentrations of 14C in fish calculated using:

Cfw , s ,14 C Bfish,14 C
Cfish, s ,14 C = (F6)
1 000

Activity concentration of 14C in freshwater fish (adapted Cfish, s ,14 C Equation (F6) Bq/kg
from Equation (F3))

Activity concentration of 14C in filtered water from river of Cfw , s ,14 C Equation (F2) Bq/m3
size, s

Concentration factor for 14C in freshwater fish Bfish,14 C 5 104 Bq/kg per [I5]
Bq/L

Conversion factor 1 000 L/m3

Deposition rate from irrigation calculated using

dirr, s= Cuw , s I irr Firr (F7)

Total daily average deposition rate of radionuclide, i, from dirr,s Equation (F7) Bq/m2
irrigation

Average concentration of radionuclide, i, in river water used Cuw ,s Equation (F1) Bq/m3
for irrigation

Daily irrigation rate I irr 0.005 m3/m2 [N2]

Fraction of the year for which irrigation occurs Firr 150 [N2]
=0.41
365
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 109

Parameter Symbol Value Units Reference

Activity concentration in terrestrial food (cereals, vegetables) calculated using:

C f ,unit,i dirr, s
C f , s ,i = (F8)
86 400

Activity concentration of radionuclide, i, in irrigated, C f , s ,i Equation (F8) Bq/kg

terrestrial food, f

Activity concentration of radionuclide, i, in terrestrial food, f, C f ,unit,i PC-CREAM output Bq/kg per
per unit deposition rate Bq s/m2

Total daily average deposition rate of radionuclide, i, from dirr,s Equation (F7) Bq/m2
irrigation

Conversion factor 86 400 s/d

Activity concentration of HTO in soil water calculated using:

I irr Cuw , s
Csw , s ,HTO = (F9)
P + 1 000 I irr

Activity concentration of HTO in soil water irrigated from Csw , s ,HTO Equation (F9) Bq/L [I4]
river of size, s

Daily irrigation rate (averaged over 150-day growing I irr 0.005 m3/m2 [N2]
season)

Average concentration of radionuclide, i, in river water used Cuw ,s Equation (F1) Bq/m3
for irrigation

Daily precipitation rate (averaged over 150-day growing P 5 L/m2 [U6]

season)

Conversion factor 1 000 L/m3

HTO activity concentration in terrestrial food (cereals, vegetables) calculated using:

= WC p RH Cam + (1 RH ) Csw , s ,HTO

C f , s ,HTO (F10)

Activity concentration of HTO in food, f C f , s ,HTO Equation (F10) Bq/kg [I5]

Fractional water content of the plant WC p Cereal: 0.12 L/kg [I5]

Vegetables: 0.92

Relative humidity RH 0.7 [I4]

HTO concentration in air moisture Cam Assumed approximately Bq/L [I4]

equal to Csw , s ,HTO

Activity concentration of HTO in soil water Csw , s ,HTO Equation (F9) Bq/L

Ratio of HTO vapour pressure to H2O vapour pressure 0.909 [I5]

Activity concentration of OBT in terrestrial food (cereals, vegetables) calculated using:

WEQ p R p C f , s ,HTO
(1 WC p )
C f , s ,OBT =
WC p
(F11)
110 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration of OBT in terrestrial food C f , s ,OBT Equation (F11) Bq/kg

Fractional water content of plant (fresh weight) WCp Cereal: 0.12 L/kg [I5]
Vegetables: 0.92

Water equivalent factor for plant WEQp Cereal: 0.56 L/kg [I5]
Vegetables: 0.51

Partition factor Rp 0.54 [I5]

Activity concentration of HTO in food, f C f , s ,HTO Equation (F10) Bq/kg [I5]

Specific activity of 14C in plant calculated using:

I irr Cuw , s
=
SA p =
SA air CDc (F12)
Fc

Specific activity of 14C in plant SAp Equation (F12) Bq/(g C) [I4]

Specific activity of 14C in air SAair Equation (F12) Bq/(g C) [I4]

Canopy dilution factor CDc 0.15 [I4]

Daily irrigation rate (averaged over growing season) Iirr 0.005 m3/m2 [N2]

Activity concentration of dissolved radionuclide i in Cuw,s Equation (F1) Bq/m3

unfiltered water

Average production rate of carbon by decomposition of Fc 0.66 (g C)/m2 [I4]

crop residues

14
C activity concentration in terrestrial food (cereals, vegetables) calculated using:
Cair,14C
C f , s ,14 C= S p= SA air S p (F13)
Sair

14
C activity concentration in food, f, irrigated from river of C f , s ,14 C Equation (F13) Bq/kg [I4]
size, s

Specific activity of 14C in air SAair Equation (F12) Bq/(g C) [I4]

Concentration of stable carbon in the plant Sp Cereal: 3.9 102 (g C)/kg [I5]
Vegetables: 30

INDIVIDUAL DOSE CALCULATIONS

Individual dose from ingestion of drinking water calculated using

H E ( water ), s ,i = Cuw , s Ding,i I water FWT Priver (F14)

Annual effective dose from the ingestion of radionuclide, i, H E ( water ), s ,i Equation (F14) Sv
in drinking water

Average activity concentration of radionuclide, i, in water Cuw ,s Equation (F1) Bq/m3

assuming complete mixing
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 111

Parameter Symbol Value Units Reference

Dose coefficient for ingestion of radionuclide, i Ding,i Dependent on radionuclide Sv/Bq [I14]

Amount of water ingested in one year I water 0.5 m3 [U6]

Drinking water treatment removal factor FWT Dependent on element [B7]

Fraction of individual water intake from river Priver 1

Dose from external exposure from freshwater sediments calculated using:

H E ( ex,riverbank ), s=
,i
Csed, s ,i rsed tsed Fgeom Dex,deposit Oriverbank (F15)

Annual effective dose from radionuclide, i, from occupancy H E ( ex,riverbank ), s ,i Equation (F15) Sv
on river banks

Activity concentration of radionuclide, i, in freshwater Csed, s ,i PC-CREAM output Bq/kg

riverbank sediment based on discharge rate of 1 Bq/s
(assumed to be the same as bed sediment) (dry weight)

Density of riverbank sediment sed 1 200 kg/m3 [I2]

Thickness of riverbank sediment tsed 0.05 m [I2]

Geometry factor for river banks for external dose Fgeom 0.2 (rounded) [A2]
coefficients

Integrated dose coefficient for irradiation from surface Dex,deposit Dependent on radionuclide Sv/s per [E2]
deposits Bq/m2

River bank occupancy Oriverbank 1.8 105 (50 hours) s Assumed value

Ingestion of freshwater fish

H E (ing,fish ), s ,r ,i = Cfish, s ,i Ding,i Flocal I fish,r (F16)

Annual effective dose from radionuclide, i, from ingestion H E (ing,fish ), s ,r ,i Equation (F16) Sv
of freshwater fish in region, r

Average activity concentration of radionuclide, i, in Cfish, s ,i Equations (F3), (F4), (F5), Bq/kg
freshwater fish (F6)

Dose coefficient for adults for ingestion Ding,i Dependent on radionuclide Sv/Bq [I14]

Fraction of food that is locally produced Flocal 0.25 See para.11

[U12]

Amount of freshwater fish ingested in a year in region, r I fish,r Dependent on region kg [F3]

Dose from ingestion of irrigated, terrestrial foods calculated using:

H E (ing,irr ), f , s ,r ,i = C f , s ,i Ding,i Flocal Fspray, s ,r I f ,r (F17)

Annual effective dose from the ingestion of irrigated, H E (ing,irr ), f , s ,r ,i Equation (F17) Sv
terrestrial food, f, containing radionuclide, i, in region, r
112 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration of radionuclide, i, in irrigated, C f , s ,i Equations (F8), (F10), (F11), Bq/kg

terrestrial food, f (F13)

Adult dose coefficient for ingestion Ding,i Dependent on radionuclide Sv/Bq [I14]

Fraction of food that is locally produced Flocal 0.25 See para.11

[U12]

Fraction of food that is spray irrigated Fspray, s ,r Small rivers: 0 [I7]

Large rivers: dependent on
region

Amount of food, f, ingested in a year in region, r I f ,r Dependent on region and kg [W2]

food

Total effective dose from ingestion of irrigated, terrestrial foods calculated using:
H E (ing,irr ), s ,r ,i = H E (ing,irr ), f , s ,r ,i (F18)
f

Total effective dose in region, r, from ingestion of irrigated, H E (ing,irr ), s ,r ,i Equation (F18) Sv
terrestrial food containing radionuclide, i

Annual effective dose from the ingestion of irrigated, H E (ing,irr ), f , s ,r ,i Equation (F17) Sv
terrestrial food, f, containing radionuclide i in region, r

Total individual dose from freshwater pathways calculated using:

H E ( fw ), s ,r ,i = H E ( water ), s ,i + H E ( ex,riverbank ), s ,i + H E ( fish ), s ,r ,i + H E (ing,irr ), s ,r ,i (F19)

Total individual dose from freshwater pathways H E ( fw ), s ,r ,i Equation (F19) Sv

Annual effective dose from the ingestion of radionuclide, i, H E ( water ), s ,i Equation (F14) Sv
in drinking water

Annual effective dose from radionuclide, i, from occupancy H E ( ex,riverbank ), s ,i Equation (F15) Sv
on river banks

Annual effective dose from radionuclide, i, from ingestion of H E ( fish ), s ,r ,i Equation (F16) Sv
freshwater fish in region, r

Total effective dose in region, r, from ingestion of irrigated, H E (ing,irr ), s ,r ,i Equation (F18) Sv
terrestrial food containing radionuclide, i

COLLECTIVE DOSE CALCULATIONS

Volumetric abstraction rate of water calculated using:

=
Virr,surface, s ,r Friver, s Firr,surface,r (F20)

Volumetric abstraction rate of surface water from river of Virr,surface, s ,r Equation (F20) m3/s
size, s, in region, r

Volumetric flow rate of river of size, s Friver,s Large: 1 000 m3/s

Small: 10

Fraction of surface water withdrawn for irrigation in region, r Firr,surface,r Dependent on region m2 a/m3 [F4]
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 113

Parameter Symbol Value Units Reference

Area irrigated by abstracted water calculated using:

Airr,surface, s ,r = 3.15 107 Airr,unit,r Virr,surface, s ,r (F21)

Area irrigated by water abstracted from river of size, s, in Airr,surface, s ,r Equation (F21) m2
region, r

Area irrigated per unit volume water withdrawn in a year in Airr,unit,r Dependent on region m2 a/m3 [F4]
region, r

Volumetric abstraction rate of surface water from river of Virr,surface, s ,r Equation (F20) m3/s
size, s, in region, r

Conversion factor 3.15 107 (rounded) s/a

Irrigation rate calculated using:

C V
dirr,surface, s ,r = uw , s irr,surface, s ,r (F22)
Airr,surface, s ,r

Irrigation rate, assumed to be the same volume of water dirr,surface, s ,r Equation (F22) Bq s/m2
with the same activity concentration abstracted from river
of size, s, in region, r

Activity concentration in unfiltered water Cuw ,s Equation (F1) Bq/m3

Volumetric abstraction rate of surface water from river of Virr,surface, s ,r Equation (F20) m3/s
size, s, in region, r

Area irrigated by water abstracted from river of size, s, in Airr,surface, s ,r Equation (F21) m2
region, r

Activity concentration in irrigated, terrestrial foods calculated using:

C=
cereal, s , r ,i Ccereal,unit,i dirr,surface, s ,r (F23)

Activity concentration of radionuclide, i, in terrestrial food Ccereal, s ,r ,i Equation (F23) Bq/kg

(represented by cereal) irrigated from river of size, s, in
region, r

Activity concentration of radionuclide, i, in terrestrial food, Ccereal,unit,i PC-CREAM output Bq/kg per
f, per unit deposition rate Bq s/m2

Irrigation rate dirr,surface, s ,r Equation (F22) Bq s/m2

Activity concentration of 14C in cereals calculated using:

Cair,14C
Ccereal, s ,r ,14=
C
S=
p SA air
Sair
(F24)
Cuw , s
=
CDc Sp
365 Airr,unit,r Fc

Activity concentration of 14C in food, f, irrigated from river of Ccereal, s ,r ,14 C Equation (F24) Bq/kg Equations
size, s (F13) & (F12)

Canopy dilution factor CDc 0.15 [I4]

114 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration in unfiltered water Cuw ,s Equation (F1) Bq/m3

Area irrigated per unit volume in region, r Airr,unit,r Dependent on region m2 a/m3 [F4]

Average daily production rate of carbon by decomposition Fc 0.66 (g C)/m2 [I4]

of crop residues

Concentration of stable carbon in the plant Sp 3.9 102 (g C)/kg [I5]

Conversion factor 365 d/a

Collective dose from ingestion of irrigated, terrestrial food calculated using:

S E (ing,irr ), s ,r ,=
i
Ccereal, s ,r ,i Ycereal,r Fspray, s ,r Airr,surface, s ,r Ding,i (F25)

Collective dose from consumption of food (represented by S E (ing,irr ), s ,r ,i Equation (F25) man Sv
major food group, cereal) in region, r

Activity concentration of radionuclide, i, in terrestrial food Ccereal, s ,r ,i Equations (F23), (F10), (F11), Bq/kg
(represented by cereal) irrigated from river of size, s, in (F24)
region, r

Yield per unit area of cereal in region, r Ycereal,r Dependent on region kg/m3 [F1]

Fraction of food that is spray irrigated Fspray, s ,r Small rivers: 0 [I7]

Large rivers: dependent on
region

Area irrigated by water abstracted from river of size, s, in Airr,surface, s ,r Equation (F21) m2
region, r

Adult dose coefficient for ingestion Ding,i Dependent on radionuclide Sv/Bq [I14]

Collective dose from drinking water calculated using:

S E ( water ), s ,i = Cuw ,i Ding,i Friver, s Tdw Amun Fdw FWT (F26)

Collective dose from drinking water for river of size, s S E ( water ), s ,i Equation (F26) man Sv

Activity concentration of dissolved radionuclide, i, in Cuw ,i Equation (F1) Bq/m3

unfiltered water

Dose coefficient for ingestion of radionuclide, i Ding,i Dependent on radionuclide Sv/ q [I14]

Volumetric flow rate of river of size, s Friver,s Large: 1 000 m3/s

Small: 10

Duration of abstraction Tdw 3.15 107 (rounded) S

Abstraction fraction for municipal water Amun 0.01

Fraction of municipal water that is drunk Fdw 0.01

Drinking water treatment removal factor FWT Dependent on element [B7]

 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 115

Parameter Symbol Value Units Reference

Collective dose from freshwater fish ingestion calculated using:

S E (ing,fish ), s ,i = Cfish, s ,i Ding,i Yfish, s Lriver, s (F27)

Collective dose from ingestion of freshwater fish from river S E (ing,fish ), s ,i Equation (F27) man Sv
of size, s

Activity concentration of radionuclide, i, in filtered water Cfish, s ,i Equations (F3), (F4), (F5), Bq/m3
from river of size, s (F6)

Dose coefficient for ingestion of radionuclide, i Ding,i Dependent on radionuclide Sv/Bq [I14]

Representative freshwater fish catch in river of size, s Yfish,s Large: 104 kg/km [F1]
Small: 10 3

Length of river of size, s Lriver,s Large: 5 102 km [I2]

Small: 1 102

Total collective dose from all freshwater pathways calculated using:

S E ( fw ), s ,r ,i = S E (ing,irr ), s ,r ,i + S E ( water ), s ,i + S E (ing,fish ), s ,i (F28)

Total collective dose in region, r, from discharges into river S E ( fw ), s ,r ,i Equation (F28) man Sv
of size, s

Collective dose from consumption of food (represented by S E (ing,irr ), s ,r ,i Equation (F25) man Sv
major food group, cereal) in region, r

Collective dose from drinking water for river of size, s S E ( water ), s ,i Equation (F26) man Sv

Collective dose from ingestion of freshwater fish of size, s S E (ing,fish ), s ,i Equation (F27) man Sv
SE(fish),s,i

GLOBAL DISPERSION

Collective doses from tritium, 14C, 85Kr and 129I were calculated by PC-CREAM (incorporates global dispersion models referenced in [U6])
116 UNSCEAR 2016 REPORT

Table A4. Parameter and equation list: Marine environment

Parameter Symbol Value Units Reference

MARINE DISPERSION

Sedimentation decay constant

K d , c ,i S c
s ,c ,i = (M38)
hc (1 + K d ,c ,i c )

Sedimentation decay constant for compartment, c ls,c,i Equation (M38) a1 [C2]

Marine sediment distribution factor for compartment Kd,c,i Dependent on radionuclide m3 t1 [I3]
c and radionuclide, i

Sedimentation rate in compartment, c Sc Local: 1.0 104 t m2 a1 (see section III.C of main
Regional: 1.0 105 text)

Depth of compartment, c hc Local: 10 m (see section III.C of main

Regional: 1000 text)

Suspended sediment load in compartment, c c Local: 2.0 104 t m3 (see section III.C of main
Regional: 1.0 107 text)

Flow rate constant, local to regional

Vl ,r
ll ,r = (M39)
Vl

Flow rate constant, local compartment to regional ll,r Equation (M39) a1 (see section III.C of main
compartment text)

Volumetric exchange between regional compartment Vl ,r 2 1010 m3 a1 (see section III.C of main
and global oceans text)

Volume of local compartment Vl 109 m3 (see section III.C of main

text)

Flow rate constant, regional compartment to global oceans

Vr , g
r , g = (M40)
Vr

Flow rate constant, regional compartment to global lr,g Equation (M40) a1 (see section III.C of main
oceans text)

Volumetric exchange between local and regional Vr , g 1015 m3 a1 (see section III.C of main
compartments text)

Volume of regional compartment Vl 1015 m3 (see section III.C of main

text)

Combined decay constant, local to regional

l ,i = li + ls ,l ,i + ll ,r (M41)

Combined decay constant for local compartment and l ,i Equation (M41) a1 [C2]
radionuclide, i
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 117

Parameter Symbol Value Units Reference

Radioactive decay constant for radionuclide, i li Dependent on radionuclide a1 [I9]

Sedimentation decay constant for local compartment ls,l,i Equation (M38) a1 [C2]

Flow rate constant, local compartment to regional ll,r Equation (M39) a1 (see section III.C of main
compartment text)

Combined decay constant, regional compartment to global oceans

r ,i =
i + s , r ,i + r , g
(M42)

Combined decay constant for regional compartment r ,i Equation (M42) a1 [C2]

and radionuclide, i

Radioactive decay constant for radionuclide, i li Dependent on radionuclide a1 [I9]

Sedimentation decay constant for regional ls,r,i Equation (M38) a1 [C2]

compartment

Flow rate constant, regional compartment to global lr,g Equation (M40) a1 (see section III.C of main
oceans text)

Activity in local compartment

l ,i ( t )
A=
Q
l ,r
( t
1 e l ,r ) (M43)

Activity of radionuclide, i, in local compartment at Al ,i ( t ) Equation (M43) Bq [C2]

time, t

Discharge rate Q 3.15 107 (rounded) Bq a1 = 1 Bq s1

Combined decay constant for local compartment and l ,r Equation (M41) a1 [C2]
radionuclide, i

Duration of discharge t 100 a

Activity in regional compartment

ll ,r Al ,i ( t )
Ar ,i ( t )
=
l ,r
(1 e l ,r t
) (M44)

Activity of radionuclide, i, in regional compartment at Ar ,i ( t ) Equation (M44) Bq [C2]

time, t

Rate constant for movement of water from local ll,r 20 a1 (see section III.C of main
compartment to regional compartment text)

Activity of radionuclide, i, in local compartment at Al ,i ( t ) Equation (M43) Bq [C2]

time, t

Combined decay constant for regional compartment l ,r Equation (M41) a1 [C2]

and radionuclide, i

Duration of discharge t 100 a

118 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Activity concentration in water

Ac ,i ( t )
Cc ,i ( t ) = (M45)
Vc

Activity concentration of radionuclide, i, in water in Cc ,i ( t ) Equation (M45) Bq m3

compartment, c, at time, t

Activity of radionuclide, i, in compartment, c, at time, Ac ,i ( t ) Equations (M43), (M44) Bq [C2]

t

Volume of compartment, c Vc Local: 1.0 109 m3 (see section III.C of main

Regional: 1.0 10 15 text)

ENVIRONMENTAL TRANSFERS

Activity concentrations in marine foods

Cc ,i B f ,i
C f , c ,i = (M46)
1 000

Activity concentration of radionuclide, i, in marine Cf,c,i Equation (M46) Bq kg1

food, f, in compartment, c

Activity concentration of radionuclide, i, in water at Cc,i Equation (M45) Bq m3

time, t

Bioaccumulation factor for marine food, f, Bf,i Dependent on food and L kg1 [I3]
concentration in marine food (fresh weight) divided element
by concentration in seawater

Conversion factor 1 000 L m3

Activity concentration in beach sediment

Cbeach,i = 0.1 tsed sed K d ,local,i Clocal,i (M47)

Activity concentration per unit area in beach Cbeach,i Equation (M47) Bq m2

sediment

Kd adjustment factor for coarser grains (i.e. sand 0.1 [S7]

rather than fine bed sediment)

Average thickness of beach sediment tsed 0.05 m [I2]

Average density of beach sediment rsed 1.2 t m3 [I2]

Marine sediment distribution factor in local Kd,local,i Dependent on radionuclide m3 t1 [I3]

compartment for radionuclide, i (L kg1)

Activity concentration of radionuclide, i, in water Clocal,i Equation (M45) Bq m3

INDIVIDUAL DOSE CALCULATIONS

External dose from beach occupancy

H E ( ex,beach ),i = Cbeach,i Fgeom Dex,deposit,i Obeach (M48)
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 119

Parameter Symbol Value Units Reference

Effective dose from radionuclide, i, from beach H E ( ex,beach ),i Equation (M48) Sv
occupancy

Activity concentration of radionuclide, i, in beach Cbeach,i Equation (M47) Bq m2

sediment

Geometry factor for beaches for external dose Fgeom 0.5 [E2]
coefficients

Dose coefficient for irradiation from surface deposits Dex,deposit,i Dependent on radionuclide Sv s1 Bq1 [E2]
m2

Beach occupancy (typical rather than critical) Obeach 3.6 105 s [R1]

Ingestion of marine food

H E (ing,marine ), f ,r=
,i (C f ,local,i f f ,local + C f ,regional,i f f ,regional ) Ding,i I f ,r (M49)

Annual effective dose from radionuclide, i, from H E (ing,marine ), f ,r Equation (M49) Sv

ingestion of freshwater fish in region, r

Average activity concentration of radionuclide, i, in C f ,local,i Equation (M46) Bq kg1

food, f, in the local marine compartment

Fraction of food, f, that is caught in the local marine f f ,local Fish: 0.25 Fish: consistent with
compartment Crustaceans: 1.0 freshwater and
terrestrial
Molluscs: 1.0
Crustaceans and
molluscs: [S7]

Average activity concentration of radionuclide, i, in C f ,regional,i Equation (M46) Bq kg1

food, f, in the regional marine compartment

Fraction of food, f, that is caught in the regional f f ,regional Fish: 0.75 Fish: consistent with
marine compartment Crustaceans: 0.0 freshwater and
terrestrial
Molluscs: 0.0 environments
Crustaceans & molluscs:
[S7]

Adult dose coefficient for ingestion Ding,i Dependent on radionuclide Sv Bq1 [I14]

Amount of marine food, f, ingested in a year in If,r Dependent on region kg [W2]

region, r

Total effective dose from ingestion of marine foods

H E (ing,marine ),r ,i = H E (ing,marine), f ,r ,i (M50)
f

Total effective dose in region, r, from ingestion of H E (ing,marine ),r ,i Equation (M50) Sv
marine food containing radionuclide, i

Effective dose from the ingestion of marine food f H E (ing,marine ), f ,r Equation (M49) Sv
containing radionuclide, i, in region, r
120 UNSCEAR 2016 REPORT

Parameter Symbol Value Units Reference

Total individual dose from marine pathways

H=
E ( marine ), r ,i
H E ( ex,beach ),i + H E (ing,marine ),r ,i (M51)

Total individual dose from marine pathways in region, H E ( marine),r ,i Equation (M51) Sv
r

Effective dose from beach occupancy H E ( ex,beach ),i Equation (M48) Sv

Effective dose from the ingestion of radionuclide, i, in H E (ing,marine ),r ,i Equation (M50) Sv
marine foods in region, r

COLLECTIVE DOSE CALCULATIONS

Collective dose from ingestion of marine fish

S E (ing,marine ),fish,c ,=
r ,i
Cfish,c ,i Yfish,r Fed,fish Ac Ding,i (M52)

Collective dose from ingestion of marine fish for S E (ing,marine ),fish, Equation (M52) man Sv
radionuclide, i

Activity concentration of radionuclide, i, in fish in Cfish,c ,i Equation (M46) Bq kg1

compartment, c

Fish catch in 1 year in region, r Yfish,r Dependent on region kg km2 [F2]

Edible fraction of catch Fed,fish 0.5 [S6]

Area of compartment, c Ac Local: 1 102 km2 (see section III.C of main

Regional: 1 10 6 text)

Adults dose coefficient for ingestion Ding,i Dependent on radionuclide Sv Bq1 [I14]

Collective dose from ingestion of shellfish

S E (ing,marine ), f ,c ,r ,=i C f ,c ,i Y f ,r Fed, f Lc Ding,i (M53)

Collective dose from ingestion of marine food, f, S E (ing,marine ), f ,c ,r Equation (M53) man Sv
(where f {molluscs,crustaceans} ) in
compartment, c, for radionuclide, i

Activity concentration of radionuclide, i, in marine C f , c ,i Equation (M46) Bq kg1

food, f, in compartment, c

Harvest of marine food, f, in 1 year per unit length of Y f ,r Dependent on region kg km1 [F2]
coastline in region, r

Edible fraction of catch for marine food, f Fed, f Crustaceans: 0.35 [S6]
Molluscs: 0.15

Length of coastline in compartment, c Lc Local: 10 km (see section III.C of main

Regional: 1 000 text)

Adults dose coefficient for ingestion Ding,i Dependent on radionuclide Sv Bq1 [I14]
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 121

Parameter Symbol Value Units Reference

Total collective dose from all marine pathways

S E ( marine ),c ,r ,i = S E (ing,marine ), f ,c ,r ,i (M54)
f

Total collective dose in compartment, c S E ( marine ),c ,r ,i Equation (M54) man Sv

Collective dose from consumption of food, f, in S E (ing,marine ), f ,c ,r Equations (M52) and (M53) man Sv
compartment, c

GLOBAL DISPERSION

Collective doses from tritium, 14C, 85Kr and 129I calculated by PC-CREAM (incorporates global dispersion models referenced in UNSCEAR 2000
Report [U6])
122 UNSCEAR 2016 REPORT

Table A5. Treatment of progeny for atmospheric and terrestrial pathways

Half-lives are given in brackets to two significant figures [I9]

Discharged Progeny Exposure pathway

parent
Plume (external irradiation Deposited material Food (ingestion)
and inhalation) (external irradiation)

90
Sr (29 a) 90
Y (64 h) Dose coefficient for Deposition and dose Dose coefficient for ingestion of
inhalation of parent rates calculated for parent accounts for progeny
accounts for progeny. parent and progeny
Activity concentration of separately
progeny assumed equal to
parent for estimating
external exposure

106
Ru (370 d) 106
Rh (30 s) Secular equilibrium Deposition and dose Dose coefficient for ingestion of
assumed. Dose rates rates calculated for parent accounts for progeny
calculated for both parent parent and progeny
and progeny separately

135
Xe (9.1h) 135
Cs (2.3 106 a) Activity concentrations of Deposition and dose rate Activity concentration calculated from
progeny in air calculated calculated for progeny deposition rate for progeny
explicitly based on transit
time

138
Xe (14 m) 138
Cs (32 min) Activity concentration of Deposition and dose rate Activity concentration calculated from
progeny in air calculated calculated for progeny deposition rate for progeny
explicitly based on transit
time

137
Cs (30 a) 137m
Ba (2.6 min) Dose coefficient for Deposition and dose Dose coefficient for ingestion of
inhalation of parent rates calculated for parent accounts for progeny
accounts for progeny. parent and progeny
Activity concentration of separately
progeny calculated at yield
of 94.6% for estimating
external exposure

210
Pb (22 a) 210
Bi (5.0 d) Ingrowth insignificant Negligible dose from Ingrowth following deposition of 210Pb
210
Po (140 d) during plume transit; progeny following to soil and plant, but activity
progeny not considered ingrowth after concentration of 210Po in terrestrial
deposition of 210Pb; not foods significantly lower than for
considered deposition of 210Po; subsequent
small doses compared to those from
210
Pb; ingrowth not considered

222
Rn (3.8 d) 218
Po (3.1 min) Dose coefficient for Negligible contributions Because of time taken for ingrowth
214
Pb (27 min) inhalation of 222Rn includes to dose from and reduced activities of long-lived
contribution from four deposition of progeny progeny compared to 222Rn,
214
Bi (20 min) immediate short-lived compared with dose negligible doses from ingestion of
214
Po (164 s) progeny [U9]. Longer lived from inhalation; not progeny in terrestrial foods; not
progeny (210Pb, 210Bi and considered considered
210
Pb (22 a) 210
Po) will not grow in
210
Bi (5.0 d) during plume transit and
210
Po (140 d) so not considered
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 123

Discharged Progeny Exposure pathway

parent
Plume (external irradiation Deposited material Food (ingestion)
and inhalation) (external irradiation)

226
Ra (1 600 a) 222
Rn (3.8 d) Ingrowth insignificant Ingrowth following Ingrowth following deposition of
218
Po (3.1 min) during plume transit; deposition of 226Ra, but 226
Ra, but radon gas largely emitted to
progeny not considered radon gas largely atmosphere and further diluted. Small
214
Pb (27 min) emitted to atmosphere doses from progeny compared with
214
Bi (20 min) and further diluted. parent; not included
Small doses from
214
Po (164 s)
progeny compared
210
Pb (22a) with parent; not
210
Bi (5.0 d) included
210
Po (140 d)

232
Th (1.4 1010 a) 228
Ra (5.8 a) Secular equilibrium Deposition rate Activity concentration of parent in air
228
Ac (6.1 h) assumed; activity calculated for each used as proxy for secular equilibrium
concentration in air progeny in secular in soil after 100 years. Only 228Ra,
228
Th (1.9 a) assumed equal to parent equilibrium 228
Th and 212Pb included
212
Pb (11 h)

234
U (2.4 105 a) 230
Th (7.7 104 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

238
U (4.5 109 a) 234
Th (24 d) Secular equilibrium Deposition rate Activity concentration of parent in air
234m
Pa (1.2 min) assumed; activity calculated for each used as proxy for secular equilibrium
concentration in air progeny in soil after 100 years. Only 234Th
assumed equal to parent included

239
Pu (2.4 104 a) 235
U (7.0 108 a)

240
Pu (6 500 a) 236
U (2.3 107 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

241
Am (430 a) 237
Np (2.1 106 a)
 124 UNSCEAR 2016 REPORT

Table A6. Treatment of progeny for freshwater pathways

Half-lives are given in brackets to two significant figures [I9]

Discharged parent Progeny Exposure pathway

Riverbank occupancy Freshwater fish Drinking water Irrigated food

90
Sr (29 a) 90
Y (64 h) Secular equilibrium
assumeda

106
Ru (370 d) 106
Rh (30 s) Secular equilibrium
assumeda
Dose coefficient for ingestion of parent accounts for progeny
137
Cs (30 a) 137m
Ba (2.6 min) 137m
Ba activity
concentration in filtered
water calculated at yield
of 94.6% a

210
Pb (22 a) 210
Bi (5.0 d) Negligible doses from
210
Po (140 d) ingrowth compared with
total dose for 210Pb; not
considered

226
Ra (1 600 a) 222
Rn (3.8 d) 222
Rn gas mainly emitted to
218
Po (3.1 min) atmosphere; Ingrowth limited during transit time in freshwater body;
contributions from doses from progeny not included
214
Pb (27 min)
subsequent progeny
214
Bi (20 min)
omitted
214
Po (164 s)
210
Pb (22 a)
210
Bi (5.0 d)
210
Po (140 d)
232
Th (1.4 1010 a) 228
Ra (5.8 a) Secular equilibrium Secular equilibrium Secular equilibrium Deposition rate of
228
Ac (6.1 h) assumeda assumed in water a assumed in parent used as
228
Th (1.9 a) untreated water a proxy for secular
equilibrium in soil
212
Pb (11 h)
after 100 years.
Only 228Ra, 228Th
and 212Pb included

234
U (2.4 105 a) 230
Th (7.7 104 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

238
U (4.5 109 a) 234
Th (24 d) Secular equilibrium Secular equilibrium Secular equilibrium Deposition rate of
234m
Pa (1.2 min) assumeda assumed in water a assumed in parent used as
untreated water a proxy for secular
equilibrium in soil
after 100 years.
Only 234Th included

239
Pu (2.4 104 a) 235
U (7.0 108 a)

240
Pu (6 500 a) 236
U (2.3 107 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

241
Am (430 a) 237
Np (2.1 106 a)

a
Dose rates calculated from parent and each progeny separately.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 125

Table A7. Treatment of progeny for marine pathways

Half-lives are given in brackets to two significant figures [I9]

Discharged Progeny Exposure pathway

parent
Beach occupancy Marine food

90
Sr (29 a) 90
Y (64 h)
Secular equilibrium assumeda
106
Ru (370 d) 106
Rh (30 s) Dose coefficient for ingestion of parent accounts for
progeny
137
Cs (30 a) 137m
Ba (2.6 min) Ba activity in local/regional compartment
137m

and on beach calculated at yield of 94.6%a

210
Pb (22 a) 210
Bi (5.0 d) Secular equilibrium assumeda Secular equilibrium assumed in water;
210
Po (140 d) bioaccumulation factors for parent and progeny
explicitly applied

226
Ra (1 600 a) 222
Rn (3.8 d) Most radon is assumed not emitted to Secular equilibrium assumed in water;
218
Po (3.1 min) atmosphere; secular equilibrium assumed a bioaccumulation factors for parent and progeny
explicitly applied
214
Pb (27 min)
214
Bi (20 m)
214
Po (164 s)
210
Pb (22 a)
210
Bi (5.0 d)
210
Po (140 d)

232
Th 228
Ra (5.8 a) Secular equilibrium assumed a Secular equilibrium assumed in water;
(1.4 1010 a) 228
Ac (6.1 h) bioaccumulation factors for parent and progeny
explicitly applied
228
Th (1.9 a)
212
Pb (11 h)

234
U (2.4 105 a) 230
Th (7.7 104 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

238
U (4.5 109 a) 234
Th (24 d) Secular equilibrium assumed a Secular equilibrium assumed in water;
234m
Pa (1.2 min) bioaccumulation factors for parent and progeny
explicitly applied

239
Pu (2.4 104 a) 235
U (7.0 108 a)

240
Pu (6 500 a) 236
U (2.3 107 a) Parent half-life sufficiently long that ingrowth not relevant at 100 years

241
Am (430 a) 237
Np (2.1 106 a)

a
Dose rates from external exposure calculated from parent and each progeny separately.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 127

REFERENCES
A1 Andersson, K.G. On the current needs in European decision support tools for contaminated
areas. Radioprotection 48(5): S57-S64 (2013).

A2 Apostoaei, A.I., S.K. Nair, B.A. Thomas et al. External exposure to radionuclides accumulated
in shoreline sediments with an application to the lower Clinch River. Health Phys 78(6): 700-
710 (2000).

B1 Beck, H.L. Exposure rate conversion factors for radionuclides deposited on the ground. EML-
378. Environmental Measurements Laboratory, Department of Energy, New York, 1980.

B2 Bergman, R., U. Bergstrm and S. Evans. Environmental transport and long-term exposure for
tritium released in the biosphere. pp.535-554 in: Proceedings of the International Symposium
on the Behaviour of Tritium in the Environment. International Atomic Energy Agency,
Vienna, 1979.

B3 Bertrand, M. Consommations et lieux dachats des produits alimentaires en 1991. INSEE

rsultats. Consommation-Mode de vie n54-55. Institut national de la statistique et des tudes
conomiques (INSEE), Paris, 1993. (French).

B4 Booth, R.S. A systems analysis model for calculation of radionuclide transport between
receiving waters and bottom sediments. pp.133-164 in: Environmental Toxicity of Aquatic
Radionuclides: Models and Mechanisms (M.W. Miller and J.S. Stannard, eds.). Ann Arbor
Science Publishers Inc, 1976.

B5 Brewer, B. Irrigation expenses threaten vineyards. Contra Costa Times, 21 January, 2001.

B6 Brown, J. and J.R. Simmonds. FARMLAND a dynamic model for the transfer of radionuclides
through terrestrial foodchains. NRPB-R273. National Radiological Protection Board, Chilton,
Didcot, 1995.

B7 Brown, J., D. Hammond and B.T. Wilkins. Handbook for assessing the impact of a
radiological incident on levels of radioactivity in drinking water and risks to operatives at
water treatment works. HPA-RPD-040. Health Protection Agency, Oxford, 2008.

B8 Burson, Z.G. and A.E. Profio. Structure shielding in reactor accidents. Health Phys 33(4): 287-
299 (1977).

C1 Calmon, P., Y. Thiry, G. Zibold et al. Radionuclide transfer in forest ecosystems. pp.333-380
in: Quantification of Radionuclide Transfer in Terrestrial and Freshwater Environments for
Radiological Assessments. IAEA-TECDOC-1616. International Atomic Energy Agency,
Vienna, 2009.

C2 CEC. Methodology for evaluating the radiological consequences of radioactive effluents

released in normal operations. National Radiological Protection Board and Commissariat
lEnergie Atomique. Document no. V/3865/79. Commission of the European Communities,
Luxembourg, 1979.

C3 CEC. Commission Recommendation of 18 December 2003 on standardised information on

radioactive airborne and liquid discharges into the environment from nuclear power reactors
and reprocessing plants in normal operation (2004/2/Euratom). Official Journal of the
European Union L2. Commission of the European Communities, Luxembourg, 2004.
128 UNSCEAR 2016 REPORT

C4 Ciffroy, P., F. Siclet, C. Damois et al. A dynamic model for assessing radiological
consequences of routine releases in the Loire river: parameterisation and
uncertainty/sensitivity analysis. J Environ Radioact 83(1): 9-48 (2005).

C5 Crawford, T.V., C.W. Miller and A.H. Weber. Atmospheric transport of radionuclides. pp.79-
140 in: Radiological Risk Assessment and Environmental Analysis (J.E. Till and H.A. Grogan,
eds.). Oxford University Press, Inc., New York, 2008.

C6 CSA. Guidelines for calculating derived release limits for radioactive material in airborne and
liquid effluents for normal operation of nuclear facilities. CSA N288.1. Canadian Standards
Association, Canada, 2014.

E1 Eaton, D. and C.E. Murphy Jr. Tritium uptake by fish in a small stream. WSRC-TR--92-193
Rev.1. Westinghouse Savannah River Co., Aiken, SC, 1992.

E2 Eckerman, K.F. and J.C. Ryman. External exposure to radionuclides in air, water, and soil.
EPA-402-R-93-081. Federal Guidance Report No. 12. U.S. Environmental Protection Agency,
Washington, D.C., 1993.

E3 Encyclopaedia Britannica. Paran River, South America. [Internet] Available from

(https://www.britannica.com/place/Parana-River) on 06.12.2016.

F1 FAO. FAO statistical yearbook 2010. Food and Agriculture Organization of the United
Nations. [Internet] Available from (http://www.fao.org/economic/ess/ess-publications/ess-
yearbook/ess-yearbook2010/en/) on 10.10.2015.

F2 FAO. FishStatJ - software for fishery statistical time series. Food and Agriculture Organization
of the United Nations. [Internet] Available from (http://www.fao.org/
fishery/statistics/software/fishstatj/en) on 08.05.2012.

F3 FAO. FAO/Stat. Food and Agriculture Organization of the United Nations. [Internet]
Available from (http://faostat3.fao.org/download/FB/BL/E) on 06.10. 2014.

F4 FAO. AQUASTAT - FAO's information system on water and agriculture. Food and
Agriculture Organization of the United Nations. [Internet] Available from
(http://www.fao.org/nr/water/aquastat/main/index.stm) on 09.07.2014.

G1 Geyh, A.S., J. Xue, H. Ozkaynak et al. The Harvard Southern California Chronic Ozone
Exposure Study: assessing ozone exposure of grade-school-age children in two Southern
California communities. Environ Health Perspect 108(3): 265-270 (2000).

G2 Golikov, V.Y., M.I. Balonov and P. Jacob. External exposure of the population living in areas
of Russia contaminated due to the Chernobyl accident. Radiat Environ Biophys 41(3): 185-
193 (2002).

H1 Hamza, W. The Nile fishes and fisheries, Chapter 14. pp.349-370 in: Biodiversity - the
Dynamic Balance of the Planet (O. Grillo, ed.) InTech, 2014.

H2 Heimann, D.C., L.A. Sprague and D.W. Blevins. Trends in suspended-sediment loads and
concentrations in the Mississippi river basin, 19502009. Scientific Investigations Report
20115200. National Water-Quality Assessment Program, U.S. Geological Survey, Virginia,
U.S.A., 2011.

H3 Hejduk, A. and K. Banasik. Suspended sediment concentration and yield in snowmelt flood
events in a small lowland river. Ann Warsaw Univ of Life Sci SGGW, Land Reclam 42(1):
61-68 (2010).
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 129

H4 Hilton, J., S. Small, D. Hornby et al. Modelling the combined impact of radionuclide
discharges reaching rivers. EA-RD-TR--P3-068/TR. https://inis.iaea.org/search/
search.aspx?orig_q=RN:34037229, 2003.

I1 IAEA. Handbook of parameter values for the prediction of radionuclide transfer in temperate
environments. Technical Reports Series No. 364. International Atomic Energy Agency,
Vienna, 1994.

I2 IAEA. Generic models for use in assessing the impact of discharges of radioactive substances
to the environment. Safety Reports Series No.19. International Atomic Energy Agency,
Vienna, 2001.

I3 IAEA. Sediment distribution coefficients and concentration factors for biota in the marine
environment. Technical Reports Series No. 422. International Atomic Energy Agency, Vienna,
2004.

I4 IAEA. Quantification of radionuclide transfer in terrestrial and freshwater environments for

radiological assessments. IAEA-TECDOC-1616. International Atomic Energy Agency,
Vienna, 2009.

I5 IAEA. Handbook of parameter values for the prediction of radionuclide transfer in terrestrial
and freshwater environments. Technical Reports Series No. 472. International Atomic Energy
Agency, Vienna, 2010.

I6 IAEA. Performance of models in radiological impact assessment for normal operation. IAEA-
TECDOC-1808. International Atomic Energy Agency, Vienna, 2017.

I7 ICID. International Commission on Irrigation and Drainage. [Internet] Available from

(http://www.icid.org/sprin_micro_11.pdf.) on 06.11.2014.

I8 ICRP. Limits for intakes of radionuclides by workers. ICRP Publication 30 (Part 1). Annals of
the ICRP 2(3-4). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1979.

I9 ICRP. Radionuclide transformations - energy and intensity of emissions. ICRP Publication 38.
Annals of the ICRP 11-13. International Commission on Radiological Protection, Pergamon
Press, Oxford, 1983.

I10 ICRP. 1990 Recommendations of the International Commission on Radiological Protection.

ICRP Publication 60. Annals of the ICRP 21(1-3). International Commission on Radiological
Protection, Pergamon Press, Oxford, 1991.

I11 ICRP. Human respiratory tract model for radiological protection. ICRP Publication 66. Annals
of the ICRP 24(1-3). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1994.

I12 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 5.
Compilation of ingestion and inhalation dose coefficients. ICRP Publication 72. Annals of the
ICRP 26(1). International Commission on Radiological Protection, Pergamon Press, Oxford, 1995.

I13 ICRP. The 2007 Recommendations of the International Commission on Radiological

Protection. ICRP Publication 103. Annals of the ICRP 37(2-4). International Commission on
Radiological Protection, Elsevier Ltd., 2007.
130 UNSCEAR 2016 REPORT

I14 ICRP. Compendium of dose coefficients based on ICRP Publication 60. ICRP Publication
119. Annals of the ICRP 41(Suppl. 1). International Commission on Radiological Protection,
Elsevier Ltd., Oxford, 2012.

I15 ICRU. Gamma-ray spectrometry in the environment. ICRU Report 53. International
Commission on Radiation Units and Measurements, Bethesda, MA, 1994.

J1 Jacob, P., H. Rosenbaum, N. Petoussi et al. Calculation of organ doses from environmental
gamma rays using human phantoms and Monte Carlo methods. Part II: Radionuclides
distributed in the air or deposited on the ground. GSF Report 12/90. National Research Centre
for Environment and Health, Neuherberg, Germany, 1990.

J2 Jacob, P. and R. Meckbach. External exposure from deposited radionuclides. pp.407-422 in:
Proceedings of the Seminar on Methods and Codes for Assessing the Off-site Consequences of
Nuclear Accidents. Volume I. EUR 13013/1 EN. Commission of the European Communities,
Luxembourg, 1991.

J3 Jacob, P., G. Prhl, I. Likhtarev et al. Pathway analysis and dose distributions. EUR 16541
EN. European Commission, Brussels, 1996.

J4 Jeon, W.-T., S.-O. Hur, K.-Y. Seong et al. Effect of green manure hairy vetch on rice growth
and saving of irrigation water. Korean J Soil Sci Fert 44(2): 181-186 (2011).

J5 Jones, K.A., C. Walsh, A. Bexon et al. Guidance on the assessment of radiation doses to
members of the public due to the operation of nuclear installations under normal conditions.
HPA-RPD-019. Health Protection Agency, Chilton, 2006.

J6 Jones, K.A., J.G. Smith, T. Anderson et al. Implied doses to the population of the EU arising
from reported discharges from EU nuclear power stations and reprocessing sites in the years
2004 to 2008. Radiation Protection No. 176. European Commission, Luxembourg, 2013.

K1 Kane, P. ECOS: sorption, dose, consumption and miscellaneous data values. Associated
Nuclear Services, Epsom UK, 1984.

K2 Kelly, G.N., J.A. Jones and P.M. Bryant. The predicted radiation exposure of the population of
the European Community resulting from discharges of krypton-85, tritium, carbon-14 and
iodine-129 from the nuclear power industry to the year 2000. Document V/2676/75.
Commission of the European Communities, Luxembourg, 1975.

K3 Killough, G.G. and P.S. Rohwer. A new look at the dosimetry of 14C released to the
atmosphere as carbon dioxide. Health Phys 34(2): 141-159 (1978).

K4 Killough, G.G. and D.C. Kocher. Global environmental transport models for tritium. CONF-
850405-7. Oak Ridge National Laboratory, 1985.

K5 Kim, S.B., W.J.G. Workman, P.A. Davis et al. Tritium concentrations in the Perch Lake
ecosystem. AECL document CTD-03700-ENA-003. Chalk River Laboratories, Ontario, 2004.

K6 Kocher, D.C. Effects of indoor residence on radiation doses from routine releases of
radionuclides to the atmosphere. Nucl Technol 48(2): 171-179 (1980).

L1 Le Grand, J., Y. Roux, R. Meckbach et al. External exposure from airborne radionuclides.
pp.385-406 in: Proceedings of the Seminar on Methods and Codes for Assessing the Off-site
Consequences of Nuclear Accidents. Volume I. EUR 13013/1 EN. Commission of the
European Communities, Luxembourg, 1991.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 131

L2 Likhtarev, I.A., L.N. Kovgan, P. Jacob et al. Chernobyl accident: retrospective and prospective
estimates of external dose of the population of Ukraine. Health Phys 82(3): 290-303 (2002).

M1 Maringer, F.J. Four decades of radioecological research on the Danube river: a reliable basis
for protecting the freshwater resources of central Europe of tomorrow. pp.P-4b-259 in:
IRPA-10. Proceedings of the 10th International Congress of the International Radiation
Protection Association on Harmonization of Radiation, Human Life and the Ecosystem. Japan
Health Physics Society, Japan, 2000.

M2 Milliman, J.D. and K.L. Farnsworth. River Discharge to the Coastal Ocean: a Global
Synthesis. Cambridge University Press, 2011.

N1 NCRP. Tritium in the environment. NCRP Report No. 62. National Council on Radiation
Protection and Measurements, Bethesda, 1979.

N2 NCRP. Screening models for releases of radionuclides to atmosphere, surface water, and
ground. NCRP Report No. 123. National Council on Radiation Protection and Measurements,
Bethesda, 1996.

N3 NCRP. Ionizing radiation exposure of the population of the United States. NCRP Report
No. 160. National Council on Radiation Protection and Measurements, Bethesda, 2009.

N4 NEA. Radiological impacts of spent nuclear fuel management options: a comparative study.
Nuclear Energy Agency, NEA/OECD, Paris, 2000.

N5 Nikitin, A.I., V.B. Chumichev, N.K. Valetova et al. The current content of artificial
radionuclides in the water of the Tobol-Irtysh river system (from the mouth of the Iset River to
the confluence with the Ob River). J Environ Radioact 96(1-3): 138-143 (2007).

N6 NRC. Final environmental statement on the transportation of radioactive materials by air and
other modes. NUREG-0170. Volume 1. Nuclear Regulatory Commission, United States, 1977.

O1 Onishi, Y. Suspended sediment concentrations for large and small rivers. Official
communication to the UNSCEAR secretariat. 2013.

O2 Ott, W.R. Human activity patterns: a review of the literature for estimating time spent indoors,
outdoors, and in transit. Proceedings of the Research Planning Conference on Human Activity
Patterns. EPA/600/4-89/004. EPA National Exposure Research Laboratory, Las Vegas, 1989.

P1 Pawar, D.H. and M.S. Palaskar. Irrigation scheduling based on soil-crop-climate integrated
approach: A case study of Upper Pengana Project in Maharashtra. Proceedings of the 4th
International R&D Conference on Water and Energy for 21st Century, Aurangabad,
Maharashtra, India, Volume 1. 2003.

P2 Petoussi-Henss, N., H. Schlattl, M. Zankl et al. Organ doses from environmental exposures
calculated using voxel phantoms of adults and children. Phys Med Biol 57(18): 5679-5713 (2012).

R1 Rommens, C., D. Laurier and A. Sugier. Methodology and results of the Nord-Cotentin
radioecological study. J Radiol Prot 20(4): 361-380 (2000).

S1 Schauer, D.A. and O.W. Linton. NCRP Report No. 160, Ionizing Radiation Exposure of the
Population of the United States, medical exposure--are we doing less with more, and is there a
role for health physicists? Health Phys 97(1): 1-5 (2009).

S2 Schimmack, W., H. Steindl and K. Bunzl. Variability of water content and of depth profiles of
global fallout 137Cs in grassland soils and the resulting external gamma-dose rates. Radiat
Environ Biophys 37(1): 27-33 (1998).
132 UNSCEAR 2016 REPORT

S3 Shahul Hameed, P., K. Shaheed and S.S.N. Somasundaram. A study on distribution of natural
radionuclide polonium-210 in a pond ecosystem. J. Biosciences 22(5): 627-634 (1997).

S4 Sheppard, S.C., P. Ciffroy, F. Siclet et al. Conceptual approaches for the development of
dynamic specific activity models of 14C transfer from surface water to humans. J Environ
Radioact 87(1): 32-51 (2006).

S5 Sheppard, S.C., M.I. Sheppard and F. Siclet. Parameterization of a dynamic specific activity
model of 14C transfer from surface water-to-humans. J Environ Radioact 87(1): 15-31 (2006).

S6 Simmonds, J.R., G. Lawson and A. Mayall. Methodology for assessing the radiological
consequences of routine releases of radionuclides to the environment. EUR 15760. European
Commission, Luxembourg, 1995.

S7 Simmonds, J.R., A.P. Bexon, S. Lepicard et al. Annex D: Radiological impact on EU member
states of radioactivity in north european waters. In: MARINA II. Update of the MARINA Project
on the Radiological Exposure of the European Community from Radioactivity in North
European Marine Waters. Radiation Protection 132. European Commission, Luxembourg, 2002.

S8 Smith, J.G. and J.R. Simmonds. The methodology for assessing the radiological consequences
of routine releases of radionuclides to the environment used in PC-CREAM 08. HPA-RPD-
058. Health Protection Agency, Chilton, Didcot, 2009.

S9 Smith, K.R., A.P. Bexon, K. Sihra et al. Guidance on the calculation, presentation and use of
collective doses for routine discharges. Radiation Protection 144. European Commission,
Luxembourg, 2007.

S10 Spraque, L.A., M.L. Clark, D.L. Rus et al. Nutrient and suspended-sediment in the Missouri
river basin, 1993-2003. Scientific Investigation Report 2006-5231. National Water-Quality
Assessment Program. U.S. Geological Survey, Virginia, 2007.

S11 Sudan. Data for 2012 (internal report). Statistical Bulletin for Animal Resources, Issue no.
21-22. Ministry of Animal Resources and Fisheries, Sudan, 2011.

T1 Titley, J.G., T. Cabianca and G. Lawson. Improved global dispersion models for iodine-129
and carbon-14. EUR 15880 EN. European Commission, Luxembourg, 1995.

U1 UN. Population division, population estimates and projections section. United Nations.
[Internet] Available from (http://esa.un.org/unpd/ppp/Figures-Output/Population/PPP_Total-
Population.htm) on 01.12.2014.

U2 UN. Demographic yearbook. United Nations. [Internet] Available from

(http://unstats.un.org/unsd/demographic/products/dyb/dyb2.htm) on 01.12.2014.

U3 UNEP. GEO-3 regions. United Nations Environment Programme. [Internet] Available from
(http://geodata.grid.unep.ch/images/regions.jpg) on 16.02.2015.

U4 UNSCEAR. Ionizing Radiation: Sources and Biological Effects. UNSCEAR 1982 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1982 Report to the
General Assembly, with annexes. United Nations sales publication E.82.IX.8. United Nations,
New York, 1982.

U5 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. UNSCEAR 1988 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1988 Report to the
General Assembly, with annexes. United Nations sales publication E.88.IX.7. United Nations,
New York, 1988.
 ANNEX A: METHODOLOGY FOR ESTIMATING PUBLIC EXPOSURES DUE TO RADIOACTIVE DISCHARGES 133

U6 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.3.
United Nations, New York, 2000.

U7 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume II: Effects. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.4.
United Nations, New York, 2000.

U8 UNSCEAR. Effects of Ionizing Radiation. Volume I: Report to the General Assembly,

Scientific Annexes A and B. UNSCEAR 2006 Report. United Nations Scientific Committee
on the Effects of Atomic Radiation. United Nations sales publication E.08.IX.6. United
Nations, New York, 2008.

U9 UNSCEAR. Effects of Ionizing Radiation. Volume II: Scientific Annexes C, D and E.

UNSCEAR 2006 Report. United Nations Scientific Committee on the Effects of Atomic
Radiation. United Nations sales publication E.09.IX.5. United Nations, New York, 2009.

U10 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources: Report to the
General Assembly, Scientific Annexes A and B. UNSCEAR 2008 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation. United Nations sales publication
E.10.XI.3. United Nations, New York, 2010.

U11 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume II: Effects. Scientific Annexes
C, D and E. UNSCEAR 2008 Report. United Nations Scientific Committee on the Effects of
Atomic Radiation. United Nations sales publication E.11.IX.3. United Nations, New York, 2011.

U12 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. Volume I: Report to the
General Assembly and Scientific Annex A. UNSCEAR 2013 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation. United Nations sales publication
E.14.IX.1. United Nations, New York, 2014.

U13 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. Report to the General Assembly
and Scientific Annexes A and B. UNSCEAR 2012 Report. United Nations Scientific
Committee on the Effects of Atomic Radiation. United Nations sales publication E.16.IX.1.
United Nations, New York, 2015.

U14 USDA. 1984 farm and ranch irrigation survey. AG84-SR-1 Special Report Series. United States
Department of Agriculture, USDA Bureau of Census, U.S. Department of Commerce, 1986.

V1 Vanoni, V.A. Sedimentation Engineering. Hydraulic Division, The American Society of Civil
Engineers, New York, 1975.

W1 Wahaab, R.A. and M.I. Badawy. Water quality assessment of the River Nile system: an
overview. Biomed Environ Sci 17(1): 87-100 (2004).

W2 WHO. Global Environment Monitoring System (GEMS/Food) - Food contamination

monitoring and assessment programme. World Health Organization. [Internet] Available from
(https://extranet.who.int/gemsfood/ ) on 14.09.2015.

W3 WHO. Guidelines for drinking-water quality. Fourth Edition. World Health Organization,
Geneva, 2011.

Z1 Zankl, M., N. Petoussi and G. Drexler. Effective dose and effective dose equivalent--the impact
of the new ICRP definition for external photon irradiation. Health Phys 62(5): 395-399 (1992).
 ANNEX B

RADIATION EXPOSURES FROM

ELECTRICITY GENERATION

135
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 137

CONTENTS
I. INTRODUCTION ................................................................................................................................. 139
A. Scope............................................................................................................................................. 140
II. BACKGROUND ................................................................................................................................... 141
A. Global trends in electricity generation technology ..................................................... 142
B. Electricity generation worldwide in 2010 ........................................................................ 144
III. ASSESSMENT APPROACH AND END POINTS .......................................................................... 145
A. Recapitulation of key features of the methodology for estimating
public exposures due to radioactive discharges ........................................................... 147
IV. RADIATION EXPOSURES ARISING FROM ELECTRICITY GENERATION BY NUCLEAR
POWER ......................................................................................................................................................... 151
A. Introduction................................................................................................................................ 151
B. Public exposure ......................................................................................................................... 152
C. Occupational exposure for the nuclear fuel cycle ........................................................ 174
D. Decommissioning..................................................................................................................... 178
V. RADIATION EXPOSURES ARISING FROM ELECTRICITY GENERATION BY COMBUSTION
OF FOSSIL FUELS ...................................................................................................................................... 182
A. Coal ................................................................................................................................................ 182
B. Natural gas .................................................................................................................................. 194
C. Oil ................................................................................................................................................... 197
D. Comparisons of exposures from fossil-fuel electricity generation ......................... 199
VI. RADIATION EXPOSURES ARISING FROM ELECTRICITY GENERATION FROM
GEOTHERMAL ENERGY........................................................................................................................... 199
VII. ASSESSING DOSES FROM THE CONSTRUCTION PHASE OF ELECTRICITY-GENERATING
TECHNOLOGIES ........................................................................................................................................ 201
VIII. COMPARISON OF RADIATION EXPOSURES FROM ELECTRICITY-GENERATING
TECHNOLOGIES ........................................................................................................................................ 211
A. Comparison of public exposures due to radioactive discharges from
the electricity-generating technologies based on the nuclear fuel
cycle, coal cycle, combustion of natural gas and oil, and use of
geothermal energy .................................................................................................................. 211
B. Comparison of public and occupational exposures .................................................... 220
C. Commentary on significance of accidents ...................................................................... 224
138 UNSCEAR 2016 REPORT

IX. RESEARCH NEEDS ............................................................................................................................. 225

X. CONCLUSIONS................................................................................................................................... 227
XI. ACKNOWLEDGEMENTS .................................................................................................................. 229
REFERENCES ............................................................................................................................................... 231
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 139

I. INTRODUCTION
1. A reliable and affordable supply of electricity is important to improve human health and welfare
worldwide, an objective recognized by both the United Nations Millennium Development Goals [U2,
U3] and Sustainable Development Goals. 1 However, policy makers and the general public also have
interest in the impacts of electricity generation on humankind and the environment.

2. Interest in exposure of the public and workers to radiation due to nuclear power dates back to the
earliest use of the technology. The first UNSCEAR Report to the General Assembly in 1958 included
data on exposures of contract employees of the United States Atomic Energy Commission and a
recognition of the need to keep track of the exposure of workers in view of the anticipated growth in the
use of nuclear technology and the associated worker population [U4]. Since then, the Committee has
periodically reviewed exposures of both the public and of workers, related to nuclear power [U5, U6,
U7, U8, U9, U11]. As a result, the Committees records of exposures of workers and the public from
nuclear power provide one of the most complete and accurate global pictures of radiation exposures
from any source of ionizing radiation.

3. The Committee has conducted evaluations of radiation exposures of the public arising from forms
of electricity generation other than nuclear power to a much more limited extent. Radiation exposures
from industrial activities outside of the nuclear sector are generally not monitored or reported in a
systematic manner; consequently, the assessment of these exposures has often relied on incomplete data
from isolated surveys or ad hoc collection of data from various reports and publications.

4. The Committees thematic priorities identified in its strategic plan 2 for 20092013 included
radiation levels of energy production and in its plan 3 for 20142019 the global impact of energy
production. The Committee decided to update its assessments of the exposures from electricity
generation, considering the principal relevant commercial technologies, both nuclear and non-nuclear.

5. The worlds mix of electricity-generating technologies changes over time in response to the
landscape of climatic, environmental, resource, political and economic challenges. Governments and
researchers may conduct various comparative studies that among other things take into account the
various implications for the public and the environment of the different technologies. Exposure to
ionizing radiation is only one of the many factors that such assessments may take into account.
However, the Committee considers that an update and extension of its past assessments of radiation
exposures of the public and workers from electricity generation could be a useful source of information
for such studies.

6. This scientific annex thus presents an analysis of the total population exposure, public and
occupational, to ionizing radiation from the different life cycle stages of electricity-generating
technologies, normalized to the electricity generated during one year for that technology. The annex is
also an update and extension of the Committees earlier evaluations of radiation exposures resulting
from discharges associated with different electricity-generating technologies. The Committee

1
Transforming our world: the 2030 Agenda for Sustainable Development. Resolution adopted by the General Assembly on
25 September 2015. General Assembly, Seventieth session, Agenda items 15 and 116 (A/RES/70/1).
2
Official Records of the General Assembly, Sixty-third session, Supplement No. 46 (A/63/46).
3
Official Records of the General Assembly, Sixty-eighth session, Supplement No. 46 (A/68/46).
140 UNSCEAR 2016 REPORT

emphasizes that the objective of this study was comparative in nature. Common approaches, available
data and balanced assumptions for assessing the main exposures and identifying the dominant
components of those exposures for different electricity-generating technologies have been used to the
extent possible.

7. Specifically, the following core questions have been addressed:

How do the individual and population exposures of humans to ionizing radiation that
result from different phases of the life cycle of each electricity-generating technology
compare with each other across the world and by region?

What are the main factors, in terms of principal sources, radionuclides, time periods and
exposed populations, contributing to the exposure for each technology?

What would be needed to improve such assessments in the future in terms of data and
research?

8. To conduct the assessment, the Committee has in parallel reviewed and revised its methodology
for estimating exposures of the public due to radioactive discharges (annex A). Other exposures, such
as occupational exposures, are assessed using other methods described here.

A. Scope
9. This annex describes an assessment of the exposures to ionizing radiation in the life cycle of
electricity-generating technologies that are currently deployed commercially, and fuelled by nuclear
energy, combustion of coal, oil or gas, geothermal energy, solar energy, wind or biomass. The
assessment includes exposures from activities in the life cycle of each technology that are relevant to the
radiological impact, from construction to decommissioning. For electricity-generating technologies
using solar energy, wind and biomass, the only activity that has been assessed in this context is the
mining of metals needed for construction. Exposures from electrical energy storage, transmission and
distribution are not considered. Moreover, the evaluation does not address energy generation other than
electricity generation, such as process heat generation or other means of distributing energy, such as
district heating.

10. The assessment considers normal operations only. Exposures from incidents and accidents are not
considered in detail here; the Committee has earlier reviewed radiation exposures in accidents (annex C
[U12]), and considered in detail the radiation exposures from the 1986 accident at the Chernobyl nuclear
power plant (annex D [U12]) and the 2011 accident at the Fukushima-Daiichi nuclear power station
(annex A [U13]) and continues to monitor developments after these accidents. The annex discusses their
significance only in drawing conclusions from this assessment for the General Assembly.

11. The assessment is limited to considering individual and population exposure of humans.
Exposures of non-human biota in the environment are not considered. Assessing impacts other than
radiation exposure from electricity generation, such as social benefits, economic issues, or non-
proliferation or security matters, is outside of the Committees remit and competence.

12. This work aims to compare exposures to ionizing radiation for different electricity-generating
technologies where individual and collective doses may be used for performing a comparative exposure
assessment. However, calculated doses are recommended only for comparative purposes and not for
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 141

estimations related to health effects. It is important to state that collective doses, as used here, are solely
an instrument to compare radiation exposures for different technologies used for electricity generation.
Collective dose is not intended as a tool for epidemiological risk assessment. Moreover, the aggregation
of very low individual doses over extended time periods is inappropriate for use in risk projections and,
in particular, the calculation of numbers of cancer deaths from collective doses based on individual
doses that are well within the variation in background exposure should be avoided. Collective doses
estimated in this annex only provide information for decision makers and researchers on radiation
exposures from different electricity generation technologies. Dose estimations for the evaluation of
implications for health should be more specific to each exposure situation.

13. The annex begins with a chapter providing background information to support the study, followed
by a chapter discussing the assessment approach and the end points for the study. That chapter also
includes a description of terminology relevant to applying the methodology for estimating public
exposures due to radioactive discharges (see annex A). The following chapters cover the radiation
exposures arising from electricity generation from each of the electricity-generating technologies: nuclear
fuel cycle, fossil fuel energy (coal, oil and gas) and geothermal energy. Next comes a chapter on assessing
occupational doses from the mining of metals for the construction phase of the electricity-generating
technologies: nuclear, coal, natural gas and the renewable technologies (biomass, solar and wind). Finally,
there are chapters that include comparisons of exposures from the principal electricity-generating
technologies, discussions of uncertainties, suggestions for future work that could improve understanding,
and concluding remarks. Note that all supporting values and calculations were manipulated with full
precision and any discrepancies in the numbers presented in tables and figures are due to rounding.

II. BACKGROUND
14. The Committees past assessments in this field have covered each stage of the nuclear fuel cycle:
resource extraction (uranium mining and milling); fuel manufacture (uranium enrichment and fuel
fabrication); power generation (nuclear power reactors); reprocessing of spent nuclear fuel to recover
uranium and plutonium for subsequent use in nuclear fuels; and the management of solid wastes
generated at the various stages. They have also included exposures from transportation activities within
and between the fuel-cycle stages [U5, U6, U7, U8, U9, U11].

15. The Committees most recent and comprehensive review of public exposures due to non-nuclear
electricity generation (coal, oil, natural gas, peat and geothermal technologies) was published in 1993
[U8], based principally on the parameters and assessments presented in the UNSCEAR 1988 Report
[U7] and the assessment methodology adopted in the UNSCEAR 1982 Report [U6]. Public exposures
were expressed in terms of the collective effective dose normalized to electricity generated, in units of
mansieverts per gigawattyear. The results from those earlier UNSCEAR assessments of normalized
collective effective doses to the public from discharges for various electricity-generating technologies
are summarized in table 1.

16. According to the UNSCEAR 1993 Report, public exposures in terms of collective effective dose per
year of practice and per unit of electricity generated due to the burning of coal and peat, and due to
geothermal sources exceeded that due to operational discharges from nuclear power generation. However,
with continued improvements in efficiency of electricity generation from fossil-fuel plants and in emission
control technology, those earlier estimates of the Committee were deemed very likely to be outdated and
potentially misleading. Reductions in exposures were expected from newer and retrofitted power plants.
142 UNSCEAR 2016 REPORT

Since its 1993 evaluation, the Committee had only updated its estimates of exposures from the generation
of nuclear power, where the most recent estimates showed a decrease for the normalized collective
effective dose to the public from power plant operation from 0.45 man Sv/(GW a) for the period 1990
1994 to 0.27 man Sv/(GW a) for the period 19982002 [U11].

Table 1. Collective effective dose to the public normalized to electricity generated due to
discharges from different electricity-generating technologies from previous UNSCEAR assessments

Normalized collective effective dose to public

(man Sv/(GW a))
UNSCEAR Report
Nuclear power Coal Peat Gas Oil Geothermal
plant operation
1982 [U6] 4.2a 2 6
1988 [U7] 2.5 b
4c
2 0.03 0.5 2
1993 [U8]d 1.34e 20f (0.03) (0.5) (2)

2000 (for 1990-1994) [U9] 0.45g

2008 (for 1998-2002) [U11] 0.27 h

a
The value for the nuclear fuel cycle was 5.7 man Sv/(GW a); for globally-circulating radionuclides after 100 years the value was
12 man Sv/(GW a); for uranium mill tailings after 100 years, the value was 0.25 man Sv/(GW a).
b
The value for the nuclear fuel cycle was 4.0 man Sv/(GW a); for mine and mill tailings and globally-circulating radionuclides
the total was 200 man Sv/(GW a).
c
The value was 6 man Sv/(GW a) for older coal-fired power plants, which were considered to constitute two thirds of the world
total and 0.3 man Sv/(GW a) for modern coal-fired power plants, which were considered to constitute one third of the world total.
d
The values reported in the UNSCEAR 1993 Report [U8] for gas, oil and geothermal technologies were not from new
assessments but taken from the UNSCEAR 1988 Report [U7].
e
The value for the nuclear fuel cycle was 3.0 man Sv/(GW a); for globally-circulating radionuclides after 100 years the value was
50 man Sv/(GW a); for uranium mine and mill tailings after 100 years, the value was 150 man Sv/(GW a).
f
From coal-fired power plants (assumed one third modern-style, one third old-style and one third Chinese-style [U8]).
g
The reported dose value is for power plant operation only.
h
The value for the nuclear fuel cycle was 0.72 man Sv/(GW a).

A. Global trends in electricity generation technology

17. Figure I shows the global trends in electricity generation and contributions made by the various
electricity-generating technologies between 1980 and 2014. Combustion of coal for electricity
generation has dominated during this period with nearly 40% of total electricity generated; at the same
time, total global electricity generation has grown by about a factor of three. The use of both natural gas
and nuclear fuels to generate electricity has grown as a percentage of the total and also in total amount
of electricity generated over the same period. The use of oil has fallen as a percentage of the total, but
only slightly in total electricity generated. The use of hydropower has increased in the total amount of
electricity generation, although it has fallen as a percentage of the total. The contribution of the others
category, which includes renewable energies (geothermal, wind, solar and biomass), has increased from
1% to 7% during the 34 years, with a faster increase in their use since about 2008.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 143

Figure I. Trends in worldwide electricity generation (GW a) from 1980 to 2014

The others category comprises geothermal, solar, wind, biomass and waste [W13]. Supporting values were expressed as full
precision and any discrepancies in the numbers presented are due to rounding
144 UNSCEAR 2016 REPORT

B. Electricity generation worldwide in 2010

18. The reference year for assessments presented in this annex is 2010. Table 2 and figure II
summarize worldwide electricity generation in 2010 [I15]. Total electricity generation worldwide in
2010 is reported as 2,452 GW a (table 2), which also shows the contribution by type for the six
geographical regions of the world as adopted by the United Nations Environment Programme (UNEP).
These regions are the same as those used in annex A. Marginally different values for electricity
generated by nuclear power in 2010 were obtained from IAEA [I6] and used in the application of the
revised methodology for estimating public exposures due to radioactive discharges (annex A). The
IAEA database [I6] provided detailed data on electricity generated in 2010 both by region and as a
function of reactor type; these data were not used to generate table 2, but were used in the more detailed
assessments discussed in chapter IV.B.2.

19. Total annual average worldwide electricity generated from nuclear power plants 19982002 was
reported as 278 GW a in table 16 and table A-5 from [U11], which can be compared to 314 GW a for 2010.

Table 2. Summary of worldwide electricity generated in 2010

Supporting values were manipulated with full precision and any discrepancies in the numbers presented are due to rounding

Type Electricity generated in 2010 (GWa) [I15, I16]

Africa Asia and Europe Latin America North West Total
Pacific and Caribbean America Asiaa
Nuclearb 1.38 66.4 137.4 3.14 106.0 314.3
Coal 29.6 567.7 150.6 6.26 237.4 0.00 991.6
Gas 22.7 146.9 177.0 35.8 122.0 39.4 543.8
Oil 9.26 36.9 12.8 20.2 6.33 27.1 112.6
Geothermal 0.17 3.19 1.28 1.13 2.00 7.77
Solar 0.06 0.68 2.62 0.03 0.47 0.00 3.86
Wind 0.23 8.80 17.4 0.14 11.9 0.00 38.5
Biomass 0.09 5.95 17.3 4.85 9.59 0.00 37.8
Hydro 12.4 128.6 105.3 80.9 72.7 0.94 400.8
Tide 0.11 0.02 0.00 0.13
Other 0.00 0.55 0.10 0.65

Total 76 965 622 153 569 67 2 452

a
The numerical entry 0.00 corresponds to no electricity generated and the entry corresponds to no data available. West Asia is
omitted in all tables generated from the assessments reported in this annex, except for the assessments on gas and oil.
b
Marginally different values, obtained from IAEA [I6], were used in the application of the revised methodology for estimating
public exposures due to radioactive discharges (annex A) for the electricity generated by nuclear power in 2010.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 145

Figure II. Breakdown of the electricity generated worldwide in 2010 by generation technology

Derived from data in table 2 (excluding contributions from Tide and Other. Supporting calculations were made to full
precision and any discrepancies in the final numbers are due to rounding

III. ASSESSMENT APPROACH AND END POINTS

20. The main end point in this annex is an assessment and comparison of collective doses, and
collective doses normalized to the electricity produced, for one year of different electricity-generating
technologies, using consistent assessment approaches. To this end, the dominant exposure pathways for
each of the electricity-generating technologies considered in this study have been assessed.

21. The Committee conducted this study by investigating sources of exposure from electricity-
generating technologies based on (a) nuclear power, (b) the combustion of coal, natural gas, oil and
biofuels, and (c) geothermal, wind and solar power. Two electricity-generating technologies (nuclear
power and combustion of coal) were investigated in detail, because a more robust database existed for
these technologies. The Committee evaluated the main sources of radioactive discharges from the life
cycle of these electricity-generating technologies. For nuclear power, these sources included uranium
mining and milling, mill tailings, power plant operation and reprocessing activities. For combustion of
coal, they were the mining for coal, power plant operation for a prototype of both a modern coal plant
and an older-style coal plant, and deposits of coal ash. These sets of sources will hereafter be called the
nuclear fuel cycle and coal cycle, respectively, for simplicity.
146 UNSCEAR 2016 REPORT

22. The Committee pursued two paths for the assessments presented in this annex. First,
UNSCEARs revised methodology for estimating public exposures due to radioactive discharges
(annex A) was used to assess components of exposure due to electricity generation from the nuclear
fuel cycle and from the coal cycle in a consistent manner. These two technologies were chosen for
substantive treatment, because reasonably reliable input data existed and, from previous assessments,
they were known to be important. The use of a common methodology for these assessments allowed
consistent comparison of estimated exposures from the two technologies. A significant end point to
these assessments was to identify and discuss dominant sources of exposure from these two electricity-
generating technologies, and the dominant radionuclides contributing to those exposures.

23. The second path was to derive reasonable estimates of public exposures arising from the other
electricity-generating technologies, and occupational exposures from all the electricity-generating
technologies considered. This permits a rough comparison of the total human exposure from each of the
electricity-generating technologies normalized to the energy generated by each technology.

24. For those electricity-generating technologies relying on power plants that burn natural gas or oil
as fuel, and power plants driven by geothermal energy, an important exposure pathway is radon gas
discharged to atmosphere. The Committees revised methodology for estimating public exposures due
to radioactive discharges (annex A) has been used for assessing atmospheric discharges from these
technologies.

25. The revised methodology (annex A) was designed for assessing public exposures from routine
releases of radionuclides to the environment during normal operations, and not for assessing occupational
exposures. Thus, occupational exposures due to electricity generation from the nuclear fuel cycle and from
the coal cycle have been assessed here primarily from data on occupational exposure from the UNSCEAR
2008 Report [U11], adjusted for electricity generated in 2010. In some cases, particularly occupational
exposures from mining for coal and uranium and other metals, relevant new data have been obtained and
estimates of dose commitments have been updated for 2010. Occupational exposures from the mining of
metal ores needed for the construction of power plants or power-generating equipment have been assessed
and compared for the various electricity-generating technologies.

26. Assessments have also been made of collective effective doses for the entire decommissioning
phase of nuclear power plants based on available data on occupational exposures from existing
decommissioned plants.

27. Exposures of the public and workers due to radioactive materials from conventional
hydroelectricity generation were not included in this annex. This is because no radioactive materials are
discharged during normal operations, and the magnitude of discharges of radioactive materials and
resulting exposures when constructing dams were deemed negligible when compared to other
exposures (for example, the mining of metals or mining for other fuels, such as coal or uranium). The
main exposure situation associated with hydroelectricity generation was considered to be occupational
exposure during the mining of metals needed for construction of the power plants. However, this
exposure was not evaluated because hydroelectric plants have considerable variation in their size,
leading to large uncertainties when adopting a standard plant design for assessment.

28. Finally, the total exposures during one year, including both public and occupational, for each
electricity-generating technology (nuclear fuel cycle, coal cycle, oil, natural gas, geothermal, solar and
wind), have been compared. Three main phases that contribute to the total exposure of the public and
workers for each technology have been considered: preparation, operation and decommissioning. For
all discharges, the same methodology has been used to estimate public exposuresa significant
improvement over the Committees earlier assessments.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 147

A. Recapitulation of key features of the methodology for estimating

public exposures due to radioactive discharges
29. The following paragraphs recapitulate some of the key features of the methodology described in
annex A that are relevant for the assessment here.

30. World-average. This term is used to qualify data that are intended to represent a value averaged
across the whole world. For example, for the default population distributions (shown in table 3 of
annex A), the world-average population for a distance band is derived from the average of the values
for the distance band within each UNEP region, weighted by the population in each region. Similarly,
world-average consumption rates for terrestrial and aquatic foods are derived from the values for each
UNEP region weighted by the populations in each region.

31. Dose. In all cases and unless otherwise stated, dose refers to the protection quantity, effective
dose. The annual effective dose is the sum of the dose from external exposure in that year plus the
committed dose from intakes by inhalation and ingestion in that year. For doses integrated over a
period, it is the sum of integrated doses from external exposure over the period and committed doses
from the integrated intakes by inhalation and ingestion over the period. Doses to the public are only
estimated for adults and the period considered for the committed dose is 50 years, i.e. adult ages from
20 to 70 years.

32. Characteristic individual dose. The individuals considered are those living in the area local to the
point of discharge with behaviour indicative of the majority of people living in that area. The dose to
these characteristic individuals is referred to here as the characteristic individual dose.

33. For discharges to atmosphere, the characteristic individuals were assumed to live 5 km from the
discharge point and to obtain 25% of their food from this distance. For discharges to rivers, the
characteristic individual was assumed to be exposed to riverbank sediment 5 km downstream from the
discharge point, to drink all of their water from the river, and to eat 25% of the freshwater fish in their
diet from the river plus 25% of their dietary grain and leafy vegetables irrigated with water from the
river. For discharges to marine environments, the characteristic individual was assumed to be exposed
to external irradiation from radionuclides in sediments from the local marine compartment using a
factor to represent average occupancy of beaches. The characteristic individual is also assumed to
consume marine foods with 25% of their dietary fish from the local marine compartment and 75% from
the regional compartment, while 100% of the dietary consumption of crustacea and molluscs were
assumed to be from the local compartment. In all cases the consumption rates used were the average for
a population. These assumptions are discussed in detail in annex A.

34. Collective doses. The standard collective dose calculated here is the integrated dose to 100 years
from one years discharge. (For the modelling of the globally-circulating radionuclides, collective doses
integrated to 500 and 10,000 years are also calculated, see paragraph 39.) The end point for all
radionuclides calculated in annex A was the collective dose per unit discharge integrated to 100 years.
For releases to atmosphere, the local and regional components of the collective doses derive from doses
to populations within a distance of 1,500 km from the discharge point and are based on four different
types of population distributions:

(a) A default population distribution based on population densities for 2010. Values of population
density are given for six different regions (as adopted by UNEP) and also a world-average value (a
population-weighted average of the population density for the UNEP regions). These are used to
assess collective doses due to discharges from the coal cycle, combustion of gas and oil, and
geothermal energy.
148 UNSCEAR 2016 REPORT

(b) The population distribution around nuclear power stations situated on the coast. Values of the
average population within distance bands around coastal nuclear power stations are given for
five regions as adopted by UNEP (but not West Asia because there was no operating nuclear power
station in this region at the time of the assessment) and also an average value for the world. The
population densities were based on data for coastal sites of actual nuclear power stations
throughout the world and the distributions are arithmetic means of the data for distance bands
around each coastal site located in the region of interest.

(c) The population distribution around nuclear power stations situated inland. Values of the
average population within distance bands around inland nuclear power stations are given for
four regions as adopted by UNEP (but not Africa or West Asia because there were no operating
nuclear power stations located inland in these regions at the time of the assessment) and also an
average value for the world. The population densities were based on data for inland sites of actual
nuclear power stations throughout the world and the distributions are arithmetic means of the data
for distance bands around each inland site located in the region of interest.

(d) A single value of population density for remote sites of 5 km-2. This is used to assess collective
doses due to discharges from uranium mines and mill tailings, which are located in areas of very low
population density. This value is also used in an alternate calculation to assess collective doses due to
discharges from geothermal sites. Values for regions of the world are not provided for these cases.

35. For releases to atmosphere, a distinction could be made between the local component of collective
dose (to populations within a distance of 100 km) and a regional component of the collective doses (to
populations between 100 and 1,500 km). For aquatic releases to rivers and lakes, it was not possible to
distinguish between local and regional components of the collective dose because of the method used to
calculate collective dose. However, for discharges to marine environments, local and regional
compartments were used to model the dispersion of the released radionuclide, and therefore local and
regional components of the collective doses could be provided. Note that the use of the word regional
here is distinct from its potential use as a qualifier meaning related to the geographical regions adopted
by UNEP.

36. Globally-circulating radionuclides. Four radionuclides (3H, 14C, 85Kr and 129I) were assumed to
become globally circulated and to continue to expose the world population for decades and, with
regards to 14C, centuries, and 129I, millions of years. These radionuclides contribute to the local and
regional component of collective dose when they are initially discharged (due to the so-called first
pass) and this component was modelled as for all other radionuclides. However, a global component
needs to be added separately for these radionuclides. Thus, an additional collective dose due to global
circulation was also modelled and results calculated. Results of the global model are provided for
discharges to atmosphere and to marine environments; for aquatic discharges to freshwater systems, the
results were assumed identical to those for a marine environment for the purposes of calculating the
global component of collective dose.

37. Integration times. The methodology applies to discharges that can be assumed to be continuous.
Account is taken of the build-up of long-lived radionuclides in the environment and the associated
continued exposure after the discharges have stopped. This is done by considering a years discharge of
a radionuclide, its dispersion in the environment and the subsequent exposure of people over many
years; the resulting dose rates are then integrated. A value of 100 years is used for the integration period
for the standard case (see further details in annex A).

38. Using the integrals it is also possible to consider that the discharges continue for many years from
the same site. This is because the integrated dose to 100 years from one years discharge is numerically
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 149

equal to the dose in the 100th year from a continuous discharge at a constant rate over a 100-year
period. The characteristic individual doses are calculated by integrating dose rates to 100 years, which
also represent the annual dose that would be received in the 100th year of discharge.

39. Collective doses can be integrated to various times, but for most radionuclides it was sufficient to
integrate to 100 years because most of the collective dose commitment (i.e. the collective dose
theoretically integrated to infinity) is delivered during this period. However, for globally-circulating
radionuclides (i.e. 3H, 14C, 85Kr and 129I), the integrated collective dose continues to increase for many
years beyond 100 years (see annex A, paragraph 14). Therefore, for globally-circulating radionuclides,
collective doses integrated to 500 and 10,000 years were also calculated.

40. The main results of the revised methodology (annex A) are for a unit discharge for one year
integrated to 100 years and can be used to estimate the impact of one-years practice by, for example,
scaling the results by the discharge of radionuclides per unit of electricity generated, or the total
discharges from a particular type of electricity generation.

41. Waste residues from uranium mining and from coal-fired power stations. These wastes, uranium
mill tailings and coal ash, are often disposed of on the surface of the ground and they contain enhanced
levels of naturally occurring radionuclides, notably 226Ra which decays into the gas 222Rn, which can be
emitted into the air for many years after the disposal occurs. The radon and its decay products give rise
to human exposure (see figure III).

42. In this case one-years practice gives rise to a continuing discharge to atmosphere for many years.
Current best practice would rehabilitate mill tailings such that emissions were reduced to background
levels (i.e. the levels that would occur from that area if the mine had not been present). Information
from Australia and elsewhere indicates that this process might take some 50 years and that monitoring
is expected to continue for some 20 to 30 years to ensure that the rehabilitation has been successful and
to monitor for any deterioration [S2, W1]. The question remains whether the site could deteriorate once
institutional control has finished, and radon emissions increase again. A base case integration time of
100 years seems reasonable with 10,000 years a cautious upper estimate.

43. Therefore, for radon discharges from land disposal of mill tailings or coal ash, allowance has to be
made for this continued discharge and this was done by multiplying the results for one years discharge
by 100 to allow for the discharge continuing for 100 years (see annex A, paragraph 17). This 100-year
multiplication is chosen for both individual and collective doses because 100 years is not much
different from a human lifetime. The Committee has also considered the effect on the integrated
collective dose of a continuing discharge for 500 and 10,000 years.
150 UNSCEAR 2016 REPORT

Figure III. The 238U and 232Th decay series

Half-life is expressed in a = year; d = day; h = hours; m = minutes; s = seconds

 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 151

44. Representative radionuclides for nuclear power plants. Discharges of radionuclides from nuclear
power plants are commonly reported by operators and regulatory bodies aggregated in groups. For
example in UNSCEAR 2008 discharges were reported as noble gases and particulates for
atmospheric discharges, or other nuclides for liquid discharges. Accurate assessments of doses can
only be carried out on a radionuclide-specific basis and so where groups of radionuclides had been
reported, it was necessary either to apportion the discharge for the groups of nuclides among the
radionuclides, or to use a representative radionuclide. The European Commission [E1] gives the
percentage of radionuclides discharged for different reactor types sited in Europe for different
groupings. For AGRs, FBRs and GCRs, 4 the discharges were split between the radionuclides
considered in the workbooks for the appropriate groupings (atmospheric noble gases, atmospheric
particulates and other liquids). The breakdown for GCRs was based on United Kingdom sites because
these were the only ones operational in 2010. For atmospheric particulate discharges from BWRs and
PWRs, 54Mn was selected as the representative radionuclide for any radionuclides not considered
explicitly in the workbooks. The grouping of other liquids did not previously include 14C and, for
many countries and most reactor types, liquid discharges of 14C are not reported. However, examination
of EC RADD [E2] shows that for European PWRs liquid discharges of 14C are significant and therefore
30% of discharges from other liquids were assumed to be 14C for PWRs. For BWRs and PWRs any
liquid discharges for radionuclides not included in the workbooks were allocated equally between 54Mn
and 58Co. For HWRs, 41Ar, 60Co and 137Cs were assumed to represent discharges for noble gases,
atmospheric particulates and other liquids respectively, based on information in [B3, B5, C2]. For
LWGRs information was taken from the EC RADD database [E2].

IV. RADIATION EXPOSURES ARISING FROM ELECTRICITY

GENERATION BY NUCLEAR POWER

A. Introduction
45. The Committee has studied radiation exposures arising from electricity generation by nuclear
power repeatedly over the years since its first publication on this subject in 1958 (see especially [U6,
U7, U8, U9, U11]). These studies have consistently shown that the major contribution to public
exposures has been through the discharges of natural radionuclides, primarily from: radon and its
progeny released during uranium mining and milling, and from mill tailings, associated with the nuclear
fuel cycle; and from carbon-14, primarily associated with reactor operation and fuel reprocessing.

46. The revised methodology for estimating public exposures due to radioactive discharges (annex A)
has been used here to assess public exposures from the following processes: uranium mining and
milling; electricity generation from nuclear power reactors; and fuel reprocessing. The assessments and
results for these processes are presented in the following section. The processesuranium enrichment,
fuel fabrication and solid waste disposaland their exposure characteristics are discussed in sections

4
The following abbreviations are used for the different nuclear power reactor types, categorized according to their coolant
systems and moderators: light-water-moderated and cooled pressurized or boiling-water reactors (PWRs and BWRs); heavy-
water-cooled and moderated reactors (HWRs); gas-cooled, graphite-moderated reactors (GCRs and AGRs); light-water cooled,
graphite-moderated reactors (LWGRs); and the liquid metal cooled fast-breeder reactors (FBRs).
152 UNSCEAR 2016 REPORT

IV.B.4 and IV.B.5, and dose estimates for these processes that were published in the UNSCEAR 2008
Report [U11] have been used here. This chapter begins with the assessment of public exposures,
followed by that of occupational exposures, and ends with considering the occupational exposures
associated specifically with the decommissioning process.

B. Public exposure

1. Mining and milling

47. Uranium ore can be extracted from the earth by physically removing it through conventional
surface or underground mining methods or by chemically dissolving the uranium out of the rock ore
through either heap leaching or in-situ leaching (ISL) (sometimes referred to as in situ recovery or ISR)
[U18]. Surface mining (also referred to as opencast, open pit or strip mining) techniques are applied to
ore bodies that are close to the surface, and are also a generally cost-effective method for extracting
large volumes of lower-grade ore, that may then be combined with other bulk extraction techniques
(such as conventional milling, leaching and extraction, or alternative techniques such as heap leaching)
which would be uneconomical for underground operations. Underground mining involves extracting
rock through a tunnel or opening in the side of a hill or mountain and is generally applied for the
extraction of higher-grade ores. In-situ leaching is generally applied to shallow deposits that exist in
non-porous shale or mudstone, or in situations where uranium can be recovered from otherwise
inaccessible or uneconomical formations [U18].

48. Uranium ores typically contain from about 0.05 to 0.3% uranium oxide (U3O8) [U18]. After
extraction from the ground, the mined uranium ore is sent to a mill, which is usually located close to the
mine. This next step in the nuclear fuel cycle, called milling, involves the extraction and purification
of uranium from the uranium ore. After a first purification process, the uranium is precipitated in a
partially refined form, known as yellowcake. The uranium concentrate, typically containing 75 to
95% uranium, is shipped to a chemical plant for further purification and chemical conversion [B4].

49. In-situ leaching is an alternative method for extracting uranium from low-grade ores or shallow
deposits that exist in non-porous shale or mudstone. An ISL plant chemically alters the uranium ore
underground before it is pumped out for processing. In the ISL process, wells are drilled into rock
containing uranium ore. An alkaline or acidic solution (known as lixiviant) is injected down the wells to
dissolve the uranium in the rock. In the case of alkaline solutions, the lixiviant is usually (a) water
mixed with oxygen and/or (b) hydrogen peroxide mixed with sodium carbonate or carbon dioxide. In
the acid leach processes (used in Kazakhstan and Australia), sulphuric acid usage can result in lixiviant
solutions with pH as low as 1 being circulated into the ore body (the process in Kazakhstan is typically
strongly acidic, while in Australia only mildly acidic). In some cases, an oxidant is added to increase
the efficiency of leaching. The lixiviant is then collected in a series of recovery wells, through which it
is pumped to a processing plant. At the processing plant, the uranium is extracted from the solution
through an ion-exchange process, and the uranium oxide concentrate (yellowcake) is then precipitated,
dried and packed.

50. After recovery of the uranium, the barren solution is re-fortified with oxidant (if required) before
being returned to the well field via the injection wells. However, a small flow (about 0.5%) is bled off to
maintain a pressure gradient in the well field and this, with some solutions from surface processing, is
treated as waste. This waste water contains various dissolved ions such as chloride, sulphate, sodium,
radium, arsenic and iron from the ore body and is re-injected into approved disposal wells in a depleted
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 153

portion of the ore body. Wells must be monitored to ensure that extraction fluids do not leave the facility
or contaminate groundwater. Waste from this process, usually filters and piping, can be disposed in a
tailings pile at a mill site or a licensed disposal facility. In-situ leaching facilities have no radon discharge
during the mining phase, and no surface tailings and little radon emission after closure [U11]. There can
however be discharge of radon during the leaching phase of the ISL mining process, which here has
been assumed to be the same as the radon discharge occurring during the milling phase [B6, B7].
Chapter IX discusses the need for further study regarding the ISL mining process.

51. In 2014, 51% of the worlds uranium was mined using ISL operations, a share that has risen
steadily mainly because of mining operations in Kazakhstan [W8]. Table 3 shows the percentage
contribution from ISL-mining to total uranium production during the period from 2008 to 2012 in each
of the UNEP regions. For the reference year of 2010 used in this annex, the total uranium production
for 15 countries is given in table 4.

Table 3. Contribution of ISL-mining to total uranium production between 2008 and 2012

Information on percentage ISL compared with other types of mining is taken from [W10]

Region Contribution to world production (%)

Total ISL Others

Asia and Pacific 51.6% 38% 13.6%

North America 21.1% 2% 19.1%
Africa 18.4% 0% 18.4%
Europe 8.6% 1% 7.6%
Latin America 0.3% 0% 0.3%

World 100% 41% 59%

Table 4. Uranium production in 2010 [O8]

The production numbers come from OECD/NEA (table 1.21, page 60, Historical uranium production [O8]). Countries that had
a label Secretariat estimate are not included here (Pakistan, Romania and India). Also excluded are a few countries that had
very small production which came from mine rehabilitation efforts only

Country (UNEP region) Uranium production (t U)

Kazakhstan (Asia and Pacific) 17 803
Canada (North America) 9 775
Australia (Asia and Pacific) 5 900

Namibia (Africa) 4 503

Niger (Africa) 4 197
Russian Federation (Europe) 3 563

Uzbekistan (Asia and Pacific) 2 874

United States (North America) 1 630
China (Asia and Pacific) 1 350
154 UNSCEAR 2016 REPORT

Country (UNEP region) Uranium production (t U)

Ukraine (Europe) 837
Malawi (Africa) 681
South Africa (Africa) 582

Czech Republic (Europe) 254

Brazil (Latin America) 148
Iran, Islamic Rep. of (Asia and Pacific) 7

Total 54 104

(a) Input to assessments for mining and milling

52. The Committees revised methodology for estimating public exposures due to radioactive
discharges (annex A) has been used to assess the collective doses and characteristic individual doses
from four categories of sources for radon discharges:

(a) Mining (underground uranium mining)

(b) Milling (natural radionuclides other than radon are also discharged)

(c) Operational mill tailings

(d) Mill tailings

53. The third category, operational mill tailings, refers to the mill tailings produced during the milling
process, and the fourth category, mill tailings, represents the tailings that are positioned in some other
place for more permanent holding. They can be eventually treated or mitigated to reduce the discharge
of radionuclides. Assessments were conducted using the methodology described in annex A for the four
categories of sources above. They were performed using the discharges normalized to energy generated
shown in table 5 and table 6, and the values for electricity generation for each region given in table 7 to
estimate characteristic individual doses to the public. As noted in the tables, many of the values for the
normalized discharges were taken from previous UNSCEAR reports. However, the estimate for the
discharge of radon from mill tailings has been updated. A value for the discharge of radon per energy
produced of 0.3 TBq/(GW a), based on a tailings area of 1 hectare and a radon emanation rate of
1 Bq s1 m2, was used in [U9]. This was lower than the previous value of 1 TBq/(GW a). Information
on emissions from Australian mine tailings [L2] is that the value of 1 Bq s1 m2 is high compared to
the range of levels that are measured. Measurements taken over a number of years at the El Sharena
mines were in the range 936 mBq s1 m2 with an average baseline value of 17 mBq which is
equivalent to 0.005 TBq/(GW a) based on a tailings area per electricity generated of 1 ha/(GW a). In
other cases where the tailings have been rehabilitated, the emissions rates are at background levels. A
study at the Ranger uranium mine found that the emission rates ranged from 0.2 to about 0.9 Bq s1 m2.
The studies related to trial landforms intended to develop their rehabilitation strategies with the aim of
reducing the long-term radon emission rates. For Olympic Dam, an emission rate of 0.5 Bq s1 m2 was
published, a value equivalent to 0.15 TBq/(GW a) while other information gives emission rates of 0.3,
0.14 and 0.04 Bq s1 m2 for the inner mine, outer mine and region. These are equivalent to 0.1, 0.04
and 0.01 TBq/(GW a) based on 1 ha/(GW a). These Australian values indicate that the UNSCEAR
2000 value (0.3 TBq/(GW a)) may be too high based on mines operating with best practice for which a
value of 0.1 or lower may be more appropriate. In the light of this, the Committee agreed that a rounded
emission rate of 0.1 TBq/(GW a) should be used in this study.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 155

Table 5. Discharges of radon from the uranium mining and milling process normalized to
electricity generated

Source Normalized discharges

(TBq/(GW a))
Mining (world average, all mines except ISL, only radon discharges) 66a
Mining (world average, all ISL mines, only radon discharges from leaching phase of 3
ISL mining process)
Milling (world average, all mines, radon discharges plus nuclides in table 6) 3b
Operational mill tailings (world average, all mines except ISL, only radon discharges) 3c

Mill tailings (world average, all mines except ISL, only radon discharges) 0.1d
a
A value of 75 TBq/(GW a) was used in earlier UNSCEAR assessments [U8], and was based on the assumption that there were
on average 300 GBq of radon released per tonne of uranium oxide [U8].
b
From UNSCEAR 1993 Report [U8].
c
From UNSCEAR 2000 Report [U9].
d
From annex A.

Table 6. Airborne discharges during the milling process (all mines including ISL) ([U8] tables 19 and 22)

Radionuclide Activity discharged per unit of electricity generated

(TBq/(GW a))
210
Po 2 105
210
Pb 2 105
226
Ra 2 105
234
U 4 104
238
U 4 104
230
Th 2 105

54. The estimate for the discharge of radon from uranium mining (non-ISL mines) has also been
updated. Total uranium production in 2010, shown in table 4 [O8], was assumed to be the amount that
was mined in that year. Values were summed for all countries to give a rounded total of 54,100 tonnes
of uranium. Similarly, total uranium production in 2010 from table 4 is assumed to equal the total
amount of uranium milled. The value for the mass of uranium required per unit of electricity generated
given in the UNSCEAR 2008 Report was 220 t/(GW a) ([U11] table 18A), which is lower than the
value of 250 t/(GW a) used earlier by the Committee [U7, U8, U9]. The trend in uranium required per
unit of electricity generated (the uranium requirement) is generally downwards, because of increasing
efficiencies in power plant operation and lower enrichment tails assays. For example, the 2014 Red
Book [O8] reported that the generic reactor uranium consumption had reduced from 175 t/GW(e) per
year at 0.30% tails assay [O6] to 163 t/GW(e) per year at 0.25% tails assay. The corresponding figures
for U3O8 are 206 and 192 tonnes, respectively.

55. The value of 220 t/(GW a) used in the UNSCEAR 2008 Report is the value used in this assessment.
Given the information in the previous paragraph, this is possibly a slight overestimate; however, it was
retained for continuity with the previous UNSCEAR assessments. This gives a value for the discharge of
radon from the mining process normalized to electricity generated of 66 TBq/(GW a), which is lower than
the value of 75 TBq/(GW a) used in earlier UNSCEAR studies [U7, U8, U9].
156 UNSCEAR 2016 REPORT

56. In modern ISL facilities, the circulating lixiviant goes directly from the well field to the header
houses and the dissolved uranium is extracted from the lixiviant through the ion exchange process. The
lixiviant is then reconstituted and returned directly to the well field in an essentially closed
(pressurized) system. In older non-pressurized systems (such as at Crow Butte, United States), the resin
column is not pressurized, so the lixiviant is exposed to normal atmospheric conditions, which results in
super-saturated radon being discharged. The discharged radon gas is vented through an exhaust stack to
the atmosphere. This discharge is most similar to that from the milling phase, not from the mining
phase. The assessments in this annex have assumed radon discharges during ISL uranium mining are
the same as those from the milling phase.

57. In order to perform assessments of the characteristic individual doses to the public for each
region, electricity generation in 2010broken down into ISL and non-ISL mining sourceswas
needed for the different regions. Total uranium production for 15 countries in 2010 is given in table 4.
Table 3 shows the contribution from ISL-mining to the total uranium production in each UNEP region
between 2008 and 2012. Using the value for uranium requirement of 220 t/(GW a), the electricity
generated from the total uranium production for each region could be estimated as shown in table 7.
Assessments for mining and milling using the revised methodology for estimating public exposures due
to radioactive discharges (annex A) used the values for total electricity generated with the normalized
discharges for milling, the total electricity generation from ISL mining with the normalized discharges
for ISL mining processes, and the total electricity generation excluding ISL mining sources with the
normalized discharges for the processes mining (non-ISL), operational mill tailings and mill tailings, to
calculate characteristic individual doses. Collective doses resulting from mining and milling were
assessed assuming a low population density (defined in annex A), and no aquatic discharges were
included in the assessment.
Table 7. Inferred electricity generation from the uranium production in each region and from ISL and non-ISL contributions

The assessment assumes 220 tonnes of uranium per gigawatt-year. Note that the total inferred electricity generation is 246 GW a, which is somewhat less than the world total nuclear electricity generation in 2010
at 314 GW a (table 2). However, this difference is not unreasonable because current reactor fuel requirements were met from primary supply (direct mine output: 78% in 2009) and secondary sources: commercial
stockpiles, nuclear weapons stockpiles, recycled plutonium and uranium from reprocessing used fuel, and some from re-enrichment of depleted uranium tailings (left over from original enrichment) [O6, W11]

Region % of world % from ISL % from other Mass of uranium Mass of uranium from Inferred electricity Inferred electricity Total inferred
production production methods from ISL (tonnes) other production generated from ISL generated from other electricity generated
methods (tonnes) (GW a) production methods (GW a)
(GW a)

Africa 18.4 0 18.4 0 9 954 0 45 45

Asia and Pacific 51.6 38 13.6 20 558 7 358 93 33 126

Europe 8.6 1 7.6 541 4 112 3 19 22

Latin America 0.3 0 0.3 0 162 0 1 1

North America 21.1 2 19.1 1 082 10 333 5 47 52

Total 100 41 59.0 22 181 31 919 101 145 246

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 157

158 UNSCEAR 2016 REPORT

2. Electricity generation from nuclear power reactors

58. The nuclear power reactors used for electricity generation that are treated here can be categorized
according to their coolant systems and moderators: light-water-moderated and cooled pressurized or
boiling-water reactors (PWRs and BWRs); heavy-water-cooled and moderated reactors (HWRs); gas-
cooled, graphite-moderated reactors (GCRs and AGRs); and light-water cooled, graphite-moderated
reactors (LWGRs). These reactor types are all thermal reactors that use the moderator material to slow
down the fast fission neutrons to thermal energies. In fast-breeder reactors (FBRs), the coolant is a
liquid metal, there is no moderator and fast neutrons induce fission.

(a) Assessments for nuclear reactor operation

59. The distribution of electricity generated by geographic region and type of location of the nuclear
power plants in the world are presented in table 8. The electricity generated in 2010 by nuclear power
was about 300 GW a according to the IAEA PRIS database [I6] (table 9). PWRs contributed most to
the total electricity generated worldwide from nuclear reactors (about 68%), followed by BWRs (about
21%) [I6]. The distribution of electricity generated by geographic region from nuclear power plants in
2010 by reactor type is shown in table 9. (The ordering of the individual reactor types in all relevant
tables and figures in this annex is according to their world share of electricity generated in 2010, as
shown in table 9.)

60. Radionuclide discharges and electricity generation data used with the revised methodology for
estimating public exposures due to radioactive discharges (annex A), for assessing discharges from
nuclear power plants in 2010 normalized to the electricity generated, are described in table 10. The
main basis for deriving the normalized discharge values was the IAEA PRIS database [I6], the EC
RADD database [E2] and the discharge data for the year 2002 in the UNSCEAR 2008 Report [U11].
Using these data the normalized discharges for 2002 were recalculated and adjusted to electricity
generation in 2010, for use with the revised methodology for estimating public exposures due to
radioactive discharges (annex A) to assess the exposures for the reference year 2010.

61. A summary of the activities of radionuclides, or radionuclide groups, discharged in airborne and
liquid effluents from nuclear reactors during routine operation in 2010, normalized to the electricity
generated in 2010, are reported in table 11. The method for determining how the aggregated nuclides
were characterized, i.e. the noble gases and particulates in the atmospheric discharges, and the other
liquids listed in the liquid discharges, has been addressed in earlier studies by UNSCEAR (e.g. [U7,
U8, U9]). The aggregated nuclides were called representative radionuclides for nuclear power plant
discharges in those studies, and the term representative radionuclides is used in this annex also.
Chapter III.A, paragraph 44 in this annex explains how the representative radionuclides, were chosen
for use in the current study with the revised methodology for estimating public exposures due to
radioactive discharges (annex A) applied to nuclear power plants.

62. The revised methodology for estimating public exposures due to radioactive discharges (annex A)
was used to assess doses due to discharges from nuclear power plants. The assessments considered
atmospheric and aquatic discharges. Collective dose estimates for discharges to atmosphere were based
on the population distributions around coastal or inland nuclear power plants, using the population
distributions generated for the different regions adopted by UNEP (see annex A). Where appropriate,
contributions from globally-circulating radionuclides were included in the assessment of collective
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 159

doses. The characteristic individual doses and collective doses were calculated separately for each of
the reactor types (PWR, BWR, HWR, LWGR, AGR, GCR and FBR).

Table 8. Number of reactors by type, location and UNEP region (2010 data)

Data from the IAEA PRIS database [I6]

Reactor type Africa Asia and Europe Latin America North Grand
Pacific and Caribbean America total
PWR total 2 55 141 2 69 269
Coastal 2 54 34 2 16 108

Inland 0 1 107 0 53 161

BWR total 0 36 19 2 35 92
Coastal 0 36 9 2 2 49
Inland 0 0 10 0 33 43

HWR total 0 24 2 2 18 46
Coastal 0 11 0 0 1 12

Inland 0 13 2 2 17 34

LWGR total 0 0 15 0 0 15
Coastal 0 0 4 0 0 4
Inland 0 0 11 0 0 11

AGR (all coastal) 0 0 14 0 0 14

GCR total 0 0 4 0 0 4
Coastal 0 0 2 0 0 2
Inland 0 0 2 0 0 2

FBR (all inland) 0 0 2 0 0 2

Grand total 2 115 197 6 122 442

160 UNSCEAR 2016 REPORT

Table 9. Electricity generation from nuclear power plants in 2010 by reactor type and UNEP region
(from [I6])

The values for the total electricity generation from nuclear power plants for each UNEP region [I9] are marginally different
from those shown in table 2 [I15]. The IAEA database [I6] provided detailed data on electricity generation in 2010 both by
region and as a function of reactor type, which was not available from the database used to generate table 2, and the more
detailed data were therefore chosen for use in the assessments discussed in this chapter

Electricity generation (GW a) % total

Reactor per
Africa Asia and Europe Latin America North West Total
type reactor
Pacific and Caribbean America Asia
type
PWR 1.47 38.47 102.47 1.57 60.89 0.00 204.88 68
BWR 0.00 19.66 12.18 0.64 31.25 0.00 63.72 21

HWR 0.00 5.34 1.22 0.76 9.76 0.00 17.09 6

LWGR 0.00 0.00 8.15 0.00 0.00 0.00 8.15 3

AGR 0.00 0.00 5.01 0.00 0.00 0.00 5.01 2

GCR 0.00 0.00 0.93 0.00 0.00 0.00 0.93 0
FBR 0.00 0.00 0.42 0.00 0.00 0.00 0.42 0

Total 1.47 63.47 130.39 2.97 101.89 0.00 300.21 100

Table 10. Key data sources and assumptions used to assess public exposures due to radioactive
discharges from nuclear power plants

Data sources Commentary

Discharges, 2002 Taken from tables A6-A12 in UNSCEAR 2008 Report [U11]. Note that WWER was
regrouped under PWR. Data for discharges from FBR are from 2001. Discharges of
particulates to air and other liquids for United Kingdom AGRs/GCRs were taken
from the EC RADD database [E2]

Electricity generation, Taken from [I6]. Note that HWLWR was regrouped under HWR and PHWR under
2002 HWR. Generation data for FBR are from 2001

Electricity generation, Taken from [I6]. Note that HWLWR was regrouped under HWR and PHWR under
2010 HWR. Includes a calculation of generation from reprocessing countries as a fraction
of all generation

Countries Countries and regions as adopted by UNEP (see annex A)

Location data Inland/coastal location of reactors, taken from [I6]

Normalized discharge data For each reactor type:a,b the normalized discharges were scaled to the electricity
generation ratio using the 2002 to 2010 electricity generation data; information
summarized in paragraph 44 was used to apportion discharge between specific
radionuclides; and location data were used to apportion between inland/coastal
locations
a
For FBRs, values of normalized discharges of 3H, 131I, 14C, liquid tritium and other liquids were taken from table 17, annex B,
UNSCEAR 2008 Report [U11] because no discharge data were available. Data for the radionuclide mix were taken from [E1].
b
For LWGRs, normalized discharge data were taken from table 17, annex B, UNSCEAR 2008 Report [U11]. EC RADD [E2]
data for European LWGR reactors were used to infer mix of noble gases, particulates and other liquids.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 161

Table 11. Estimated normalized discharges from nuclear power plants 2010

Estimated normalized discharges per unit of electricity generated (TBq per GW a)

Reactor Discharges to atmosphere Aquatic discharges
type
Noble Tritium 131
I 14
C Partic- 35
S Liquid Other
gases ulates tritium liquid
PWR 5.8 100 1.5 100 8.0 105 8.3 102 3.6 105 0 1.8 101 3.8 103
BWR 1.8 101 1.3 100 4.2 104 1.3 101 1.8 103 0 8.2 101 2.1 103
HWR 3.5 101 2.0 102 2.3 105 6.0 101 1.7 105 0 1.7 102 3.1 102

LWGR 4.6 102 2.6 101 9.9 103 1.3 100 2.7 103 0 7.8 101 2.0 103
AGR 1.9 101 4.0 100 3.2 105 1.4 100 2.2 105 6.6 102 4.1 102 8.1 101
GCR 1.7 103 5.0 100 0 5.5 100 3.0 104 3.7 101 4.7 100 1.2 100

FBR 4.4 101 4.9 101 2.0 104 1.2 101 1.4 104 0 1.7 100 2.3 102

3. Fuel reprocessing
63. Reprocessing of used nuclear fuel has been practised for several decades in a number of countries,
mainly for extracting and recycling fissile materials. Current recycling practices are primarily focused
on the conversion of fertile 5 238U to fissile plutonium, for which a significant amount of the plutonium
recovered from used fuel has been recycled into mixed oxide (MOX) fuel. Table 12 shows the current
location of reprocessing facilities in the world, with operations in France, the Russian Federation and
the United Kingdom dominating. Not shown in the table is the Rokkasho facility in Japan, which is
expected to start operations in 2018 and to have a commercial reprocessing capacity of 800 tonnes per
year. As of 2015, about 31% (90,000 tonnes of 290,000 tonnes) of used fuel from commercial power
reactors has been reprocessed [W12].

64. Characteristic individual doses and collective doses were calculated for Sellafield in the United
Kingdom, La Hague in France and Ozersk (Mayak) in the Russian Federation, which are also the three
main reprocessing facilities (table 12), using available data on discharges. The doses due to discharges
from fuel reprocessing facilities were assessed using the revised methodology for estimating public
exposures due to radioactive discharges (annex A).

65. Data for both atmospheric and aquatic discharges were obtained for the reprocessing facilities at
Sellafield, United Kingdom and La Hague in France for the year 2010 [E2] and for atmospheric
discharges only at Ozersk (Mayak), Russian Federation [F2]. These data include discharges from other
activities on site but it could be assumed that most discharges are related to reprocessing activities.
Discharges normalized to electricity generation were calculated for La Hague based on the assumption
that discharges from the reprocessing facility were related to the reprocessing of fuel equivalent to that
required to power French PWRs in 2010, 47 GW a [I6, W9]. However, for Sellafield and Mayak it was
not possible to relate the discharges to electricity generated.

5
Fertile material is a material that is not fissionable by thermal neutrons, but can be converted into a fissile material by neutron
absorption and subsequent nuclei conversions.
162 UNSCEAR 2016 REPORT

66. The discharges for reprocessing facilities are shown in table 13. Consequently, characteristic
individual doses and collective doses (as defined in section III) could be assessed for all three of these
facilities, all located in the region named Europe (as adopted by UNEP). However, because discharges
normalized to electricity generation were only available for La Hague in France, the characteristic dose
normalized to electricity generated could be assessed only for this facility. Collective doses normalized
to electricity generation for Europe and the whole world were also based on La Hague, but took into
account the fraction of electricity generated by all French PWRs in 2010 compared to Europe and the
whole world, 0.55 and 0.24 respectively, for the calculation of the doses due to the first pass.

Table 12. World commercial reprocessing capacity [N1, O2, W12]

Fuel type Facility Commercial reprocessing capacity

(tonnes per year)
LWR fuel France, La Hague 1 700

United Kingdom, Sellafield (THORP) 600

Russian Federation, Ozersk (Mayak) 400

Total LWR approximately 2 700

Other nuclear fuels United Kingdom, Sellafield (Magnox) 1 500

India (PHWR, 4 plants) 330

Japan, Tokai (MOX) 40

Total other approximately 1 870

Total civil capacity approximately 4 570

Table 13. Discharges from reprocessing facilities [E2]

Discharged Discharges to atmosphere Aquatic discharges

radionuclide Sellafield La Hague Mayak Sellafield La Hague Mayak
DISCHARGES PER UNIT OF ELECTRICITY GENERATED (TBq/(GW a))
3
H 3.8 104 6.7 106
14
C 1.1 104 5.0 103
41
Ar
54
Mn 1.4 10 0

58
Co 7.0 10 2

60
Co 4.9 103 4.4 101
85
Kr 1.5 108
90
Sr 9.1 10 1

106
Ru 4.8 102
1.4 10 3

129
I 3.1 100 9.3 102
131
I 1.6 101 7.9 100
135
Xe
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 163

Discharged Discharges to atmosphere Aquatic discharges

radionuclide Sellafield La Hague Mayak Sellafield La Hague Mayak
134
Cs 3.6 103 5.1 101
137
Cs 3.0 103
7.3 102

239
Pu 1.3 103
6.3 100

241
Am 1.3 101

AVERAGE DISCHARGE RATE OVER A YEAR (Bq/s)

3
H 3.1 106
1.8 106 4.4 107 3.2 108
14
C 8.7 103 5.1 105 1.4 105 2.3 105
41
Ar 1.2 106
54
Mn 6.4 101

58
Co 3.3 100

60
Co 2.3 101 3.4 101 3.1 103 2.0 103
85
Kr 1.4 109 7.1 109
90
Sr 1.3 100
4.2 101
3.2 104
4.2 103

106
Ru 2.4 101
2.2 10 0
5.7 100
3.7 104
6.5 104

129
I 3.1 102 1.4 102 8.7 103 4.3 104
131
I 1.2 101 7.7 100 2.5 101 3.7 102
135
Xe 1.8 105

134
Cs 1.7 101
3.5 103
2.4 103

137
Cs 3.0 100 1.4 101 2.1 101 1.5 105 3.4 104
239
Pu 6.2 101 5.9 102 1.3 101 4.2 103 2.9 102
241
Am 1.2 100
1.1 104
6.2 102

(a) Resultspublic exposures from mining and milling, electricity generation from
nuclear power reactors and fuel reprocessing
67. Characteristic individual doses to the publicfor the nuclear fuel cycle processes: mining and
milling; electricity generation from nuclear power reactors; and fuel reprocessingnormalized to the
electricity each process generated in 2010 are summarized in table 14. Doses for the characteristic
individual from mining and milling are shown for both non-ISL and ISL mines. The characteristic
individual was assumed to be located 5 km from the source. The doses represent the radiation exposure
from all discharges in a year and, in the case of mill tailings, the doses are associated with those
emissions for a period of 100 years (see chapter III).

68. The largest estimated characteristic individual doses normalized to electricity generation in 2010
for all discharges come from mining and milling activities in non-ISL mines in all regions, followed by
operational discharges from nuclear power plants. Estimated characteristic individual doses from ISL
mines are about an order of magnitude smaller than those from non-ISL mines, owing to the differences
in radon discharges from the two processes. Estimated characteristic individual doses from mining and
milling are primarily associated with radon exposures, while those from nuclear power plant operations
reflect differences in the type and relative proportion of the different reactor types and thus different
164 UNSCEAR 2016 REPORT

radionuclide discharge mixes within a region, and to a lesser extent, variations in food consumption
patterns across regions. Estimated characteristic individual doses for a region normalized to electricity
generated that are associated with reprocessing are only reported for the region Europe, where the
French, British and Russian reprocessing facilities are located. All of the estimated characteristic
individual doses are very low.

Table 14. Summary of characteristic individual doses to the public normalized to electricity
generated in 2010 for mining and milling, electricity generation from nuclear power reactors, and
fuel reprocessing (mSv/(GW a))

Discharge type and source Characteristic individual doses a,b to the public (mSv/(GW a))
Africa Asia and Europe Latin North
Pacific America America
ATMOSPHERIC DISCHARGES
Mining and milling - non-ISL mines
c
6.9 103 6.9 103 6.9 103 6.9 103 6.9 103
Mining and milling - ISL mines 5.5 104 5.5 104 5.5 104

Power plants 1.3 105 5.1 105 7.2 105 1.1 104 5.8 105
Reprocessing 5.7 105

AQUATIC DISCHARGES
Mining and milling - non-ISL mines
Mining and milling - ISL mines
Power plants 7.0 10 5
1.5 104
1.2 10 3
1.5 10 4
5.0 105

Reprocessing 6.9 104

TOTAL FOR ALL DISCHARGES

Mining and milling - non-ISL mines
c
6.9 103 6.9 103 6.9 103 6.9 103 6.9 103
Mining and milling - ISL mines 5.5 104 5.5 104 5.5 104

Power plants 8.2 105 2.0 104 1.3 103 2.6 104 1.1 104
Reprocessing 7.5 104
a
The characteristic individuals are those living 5 km from the points of discharge with behaviour indicative of the majority of
people living the area.
b
It is only appropriate to present the characteristic individual doses for discharges normalized to electricity generated and not
summations of the various individual dose values, because the same individuals cannot be exposed to all discharges from the
various sources in each region. Because collective doses are the sum of all individual doses, they can be presented both as total
collective dose and as collective dose normalized to electricity generated.
c
The radon emissions from mill tailings produced per unit of electricity generated are assumed to continue for 100 years.

69. Collective doses to the public for discharges from mining and milling, power plant operation and
reprocessing in 2010 are summarized in table 15, and in table 16 normalized to the electricity generated
by each process for that year. The local component (for the region-averaged or world-averaged
population within 100 km) and the regional component (for the region-averaged or world-averaged
population between 100 km and 1,500 km) of the collective doses due to the atmospheric releases are
presented separately, integrated to 100 years (first pass). Radon emissions from mill tailings produced
per unit of electricity generated are assumed to continue for 100 years. The global components of
collective doses resulting from the globally-circulating radionuclides (3H, 14C, 85Kr and 129I) are also
shown, integrated to 100, 500 and 10,000 years. For ease of comparison, table 17 shows the final
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 165

aggregated values for the total collective dose to the world public and for the collective dose
normalized to electricity generated for each of the processes mining and milling, power plant operation
and reprocessing.

70. The results in tables 15 and 16 indicate that for discharges from both power plant operation and
reprocessing facilities, the global component of the collective doses due to globally-circulating
radionuclides integrated to 100 years and longer exceed the local and regional components of the
collective doses due to the initial discharge (first pass) integrated out to 1,500 km. It is important to
recognize that global components are summed over the population of the entire world, which is taken as
1010 people to account for growth over the next 100 years (called whole world population in the
tables 15 and 16). This component thus is the sum of a large number of very small doses with per caput
values of the order of 108 Sv, and per caput values normalized to electricity generated of the order of
1010 Sv/(GW a). Values for the collective doses in both tables 15 and 16 for the first-pass calculation
are the doses integrated to 100 years and to a distance of 1,500 km for the local and regional
components due to atmospheric discharges. Based on the results of the assessment, doses beyond this
distance were assumed to be negligible.

71. For the nuclear industry, a significant fraction of the collective dose is due to radon discharges
from mining and disposal of mill tailings. The importance of carbon-14 discharges from nuclear power
plants and reprocessing is also evident. Table 18 shows the contribution from individual radionuclides
to the collective doses to the public, and collective doses per unit of electricity generated, from
reprocessing integrated to 100 years, not including contributions from globally-circulating
radionuclides. Figure IV is derived from the data in table 18, showing collective doses per unit of
electricity generated for individual radionuclides discharged from reprocessing. The importance of the
carbon-14 contribution to dose is evident.

72. Figure V shows collective doses per unit of electricity generated due to the globally-circulating
radionuclides (tritium, 14C and 85Kr) integrated to 100, 500 and 10,000 years, for each reactor type
considered. Besides indicating the variation in the dose contribution from globally-circulating
radionuclides for the different reactor types, the graphs also illustrate the importance of 14C, whose
contribution to collective dose increases with integration time. The global circulation for 129I is not
included in figure V because it is only discharged during the reprocessing phase.

73. Figure VI compares the collective doses to the public from all discharges, atmospheric and
aquatic, from all nuclear power plants by region and by reactor type, not including contributions from
globally-circulating radionuclides. The data indicate that HWRs and PWRs are the reactor types that
contribute most to the collective doses. Table 19 shows the total collective dose for all nuclear reactor
types summed and by region, with North America and Europe showing the highest values.
 166 UNSCEAR 2016 REPORT
Table 15. Summary of collective doses to the public due to mining and milling, electricity generation from nuclear power reactors, and fuel reprocessing (in 2010)

Discharge type and source Collective dosea, b (man Sv)

Africa Asia and Pacific Europe Latin America North America World
LOCAL COMPONENT ATMOSPHERIC RELEASES
Mining and milling No data by region 1.0 101

Power plants

Inland 0 1.2 100 2.0 100 1.7 101 2.2 100 5.3 100
Coastal 8.2 103 2.7 100 1.2 100 1.5 102 1.2 101 3.8 100
Reprocessing 1.3 100 2.1 100

REGIONAL COMPONENT ATMOSPHERIC DISCHARGES

Mining and milling No data by region 4.3 101

Power plants
Inland 0 3.0 100 6.7 100 1.3 101 5.8 100 1.6 101

Coastal 1.6 103 6.8 100 3.5 100 2.9 102 1.5 101 8.4 100
Reprocessing 4.3 100 4.4 100

LOCAL AND REGIONAL COMPONENT AQUATIC DISCHARGES

Mining and milling No data by region

Power plants

Inland 0 1.4 100 1.4 101 8.1 101 1.8 101 3.4 101
Coastal 9.3 105 2.5 102 6.8 102 3.5 104 1.3 103 3.9 101
Reprocessing 2.6 10-1 1.1 100

LOCAL AND REGIONAL COMPONENTS (ATMOSPHERIC AND AQUATIC DISCHARGES)

Mining and milling No data by region 5.3 101

Power plants 9.9 103 1.5 101 2.8 101 1.2 100 2.6 101 6.8 101
Reprocessing 5.8 100 7.6 100

TOTAL LOCAL AND REGIONAL COMPONENTSc 9.9 103 1.5 101 3.3 101 1.2 100 2.6 101 1.3 102
 GLOBAL COMPONENT GLOBALLY-DISPERSED RADIONUCLIDES

Collective dose (man Sv)

Source and integration time
Whole world population
Mining and milling integrated to
100 years
500 years

10 000 years

Power plants integrated to

100 years 5.3 102
500 years 1.0 103

10 000 years 5.0 103

Reprocessing integrated to
100 years 2.6 102

500 years 4.9 102

10 000 years 2.4 103

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 167

a
Local and regional components of the collective doses (due to the first pass) are integrated to 100 years. The local and regional components are explained in chapter III. Per caput values are not given in the table, but are
discussed in the text.
b
For all tables in the annex, the calculations resulting from assessments are made to full precision and any discrepancies in the final sum of numbers in the tables are due to rounding.
c
Total of the local and regional components due to mining and milling, power plants and reprocessing from the first-pass. The totals for each region include only discharges from power plants, except for Europe, which
also has a value for reprocessing.
Table 16. Summary of collective doses to the public due to mining and milling, electricity generation from nuclear power reactors, and fuel reprocessing (in 2010)

168 UNSCEAR 2016 REPORT

normalized to electricity generated

Discharge type and source Collective dosea per unit of electricity generated (man Sv/(GW a))
Africa Asia and Pacific Europe Latin America North America World-averageb
LOCAL COMPONENT ATMOSPHERIC DISCHARGES
Mining and milling No data by region 4.1 102

Power plants

Inland 0 8.0 101 2.2 102 2.2 101 2.5 102 2.9 102
Coastal 5.6 103 4.3 102 3.2 102 6.8 103 8.5 103 3.3 102
Reprocessing 1.2 102 8.6 103

REGIONAL COMPONENT ATMOSPHERIC RELEASES

Mining and milling No data by region 1.7 101

Power plants
Inland 0 2.0 100 7.2 102 1.7 101 6.5 102 8.9 102

Coastal 1.1 103 1.1 101 9.4 102 1.3 102 1.1 102 7.2 102
Reprocessing 4.0 102 1.8 102

LOCAL AND REGIONAL COMPONENTS (AQUATIC DISCHARGES)

Mining and milling No data by region

Power plants

Inland 0 9.3 101 1.5 101 1.1 100 2.0 101 1.9 101
Coastal 6.3 105 4.1 104 1.8 103 1.6 104 9.6 105 3.3 103
Reprocessing 3.0 103 2.1 103

LOCAL AND REGIONAL COMPONENTS (ATMOSPHERIC AND AQUATIC DISCHARGES)

Mining and milling No data by region 2.2 101

Power plants 6.7 103 2.4 101 2.1 101 3.9 101 2.6 101 2.3 101
Reprocessing 5.5 102 2.9 102

TOTAL LOCAL AND REGIONAL COMPONENTSc 6.7 103 2.4 101 2.6 101 3.9 101 2.6 101 4.3 101
 GLOBAL COMPONENT GLOBALLY-DISPERSED RADIONUCLIDES

Source and integration time Collective dose per unit of electricity generated (man Sv/(GW a))
Whole world population
Mining and milling integrated to
100 years
500 years

10 000 years

Power plants integrated to

100 years 1.8 100
500 years 3.4 100

10 000 years 1.7 101

Reprocessing integrated tod

100 years 1.2 100

500 years 2.1 100

10 000 years 1.0 101

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 169

a
Local and regional components of the collective doses (due to the first pass) are integrated to 100 years. The local and regional components are explained in chapter III. Per caput values are not given in the table, but are
discussed in the text.
b
World-average is used to qualify the calculations and data that are intended to represent a value averaged across the whole world. In this case the world average is the average of the population in each annulus for all
nuclear sites for which data are available (annex A). Similarly, consumption rates for terrestrial and aquatic foods are average values for the world population. The world-average values can be used for comparative
purposes including for comparison with the previous versions of the methodology for estimating public exposures due to radioactive discharges (annex A).
c
Total of the local and regional components due to mining and milling, power plants and reprocessing from the first-pass. The totals for each region include only discharges from power plants, except for Europe, which
also has a value for reprocessing.
d
The normalized values were calculated as described in paragraphs 65 and 66.
170 UNSCEAR 2016 REPORT

Table 17. Comparison of the worldwide collective dose, and associated collective dose normalized
to electricity generated, from mining and milling, power plant operation and reprocessing

Nuclear fuel cycle Collective dose Normalized collective dosea

(man Sv) (man Sv/(GW a))
Local and regional component from mining and 130 0.43
milling, power plants and reprocessing (first pass)b

Local and regional component (integrated to

100 years) plus global component integrated to
100 years 910 3.0
500 years 1 700 5.5
10 000 years 7 600 25
a
These values are averages for the whole world. The world average is the average of the population in each annulus for all
nuclear sites for which data are available (annex A). Similarly, consumption rates for terrestrial and aquatic foods were average
values for the world population. The results can be used for comparative purposes including for comparison with the previous
versions of the methodology for estimating public exposures due to radioactive discharges (annex A).
b
Local and regional components of the collective doses (due to the first pass) are integrated to 100 years. The local and regional
components are explained in chapter III.

Table 18. Radionuclide contributions to the local and regional components of collective doses to
the public from reprocessing integrated to 100 years

Collective dose per unit of electricity Collective dose

Radionuclide generated (man Sv/(GW a)) (man Sv)
Atmosphere Aquatic Total Atmosphere Aquatic Total
3
H 1.3 103 3.6 106 1.3 103 7.1 101 8.0 104 7.1 101
14
C 2.0 102 1.7 103 2.2 102 4.0 100 5.4 101 4.5 100
41
Ar 0 0 0 1.7 103 0 1.7 103
54
Mn 0 8.4 107 8.4 107 0 1.7 104 1.7 104
58
Co 0 1.8 108 1.8 108 0 3.6 106 3.6 106
60
Co 3.4 107 6.1 105 6.1 105 1.6 104 3.0 102 3.0 102
85
Kr 4.5 103 0 4.5 103 1.1 100 0 1.1 100
90
Sr 0 5.8 107 5.8 107 3.0 102 9.8 104 3.1 102
106
Ru 4.2 107 9.0 105 9.0 105 1.2 103 2.8 102 2.9 102
129
I 8.3 104 2.3 105 8.5 104 5.1 101 5.3 103 5.2 101
131
I 9.6 107 1.3 109 9.6 107 1.1 103 2.6 107 1.1 103
135
Xe 0 0 0 3.6 104 0 3.6 104
134
Cs 2.0 107 1.7 106 1.9 106 3.9 105 8.1 104 8.5 104
137
Cs 1.9 107 1.7 105 1.7 105 6.5 103 1.8 102 2.5 102
239
Pu 3.7 106 8.3 105 8.7 105 1.6 101 2.5 101 4.2 101
241
Am 0 5.7 105 5.7 105 1.2 102 2.0 101 2.2 101

Total 2.7 102 2.1 103 2.9 102 6.5 100 1.1 100 7.6 100
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 171

Table 19. Local and regional components of collective dose integrated to 100 years summed over
all reactor types

Africa Asia and Europe Latin North World

Pacific America America

Collective dose (man Sv) 9.9 103 1.5 101 2.8 101 1.2 100 2.6 101 6.8 101

Figure IV. Contribution of radionuclides to the local and regional components of collective dose to
the public from reprocessing integrated to 100 years and normalized to electricity generation

See data in table 18

172 UNSCEAR 2016 REPORT

Figure V. Collective doses from globally-circulating radionuclides per unit of electricity generated
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 173

Figure VI. Local and regional components of the collective doses to the public due to total estimated
discharges from nuclear power plants (2010) integrated to 100 years, by UNEP region and by reactor type

4. Uranium enrichment and fuel fabrication

74. The discharges from the processes of conversion, uranium enrichment and subsequent fuel
fabrication needed for different reactor types have generally been determined in earlier studies to be
relatively small, consisting mainly of radionuclides within the decay chains of uranium isotopes. The
collective dose to the public estimated on the basis of the electricity generated, derived in the
UNSCEAR 2000 Report [U9] and re-used in the UNSCEAR 2008 Report [U11], were used here to
obtain an updated value based on the 2010 electricity generation from nuclear power. This value for
uranium enrichment and fuel fabrication facilities is 0.003 man Sv/(GW a). The Committee earlier
estimated the average annual collective dose to the public from all plants for the period 19982002 to
be 0.8 man Sv [U11]. Updating this value to the electricity generated from nuclear power in 2010 from
table 2, the average annual collective dose to the public due to uranium enrichment and fuel fabrication
in the year 2010 is about 0.9 man Sv.
174 UNSCEAR 2016 REPORT

5. Solid waste disposal

75. Solid wastes arise at various stages in the nuclear fuel cycle. These include low- and intermediate-
level wastes, mainly from reactor operations, high-level wastes from fuel reprocessing, and spent fuel
for direct disposal. The activity concentrations of these wastes can range from just above natural
background levels, as in mill tailings, to much higher levels, such as in spent reactor fuel.

76. Low- and intermediate-level wastes are generally disposed of by shallow burial in trenches or
concrete-lined structures, but more advanced disposal sites also exist. Before disposal, all such material
is manipulated and transported, which gives rise to both occupational and public exposures. For the
nuclear fuel cycle, doses to members of the public from the transport of radioactive material have been
earlier estimated using the factor of 0.1 man Sv/(GW a) for the collective dose per unit of electricity
generated [U11]. Wastes from the nuclear fuel cycle also include large quantities of depleted uranium
from uranium enrichment operations [U18]. High-level wastes and spent fuel are presently retained in
interim storage facilities until adequate methods for disposal have been devised and disposal sites
selected [U11].

77. Estimates of the doses due to solid waste disposal have been based on the projected eventual
migration of radionuclides from the burial site into groundwater. The collective dose due to low- and
intermediate-level waste disposal normalized to electricity generated is estimated to be approximately
0.5 man Sv/(GW a), due almost entirely to 14C. The average worldwide per caput annual effective dose
would be about 1 nSv per year of practice [U11].

C. Occupational exposure for the nuclear fuel cycle

78. The assessment of occupational exposures for the nuclear fuel cycle considers the practices of
mining, milling, uranium enrichment and conversion, fuel fabrication, reactor operation, fuel
reprocessing and research.

79. Collective doses due to occupational exposure for each practice in the nuclear fuel cycle that were
reported in the UNSCEAR 2008 Report [U11] for the years 20002002 were used as the starting point
to estimate the 2010 occupational exposures. This approach was taken since data were not readily
available on the total number of, or the average annual effective dose to, monitored workers worldwide
in 2010. As reported in the UNSCEAR 2008 Report, the Committee used the amount of electricity that
can be generated from each practice (i.e. how much electricity can be generated from the uranium that
is mined, milled, converted, and so on) to normalize collective doses to the electricity generated.

80. Table 20 summarizes the data on occupational exposure for the nuclear fuel cycle in the period
20002002 that are relevant for estimating occupational exposure for the nuclear fuel cycle in 2010.

81. The ratio from the 20002002 data for the average annual collective dose per unit of electricity
generated was combined with the electricity generation for 2010 to estimate the annual collective dose
for 2010. Similarly, the ratio obtained from 2002 for the average annual collective dose per unit mass of
uranium mined, which can be further transformed to the collective dose per unit of electricity generated
is assumed appropriate for the mining process. Actual data from the Canada, the United Kingdom and
the United States for 2010 were obtained on the number of monitored workers and their average annual
effective dose, and were compared with the respective country values for 20002002 in the UNSCEAR
2008 Report [U11]. These comparisons showed reasonable agreement with the scaled average annual
effective dose, giving some confidence in the assumptions used.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 175

Table 20. Worldwide average annual individual and collective doses to workers due to the commercial
nuclear fuel cycle

From table 72 in the UNSCEAR 2008 Report [U11]; for the years 20002002a

Practice Monitored Average annualb Average annualb collective Average annualb

workers collective dose dose per unit of electricity effective dose to
(thousands) (man Sv) generated monitored workers
(man Sv/(GW a)) (mSv)
Mining 12 22 0.1 1.8
Milling 3 3 0.02 1.0

Enrichment 18 2 0.02 0.1

Fuel fabrication 20 31 0.1 1.6
Reactor operation 437 617 2.5 1.4

Reprocessing c,d
76 68 0.9
Research 90 36 0.1 0.4

Total 656 779 2.8 1.2

a
Some values in the table have been corrected since table 72 was published in the UNSCEAR 2008 Report [U11]; for the years
20002002.
b
The words average annual used in the table denote the annual dose averaged over the 2000-2002 year span.
c
Also includes the reprocessing of some fuel associated with military application.
d
The average annual collective dose per unit of electricity generated was not possible to obtain, because there were no data
readily available on the amount of electricity generated from the reprocessed fuel.

82. Table 21 shows the current results for 2010 for the worldwide average annual collective dose
compared to the 20002002 results. The procedure for estimating the values for each practice for 2010
is explained below.

Uranium mining. The amount of uranium ore mined was assumed equal to the amount
produced [U11]. Total uranium production in 2010 for 16 countries is shown in table 4 [O8].
Values were summed for all countries to give 54,100 t of uranium. Of this amount, 31,900 t
was produced from open pit and underground uranium mines and 22,200 t was produced from
in-situ leaching (ISL) processes. Assuming the same collective dose per unit mass of uranium
extracted of 0.623 man Sv/kt as for 20002002 [U11], then the collective dose (rounded) is
20 man Sv. This value excludes uranium produced from ISL.

Uranium milling. As for uranium mining, the total uranium production in 2010 was taken from
table 4 [O8] and assumed to equal the total amount of uranium milled, 54,104 t. Applying the
same simplifying assumptions used in the UNSCEAR 2008 Report [U11] (i.e. that all milled
uranium is used in LWRs and that the uranium requirement is 220 t/(GW a) ([U11]; table 18-A)),
the equivalent amount of electricity for 2010 is 246 GW a. The collective dose (rounded) for
2010 is then 5 man Sv. While ISL extract is not milled, it does go through some treatment to
produce yellowcake. Thus the full amount of uranium mined, 54,100 t, is used here.

Uranium conversion and enrichment, and fuel fabrication. UNSCEAR 2008 Report (table 15)
shows the worldwide installed capacity for fuel cycle installations taken from the IAEA
Nuclear Fuel Cycle Information System, NFCIS. Compiling the data from NFCIS for 2010
and comparing with table 15 from the UNSCEAR 2008 Report shows little has changed in the
176 UNSCEAR 2016 REPORT

capacity of these processes. All three processes show decreased capacities of between 8% and
9%. Using the assumption that these practices give the same collective dose per unit of
electricity generated in 2010 as in 2002, the average annual collective dose for 2010 in
table 21 has been adjusted to a lower value by 8%.

Reactor operation. The normalized average annual collective dose for 20002002 was
2.5 man Sv/(GW a). Instead of using this value directly, as explained above, data from the
OECD/NEA [O4] indicated a general decrease in the average annual collective doses. The
reported average annual collective dose for operating nuclear power plants fell by 20%
between 2002 and 2010, attributed to additional operating experience and the global exchange
of best radiation protection practices. Using this, the normalized average annual collective
dose for 2010 can be taken as 2.0 man Sv/(GW a). Assuming the total worldwide electricity
generation in 2010 is 314 GW a, an average annual collective dose of 628 man Sv was
calculated, about the same as that reported for 2002, as shown in table 21.

Reprocessing. In the absence of updated data on occupational exposures during fuel

reprocessing in 2010, it was assumed that reprocessing activities have remained more or less
constant and the same collective dose was assigned as previously, i.e. no change. Data from
UNSCEAR [U11] were not presented in terms of dose per unit of electricity generated.

Research. In the absence of updated data on occupational exposures obtained during research
for 2010, it was assumed that research activities have remained more or less constant and the
same collective dose was assigned as previously. However, it should be noted that a slight
downward trend in collective dose was observed over the previous two time periods in
UNSCEAR [U11]. Data from UNSCEAR [U11] were not presented in terms of dose per unit
of electricity generated.
Table 21. Worldwide levels of exposure of workers due to the commercial nuclear fuel cycle for 2010 and 20002002

Practice Average annual collective effective Average annual collective dose Average annual collective dose per unit of Remarks
dose 20002002 (man Sv) 2010 (man Sv) electricity generated (man Sv/(GW a))
(From [U11] for years 20002002) (Estimates from this study) (From [U11] for years 20002002)

Mining 22 20 0.1 U production 2010=54.1 kt; of this 31.9 kt was the amount of
U produced excluding ISL production (35 ktU average in
19982003)

Milling 3 5 0.02 246 GW a calculated using same simplifying assumption

from UNSCEAR 2008 Report, table 18-A, that all milled
uranium was used in LWRs, and that uranium requirement
was 220 t/(GW a). Doses resulting from subsequent
remediation activities were not accounted for

Enrichment 2 2 0.02 8% lower capacity than in 20002002 (not noticeable

because numbers are rounded)

Fuel fabrication 31 29 0.1 8% lower throughput than in 20002002

Reactor operation 617 628 2.5 314 GW a total electricity generated by nuclear power plants
in 2010 [I15, I16]

Reprocessing 68 68 Assumed constant collective dosei.e. no change. Data were

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 177

not presented [U11] in terms of dose per unit of electricity
generated

Research 36 36 0.1 Assume constant collective dosei.e. no change

Total 779 788 2.8

178 UNSCEAR 2016 REPORT

D. Decommissioning
83. When a power company decides to close a nuclear power plant permanently, the facility must be
decommissioned by safely removing it from service and reducing residual radioactivity to a level that
permits release of the property and termination of the operating licence. As more commercial nuclear
power reactors reach the end of their operating licence, there is a commensurate increase in
decontamination and decommissioning activities that involve radiation exposure. These activities
include decontamination of structures and components, dismantling of components and demolition of
buildings, remediation of any contaminated ground, and removal of the resulting waste. As of 2015,
156 commercial and prototype reactors located in 19 countries with a total of 60.9 GW(e) capacity were
permanently shut down [I8]. Most of these reactors were PWRs, GCRs and BWRs. The current
assessment concentrated on occupational exposures during the decommissioning process.

84. The decommissioning process begins when a power company decides to permanently cease
operations. The operator of a nuclear facility may choose between three decommissioning strategies:
(a) immediate dismantling, (b) deferred dismantling after a safe storage period, and (c) entombment of
the facility. Under immediate dismantling, decommissioning activities begin shortly after the
permanent cessation of operations. This strategy implies prompt completion of the decommissioning
project (approximately 10 to 20 years) and involves the removal of all radioactive material from the
facility to another new or existing licenced facility.

85. Under deferred dismantling (sometimes called safe storage, safe store or safe enclosure), a nuclear
facility is maintained and monitored in a condition that allows the radioactivity to decay and thus
reduce occupational exposure. Parts of a facility containing radioactive contaminants are safely stored
and maintained for upwards of 40 to 60 years until they can subsequently be decontaminated to levels
that permit parts of the facility to be dismantled and released for unrestricted use.

86. Under entombment, radioactive contaminants are encased in a structurally long-lived material
such as concrete until radioactivity decays to a level permitting the unrestricted release of the facility,
or release with restrictions imposed by the regulatory body.

87. As of December 2015, the International Atomic Energy Agency/Nuclear Energy Agency
Information System on Occupational Exposure (ISOE) database (http://www.isoe-network.net/)
contained data on doses to workers from 84 reactors that were shut down, in some stage of
decommissioning or fully decommissioned. These reactor units were generally of different types and
sizes, and at different phases of their decommissioning programmes. At least 15 reactors had been fully
dismantled, over 50 reactors were being dismantled, over 50 reactors were in deferred dismantlement,
three had been entombed, and for others the decommissioning strategy was not specified yet.

88. Table 22 provides the average annual collective dose for occupational exposure per unit for up to
71 permanently shut down reactors by country and reactor type for 20082013, based on data recorded
in the ISOE database and supplemented by individual country reports. Different decontamination and
decommissioning strategies were being employed and each facility shown in the database was in a
different stage of decommissioning, making definitive trends difficult to deduce from this information.
Depending on the phase of decommissioning, there may be little or no worker exposure one year and
the next year collective dose to workers may rise as much as 100-fold (adapted from [O5]). Considering
the average collective doses for each of the reactor types, however, and the total average for all the
units, the average values demonstrate some stability over these six years, with the total average for all
units being about 0.06 man Sv per year and reactor.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 179

89. Table 23 shows data from the whole decommissioning period for nuclear power plants that
have been immediately decontaminated and decommissioned. The dose values given are the
integrated doses for the whole decommissioning process. There may be some additional dose due to
spent fuel storage until a permanent repository is opened, but these occupational exposures can be
considered negligible compared to active decommissioning. For example, for five of the
decommissioned units that reported to the REIRS database (Radiation Exposure and Information
Reporting Systems) ([U24]; table 3.1), collective doses from occupational exposure due to spent fuel
storage ranged from 0 to 1.86 103 man Sv annually. The values for electricity supplied were the
total integrated electricity generation from the time the unit went on the grid until it ceased operation.

90. The relative quantity needed for the comparative study conducted here was an estimate of the
collective doses from occupational exposure for decommissioning, integrated over the period of
decommissioning. The fact that the annual collective dose varies from year to year is not relevant if a
reasonable average can be obtained. Although limited, the Committee considered that the data on
integrated doses for nuclear reactors shown in table 23 were sufficient for its use here.

91. The average total collective dose associated with the strategy of immediate decontamination and
decommissioning of the commercial nuclear power plants shown in table 23 was less than 10 man Sv
per reactor. While the alternative decommissioning strategies, deferred dismantling and entombment,
may result in some reduction in the total collective dose per reactor from decommissioning, it may be
another 10 to 20 years before any definitive conclusions can be documented.

92. Table 23 also shows the collective dose from occupational exposure normalized to the total
integrated electricity supplied for each of the reactors, with an average for these 8 reactors of
1.80 man Sv/(GW a).
 180 UNSCEAR 2016 REPORT
Table 22. Number of units and average annual collective dose from occupational exposure per reactor by country and reactor type for definitely shutdown reactors, 20082013

The columns headed dose give the average annual collective dose from occupational exposure in man-millisieverts per reactor (man mSv)

2008 2009 2010 2011 2012 2013

No. Dose No. Dose No. Dose No. Dose No. Dose No. Dose
PWR France 1 23.2 1 62.1 1 117.2 1 264.1 1 275.6 1 189.3

Germany 5 160.0 5 128.0 3 278.6 3 126.3 3 114.4 4 77.7

Italy 1 1.1 1 1.7 1 3.2 1 1.8 1 3 1 5.2

Spain 1 134.7 1 244.0 1 53.0 1 190.0 1 307.9 1 468.9

United States 10 7.1 8 1.5 8 2.4 6 49.4 6 127.1 12 47.3

Average 18 57.2 16 60.0 14 73.5 12 94.4 12 141 19 81.1

VVER Bulgaria 4 31.0 4 29.4 4 11.3 4 9.2 4 10.1 4 3.3

Germany 5 27.0 5 20.0

Russian Federation 2 78.0 2 84.0 2 77.6 2 66.3 2 79.2 2 49.6

Slovak Republic 1 48.2 2 106.0 2 12.4 2 10.0 2 4.3

Average 12 38.6 13 46.0 8 28.2 8 23.7 8 25.9 6 18.7

BWR Germany 3 179.0 3 138.0 1 427.1 1 289.5 1 88.2 1 72.0

Italy 2 29.1 2 61.8 2 60.3 2 15.1 2 18.4 2 34.2

Japan 2 123.8 2 48.4 2 41.2 2 64.2

The Netherlands 1 0.3 1 0.6 n/a n/a 1 10

Sweden 2 39.1 2 27.0 2 6.2 2 27.2 2 20 2 3.45

United States 3 13.4 4 5.1 5 21.8 4 30.7 4 59.4 5 55.7

Average 11 64.9 12 51.1 12 76.3 12 50.3 11 44.1 12 46.2

 2008 2009 2010 2011 2012 2013
No. Dose No. Dose No. Dose No. Dose No. Dose No. Dose
GCR France 6 2.8 6 8.8 6 1.3 6 2.4 6 7.4 6 8.2

Germany 2 13.0 2 17.0

Italy 1 2.9 1 0 1 1.7 1 10.4 1 0.2 1 2.2

Japan 1 20.0 1 20.0 1 50 1 50 1 70 1 10

Spain 1 0 1 0 1 0

United Kingdom 16 48.0 16 42.0 16 55 16 49 19 56 19 57.3

United States 1 0 1 0

Total number, 26 32.1 26 30.0 24 39.1 25 34.4 28 42.1 28 41.1

average dose

HTGR Germany 1 0 1 0 1 0 1 0 1 0

FBR United States 1 80.14 1 77.9 1 294.9 1 2 1 0.1

LWGR Lithuania 1 188.4 1 144.7 2 236.2 2 304.8 2 264.9 2 304.8

LWCHWR Japan 1 431.3 1 114.6 1 111.6 1 126.6 1 148.8 1 134.1

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 181

All Units Average 70 52.3 71 47.7 63 59.9 62 61.5 64 66.3 71 57.8

This table is adapted from table 4 in the 2012 ISOE annual report [O5]and table 5 from the 2010 ISOE annual report [O4]. Additional information was obtained from the individual country reports and the ISOE database,
and the United States Nuclear Regulatory Commission Radiation Exposure and Information Reporting System (REIRS) [U24] database for United States definitely shut reactors. Data for 2013 were obtained from the
2013 ISOE country reports [O7] and the ISOE database [I20].
Dashes () represent missing or partial data in the data source. The acronyms for the reactor type used in tables 22 and 23 differ somewhat from those in the other tables in this annex (see paragraph 58). The acronyms in
this table and table 23 are the designations used by the Nuclear Energy Agency.
The following explains the meaning of the acronyms that differ from previous tables:
LWCHWR (Light Water Cooled Heavy Water Reactor) also known as a CANDU reactor is the Canadian designed system. IAEA would refer to this as a PHWR.
VVER (water, water, energetic reactor) is a Soviet designed PWR.
HTGR, (also called HTGC, high temperature gas cooled) reactor. More of a prototype reactor but nuclear power plants include Peach Bottom and Fort Saint Vrain in the United States; Dragon reactor in the United
Kingdom, and THTR-300 and AVR in Germany.
182 UNSCEAR 2016 REPORT

Table 23. Collective dose, and collective dose normalized to electricity supplied, due to the complete
decommissioning process of immediately decontaminated and decommissioned commercial nuclear
power plants

Reactor gross electrical capacity and total electricity supplied were obtained from the IAEA PRIS database [I7]. Collective
doses were obtained from the ISOE database [I20] and REIRS reports [U24]

Gross Collective Electricity Normalized

Reactor
Country Plant Capacity dose supplied collective dose
type
(MW(e)) (man Sv) (GW a) (man Sv/ GW a)

PWR United States Haddem Neck 603 8.457 12.1 0.70

Maine Yankee 900 6.195 13.6 0.46

Rancho Seco 917 2.345 5.1 0.46

San Onofre 1 456 3.002 5.8 0.52

Trojan 1 155 2.973 9.6 0.31

Yankee Rowe 180 6.467 3.9 1.66

AVERAGE 4.907 8.34 0.69

BWR United States Big Rock Point 71 5.703 1.45 3.93

HTGR United States Fort St. Vrain 342 3.961 0.62 6.39

All units Average 4.900 1.80

V. RADIATION EXPOSURES ARISING FROM ELECTRICITY

GENERATION BY COMBUSTION OF FOSSIL FUELS

A. Coal

1. Introduction
93. Coal is a family name for a variety of solid organic fuels and refers to a whole range of
combustible sedimentary rock materials spanning a continuous quality scale. For convenience, this
continuous series is often divided into two main categories: hard coal (which includes anthracite and
bituminous coal) and brown coal (which includes sub-bituminous coal and lignite). The International
Energy Agency [I15] makes use of two broad categories of coal; hard coal as having a gross calorific
value not less than 23.9 GJ/t (5,700 kcal/kg) and brown coal with a gross calorific value less than
23.9 GJ/t (5,700 kcal/kg). Often coal data are presented in tonnes of coal equivalent (TCE) where one
tonne of coal is equivalent to 7 million kilocalories. This description standardizes the carbon content
and heat value of a particular type of coal. The International Energy Agency estimates that there are
equal recoverable global reserves of both hard coal and brown coal [I14].
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 183

94. Combustion of coal for electricity generation is the largest contributor to worldwide electricity
generation (figure I). Coal has maintained a share of about 40% of the total electricity generated for
some decades, while the total electricity generation from coal and other sources has increased steadily
as shown in figure I [I15]. The UNEP geographic regionAsia and the Pacificaccounted for 57% of
the total electricity generation from coal in 2010, with China accounting for about 70% of this [U17]. In
contrast, the geographic regionWest Asiaused no coal for electricity generation in 2010 (table 2).

95. Electricity generation is not the only end-product of coal combustion, others being for example
steel production and cement manufacturing. The largest 10 coal-producing countries are shown for the
years 20082012 in table 24. China was the largest producer with a 30% increase in production over
this five-year period. The total world primary coal production for this period had a 16% increased
production in 2012 compared with 2008, and about a 60% increased production during the 10-year
period since 2002 [U17].

96. Coal contains naturally occurring radionuclides from the uranium and thorium series (figure III)
and potassium-40. The concentration of naturally occurring radioactive material in coal depends on the
characteristics of the geological formation of the coal seams where it originated. Table 25 provides a
representative overview of the range in concentrations, where large variations in the activity
concentrations can be noted.

Table 24. Total primary coal production 20082012 (million metric tonnes, Mt)

Source: United States Energy Information Administration, International Energy Statistics 20082012, open source data [U17]

Country 2008 2009 2010 2011 2012

China 2 811 2 995 3 230 3 518 3 645

United States 1 063 975 984 994 922

India 517 558 562 575 589
Australia 392 408 424 402 421

Russian Federation 305 276 322 322 354

South Africa 252 249 255 253 259
Indonesia 249 291 325 360 443
Poland 143 135 133 139 144
Kazakhstan 111 101 111 116 126
Colombia 74 73 74 86 89

World 6 778 6 896 7 257 7 660 7 881

184 UNSCEAR 2016 REPORT

Table 25. Example ranges and/or averages of radionuclide activity concentrations in coal (Bq/kg)
(table VII from IAEA [I2])
Country U
238
Th
230 226
Ra 210
Pb Po
210
Th
232 228
Ra 40
K

Australia 8.547 2168 1924 2033 1628 1169 1164 23140

Brazila 72 72 72 62 62

Egypt 59 26 8 8

Germany 10145 1063 10700

32a 21a 225a

Lignite <158 <158 <4220

(Former 10a 8a 22a

Democratic
Republic of
Germany)

Greeceb 117390 44206 59205 941

Hungary 20480 1297 30384

Italyc 233 184 21815

Poland <159 <123 <785

18d 11d

Romania <415 <557 <510 <580 <170

80a 126a 210a 262a 62a

United 719 8.525.5 7.821.8 719 55314

Kingdom

United States 6.373 8.959.2 12.277.7 3.351.8 3.721.1

a
Average.
b
Lignite.
c
Lignite, average.
d
Average for all coal seams.

97. When coal is combusted the majority of the non-combustible matter remains in the bottom and fly
ash. Because radionuclides are present in the mineral constituents, they also tend to remain in the ash,
where the concentration of the radionuclides becomes enhanced [I2, S4, Z1]. However, radon is a noble
gas and all the radon present in the coal is emitted through the flue-gas stack. As coal-fired power
plants have modernized, the cleaning and filtering systems have become increasingly efficient and
larger fractions of the particulate matter in the fly ash have been captured and removed from the
discharge to atmosphere. The coal ash that is collected is either recycled for beneficial use or it is
disposed of in landfills or impoundments. 6

98. The components of the coal cycle that can lead to radiation exposures of the public are
(a) discharges to atmosphere of radon from coal-mining activities and other wastes produced during
mining, (b) radioactive discharges from the operation of coal-fired power plants, and (c) radiation
exposures from the recycling and use of, or disposal of, coal ash. Occupational radiation exposures

6
Coal ash can be disposed as a coal slurry in impoundments, a dammed reservoir that contains the coal slurry.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 185

occur during (a) the mining for coal, and (b) work performed associated with power plant operation and
disposal operations. In earlier studies, occupational exposures from work performed at the power plant
were determined to be small [C5, G4, G5] in relation to the occupational radiation exposures incurred
during coal mining. For this reason, combined with the relatively small number of workers involved
compared to coal mining, this exposure source was not considered further in the current study. The
following sections consider the other exposure sources and pathways.

2. Public exposure

(a) Coal mining

99. Radon emissions from coal mining contribute to exposure of the public from the coal cycle.
Radon gas is emitted from the mines as a result of the coal mining activities, primarily through venting
of the mines. Although data that could enable the assessment of this impact have earlier been scarce,
new measurements on radon emissions from coal mining in China have become available. Radon
emissions from 23 large-, medium- and small-sized coal mines in eight different provinces in China
have been assessed, based on the monitoring results of radon activity in the mines. The large- and
medium-sized coal mines are equipped with ventilation systems that are described as good or
relatively good. The smaller, privately owned coal mines generally are poorly ventilated, with radon
concentrations 15 to 30 times higher than in the larger mines [L1]. The radon emissions have been
normalized to the unit of coal output and per unit of electricity generated for each coal mine and for
each size-class of coal mine. Typical values were obtained as shown in table 26 [W2].

Table 26. Radon emissions during one year from large-, medium- and small-sized coal mines in
China [W2]

Type of coal mine Coal output Number of Radon emissions Radon emissions per unit of
(t/a) coal mines per unit of coal electricity generated
output (Bq/t) (TBq/(GW a))a
Large-sized 1.6 109 7 1.9 105 0.57
Medium-sized 5.7 10 8
4 1.3 106
3.6
Small-sized 3.9 108 12 3.5 106 11

Typical value 9.3 105 2.8

a
From the data in table 26, obtained from the Chinese delegation to the Committee, the coal production per unit of electricity
generation could be calculated as between 2.8 and 3.1 106 t/(GW a).

100. The radioactive content of coal-mine tailings do not differ significantly from background
concentrations of natural radionuclides in soil; this situation is different from uranium mining and mill
tailings, which have higher concentrations of the natural radionuclides [U9]. This exposure pathway is
thus not considered further in this analysis.

101. The typical value for the radon emission (based on discharges of radon during one year)
normalized to electricity generated obtained from the Chinese data, 2.8 TBq/(GW a), has been used
with the electricity generation data from table 2 and the Committees revised methodology for
estimating public exposures due to radioactive discharges (annex A) to assess the radiological impact
from this source. The default population distribution (see section III.A) was used and the radon was
assumed to be released on its own with account taken of its short-lived progeny (218Po, 214Pb, 214Bi, and
214
Po) in estimating the doses from inhalation.
186 UNSCEAR 2016 REPORT

(b) Electricity generation from combustion of coal

102. Coal combustion in power plants results in the direct release of gaseous radionuclides, and the
production of ash with enhanced concentrations of natural radionuclides relative to those of coal [S4].
A fraction of the ash produced is released to atmosphere, dependent on the particulate collection
devices used in the power plant. Modern coal-fired power plants have more efficient particulate
collection devices than older coal-fired power plants.

103. To assess the doses due to atmospheric releases from coal-fired power plants, a representative
case was considered. The coal (e.g. hard coal) was assumed to have an energy equivalent of 24 GJ/t
with a power generation efficiency of 38% [I13]. Although the values of 238U concentrations shown in
table 25 show wide variations, an average 238U concentration of 20 Bq/kg was chosen, based on typical
values for Chinese coal reported as 1025 Bq/kg [I10]. China is the largest producer of coal worldwide,
so the value of 20 Bq/kg for the 238U concentration in coal was considered reasonable as a typical
average global value. Other studies have also found a normal range of 2025 Bq/kg observed globally
for good quality coal [S4, T1, U6]. A value of 20 Bq/kg for the 238U concentration in coal had also been
used in the UNSCEAR 1988 Report [U7]. It should also be noted that the resulting individual and
collective doses calculated by the revised methodology for estimating public exposures due to
radioactive discharges (annex A) are directly proportional to the 238U concentration in coal.

104. Two different release rates were evaluated to represent the discharge characteristics of older coal-
fired power plants and modern coal-fired power plants (table 27). The relative distribution among the
different radionuclides in table 27 for the older plants were taken from Hedvall and Erlandsson [H2]
while the same for the modern plants were based on Zeevaert et al. [Z1]. The key difference between
the two data sets is that 222Rn concentrations in the discharges are 100 times larger than 226Ra in older
plants and 1,500 larger than 226Ra in modern plants that have more efficient filtering systems. However,
in both cases the value for the discharge of lead and polonium particles has been chosen as twice that of
uranium and radium [C4, T1]. Because radon is a noble gas, it is unaffected by filters. As seen from
table 27, these releases are not in secular equilibrium. The distribution of older versus modern coal-
fired power plants in the various regions of the world is not well-characterized so the results from both
cases are included to illustrate the difference in the resulting doses.

105. The assessments presented in this annex using the Committees revised methodology for
estimating public exposures due to radioactive discharges (annex A) do not include dose contributions
from 40K or the 232Th decay chain. The absorption of 40K in humans is homeostatically controlled and
therefore any doses from 40K discharges have been neglected. The contribution of the 232Th decay chain
to exposures due to electricity generation from coal combustion has been studied by Zeevaert et al.
[Z1]. They found that the concentrations of the radionuclides in the 238U and 232Th decay chains in flue
gases discharged were similar. The largest contribution to individual doses came from deposition, with
consumption of food crops dominating. In this category, the contribution from the 238U decay series was
one order of magnitude larger than the contribution from the 232Th decay series. Although inclusion of
the 232Th decay series may lead to somewhat higher doses from electricity generation due to coal
combustion, it was excluded from the analysis in this annex.

106. Aquatic discharges from coal-fired power plants may lead to additional exposure of local
population groups, but this contribution has been found to be small and site-specific [L2, Z1], and were
therefore not considered here.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 187

Table 27. Releases to atmosphere (during one year) normalized to electricity generated from coal-
fired power plants, based on a representative coal containing 20 Bq/kg of 238U

Radionuclide Normalized release (TBq/(GW a))

Older coal-fired power plantsa Modern coal-fired power plantsb

222
Rn 0.07 0.07
210
Po 1.4 103 9.3 105
210
Pb 1.4 103 9.3 105
226
Ra 0. 7 103 4.7 105
234
U 0. 7 103 4.7 105
238
U 0. 7 103 4.7 105
230
Th 0. 7 103 4.7 105
a
Relative radionuclide distribution source: [H2].
b
Relative radionuclide distribution source: [Z1].

(c) Radon discharges from coal ash disposed in landfills

107. Coal ash is also referred to as coal combustion residuals (CCR) and coal combustion products
(CCP) depending on the industry and the country, and is one of the larger streams of industrial waste
generated in the United States [U21]. Coal ash includes a number of by-products from combustion of
coal including: fly ash; (furnace) bottom ash; fluidized bed combustion ash; boiler slag; semi-dry
absorption product; and flue gas desulphurization gypsum.

108. The coal ash collected from a power plant is either recycled for beneficial use or it is disposed of
in landfills or impoundments. Figure VII shows a time history of fly ash production and the fraction
used for construction in the period 19662010 in the United States [A1]. Approximately 40% was used
commercially, leaving the remaining 60% for disposal. This does not include bottom ash, but the
proportion sent for disposal was similar [A2].

109. Conventional back-filling or earthmoving operations are used to dispose of the coal combustion
residues (CCR) in dry or slightly moist conditions. Impoundments represent a wet disposal method
where the CCR is mixed with water at the power station and conveyed hydraulically through pipelines
to artificial lagoons where the slurry is discharged. According to the Electric Power Research Institute
[E3], about 60% of the disposed fly ash in the United States is managed dry in landfills, and 40% is
managed wet in impoundments, and it noted that there was a long-term trend toward increased use of
dry management practices [E3].

110. Earlier studies concluded that the individual doses to the public due to radioactive discharges
from deposited coal ash were below values relevant from an individual radiation protection viewpoint
(e.g. [U23]). The main reason was that the fraction of radon emanating from coal ash is small. This
pathway was however considered relevant in the current assessment for collective doses given that the
amount of coal ash produced is large and could be significant to the comparisons of collective doses.
Also, a similar pathway was considered from the uranium mill tailings in the nuclear fuel cycle
motivating a comparison between these two pathways.
188 UNSCEAR 2016 REPORT

Figure VII. Fly ash production and use in the United States (19662010) [A1]

(Blue = production, Red = used as a construction material in a variety of applications)

111. Method for treating the radon discharge from coal ash deposits. The amount of radon released to
atmosphere from coal ash depends on the activity concentration of radon in the ash, which is related to
the activity concentration of 226Ra, and the extent to which the radon produced in the ash is emitted to
atmosphere. A recent comprehensive review of radon emanation measurements for mineral, rock, soil,
mill tailings and fly ash found the radon emanation fraction (the ratio of the radon emitted to that
generated in the mill tailings (or ash deposits) per unit volume) for mill tailings (mostly uranium mill
tailings) was 0.17 and for fly ash was 0.03 [S1]. In another study [S6] the radon emanation fraction for
coal ash was found to be between 0.08 and 0.13 [S6]. In summary, the first cited study estimated a
radon emanation from coal fly-ash of about 20% of the radon emanation from uranium mill tailings,
and the second study estimated a value of about 50%. A representative average concentration of 226Ra
in coal ashes (fly- and bottom-ash) was taken as 100 Bq/kg [M2].

112. The 222Rn flux (Bq/(m2 s)) was studied using data from a repository for uranium residues in India
[I5] and a linear relationship was found between the radon flux and the activity concentration of 226Ra
in the source:

222
Rn flux (Bq/(m2 s)) = 8.3 10-4 kg/(m2 s) 226Ra concentration (Bq/kg) (1)

113. Because the relationship between the radon flux and the radon emanation factor is linear, the
constant in the above equation can be assumed to be proportional to the ratio of the radon emanation
factors for different sources. Therefore, for fly ash the constant above could be reduced to between
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 189

20 and 50% of its value (comparable with the values from the studies cited above for the radon
emanation from coal ash as compared to uranium mine tailings) to obtain an estimate of the 222Rn flux
from coal ash based on the 226Ra concentration in coal ash. A value of 20% was used in this analysis,
giving a linear constant in the above relationship of 1.7 104 kg/(m2 s).

114. According to a survey in the United States regarding the disposal of solid waste from coal-fired
power plants, a value for the area needed for disposing of solid wastes normalized to electricity
generated was reported as 211 m2/(GW h) [A9, F3]. A central value of 6 m2/(GW h) was used for this
analysis. Multiplying the radon flux during one year, obtained by using equation (1), with the linear
constant adjusted for coal ash and a 226Ra concentration of 100 Bq/kg, with the area needed for
disposing of solid wastes normalized to electricity generated, gave the source term for the annual
discharge of 222Rn from deposited coal ashes of about 0.03 TBq/(GW a).

115. A similar value was reached by an independent method. This assessment used an annual discharge
of 0.1 TBq/(GW a) as the source term for radon discharges from uranium mill tailings, one part of the
nuclear fuel cycle. Using the source term 0.1 TBq/(GW a) for radon discharge from uranium mill
tailings and reducing it to 20 to 50% gives an estimate for the annual discharge of radon from coal ash
of 0.020.05 TBq/(GW a), agreeing well with the value 0.03 TBq/(GW a) obtained from the alternate
assessment above.

116. The amount of ash placed in disposal was assumed to be 60% of the total ash produced in coal
power plants, based on data cited earlier from the United States and the United Kingdom. The larger six
coal-producing countries in table 24 are also countries with large land mass and therefore the value of
60% from the United States was assumed appropriate for use in this assessment. The remaining 40% of
the ashes were assumed to be used in a variety of building-related activities and the resulting radiation
exposures are discussed in subsection (e).

117. It was explained in paragraph 43 that radon discharges from one gigawatt-year of coal ash
deposits were assumed to continue for 100 years and that allowance had to be made in the methodology
for this continued discharge. This was done by multiplying the results for one years discharge by 100
to allow for the discharge continuing for 100 years (see annex A, paragraph 17). Taking into account
this multiplication by 100 years and that 60% of the total ash produced in coal-fired power plants was
assumed placed in disposals, gave a value for the radon discharge from coal ash deposit normalized to
electricity generation of 1.8 TBq/(GW a).

(d) Resultspublic exposures due to discharges from coal mining, coal-fired

power plants and ash deposits
118. The annual discharges to atmosphere obtained as described above were used to derive individual
and collective doses using the revised methodology for estimating public exposures due to radioactive
discharges (annex A). The results of the calculation of the characteristic individual doses per unit of
electricity generated integrated to 100 years due to atmospheric discharges from the three sources of
public exposures from the coal cycle are shown in table 28. These three sources are (a) discharges from
coal-fired power plants, (b) discharges from coal ash disposed in landfills and (c) discharges from coal
mining. The radon discharges from one gigawatt-year of coal ash deposits were assumed to continue for
100 years. The results are presented for both older coal-fired and for modern coal-fired power plants. If
the coal-fired power plants are older, then the individual doses from all three routes are similar, with
those from the discharges from power plants being the highest. For newer power plants, the doses due
to discharges from the plants are about 10% of those from the other two sources.
190 UNSCEAR 2016 REPORT

119. The values for each region that are shown in table 28 for the characteristic individual doses per
unit of electricity generation due to radon discharges from coal mining and from coal ash deposits
remain constant across the regions. This is because the discharges from mining and from ash deposits
are from radon gas and therefore only result in inhalation doses. However, the discharges from coal-
fired power plants include the other radionuclides from the 238U series, as shown in table 27, and result
in doses due to deposition and food consumption, which do have variations with geographical region.
As described above, the 232Th series was not considered in these calculations, and earlier assessments
by [Z1], imply that this contribution is less than 10% of the contribution from 238U.

120. The results also show a difference in the characteristic individual doses per unit of electricity
generated due to coal-fired power plant discharges between the older and the modern coal-fired plants
with the values for the modern plants being about 10% of those for the older plants. It could be
expected that the values resulting from discharges from ash would differ between the older and the
modern coal-fired power plants, but this is not reflected in table 28. This is because the actual
difference in the amount that remains in the ash for an older versus a modern coal-fired power plant is
very small, of the order of about 1%, and for both cases most of the radionuclides stay in the ash. This
difference therefore does not show up in the results.

121. The collective doses to the public integrated to 100 years due to electricity generation from the
coal cycle for each region and for the world as a whole are shown for the total electricity generated
from coal in the year 2010 in table 29. Both cases, assuming (a) that all coal-fired plants are older and
(b) that all coal-fired plants are modern, are compared. Considering the three discharge sources (from
mining, power plants and ash deposits), the contribution to the collective dose from the discharges from
coal-fired plants dominates for the older coal-fired plants, but is the smallest term for the modern plants
where the contribution from mining is largest. To facilitate comparison, table 30 compares only the
terms from the older versus modern coal-fired power plants.

Table 28. Summary of characteristic individual doses per unit of electricity generated integrated to
100 years due to atmospheric discharges from coal-fired power plant sources (mSv/(GWa))

Source Africa Asia and Europe Latin North

Pacific America America
From mining 2.4 104 2.4 104 2.4 104 2.4 104 2.4 104

From older coal plants 3.8 104 4.0 104 4.2 104 3.9 104 4.2 104

From modern coal plants 3.1 105 3.2 105 3.4 105 3.2 105 3.4 105

From asha 1.5 104 1.5 104 1.5 104 1.5 104 1.5 104

a
The annual discharges of radon from coal ash produced per unit of electricity generated are assumed to continue for 100 years.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 191

Table 29. Summary of collective doses to the public integrated to 100 years for the total electricity
generated from combustion of coal (in 2010) (man Sv)

Discharge type and source Africa Asia and Europe Latin North World
Pacific America America
LOCAL COMPONENTa ATMOSPHERIC DISCHARGES
Mining 1.0 100 7.1 101 8.6 100 3.8 101 3.4 100 7.0 101
Older coal plants 1.8 100 1.3 102 1.9 101 7.1 101 7.4 100 1.4 102
Modern coal plants 1.5 101 1.1 101 1.5 100 5.6 102 5.8 101 1.1 101

Ashb 6.7 101 4.6 101 5.5 100 2.4 101 2.2 100 4.5 101
Total (assuming all older coal plants) 3.5 100 2.5 102 3.3 101 1.3 100 1.3 101 2.5 102
Total (assuming all modern plants) 1.9 100 1.3 102 1.6 101 6.8 101 6.2 100 1.3 102

REGIONAL COMPONENTc ATMOSPHERIC DISCHARGES

Mining 4.5 100 3.0 102 3.7 101 1.6 100 1.5 101 3.0 102
Older coal plants 8.7 100 6.4 102 9.0 101 3.4 100 3.5 101 6.4 102
Modern coal plants 6.9 101 5.0 101 6.9 100 2.6 101 2.7 100 5.0 101
Ashb 2.9 100 2.0 102 2.4 101 1.0 100 9.4 100 1.9 102
Total (assuming all older coal plants) 1.6 101 1.1 103 1.5 102 6.0 100 5.9 101 1.1 103
Total (assuming all modern plants) 8.0 100 5.5 102 6.7 101 2.9 100 2.7 101 5.5 102

TOTAL LOCAL AND REGIONAL COMPONENTS ATMOSPHERIC DISCHARGES

Mining 5.5 100 3.7 102 4.5 101 2.0 100 1.8 101 3.7 102
Older coal plants 1.1 101 7.7 102 1.1 102 4.1 100 4.3 101 7.8 102
Modern coal plants 8.3 101 6.0 101 8.3 100 3.2 10-1 3.3 100 6.0 101
Ashb 3.5 100 2.4 102 2.9 101 1.3 100 1.2 101 2.4 102
Total (assuming all older coal plants) 2.0 101 1.4 103 1.8 102 7.4 100 7.2 101 1.4 103
Total (assuming all modern plants) 9.9 100 6.8 102 8.3 101 3.6 100 3.3 101 6.7 102
a
The local component of collective dose is calculated for people living between 0 and 100 km from the discharge point.
b
The discharges of radon from coal ash produced in 2010 from electricity generation are assumed to continue for 100 years.
c
The regional component of collective dose is calculated for people living between 100 and 1,500 km from the discharge point.
192 UNSCEAR 2016 REPORT

Table 30. Comparison of collective doses to the public integrated to 100 years from coal-fired
power plants given alternative plant design (manSv)

Collective doses to the public integrated to 100 years (man Sv)

Doses from coal-fired power Africa Asia and Europe Latin North World
plants Pacific America America
LOCAL COMPONENTa ATMOSPHERIC DISCHARGES
Older coal plants 1.8 100 1.3 102 1.9 101 7.1 101 7.4 100 1.4 102

Modern coal plants 1.5 101 1.1 101 1.5 100 5.6 102 5.8 101 1.1 101

REGIONAL COMPONENTb ATMOSPHERIC DISCHARGES

Older coal plants 8.7 100 6.4 102 9.0 101 3.4 100 3.5 101 6.4 102

Modern coal plants 6.9 101 5.0 101 6.9 100 2.6 101 2.7 100 5.0 101

TOTAL LOCAL AND REGIONAL COMPONENTS ATMOSPHERIC DISCHARGES

Older coal plants 1.1 101 7.7 102 1.1 102 4.1 100 4.3 101 7.8 102

Modern coal plants 8.3 101 6.0 101 8.3 100 3.2 101 3.3 100 6.0 101
a
The local component of collective dose is calculated for people living between 0 and 100 km from the discharge point.
b
The regional component of collective dose is calculated for people living between 100 and 1,500 km from the discharge point.

(e) Coal ash in building materials

122. Coal combustion products (CCPs) are used as ingredients in a variety of construction materials
including concrete, grouting, fill material, lightweight aggregate, road construction or maintenance
materials, and in soil stabilization [U1].

123. Fly ash and furnace bottom ash are the main CCPs and the extent to which they are utilized varies
by country. Approximately 40% of CCPs produced in the United States are utilized in construction
materials [A2], and the percentage utilization was around 45% in the European Union [E2] for fly ash
and bottom ash CCPs produced in 2010. China is the largest producer of electricity from combustion of
coal and consequently the largest producer of coal ash. According to Tang et al. [T2], China produces
proportionately more coal ash because it has a very low coal washing rate (51%). In 2010, the fly ash
utilization rate in China was over 65% [T2]. In contrast, in 2009 less than 20% of CCPs produced in
Australia were utilized [H3].

124. As noted previously, the activity concentration of natural radionuclides in coal ash is increased
relative to coal by a factor of 510. However, the concentration is again diluted when the fly ash is
mixed with other materials to form construction material. The presence of fly ash in building materials
may increase indoor exposures to gamma radiation from 226Ra, 232Th and 40K contained in the coal ash,
and exposures due to inhalation of radon emanating from the building construction material such as
concrete [D3, F1, T4].

125. The relationship between coal ash in construction material and exposure is not straightforward.
The use of coal ash has been shown to either increase, decrease or have essentially no effect on
exposures relative to traditional construction materials [T4]. This reflects differences in the source of
the coal, the natural radionuclide content of the traditional construction materials, and to the changes in
structural properties of the construction material. For example, the use of fly ash and furnace slag in
concrete has been shown to reduce radon exhalation rates because it alters the porosity and chemistry of
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 193

the construction material [D3, D4]. As of the year 2016 there are no consistent or internationally
accepted limits for radionuclides in building materials.

126. A population-weighted average annual effective dose of 0.41 mSv from indoor exposure to
gamma rays mainly determined by construction materials was reported by UNSCEAR [U9]. The results
for individual countries generally ranged between 0.3 and 0.6 mSv. A more recent analysis by de Jong
et al. [D3] calculated an annual effective dose of 0.32 mSv for the Netherlands, which is at the lower
end of the range given by UNSCEAR [U9]. In both evaluations, the results included dwellings where
fly ash was used in the construction material.

127. Collective doses either for geographical region or for the world as a whole due to the use of coal
ash in building materials have not been analysed further in this study. Although the individual doses
that have been assessed are very low, some contribution to the total collective dose can be expected
from the use of coal ash in construction materials. However, the various factors causing variability in
the exposure characteristics cause major uncertainty in the assessments.

3. Occupational exposures from coal mining

128. Previous UNSCEAR studies have estimated the collective doses due to occupational exposure
from coal mining. In the UNSCEAR 1988 Report [U7], based on exposure data from British coal
mining and the worldwide coal production rate, the worldwide collective dose was estimated as
2,000 man Sv. The UNSCEAR 2000 Report [U9] updated this value to an estimated 2,600 man Sv. The
UNSCEAR 2008 Report estimated the worldwide collective dose to workers from coal mining in
2002-2004 at 16,560 man Sv in a year [U11]. Most of the data on occupational exposure from coal
mining that were used in the UNSCEAR 2008 Report came from China. A study on radiation levels in
China reported the annual average collective dose to underground coal miners in 20032004 as
16,500 man Sv [L1]. In addition, occupational exposures from coal mining in the United Kingdom in
2002 were reported in the UNSCEAR 2008 Report [U11] as 3 man Sv for 5,000 coal miners and an
average individual dose of 0.6 mSv.

129. The worldwide collective dose to coal miners in 2010 could be estimated from the 2002 collective
dose value assuming that the increase in electricity generation from coal in the period 20022010 was
proportional to the increase in occupational exposure due to coal mining for the same time span. This
assumed that the same collective dose normalized to electricity generation that was assessed for
2002-2004 [U11] was applicable for 2010 and that the efficiency of coal mining remained the same, i.e.
the same number of coal miners was needed to obtain the same amount of coal. The electricity
generated from coal-fired power plants in the reference year of 2010 was 992 GWa (table 2). The
corresponding number for the period of 20022004 considered in the UNSCEAR 2008 Report was
720 GWa [I12]. This gave a rounded worldwide estimate of the collective dose to coal miners of
23,000 man Sv, as a first approximation.

130. Table 31 shows results from a Chinese study [L1], together with data on the number of coal
miners in three categories of mines in China for 2010 [W2]. No new values of the average annual
effective dose to coal miners have been published since the 2010 values in table 31.

131. Chinese coal production between 2004 and 2010 increased by approximately 50%. During this
period, there was an increase in the number of underground miners working at medium- and large-sized
coal mines and a significant decrease in the number working in small-sized township and privately-
owned coal mines [W2]. The last column in table 31 uses the average annual effective dose from 2004
and the number of coal miners in 2010 to estimate the collective dose in 2010 in China for the three
194 UNSCEAR 2016 REPORT

coal-mine categories: small, medium and large. This can be compared to the collective dose in
2002-2004, also shown in the table. (The number of bone-coal miners in China in 2010 was not
available, although this group represented a small fraction of the total in the 2004 data.) This approach
gave an estimated collective dose in 2010 for coal miners in China of about 10,000 man Sv.

132. During the period 20022004, China accounted for almost 90% of the underground coal miners in
the world. Assuming that the proportion of underground coal miners from countries worldwide was about
the same in 2010 as in 20022004, the total world collective dose to coal miners was estimated to be about
11,000 man Sv (compared to 23,000 man Sv obtained using the simpler approach above). This can also be
compared to the value 16,560 man Sv that was reported in the UNSCEAR 2008 Report [U11]. The
implied decrease would be due to more efficient coal-mining technology for underground coal extraction
and to the closing of many of the smaller, less efficient coal mines with poor ventilation [W2].

133. The dose contribution for thorium and its decay products and aerosols containing long-lived
alpha-emitting radionuclides were not assessed in this study [L1]. In the past, the effective dose from
aerosols containing long-lived alpha-emitting radionuclides was roughly estimated by using the
contents of radioactive material in the coal and the concentrations of dust in underground coal mines.
The estimation indicated that the annual individual dose ranges from several to ten microsieverts, which
has been assumed to be of little significance in this comparative study.

Table 31. Annual doses to underground coal miners in China [L1]

The information in the table for number of miners for the year 2010 was obtained from an official communication from the
Chinese delegation to the Committee

Type of coal Average annual Number of Number of Collective Collective dose in 2010,
mine effective dose miners in miners in dose using average annual
(2004) 2004 2010 20022004 effective dose from
(mSv) (millions) (millions) (man Sv) 2004 and number of
miners in 2010 (man Sv)

Large-sized 0.32 1 1.26 315 403

Medium-sized 0.63 1 1.31 630 825

Small-sized 3.78 4 2.27 15 100 8 581
Bone-coal 11.3 0.05 Not available 567 Not available

Total 2.75 (weighted 6.05 4.84 16 612 9 809

average)

B. Natural gas
134. According to table 2 the global combustion of natural gas produced 544 GW a of electrical energy
in 2010. This amount of electrical energy was second only to that due to the combustion of coal at
992 GW a and was 1.7 times the amount of electricity generated by nuclear power plants. The
generation of 1 kW h of electrical energy has been estimated to require the combustion of 0.286 m3 of
natural gas [U20].

135. Natural gas is essentially methane (CH4) with trace amounts of other materials. The most
important radionuclide released during the combustion of natural gas is 222Rn. Several authors have
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 195

reported on the concentration of 222Rn in natural gas as measured at the wellhead or at various other
locations in the gas-delivery system. In developed countries, natural gas is typically not used directly
from the wellhead, but is processed to remove moisture and refrigerated to condense and remove higher
chain gases, including ethane (C2H6), propane (C3H8) and butane (C4H10). The condensation process is
important, because radon tends to condense with ethane and propane [D6, G2]. Gesell [G2] studied nine
gas-processing plants and found that, on average, the 222Rn content of gas ready to be sold for
combustion (the sales gas) was 34% of that at the wellhead (the input gas) with a range of 4 to 90%.
Further, natural gas may be transported by pipeline over long distances and is also typically stored at
locations near its end use. Because the half-life of 222Rn is 3.8 days, appreciable decay can occur
between the production and combustion of natural gas.

136. Measured values of 222Rn in natural gas are summarized in table 32. The range of individual
values was large; the weighted average of 879 samples was 625 Bq/m3. Most of the values in table 32
were for samples taken at the wellhead. Considering the processes that reduce the concentration of
radon from the gas at the wellhead to the gas sold for combustion [V2], a value of 300 Bq/m3 was
considered to be a reasonable estimate.

Table 32. Summary of reported concentrations of 222Rn in samples of natural gas

Unless noted otherwise, the samples were taken at or near the wellhead. Concentration values are in units of Bq/m3

Source Area Number of Minimum Maximum Mean

samples value value
[B9] NW New Mexico 307 7.4 5 880 910
SW Colorado
[M4] NW New Mexico 42 11.8 2 130 610
SW Colorado
[G2] Texas, Oklahoma, Louisiana 15 37 4 400 1 330
[V1] Netherlands, Germany, North Sea, ~200 33 1 650 74a
Borneo, Nigeria
[K2] N. Germany 196 4 000 580
[V2] British Columbia ~32 7 921 272
[O9] Ireland 8b 116 918 638
[A4] c Syrian Arab Republic 36 15 1 142 400
[R4] Pennsylvania 21 37 2 920 1 370
[P2] Pennsylvania 22 110 5 476 1 770

Weighted average 879 625

a
Average results for the Netherlands only.
b
Same field sampled at various times.
c
Taken at many locations, including those in gas-processing plants.
196 UNSCEAR 2016 REPORT

137. The calculation of the activity of 222Rn released normalized to electricity generated is thus given as

m3 Bq h kW TBq TBq
0.2860 300 3 8760 10 6 10 12 = 0.75 (2)
kW h m a GW Bq GW a

This value for the release of 222Rn normalized to electricity generated was used with UNSCEARs
revised methodology for estimating public exposures due to radioactive discharges (annex A). The
2010 data on electricity generation from natural gas for each region, shown in table 2, were used, and
the default population distribution based on population densities for 2010 was used for the collective
doses (see paragraph 34). The resulting characteristic individual and collective doses are compared with
those for other electricity-generating technologies in table 33 (for characteristic individual doses) at the
end of this chapter and in table 46 (for collective doses) in chapter VIII.

138. Other radionuclides can appear in natural gas. One of them is 220Rn, but its half-life (55.6 s) is too
short for it to appear at the point of end use of the gas for combustion. If water is co-produced with
natural gas, the water may contain soluble amounts of radium. Radium-226 is the parent of 222Rn and is
a decay product of 238U; 224Ra is the parent of 220Rn and is a decay product of 228Ra and ultimately of
232
Th (see figure III). The progeny of the three radium isotopes and 222Rn can also be present in natural
gas, but typically in small amounts. Lead-210 is a longer-lived (half-life of 22.2 years) decay product of
226
Ra and 222Rn, and van der Heijde et al. noted that the activities of 210Pb and its decay product 210Po in
natural gas are more than can be accounted for by the decay of 226Ra and 222Rn [V1]. These authors
concluded that the additional 210Pb and 210Po had been produced from the natural gas reservoir. In
general, these radionuclides other than radon do not reach the end point of combustion of natural gas,
but they typically plate out within the gas-distribution system fairly close to the point of withdrawal
and/or treatment. Because the radionuclides are contained within equipment, workers can incur external
exposure to radiation. Organo and Fenton [O9] concluded after investigations in Ireland that workers at
offshore locations would be exposed at most to around 100 Sv in a year. The Pennsylvania
Department of Environmental Protection [P2] measured the external ambient dose rate at several
locations within gas production and processing sites and estimated a maximum annual average dose of
270 Sv. The Committee has assumed a reasonably realistic estimate of 100 Sv in a year.

139. A problem in estimating the collective effective dose to workers is determining the number of
workers in the industry apportioned to the amount of natural gas used to generate electricity. An
indirect method was used to make this determination. According to the United States Bureau of Labor
Statistics [U16], there were 89,000 production and non-supervisory workers in the oil- and gas-
extraction subsector. Most of these workers were not working directly in support of natural gas
production for electricity generation, but there was no further breakdown in the data. In order to
estimate the fraction of workers supporting the generation of electrical energy by combustion of natural
gas, data from [I17] for the OECD countries for 2010 were used. In terms of energy produced, these
data indicated that 12% of the energy produced from oil and natural gas together was used for
electricity generation due to combustion of natural gas. Thus, it was estimated that 10,700 workers were
employed in the United States to produce natural gas for the generation of electricity.

140. According to [U19] 209.2 billion cubic metres of natural gas were used in 2010 to produce
electricity in the United States. With the use of previously mentioned conversion factors this was
equivalent to 83.5 GW a of electrical energy. Thus, the normalized collective dose for workers based on
the data originating in the United States was 0.013 man Sv/(GW a).
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 197

C. Oil
141. The combustion of oil for electricity generation accounted for less than 5% of the worlds
electricity generation in 2010, but accounted for more than 30% of electrical energy generated in West
Asia. The generation of 1 kW h of electrical energy is estimated to require the combustion of 0.278 L of
petroleum [U20]. Most of the petroleum combusted to generate electricity in 2010 in the OECD
countries consisted of fuel oil [I17], which has an approximate density of 0.95 kg/L.

142. Petroleum is produced from underground reservoirs in the presence of geological formations that
contain the primordial radionuclides 40K, 238U and 232Th; the latter two are parents of chains that
produce a series of radionuclides usually present in secular equilibrium (see figure III). A process that
can perturb the secular equilibrium is the dissolution of some members of the chain in water, which at
depth can be at high temperature. Water containing dissolved radium is typically co-produced with oil.
Radium-226 is a member of the chain headed by 238U; 226Ra is the parent of 222Rn and an additional
series of decay products including 210Pb and 210Po. Radium-228 is a member of the chain headed by
232
Th; 228Ra is the parent of 224Ra and 220Rn. Both 226Ra and 224Ra are parents of additional short-lived
beta or gamma-emitting radionuclides.

143. Apparently some power plants can burn crude (or heavy) oil with its associated natural gas [A5], but
it is more typical that associated natural gas and water produced in the process must be removed from
petroleum before delivery to a pipeline or refinery [S5]. Radon is more typically associated with natural
gas and the soluble isotopes of radium are typically associated with the produced water. As produced
water is withdrawn from depth, it cools and dissolved minerals, including radium, can form precipitates
and deposit on the production tubing and on various other production equipment. These deposits are
known as scale or sludge. With time the radium contained within scale and sludge will come into
equilibrium with gamma-emitting decay products and give rise to external exposure of workers.

144. There are many studies of the presence of radionuclides in scale, sludge, (e.g. [I1, R3]) and
produced water [S5], and of radon in natural gas (see section V.B above), but there are few
measurements of radium or radon in crude oil or its subsequent products. Crude oil generally contains
some remaining water within a water-in-oil emulsion and this water contains dissolved radium [S5].
Radon is also quite soluble in oil [B2].

145. The report by Bell et al. [B2] appears to be the only one that reports measurements of both 222Rn
and 226Ra in the same samples of crude oil. Because the half-life of 222Rn is 3.82 days, the samples had
to be delivered quickly and correction made for decay in transit. Seven samples of crude oil from wells
in Texas and Oklahoma were measured. The results for 222Rn varied from a minimum of 3.2 Bq/kg to a
maximum of 17 Bq/kg and had an average of 7 Bq/kg. The results for 226Ra had a minimum of
0.2 Bq/kg, a maximum of 13 Bq/kg and an average of 0.7 Bq/kg. The ratio of 222Rn to 226Ra had a
minimum of 4, a maximum of 38, and an average of 15. Thus, most of the 222Rn in fresh samples of
crude oil was unsupported, that is, it was far in excess of the activity that was in equilibrium with 226Ra.
Except for those power plants that burn crude oil directly, this excess 222Rn would decay or be removed
in the refining and/or transportation processes.

146. Hamlat et al. [H1] reported concentrations of 226Ra in an unspecified number of samples of crude
oil collected in Algeria. The range of values was from 6 Bq/kg to 20 Bq/kg; these values are reasonably
consistent with those reported by Bell et al. [B2]. Al-Saleh and Al-Harshan [A7] measured
radionuclides in 14 petroleum-product samples, including crude oil drawn from the Riyadh City
Refinery. The detection limit for 226Ra was 0.014 Bq/kg; the only products with detectable amounts of
226
Ra were sweet naphtha (0.650.40 Bq/kg) and flushing oil (0.450.20 Bq/kg).
198 UNSCEAR 2016 REPORT

147. The limited amount of data indicated above suggests a cautious approach in estimating the release
of radionuclides from the combustion of oil products. For this procedure the Committee estimated that
there was a concentration of 226Ra in all oil products of 1 Bq/kg. Further, it assumed that unsupported
amounts of 222Rn decayed or otherwise were eliminated in the process, so that the concentration of
222
Rn in all oil products was also 1 Bq/kg.

148. Then, the normalized release of 222Rn was calculated as

L kg Bq h kW TBq TBq
0.278 0.95 1 8760 10 6 10 12 = 0.002 (3)
kW h L kg a GW Bq GW a

This value for the release of 222Rn normalized to electricity generated was used with the revised
methodology for estimating public exposures due to radioactive discharges (annex A). The 2010 data
on electricity generation from oil for each region, shown in table 2, were used, and the default
population distribution based on population densities for 2010 was used for the collective doses (see
paragraph 34). The resulting characteristic individual and collective doses are compared with those for
other electricity-generating technologies in table 33 (for characteristic individual doses) at the end of
this chapter and in table 46 (for collective doses) in chapter VIII.

149. There may be other radionuclides emitted by the combustion of oil products; Al-Masri and
Haddad [A5] reported values of about 60 Bq/kg of 210Po in heavy oil fuel used in three power plants in
the Syrian Arab Republic. The same authors [A6] also reported that soil in the vicinity of the three
power plants contained enhanced levels of 210Pb and 210Po. Other authors have not reported
measurements of 210Pb and 210Po associated with oil-fired power plants.

150. As mentioned above, 226Ra and 228Ra and their decay products accumulate in scale and sludge
around wellheads and processing equipment. This gives rise to occupational exposure, when workers
are in the vicinity of these elements. Kvasnicka [K4] estimated that the maximum effective dose on
offshore platforms could be managed to be below 1 mSv in a year. It is expected that the doses to
workers on offshore platforms would be higher than on onshore platforms, because of more substantial
problems with scale formation and the limited space in the working environment on the platforms.
Hamlat et al. [H1] measured ambient dose rates in the vicinity of onshore oil and natural gas production
equipment. They estimated that annual effective doses around oil extraction equipment might range
from 40 to 600 Sv for normal activities. A value of 300 Sv was adopted for this analysis.

151. There were no direct data on the number of workers in the oil industry apportioned to the
combustion of oil for the generation of electricity, so an indirect method was used. According to the
United States Bureau of Labor Statistics [U16], there were 89,000 production and non-supervisory
workers in the oil- and gas-extraction subsector in the United States in 2010. In order to estimate the
fraction of workers who were supporting the generation of electrical energy by the combustion of oil,
data from [I17] were used. In terms of energy produced these data indicated that only 1.7% of energy
produced from oil and natural gas was used for the generation of electricity by the combustion of oil in
2010. Thus, it was estimated that 1,520 workers in the United States could be apportioned to generation
of electrical energy by the combustion of oil.

152. According to [U19] 7.02 109 kg of oil were used in 2010 to generate electricity in the United
States. Using the conversion factors given above, this is equivalent to 3.03 GW a of electricity. Thus,
the collective dose to workers normalized to the electricity generated, based on data from the United
States, was estimated to be 0.15 man Sv/(GW a).
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 199

D. Comparisons of exposures from fossil-fuel electricity generation

153. Table 33 shows characteristic individual doses integrated to 100 years per unit of electricity
generated for the UNEP regions. These doses are due to atmospheric discharges from the coal cycle and
the combustion of gas or oil in power plants generating electricity. Although the doses resulting from
the modern coal plants are about 10% of the doses from the older plants, in total the characteristic
individual doses per unit of electricity generated from both the older and the modern coal plants are
larger than from the gas or the oil combustion technologies. See chapter VIII for further comparisons
between these and other electricity-generation technologies.

Table 33. Summary of characteristic individual doses per unit of electricity generated integrated to
100 years due to atmospheric discharges from the coal cycle and the combustion of gas and oil
(mSv/(GW a))

Showing results from both older and modern coal plants

Electricity Release Africa Asia and Europe Latin North West Asia
generation Pacific America America
technology
Coal

- From mining Radon 2.4 104 2.4 104 2.4 104 2.4 104 2.4 104 0

- From older Natural 3.8 104 4.0 104 4.2 104 3.9 104 4.2 104 0
coal plants radionuclides
including
- From 3.1 105 3.2 105 3.4 105 3.2 105 3.4 105 0
radon
modern
coal plants

- From ash Radon 1.5 104 1.5 104 1.5 104 1.5 104 1.5 104 0

Gas Radon 6.3 105 6.3 105 6.3 105 6.3 105 6.3 105 6.3 105

Oil Radon 1.7 107 1.7 107 1.7 107 1.7 107 1.7 107 1.7 107

VI. RADIATION EXPOSURES ARISING FROM ELECTRICITY

GENERATION FROM GEOTHERMAL ENERGY
154. Geothermal energy is derived from the heat of the earth at locations where magma is closer than
normal to the earths surface. The generation of electricity from geothermal sources presently depends
upon the use of steam or hot water from wells drilled into underground reservoirs. As shown in table 2
[I17], geothermal energy in the reference year 2010 generated 7.77 GW a of electricity and accounted
for 0.3% of the total worlds electricity generation in 2010.

155. Electricity generation from geothermal sources first began in 1904 at Larderello, Italy, with an
experimental 10 kW generator. At the present time the installed capacity is in excess of 9,000 MW in
25 countries [L3]. The larger power plants are at The Geysers, California; Larderello, Italy; Cerro
Prieto, Mexico; Leyte, Philippines; Salton Sea, California; Hellisheidi, Iceland; Tiwi and Malitbog,
Philippines; and Wayang Windu and Darajat, Indonesia [K1].
200 UNSCEAR 2016 REPORT

156. Different approaches can be taken to generate electricity from geothermal sources. If dry steam is
available, the simplest is to run the steam through a turbine and exhaust to the atmosphere. Efficiency is
improved if the dry steam is condensed, with heat dissipated through cooling towers. Most reservoirs,
however, do not produce dry steam, but some combination of steam and hot water or hot water alone.
In this situation, it is necessary to separate steam from water, or, if hot water alone is produced, to flash
the water to steam through single, double, or even triple passes. A binary process is in use at some sites
that use water of lower temperature; hot water is run through a primary loop and a fluid with a lower
boiling point through a secondary loop. As of the year 2016, most electrical energy generation is
through the flash process with single or double flash; the other major technology is the use of dry steam
with condensation [M3].

157. During the 1970s there was substantial interest in evaluating the potential impact of the release of
222
Rn from gases vented during the operation of geothermal electricity-generating plants. The more
complete evaluations were reported for The Geysers and the Larderello Plants. Anspaugh [A10]
reported that 19.3 TBq/a were being released from The Geysers during the operation of 11 dry steam
condensing units with 502 MW of electrical power; the daily rate was noted to be equivalent to the
average daily background emission of 222Rn from about 40 km2 of soil. For the Larderello geothermal
complex, the emission rate per unit energy was 38.5 TBq/(GW a) according to George et al. [G1]. This
was based on measurements performed by DAmore [D1], who calculated that 300 GBq/d were being
released from the Larderello geothermal complex, which had an installed capacity of about 420 MW
[P1]. Thus, the emission rate per unit energy at Larderello was 260 TBq/(GW a).

158. The above represent the two situations where data were complete enough to estimate emissions
per unit of electricity generated. Based on this limited set of data, the Committee assumed an average
value of 150 TBq/(GW a). There are measurements of radon in steam or hot water from other locations,
and samples of such data are given in table 34; at the time of the assessment no data were available for
the amount of 222Rn that might be released from geothermal sites using binary circle systems. A reason
that data are scarce on radon related to geothermal energy is that, following the measurements in the
1970s, it was generally concluded that this source of radon was not significant compared to the natural
exhalation of radon from soil. Measurements of 222Rn and 220Rn in geothermal fluids do continue, but
are mainly related to understanding of the dynamics of resources [W6].

Table 34. Reported concentrations of 222Rn in fluids produced from geothermal wells

Location Activity of 222Rn per unit of Reference

produced fluid (Bq/kg)
The Geysers, weighted average 520 [A10]
The Geysers 620 [K3]
Larderello 1 280 [K3]
Larderello, weighted average 3 100 [G1]

Salton Sea, California 110 [K3]

East Mesa, Imperial Valley, CA 1.1 [K3]
Wairakei, New Zealand 630 [W6]

159. The value of 150 TBq/(GW a) for the release of 222Rn normalized to electricity generation from
geothermal energy was used with the electricity generation data by region from table 2 and with
estimates of population density to give an estimate of the characteristic individual and collective doses.
There is substantial uncertainty regarding the population density around geothermal power plants, as it
is necessary to locate the plants within the narrow confines of the resource. The Geysers and the Cerro
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 201

Prieto plants are in isolated locations, but large metropolitan areas are within 1,500 km. The other
extremes are geothermal power plants located on islands. Because of these discrepancies and the lack of
specific population data on each power plant, the Committee made two calculations for geothermal: one
with the assumption of low-density population, as in the assessments for uranium mining and milling,
and one with the default population density.

160. Table 46 in chapter VIII compares the collective doses to the public for the discharges from the
nuclear fuel cycle, coal cycle, gas, oil and geothermal electricity-generating technologies. As explained
in the previous paragraph, two values are given for geothermal energy. The normalized discharge of
radon for geothermal electricity generation is the largest of the electricity-generating technologies
shown in table 46. However, the impact of geothermal electricity generation remains small because of
the limited use of this technology. Further, there is substantial uncertainty regarding the discharge of
radon from geothermal power plants; data are available for only two plants, and the estimated
discharges from these two plants vary substantially.

161. There are even fewer data on occupational exposure in the geothermal industry. Anspaugh [A10]
reported no evidence of increased external gamma exposure, but did note increased levels of radon and
short-lived progeny at several locations normally inaccessible except during maintenance. In general,
moderate to severe disequilibria were found, and an average excess amount of exposure was
0.05 Working Levels 7 (WL) (1 J m3). Razzano and Cei [R1] indicated that 36 persons work over
three shifts to operate the Larderello complex. From data in the same paper, the electricity generated in
2013 at Larderello was 5,659 GW h with 767 MW of efficient installed capacity. The combination of
these data leads to an annual occupational exposure of 5 man WLM/(GW a). According to [I11]
1 WLM is equivalent to an effective dose of 1020 mSv, but the higher values are associated with
mines and homes. A more reasonable value for outdoor exposure is 10 mSv, which would also be
consistent with [U10]. Thus, the normalized collective dose to workers in geothermal electricity
generation would be 0.05 man Sv/(GW a).

VII. ASSESSING DOSES FROM THE CONSTRUCTION PHASE OF

ELECTRICITY-GENERATING TECHNOLOGIES
162. All electricity-generating technologies considered by the Committee have a construction phase
during which the facilities and infrastructure are established. Although in general there are no
significant radiation exposures associated with the construction, there may be radiation exposures
associated with obtaining the materials for construction such as in mining activities.

163. Industries not directly linked to electricity generation, such as mining, milling and processing of
metal ores, contribute to occupational and public exposure because of the presence of natural
radionuclides [U11, U22]. While public exposures from the mining industry may be negligible on an
individual basis, typical occupational doses in the mining industry may be up to a few millisieverts in a
year. The collective dose from occupational exposures during the mining of metals was evaluated for
this study on the basis of the dose per unit mass of metal ore mined for each electricity-generating
technology, as discussed in the UNSCEAR 2008 Report [U11], and on recent data from 2012 on
radiation exposure of miners of rare earth metals in China [W3].

7
A working level (WL) is a unit of potential alpha energy per unit of air. In the SI system of units, one WL is equal to 2.08 105 J m3.
A working level month is a unit of exposure to 1 WL for a month. For occupational personnel one month is considered to be 170 hours.
202 UNSCEAR 2016 REPORT

164. Aggregate is also used extensively in the construction of electricity-generating plants. Aggregate
in building and construction refers to the material used for mixing with cement, bitumen, lime, gypsum
and other additives to form concrete or mortar. Commonly used aggregates include sand, crushed or
broken stone, gravel, and blast-furnace slag. In some cases, the amount of aggregate used is similar to
the total mass of all other materials.

165. The relative volumes of concrete required to construct various nuclear and coal-fired power plants
are tabulated in tables 35 and 36. Typical radionuclide concentrations in raw and processed materials
and in wastes of the mineral processing industry, including the cement industry, were reviewed in the
UNSCEAR 2000 Report (see table 27 in [U9]). Typical radionuclide concentrations in metal ores were
reported to be about three orders of magnitude higher than of materials in the cement industry. For this
assessment, only the occupational exposures from the mining of metal ores needed for metals used in
construction of electricity-generating technologies were evaluated.

Table 35. Estimates of the amounts of construction materials used in nuclear power plants

Plant Mass of metals (t) Volume of concrete (m3)

PWR, 1 000 MW(e) [D5] 61 000 169 000
BWR, 1 000 MW(e) [D5] 66 000 200 000
PWR, 1 000 MW(e) [B8]a 38 000 (estimated) 75 000 (estimated)

Nuclear power plant [W5] 36 000 75 000

a
The amount presented for specific metals: Al, Cr, Cu, Ag, Fe, Pb, Mg, Mn, Mo, Ni, Sn, Zn.

Table 36. Estimates of the amounts of construction materials used in coal-fired power plants

Planta Mass of Comments Volume of

metals (t) concrete (m3)
Coal-fired power plant [S7] 50 000 steel Amount presented in generic terms as "steel" 380 000
and expressed in kg/MW(e)

Coal-fired power plant [W5] 41 000 metals Amount presented for specific metals: Al, Cr, 31 000
Cu, Mn, Mo, Ni, steel (low alloy and stainless)
and V and expressed in t/GW(e)
a
1,000 MW(e) plant assumed.

166. The mass of metal ore required to produce the metals in the building materials has been estimated
from information on the total metal inventory required and from the metal ore grade used. It was
assumed that steel was the primary metal used. Iron is alloyed with carbon to produce plain carbon
steel. Alloy steels also contain various alloying metals. The category of steel used in the different
electricity-generating technologies and the contents of alloying metals were considered. Plain carbon
steels were assumed to contain iron and 0.160.59% carbon. Low-alloy steels were assumed to contain
less than 0.25% carbon, as well as small amounts of nickel, chromium, molybdenum, manganese and
silicon. High-alloy stainless steels were assumed to contain molybdenum, chromium and/or nickel and
other elements. These assumptions are based on the information in [C1, G3].

167. For high-alloy stainless steel, type S316 was used as the basis for this assessment, because it is
widely used in industrial applications. Type S316 contains around 2% by mass of molybdenum [A3,
S3]. Other steel types may contain different levels of molybdenum, or none at all, which could
influence the results. This is because molybdenum is a low-grade metal ore and therefore extraction of
molybdenum may make a larger contribution to occupational exposures [A3, S3].
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 203

168. Based on the above, the composition of typical steels for each category (plain, low-alloyed and
high-alloyed stainless) was assumed as shown in table 37. If the type of steel was not known, the
proportions given in table 38 were assumed.

Table 37. Composition of different types of steel (based on [C1, G3])

Material Composition (% by weight)

PLAIN STEEL
Iron 99.62

Carbon 0.38

LOW-ALLOY STEEL
Iron 99.63
Molybdenum 0.33

Carbon 0.05

HIGH-ALLOY STAINLESS STEEL

Iron 65.35

Chromium 17.00
Nickel 12.00

Molybdenum 2.50

Manganese 2.00
Other non-metallic elements 1.07

Carbon 0.08

Table 38. Assumed mix of steel types used in construction where details were not specified
(based on [C1, G3])

Steel type Proportion of total (%)

High-alloy stainless steel 10
Low-alloy steel 30
Plain steel 60

169. Assumptions on ores. The amount of ore required for the construction of an electricity-generating
facility also depends on the ore grade. The assumptions made regarding ore grades are shown in
table 39 [A8, C3, M1].
204 UNSCEAR 2016 REPORT

Table 39. Metals and their assumed ore grades

The data presented here are typical values taken from [A8, C3, M1], and the numbers have been rounded

Resource Ore grade at mine (%)

Tin 79
Manganese 47
Zirconium 47

Iron 46
Magnesium 41
Tantalum 31

Chromium 26
Aluminium 11
Zinc 7

Lead 6
Nickel 2

Copper 1.4
Vanadium 0.3
Molybdenum 0.18

Silver 0.01
Gold 0.0005

170. Recycling of metals. Recycling is an important part of the metal industry from both an economic
and energy-saving point of view. In the United States, approximately two thirds of the steel produced in
2008 was made from recycled material [U15]. Around 43% of the total crude steel production
worldwide is made from recycled steel, with the main sources for steel recycling typically being
discarded cars, household appliances and steel cans, as well as old buildings and structures [W4].
Typical building construction uses approximately 6065% recycled metal [B1], while in the United
Kingdom about 94% of construction steel is recovered [T3].

171. Special rules apply to the re-use of scrap steel and other metals from nuclear power plants;
authorities establish limits on the activity concentrations of radionuclides in materials that can be
released for recycling. There is, however, no indication that there are significant differences in the
degree of recycling between the various electricity-generating technologies.

172. The effect of recycling metals has not been included in the present study, because of insufficient
data for large portions of the world. By not explicitly considering recycling, the collective dose per unit
of electricity generated due to the extraction and processing of the metals used in the construction of a
facility is likely to be overestimated by around a factor of two on average, and possibly a factor of five
or more for some regions of the world with developed recycling processes. This should not affect the
relative comparisons of occupational exposures from mining metals between technologies, but would
affect comparisons of the total exposures (public and occupational) among the electricity-generating
technologies.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 205

173. Dose assessment. To estimate the radiation exposure of workers in the mining industry
(occupational exposure) from ore extraction and processing, the following assumptions were made:

(a) The ores are extracted only through underground mining. The use of above-ground mining
would give rise to smaller radiation doses than underground mining [U11]. Consequently, this
assumption leads to an overestimation of collective dose.

(b) The presence of several metals in an ore was not considered. Each ore was assumed to contain
only one metal type, and thus the contribution to occupational doses from the extraction and
processing of the various minerals was evaluated independently and summed. In reality some
metals will be extracted together and consequently the resulting collective doses may have been
overestimated.

(c) Occupational exposures from mining and processing of raw materials were inferred from data
on uranium mining and processing, and on mining other minerals. Mining data for underground
copper mines in Australia (Mount Isa), Canada (Kidd Creek), South Africa (Palabora), and
Portugal (Neves Corvo) show annual production rates of 15 kilotonnes of ore per employee [I19].
The denoted employees included all workers at the mines. Based on these data, the Committee
selected a mid-range value of 3 kt of metal ore per employee for use in this assessment (table 40).

(d) The average annual effective dose to non-uranium miners, including those involved in mineral
processing, given in the UNSCEAR 2008 Report [U11] was 3.0 mSv, with a range of 1.3 to
5.0 mSv. Using an annual average value of 3.0 mSv for all workers associated with the mining
industry, and the value of 3 kt of metal ore per employee given in (c), the collective dose per unit
mass of ore extracted was estimated at 1 109 manSv/kg (table 40). The data for (c) and (d) were
obtained from different worker populations and this introduces some uncertainty in the estimation
of the collective dose per unit mass of ore extracted.

(e) In a recent study in China [W3] data were collected on radiation levels, and dose assessments
were performed for workers of the rare earth industry in the Baotou area and Sichuan Province,
two large mining districts in China. These regions account for about 77% of the total annual output
of rare earth mines in China; the reported annual production rates were 0.01 and 0.02 kt of rare
earth metal per mining employee, respectively.

(f) The Chinese study has estimated the collective dose per unit mass of rare earth metal
extracted. The typical value for the whole rare earth industry that was obtained in the Chinese
study on radiation exposure of Chinese workers in mining, crushing, beneficiation and refining is
about 150 109 manSv/kg. This value relates to the mass of metal extracted, which differs from
the value 1 109 manSv/kg given in (d) which relates to mass of metal ore. The Committee used
the latter value in the analysis in order to account for the different metal requirements with
associated metal ore grades of each electricity-generating technology.

(g) As a check, an alternative estimate was obtained by comparing with the data on uranium
mining. The collective dose per unit mass of uranium extracted given in the UNSCEAR 2008
Report [U11] was 1 manSv/kt (based on 20002002 data). Assuming (from information in [I4,
O8]) an average grade for uranium ore of 0.3%, the collective dose per unit mass of uranium ore
extracted is 3 10-9 man Sv/kg.

174. Although the assumptions in paragraph 173 affect the absolute values of the assessed collective
dose, the relative importance of the different electricity-generating technologies with regards to the
construction phase remains.
206 UNSCEAR 2016 REPORT

Table 40. Collective effective dose per unit mass of ore mined used in this study

Estimate Annual production rate Average annual Collective dose per unit mass of
based on of ore per person (kt) occupational dose (mSv) metal ore extracted (man Sv/kg)
[U11] 3 3 1 109

175. Nuclear power plants. For the current assessment, the assumed design lifetime of a nuclear power
plant was 40 years and the capacity factor 8 was 90% (table 41).

176. Three different studies of the use of materials in nuclear power plants have been published. Dones
et al. [D5] studied the material used in two Swiss nuclear power plantsthe pressurized-water reactor
(PWR) at Gsgen and the boiling-water reactor (BWR) at Leibstadt. Both were of the 1,000 MW(e)
type, but the capacity of the Leibstadt plant was increased to 1,190 MW(e) following construction. For
the latter, the total amount of metals used was 66,000 tonnes, of which two thirds were unalloyed
steel/iron. In addition, 200,000 m3 of concrete were used. Bryan and Dudley [B8] considered the
quantities of different materials used in a typical 1,000 MW(e) PWR with river water cooling. The total
mass of metals used was estimated at 38,000 tonnes but the proportions of each metal were not
specified. The total amount of concrete used was estimated as 210,000 tonnes (~87,260 m3). White and
Kulcinski [W5] gave the amount of material needed for four power plants, including a nuclear power
plant, where values of 36,000 tonnes of metals and 180,000 tonnes (~75,000 m3) concrete were given.
The study focused on the energy payback ratio 9 and CO2 emissions, and it was not in that sense a full
life cycle assessment. Furthermore, the study did not consider potential radiation exposure. The data
from the three studies are presented in table 35.

Table 41. Assumed plant sizes, life times and capacity factors for the different electricity-
generating technologies
*
Assumed to be the same as coal

Electricity-generating Typical plant gross capacity Lifetime Capacity factor

technology (MW(e)) (years) (%)
Nuclear [I3] 1 000 40 90
Coal [F3, I18] 1 000 40 80
Natural gas [O1, S8] 505 25 80
Solar PV [F3, N3] 30 9.4
Wind turbine [D2, N4, O10] 25 20 24
Biomass [W5] 1 000* 20 75
Geothermal [R5, S9] NA 30 NA

8
Capacity factor (net) is the ratio of the net electricity generated, for the time considered, to the energy that could have been
generated at continuous full-power operation during the same period [U25].
9
Energy payback ratio is the ratio of total electrical energy produced during a systems normal lifespan, divided by the electrical
energy required to build, maintain and fuel it.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 207

177. Example of a dose calculation for a nuclear power plant. Specifications of the typical plant gross
capacity, lifetimes and capacity factors (can also be called load factor) used in the calculations are
shown in table 41. Table 42 shows an example of the details of a dose calculation for a nuclear power
plant. The collective dose was estimated based on occupational exposures related to metal ore mining
and processing of the metal ore for metals needed to construct the power plants.

Table 42. Example calculation of collective dose from occupational exposure during construction
phase of a nuclear power plant

Type: PWR Gsgen, 970 MW(e), capacity factor 90% [D5]

Resource/Elements Metal needed for Ore grade at Amount of ore needed Collective dose
construction (kg/kWh) mine (%) at mine (kg/kWh) (manSv/(GW a))

Stainless steel

Iron 4.7 105 46 1.0 104 8.9 104

Manganese 1.4 106 47 3.1 106 2.7 105
Chromium 1.2 105 26 4.8 105 4.2 104
Molybdenum 1.8 106 0.18 9.9 104 8.7 103

Nickel 8.6 106 2 4.3 104 3.8 103

Low-alloyed
Iron 1.8 105 46 3.9 105 3.5 104

Molybdenum 5.9 108 0.18 3.3 105 2.9 104

Unalloyed steel

Iron 1.1 104 46 2.4 104 2.1 103

Other

Copper 4.7 106 1.4 3.3 104 2.9 103

Aluminium 6.3 107 11 5.6 10-6 4.9 105

Total 1.9 102

178. Coal-fired power plants. There are a number of different designs of coal-fired power plants.
However, the amounts of material used in their construction do not differ significantly. Spath et al. [S7]
conducted a full life cycle assessment of a coal-fired power plant covering the construction and
decommissioning of the plant. The amount of steel used in the construction of a 1,000 MW(e) plant was
50,000 tonnes, but no data were given on the other metals. Also, the proportions of plain carbon steel,
low-alloy steel and stainless steel were not specified. White and Kulcinski [W5] described a total
amount of 41,000 tonnes of metals in their study. In both studies [S7, W5], the analysed units had an
assumed capacity factor of 80% and a lifetime of 40 years. The information from these studies is
summarized in table 36.

179. Natural gas power plant. The designs of natural gas power plants also vary. Spath and Mann [S8]
presented a full life cycle assessment for a 505 MW(e) natural gas power plant covering its construction
and decommissioning. The study is similar in approach to the one they used for coal-fired power plants
[S7]. Again, potential radiation exposure was not considered and no specific data were provided on the
particular metals used to construct the plant. The amount of steel used was 31 t/MW(e) and the amount
of concrete used was 98 t/MW(e). The Committee assumed that the average capacity of a natural gas
208 UNSCEAR 2016 REPORT

power plant was 505 MW(e) to estimate the amount of each metal using the assumptions shown in
table 37 and table 38.

180. Solar energy power plant. Solar energy is by far the largest energy resource available on earth.
Two different technologies contribute to solar electricity generation: solar photovoltaics (PV) and
concentrating solar power (CSP).

181. Solar PV systems convert direct and diffused solar radiation into electricity through a
photovoltaic process using semiconductor devices. Solar PV systems can be used anywhere in the
world on suitable land and on buildings. Solar PV technology is also very adaptable to being used in a
modular fashion, which means that systems can be installed close to centres of demand.

182. Silicon-based PV solar cells contain small amounts of rare earth elements. Crystalline-silicon PV
cells are the most common PV cells in use today [U14].

183. Concentrating solar power (CSP) systems are designed to produce high-temperature heat for
electricity generation or for co-generation of electricity and heat. CSP systems are capable only of
exploiting direct normal irradiation, which is the energy received directly from the sun (i.e. not
scattered by the atmosphere) on a surface tracked perpendicular to the suns rays. Areas suitable for
CSP development are those with strong sunshine and clear skies, usually arid or semi-arid areas.
Parabolic mirrors or troughs are used in CSP 10 technology. The parabolic mirrors are designed to
concentrate solar radiation onto linear heat collection elements [N2].

184. A solar energy power plant does not have a fuel cycle in the same way that most other energy
technologies do, consequently any potential human exposure to ionizing radiation is related to the use
of natural resources for manufacturing the centralized CSP plants or the PV solar cells. The largest
contribution to the collective dose is from the acquisition of raw materials, especially minerals used for
manufacturing the CSP plants or PV solar cells.

185. Silicon-based PV solar cells contain small amounts of rare earth elements, which are of particular
interest because the ores from which they are obtained have relatively high contents of uranium and
thorium. Thus, the mining of the ores and their subsequent processing lead to occupational radiation
exposures.

186. Because relevant data on the use of metals and other natural resources were only readily available
for the PV solar systems, the assessment was only carried out for this type of solar energy production.

187. The metals required to construct a solar PV system and amount of ore required are shown in
table 43. The results are based on data for a life cycle assessment of a 3 kWp 11 multicrystallinesilicon
PV solar panel mounted on a slanted roof in Switzerland [J1]. The solar panel efficiency was 9.4%. The
solar PV system lifetime was assumed to be 30 years, with the exception of the power inverter that
changes direct current to alternating current. In this case a lifetime of 15 years was assumed. The parts
included were mainly PV panels, mounting structures, inverter and electric installations. Metals used
for the solar panels included small amounts of rare earth elements, which are derived from ores with
high contents of thorium or uranium.

188. The amount of ore required per unit of electricity generated depends on the ore grade. In table 43,
the values were based on published ore grades [A8, C3, J1].

10
Also called concentrated solar power and concentrated solar thermal (CST).
11
The symbol, kWp, stands for the peak power (kW) of the solar PV systemthe basic unit for the characterization of the
capacity of PV plants measured in a standardized test at a temperature of 25 C and an irradiation of 1,000 W/m2.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 209

Table 43. Ore usage at mine for the production of energy by a PV solar system mounted on a
slanted roof in Switzerland

Figures are for production of 1 kWh by a 3 kWpa PV solar system; ore grade values provided in table 38

Resource Amount of metals required per unit Amount of ore required per unit
electricity producedb electricity produced
(kg/kW h) (kg/kW h)
Aluminium 1.7 103 1.5 102
Chromium 2.6 106 1.0 105
Copper 4.2 104 3.0 102

Iron 1.3 103 2.7 103

Lead 1.3 106 2.2 105
Magnesium 5.4 105

Manganese 4.4 106 9.3 106

Molybdenum 4.2 106 8.3 103
Nickel 5.0 108 2.5 106
Silver 2.9 106 2.9 102

Tantalum 7.8 107 2.5 106

Tin 1.6 106 2.1 106
Zinc 3.1 106 4.4 105
a
The symbol, kWp, stands for the peak power (kW) of the solar PV systemthe basic unit for the characterization of the
capacity of PV plants measured in a standardized test at a temperature of 25 C and an irradiation of 1,000 W/m2.
b
Calculated based on data in Jungbluth et al. [J1].

189. Wind power plant. The kinetic energy of wind is exploited in wind turbines for electricity
generation. Wind speeds suitable for electricity generation range from 4 to 25 metres per second. These
are attainable practically all over the world, with the exception of some equatorial regions. Wind power
is exploited not only onshore but also offshore, where wind speeds are higher and the wind is typically
available more regularly and for longer periods of time. The depth of water and distance from centres of
demand onshore are major factors influencing the siting of offshore developments [O3].

190. An assessment of materials used to build an onshore wind power unit was conducted by White
and Kulcinski [W5]. The assessment was for a three-blade wind turbine and assumed a 25 MW(e) unit,
with a capacity factor of 24% and a lifetime of 20 years. The data on construction materials is
summarized in table 44.

Table 44. Estimates of the amounts of construction materials used in wind power plants

Data in White and Kulcinski [W5] and references therein. Amount presented is for low-alloy and stainless steel and expressed
in t/GW(e)

Planta Mass of metals (t) Volume of concrete (m3)

Wind power plant [S7] 85 000 steel 130 000
a
25 MW(e) plant assumed.
210 UNSCEAR 2016 REPORT

191. Biomass power plant. Biomass fuel includes straw, wood or wood residues from forests, and
wood waste from wood-processing plants, such as sawmills or pulp and paper mills. Biomass fuel can
contain varying amounts of radionuclides associated with past emissions or accidents. One example of
this is fuel containing varying amounts of 137Cs in wood that has been obtained from forested regions
affected by fallout from past nuclear power plant accidents. The 137Cs can be emitted in the flue gases
or concentrated in the ash from the combustion of biomass containing 137Cs and possibly lead to
exposures. In some areas filters are currently in use to decrease the 137Cs in the flue gases, and
treatment or proper burial of the ashes can reduce or prevent exposures. Exposures via these pathways
are not treated further here since the 137Cs originates from nuclear power plant accidents and are
concentrated in the forested areas that have received significant fallout from past accidents.

192. In general terms, biomass fuel has a lower energy density than coal and is more challenging to
handle. As a consequence, the fuel handling equipment is heavier than that for a coal-fired power plant of
equivalent capacity. Technically, biomass power plants can be as large as coal-fired power plants in terms
of installed capacity, but, in general, biomass units are not built as large as coal units, mostly because of
the significant difficulties with supply and storage associated with materials of lower energy density [R2].

193. The design of a biomass power plant is not very different from that of a coal-fired power plant.
The main difference is that the ash handling equipment is generally smaller in a biomass plant,
depending on the type of fuel used (biomass typically generates less ash than coal). However, this
difference may be neglected to a first approximation, and therefore the Committee assumed that the
materials required per kilowatt to build a biomass power plant were the same as those given by White
and Kulcinski for a coal-fired power plant [W5]. For the biomass power plant, the capacity factor was
assumed to be 75% and plant lifetime 20 years.

194. Summary. Table 45 shows occupational collective effective doses normalized to energy
production due to mining for ores and the processing needed for construction of the electricity-
generating plants or devices, as estimated in this study. Electricity-generating technologies using coal,
natural gas and biomass have the lowest collective doses normalized to electricity generated, followed
by slightly higher values for nuclear power. Wind power shows a larger collective dose by about a
factor of 10 compared to the lowest values, and solar power shows the largest value by about a factor
of 80. The differences between the various electricity-generating technologies are connected to various
plant facilities using different types and amounts of steel and metals. Another reason is the differences
in the capacity factors and lifetime of the plants, with a higher capacity factor and longer lifetime
implying a lower collective dose per unit of electricity generation.

Table 45. Collective effective dose normalized to unit of electricity generation for construction of
electricity-generating plants or devices

Electricity-generating technology Normalized occupational collective dose due to mining

and processing of ores needed for construction
(manSv/(GW a))
Nuclear 0.02
Coal 0.01

Natural gas 0.01

Solar PV 0.8
Wind 0.1
Biomass 0.01
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 211

VIII. COMPARISON OF RADIATION EXPOSURES FROM

ELECTRICITY-GENERATING TECHNOLOGIES
195. Sources of radiation exposure from electricity-generating technologies based on the (a) nuclear
fuel cycle, (b) coal cycle, the combustion of natural gas, oil and biofuels, (c) geothermal energy,
(d) wind power and (e) solar energy have been investigated in this annex. Two electricity-generating
technologies, the nuclear fuel cycle and the coal cycle have been substantially investigated using the
same methodology, the Committees revised methodology for estimating public exposures due to
radioactive discharges (annex A). This same methodology was also used to investigate public exposures
derived with more rudimentary assessments based on less available data for the electricity-generating
technologies that employ combustion of oil and natural gas, and geothermal energy.

196. Occupational exposures for all of these technologies were also estimated, relying mainly on data
from dosimetric records of worker exposures. In addition, new assessments on occupational exposures
from (a) decommissioning of nuclear power reactors and (b) the mining of rare earth metals needed for
the construction phase in different electricity-generating technologies have been presented, adding for
the first time solar energy, wind power and combustion of biomass to the electricity-generating
technologies assessed by the Committee.

197. This chapter compares the results for the different electricity-generating technologies investigated
by applying the revised methodology for estimating public exposures due to radioactive discharges
(annex A). This is followed by a comparison of all results for both occupational and public exposures.

A. Comparison of public exposures due to radioactive discharges

from the electricity-generating technologies based on the nuclear
fuel cycle, coal cycle, combustion of natural gas and oil, and use of
geothermal energy
198. Table 46 shows results from applying the revised methodology for estimating public exposures
due to radioactive discharges (annex A). The results represent the sum of all discharges assessed in this
annex for the electricity-generating technologies based on the (a) nuclear fuel cycle, (b) coal cycle and
combustion of natural gas and oil, and (c) use of geothermal energy.

199. The table provides estimates of collective doses to the worldwide public, and associated collective
doses normalized to electricity generation in 2010, integrated to 100 years. Collective doses normalized
to electricity generation are the sum of the collective doses for each process in the nuclear fuel cycle, or
the coal cycle, divided by the total electricity generated in 2010 for that cycle. For the coal cycle, the
doses were estimated assuming all discharges were either from older coal plants or from modern ones.
Both results are shown and represent a range of values for the coal cycle. Although the assessments of
the nuclear fuel cycle and coal cycle were more substantive than those for the combustion of oil and
natural gas, and geothermal energy, the Committee considered that comparing the magnitudes of the
total and normalized collective doses was still valid.
212 UNSCEAR 2016 REPORT

Table 46. Collective dose to the worldwide public, and associated normalized collective dose for 2010,
integrated to 100 years a, b

Except where otherwise specified, the collective doses given are for the local and regional components. Shown also is the
percentage of total world electricity generation in 2010 for each electricity-generating technology and the discharges for
222
Rn normalized to the electricity generation in 2010

Electricity-generating Collective Normalized % of total world Normalized 222Rn

technology dose collective dose electricity discharges
(man Sv) (man Sv/(GW a)) generation in (TBq/(GW a))
2010
NUCLEAR FUEL CYCLE
Nuclear, total from mining and 130 0.43 13 Uranium mining 66
milling, power plants and Milling 3
reprocessing, excluding global
component Operational mill
tailings 3
Mill tailingsc 10

Adding global component

integrated to
100 years 910 3.0
500 years 1 700 5.5

10,000 years 7 600 25

COAL CYCLE
Coal, older coal plants 1 400 1.4 40 Coal mining 2.8

Coal, modern coal plants 670 0.7 Power plants 0.07

Ashc 1.8

OTHERS
Natural gas 55 0.10 22 0.75

Oil 0.03 0.000 3 4.6 0.002

Geothermal (low-density 5160 120 0.3 150

population default
population)

a
Projections of any health effects using collective doses in the table are not recommended.
b
All estimates are calculated based on best estimates; site- and location-specific collective doses are not presented.
c
The values of the normalized 222Rn discharges (TBq/(GW a)) shown in table 46 for uranium mine mill tailings (Mill tailings)
and for coal ash deposits (Ash) were multiplied by 100 to account for radon emanating for 100 years from these surfaces. The
value for coal ash deposits was also multiplied by a factor of 0.6 since only 60% of the ashes produced are deposited.

200. Excluding the global component from the globally-circulating nuclides, electricity generation
from the coal cycle gave the highest collective dose to the public integrated to 100 yearsboth for the
total generation in 2010 and when normalized to unit of electricity generatedfor both older and
modern coal plants. When the global component resulting from the globally-circulating radionuclides
originating from the nuclear fuel cycle was taken into account, integrated to 100 years, the total
collective dose from the nuclear fuel cycle was of the same order as from the coal cycle, about in the
middle range of the values for modern versus older coal plants. Because of ongoing global exposures
from the globally-circulating radionuclide 14C discharged from nuclear power plants and from
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 213

reprocessing facilities, the collective dose from the nuclear fuel cycle slowly increases over centuries as
shown in figure V, in section IV. This is also evident in the doses shown in table 46 for the global
circulation integrated to 100, 500 and 10,000 years. The total collective dose normalized to electricity
generated for the nuclear fuel cycle, including the globally-circulating nuclides and integrated to
10,000 years, is 25 man Sv/(GW a) as shown in table 46. Comparing with earlier UNSCEAR
assessments, the UNSCEAR 2000 Report [U9] estimated the same quantity as 40 man Sv/(GW a). As
described earlier, the collective dose from globally-circulating radionuclides is the sum of very small
doses to the entire worlds population. The local and regional components of collective dose assessed
for the nuclear fuel cycle excluding globally-circulating radionuclides, and for the coal cycle, are for
the local and regional populations exposed to the respective discharges of the source (see
paragraph 34).

201. The magnitude of collective doses resulting from the coal cycle, for both the older and the
modern coal plants, is due in large part to (a) the amount of electricity generated from coal (40% of
the worlds electricity generation in 2010), (b) the greater number of coal mines compared to
uranium mines and (c) the contribution to the collective dose from the discharge of radon and other
radionuclides from the 238U series arising from the combustion of coal at the power plants. Another
important factor is that a larger population base lives near coal plants and coal mines compared to
nuclear plants and uranium mines.

202. Natural gas accounted for 22% of the worlds electricity generation in 2010, which led to a larger
contribution to the collective dose than oil, but was still small in relation to coal. Combustion of oil for
electricity generation contributed only marginally to collective doses. These relatively small doses from
natural gas and oil combustion were however the dominant contribution to collective dose for West
Asia where, as shown in table 2, natural gas and oil accounted for most of the electricity generation in
that region in 2010.

203. Discharges of 222Rn normalized to the electricity generated are also shown in table 46; the largest
value is estimated for geothermal energy. This finding is based on few existing data. However, because
of the process for generating electricity from geothermal energy discussed in section VI, it is reasonable
that the impact of 222Rn discharges from gases vented during the operation of geothermal electricity-
producing plants is relatively significant. The resulting doses from geothermal energy shown in table 46
are assessed using two different population densities, low-density and the default population density
(see chapter III), where the available population data discussed in chapter VI suggest the low-density
population distribution may be more realistic. The range in results for the collective dose demonstrates
the strong dependence on the population data. However, the impact of geothermal electricity generation
remains small because of the limited use of this technology.

204. The next largest value for the normalized 222Rn discharge is for uranium mining, which is a factor
of about 20 larger than the value for coal mining. This is also reasonable because uranium mining
occurs where there are high levels of uranium and therefore of its progeny products such as 222Rn.
Shown in table 47 is a comparison of estimated doses to the public from the mining of coal and the
mining of uranium. The assessments resulted in a larger estimate for the characteristic individual dose
per unit of electricity generated for uranium mining (non-ISL) than for coal mining, consistent with the
larger normalized 222Rn discharge. However, values for both the worldwide collective dose (essentially
comprising local and regional components alone) and of the associated values normalized to unit of
electricity generated in 2010 are larger for coal mining than uranium mining. This is because of the
larger number of coal mines in the world and because the population density is higher around coal
mines compared to uranium mines.
214 UNSCEAR 2016 REPORT

Table 47. Comparison of doses to the public from mining of coal and mining of uranium

Coal mining Uranium mining

non-ISL ISL
Characteristic individual dose per unit of electricity generated 2.4 107
5.5 106
2.5 107
(Sv/(GW a))
Worldwide collective dose (man Sv) 3.7 102 4.0 101 1.3 100
World-average collective dose per unit of electricity generated 3.8 101 2.8 101 1.3 102
(man Sv/(GW a))

205. Figure VIII shows graphically the contribution each electricity-generating technology considered
makes to the worldwide collective dose to the public. Two distinct values are presented for the nuclear
fuel cycle: the first is the local and regional component for the population exposed directly to the
discharges, and the second includes the global component for the population of the world as a whole
due to globally circulating radionuclides. The values for both the coal cycle and geothermal energy are
deemed to represent boundary values, assuming either all modern or all older coal plants for the coal
cycle, and for geothermal energy assuming two different population distributions. Figure IX shows the
same sources as figure VIII but for the total collective dose normalized to electricity generated. The
values for the nuclear fuel cycle without global circulation are slightly less than for the coal cycle,
which lies somewhere between the two values given for the modern and older coal plants. Including the
global circulation to the nuclear fuel cycle results in the larger value for these two technologies.
Comparing these data with figure VIII shows how the magnitude of electricity generated for each
technology results in the collective dose to the public. Figure IX also shows the magnitude of the upper
bound for the collective dose normalized to electricity generation for geothermal energy.

206. Table 48 breaks down the comparison of collective dosesand collective doses normalized to
electricity generated in 2010for each of the source components of the nuclear fuel cycle and coal
cycle technologies that have been assessed. These components are coal mining, discharges from coal
combustion during power plant operation and from coal ash deposits associated with the coal cycle for
electricity generation. For the nuclear fuel cycle, these components are uranium mining and milling
including mill tailings, discharges during electricity generation from the nuclear power plant and
discharges from reprocessing. Collective doses normalized to electricity generated in 2010 for each of
the source components for the nuclear fuel cycle shown in table 48 use the electricity produced as a
result of each process in the normalizations.

207. Although the differences in collective doses normalized to electricity generation for the various
sources are not very large, (with the exception of old-style coal plants, which have the largest value),
the total collective doses in 2010 show significant differences. For both sets of results, the largest
values are from old-style coal combustion power plants. Collective doses for the different components
associated with the coal cycle are all larger than collective doses for the components associated with the
nuclear fuel cycle, except for modern coal plants compared to nuclear power plants. The categories
shown in table 48 make clear the importance of radon discharges from coal mining and from coal ash
deposits, whichwith the assumptions made in this studymake significant contributions to the total
collective doses even when the coal combustion power plant is modern. The differences in the dose
values for the old-style compared with the modern coal combustion power plant are due to the nuclides
in the 238U decay chain other than radon, because the radon discharge from the flue-gas stack is the
same in both cases. Table 49 shows this difference between modern versus old-style coal combustion
power plants regarding dose contribution per radionuclide from the 238U decay chain.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 215

208. Table 49 shows local and regional components of collective doses assessed from discharges of
radionuclides in the 238U decay chain from the flue-gas stack of a coal plant during the combustion of
coal for electricity generation in a modern compared to an old-style coal plant. The 222Rn discharges are
the same for both types of coal plant; consequently the doses from 222Rn are also the same in table 49. It
is clear from the breakdown of the radionuclides shown in table 49 that the large differences in the
doses assessed for the old-style versus modern coal plants are due to the other radionuclides shown,
with 210Pb and 210Po giving the largest contribution.

Figure VIII. Local and regional components of the collective doses to the public integrated to
100 years, for the electricity-generating technologies based on the nuclear fuel cycle, coal cycle,
combustion of gas and use of geothermal energy

Data from table 46

216 UNSCEAR 2016 REPORT

Figure IX. Local and regional collective doses to the public integrated to 100 years, normalized to
electricity generated for the technologies based on the nuclear fuel cycle, coal cycle, combustion of
gas and use of geothermal energy

Table 48. Comparison of collective doses to the public, and collective doses normalized to
electricity generation in 2010, integrated to 100 years, to the world-average population within a
1,500 km radius of each source for the electricity-generating technologies based on the coal cycle
and the nuclear fuel cycle

Coal Nuclear
Source Collective Normalized Source Collective Normalized
dose collective dose dose collective dose
(man Sv) (man Sv/(Gw a)) (man Sv) (man Sv/(Gw a))
Coal mining 370 0.4 Uranium mininga 53 0.2
and milling
Older coal plants 780 0.8 NPP generation 68 0.2

Modern coal plants 60 0.1

From coal ash deposits 240 0.2 Reprocessing 7.6 0.03

a
Of the 53 man Sv for uranium mining and milling, 40 man Sv is from mining only.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 217

Table 49. Local and regional components of collective dose, integrated to 100 years, due to
discharges from modern and old-style coal combustion power plants

Collective dose from modern coal plants Collective dose from old-style coal plants
Radionuclide (man Sv) (man Sv)
Local Regional Local Regional
210
Pb 2 9 28 135
210
Po 3 14 46 216
222
Rn 2 8 2 8
226
Ra 1 7 20 98
230
Th 2 8 26 122
234
U 0.5 2 7 34
238
U 0.4 2 6 29

Totala 11 50 135 641

a
Slight differences between the totals shown here and those in table 48 are due to rounding performed on the values in table 48.
The values in table 49 are left as generated in the assessment to avoid inconsistent rounding of individual radionuclides for the
purpose of attaining the same total rounded value as shown in table 48. It should also be recalled that in this table, as in all tables
in the annex resulting from assessments, the calculations are done to full precision and any discrepancies in the final sum of
numbers in the tables are due to rounding.

209. Doses assessed from the radon emanation from coal ash versus uranium mill tailings, using 222Rn
discharges normalized to electricity generated given in table 46, are compared in table 50. As described
in chapter IV and more extensively in annex A, a low population density was used in the assessments
on uranium mill tailings whereas a default population density was used in the assessment on coal ash.
The characteristic individual dose per unit of electricity generated is larger for uranium mill tailings
compared to coal ash, reflecting the greater activity concentrations of naturally occurring radionuclides,
including radon, in mill tailings. However, when the total electricity generation is taken into account,
collective doses assessed from coal ash are greater than those from mill tailings. The results show the
effect of the continued emanation of radon centuries into the future.

Table 50. Collective dose per unit of electricity generated to world-average population within
1,500 km and integrated to 100 years, also integrated for two different times since disposal, and
characteristic individual dose per unit of electricity generated from emanation of radon from coal
ash and mill tailings

Collective dose per unit of electricity Characteristic individual dose per unit of
Time since generated (man Sv/(GW a)) electricity generated (Sv/(GW a))
disposal
Coal ash Mill tailings Coal ash Mill tailings
100 years 0.2 0.04 1.5 10 7
8.4 107

500 years 1.2 0.2

210. Significant radionuclides from nuclear power plant discharges. The importance of 14C and the
globally circulating radionuclides, and 222Rn and the 238U series, is discussed in other sections. The
following concentrates on comparing other significant radionuclides discharged from different nuclear
power plant types.
218 UNSCEAR 2016 REPORT

211. Table 51 shows the world-average collective dose per unit of electricity generated for the nuclear
power plant types assessed in this annex. The table illustrates the relative importance to the world-
average collective dose per unit of electricity generated of 90Sr, 137Cs and 35S from the GCR-type
reactor, and tritium from the HWR reactor.

Table 51. Local and regional components of collective doses, integrated to 100 years, normalized
to electricity generation in 2010, shown for all nuclear power plant types considered

See chapter IV for details on the power plant types

Discharged World-average collective dose per unit of electricity generated for each nuclear power plant type
radionuclide (man Sv/(GW a))
PWR BWR HWR LWGR AGR GCR FBR
3
H 7.2 102
8.6 10 3
1.7 10 0
1.2 101
1.9 102
3.3 102
2.3 101
14
C 4.6 102 3.2 102 1.4 101 3.1 101 3.5 101 1.3 100 2.8 102
35
S 0 0 0 0 4.5 102 2.4 101 0
41
Ar 4.5 105
4.6 10 5
1.3 103
3.4 103
8.4 104
7.0 102
0
54
Mn 2.3 105
1.0 10 4
0 4.0 105
0 0 1.8 104
58
Co 3.7 105 4.0 105 0 5.9 106 0 0 1.7 104
60
Co 2.0 104 5.6 103 1.5 104 8.8 103 7.0 102 3.5 103 2.3 103
65
Zn 0 5.0 104 0 1.6 104 7.0 105 0 0
85
Kr 2.2 106
1.4 10 5
0 0 0 0 1.7 104
90
Sr 7.0 105 1.2 103 0 1.2 102 3.4 105 4.7 101 0
106
Ru 3.7 105 1.0 106 0 0 0 0 0
131
I 6.0 105
3.2 10 4
1.7 105
7.5 103
2.5 105
0 1.5 104
133
Xe 1.2 104 1.5 104 0 1.1 102 0 0 0
135
Xe 3.4 105 3.4 104 0 2.4 103 0 0 0
138
Xe 1.5 109 3.1 107 0 1.1 106 0 0 0
134
Cs 3.2 104
2.6 10 4
0 3.4 104
7.4 104
3.8 102
1.1 102
137
Cs 4.2 104 1.0 103 4.6 102 6.6 103 6.6 104 6.2 101 7.7 103

Total 1.2 101 5.0 102 1.9 100 4.8 101 4.9 101 2.8 100 2.8 101

212. Figure X shows estimated collective doses by world region and totalled for the world by nuclear
reactor type, with the values taken from table 51. Both collective doses, and collective doses
normalized to electricity generated in 2010 are shown. The collective doses show a different profile
than that shown in the figure displaying the collective doses normalized to electricity generated. Two
nuclear power plant types, GCRs and HWRs, dominate the collective doses normalized to electricity
generated. However it is the power plant types PWRs and HWRs that dominate the contribution to the
collective doses; this is because of a combination of the number of reactors that exist in each region
coupled with the discharged amounts of significant radionuclides for each reactor type.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 219

Figure X. Local and regional collective doses to the public integrated to 100 years, and associated
values normalized to electricity generation, by world region and nuclear reactor type

Results shown for collective doses from all discharges (man Sv) and for collective dose from all discharges normalized to
electricity generation (man Sv/(GW a)) in 2010. Note that the order of the individual reactor types is according to their world
share of electricity generated in 2010, as shown in table 9
220 UNSCEAR 2016 REPORT

B. Comparison of public and occupational exposures

213. Table 52 summarizes the results from all occupational and public exposures that have been
presented in this annex. The table shows collective doses normalized to electricity generated in 2010,
followed by collective doses for the same year. The local and regional components of doses to the
public are for world-average populations out to 100 km and between 100 km and 1,500 km from a
facility respectively, integrated over 100 years. Other conditions are as stated earlier in this chapter.
This section considers the total exposure from each electricity-generating technology, i.e. public plus
occupational exposures.

214. The occupational exposures from coal mining dominate collective doses, so that the total
exposure of the public and occupational exposures combined is largest from the coal cycle. This is true
even considering the public exposures from the nuclear fuel cycle including the global circulation
integrated out to 10,000 years. The next largest is the total exposure from the nuclear fuel cycle, where
public exposure including global circulation gives the highest dose followed by occupational exposures.
The same trend is observed for the associated collective doses normalized to electricity generated, with
the exception of the public exposures from the nuclear fuel cycle normalized to electricity generated
integrated out to 10,000 years, which gave the largest value.

215. The comparisons of public doses shown in table 52 that have been assessed using the revised
methodology for estimating public exposures due to radioactive discharges (annex A) have been described
in section A of this chapter. They are provided in this table to facilitate comparison with occupational
exposures and view the entire exposure, public and occupational, from each electricity-generating
technology. One comparison of interest noticeable in table 52 is that collective doses to the public
resulting from radon discharges from natural gas power plants are about the same as collective doses to
the public resulting from the discharges from operation of nuclear power plants (55 versus 68 man Sv,
respectively). However, the collective dose to the public normalized to electricity generated in 2010 is
about a factor of four larger for the nuclear fuel cycle than for the combustion of natural gas (not including
global circulation). The relatively large actual collective dose assessed for combustion of natural gas
results from the larger amount of electricity produced by natural gas than nuclear power in 2010.

216. The assessments on collective dose provided in the annex can be put in perspective by
considering natural exposure to background radiation from 222Rn. All doses to the public from natural
gas, oil, and geothermal energy are calculated from the release of 222Rn. A significant fraction of the
doses from the coal and nuclear cycles are also due to the release of 222Rn. Radon-222 occurs
naturally in nature and is a major fraction of the dose to man from all naturally occurring sources of
radiation. Wilkening et al. [W7] have estimated that the average exhalation rate of 222Rn from soil is
1.6 102 Bq (m2 s)1. As a result of this exhalation from soil and the additional exhalation of 222Rn
from building materials, the Committee [U9] has previously estimated that the average effective dose
from 222Rn is 1.15 mSv a1. For the world population considered here, the collective dose due to
exposure to naturally occurring radon is 11,500,000 man Sv a1, a value much larger than any given
in table 46.

217. Total occupational collective doses are significantly greater than public collective doses for the
coal cycle and combustion of oil, from a factor of about ten for coal plants to about five hundred for
combustion of oil, although the values for combustion of oil are in comparison small. For natural gas
and geothermal energy, total public collective doses were assessed to be larger than total occupational
collective doses. Total occupational collective doses for the nuclear fuel cycle are of about the same
order as the total collective dose to the public when including the global circulation to 100 years.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 221

218. The largest occupational collective dose normalized to energy generated in 2010 resulting from
the mining for metals for construction materials was from solar photovoltaic (PV) technology, which
was a factor of forty and eighty larger than for the nuclear fuel cycle and coal cycle, respectively. This
was followed by the occupational collective dose for wind power, which was also larger than the values
for the nuclear fuel cycle and coal cycle. These differences come from the different metal requirements
for solar PV and wind power technologies, discussed in chapter VII.
 222 UNSCEAR 2016 REPORT
Table 52. Comparison of the public and occupational exposures assessed in this annexa,b

Modern Older coal

Nuclear fuel cycle Coal Natural gas Oil Geothermal Solar PV Wind Biomass
coal plant plant
COLLECTIVE DOSES NORMALIZED TO ELECTRICITY GENERATION (man Sv/(GW a))
Public U mining and 0.2 Coal mining 0.4 0.4
milling (Rn discharge)

Nuclear power 0.2 Coal plant 0.1 0.8

plant operation operation

Reprocessing 0.03 Ash, radon 0.2 0.2

emanation

Total public (not including globally 0.43 0.7 1.4 0.1 0.000 3 120
circulating radionuclides)

Total public (including 100 years 3.0

globally circulating
500 years 5.5
radionuclides)
10 000 years 25

Occupational 2.7c 11 11 0.01 0.15 0.05

Occupational - decommissioning, nuclear 1.8

Occupational - mining for construction 0.02 0.01 0.01 0.01 0.8 0.1 0.01

COLLECTIVE DOSES (man Sv)

Public U mining and 53 Coal mining 370 370
milling (Rn discharge)

Nuclear power 68 Coal plant 60 780

plant operation operation

Reprocessing 8 Ash, radon 240 240

emanation

Total public (not including globally 130 670 1 400 55 0.03 5160
circulating radionuclides)
 Modern Older coal
Nuclear fuel cycle Coal Natural gas Oil Geothermal Solar PV Wind Biomass
coal plant plant
Total public (including 100 years 910
globally circulating
500 years 1 700
radionuclides)
10 000 years 7 600

Occupational 788 11 000 11 000 7 17 0.40.8

Decommissioning (occupational) per nuclear 5

reactor

Occupational - mining for construction 6 7 7 3 3 4 0.4

a
Projections of any health effects using collective doses in the table are not recommended.
b
All estimates are calculated based on best estimates; site- and location-specific collective doses are not presented.
c
From UNSCEAR 2008 Report.

ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 223

224 UNSCEAR 2016 REPORT

C. Commentary on significance of accidents

219. While this annex has focused on comparing the exposures from normal operations of the various
electricity-generating technologies, a commentary is needed regarding the risk of serious accidents that
give rise to radiation exposure. This is clearly only significant for nuclear power. While it is beyond the
competence of the Committee to assess the probability of any future accident, the Committee has
assessed information on past accidents that have exposed the public and workers, notably the 1986
accident at the Chernobyl nuclear power plant in the former Soviet Union (UNSCEAR 2008 Report,
annexes C and D [U12]). Moreover since that report, the Committee has conducted an assessment of
the levels and effects of radiation exposure due to the 2011 nuclear accident at the Fukushima Daiichi
nuclear power station (FDNPS). The UNSCEAR 2013 Report [U13] gave estimates of the collective
doses to the population of Japan due to the FDNPS accident. These estimates were compared with the
previous estimates by the Committee for populations of European countries exposed to radiation
following the 1986 Chernobyl accident in the former Soviet Union. This comparison is shown in
table 53.

Table 53. Estimates of the collective effective doses from the FDNPS and Chernobyl accidents [U13]

Accident Collective effective dose (thousand man Sv)

Over first year Over ten years Up to age 80 yearsa
Fukushima Daiichi nuclear power station 18 36 48

Chernobyl Unit 4 400

a
Summing the dose to all exposed individuals integrated from their age at the time of the accident until they reach age 80 years.

220. Comparing radiation doses alone, it is clear that serious accidents give rise to collective doses that
are very many times greater than collective doses due to normal operations. For example, the collective
dose from the single accident at Chernobyl Unit 4 is more than 400 times the annual global collective
dose to the public from all nuclear power. Such a comparison of collective doses can be made only in
order to gain perspective on the magnitude of the radiological impact. However, great care must be
taken when assimilating these comparisons, because there are obviously many major non-radiological
differences between accidents and normal operations for electricity generation. While collective doses
from such accidents have been much larger than those from annual normal operations, the distribution
of doses is more localized geographically, whereas the collective doses from normal operations for
electricity generation are population-averaged over geographical regions or the world as a whole.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 225

IX. RESEARCH NEEDS

221. The use of electricity is ubiquitous throughout the world and the demand for electricity continues
to grow. Because of the prevalence of different types of electricity-generating technologies coupled
with the diversity inherent in the countries of the world, standard data on exposures to radioactive
discharges or materials is non-uniform or non-existent. Thus, any study comparing different electricity-
generating technologies will suffer from inconsistent or incomplete databases. This annex has used
consistent methodologies together with sound judgement on the use and interpretation of available data
to assess public and occupational exposures from electricity-generating technologies. Some specific
areas of interest that would improve understanding or that would increase the certainty of this work are
included in this section.

222. Data. Consistent data on radioactive discharges across all electricity-generating technologies,
including how they change with time and changing practices across all electricity-generating
technologies, would improve the ability to assess and compare these technologies. Data on occupational
exposures and their change with time and practice for each technology would also help. The following
lists some specific needs that have been identified:

(a) Decommissioning of coal power plants generates NORM wastes, the magnitude of which
should be quantified (NORM wastes are naturally occurring radionuclides such as lead and
polonium which in this case become plated out in pipes and the boiler, and other mechanical parts
of the plant [M5]). Occupational exposures received during decommissioning of coal power plants
need to be monitored in a representative manner in order to allow assessment of this exposure route
for workers. This is also true regarding occupational exposures of workers decommissioning
nuclear power plants.

(b) The combustion of natural gas and oil for electricity generation causes naturally occurring
radionuclides to accumulate in the pipes. Assessment of any occupational or public exposures from
this source is dependent on representative data on the activity concentrations that occur. There are
some data available on this, however the measured external gamma exposure rates are of more
significance.

(c) In order to assess the occupational doses resulting from the management of radioactive waste
generated in nuclear power plants, representative data on doses received by workers needs to be
collected or compiled [M5].

(d) Information on the proportion of nuclear fuel reprocessed for peaceful (i.e. electricity
generation) versus non-peaceful (i.e. military applications) has only been available for the La
Hague, France reprocessing facility and therefore the assessment of the doses due to discharges
from reprocessing activities was compromised by the lack of open information. Also, some of the
occupational doses at reprocessing facilities, such as Sellafield in the United Kingdom, are from
the reprocessing of historical wastes so it is very hard to relate occupational doses to just nuclear
fuel that has been reprocessed, or to electricity generation. Actual data on reprocessing activities
related to electricity generation are needed to improve the exposure assessments.

(e) Storage, transmission and distribution of electricity are not included in the assessments of dose
from electricity generation presented in this annex. These infrastructures are used regardless of the
type of electricity generation (although there are differences between centralized and distributed
grid systems that are optimal for different technologies), and these kinds of dose assessments could
be included in any future update of the evaluations presented in the annex.
226 UNSCEAR 2016 REPORT

223. Databases could be consistent and standardized; these types of efforts are currently conducted by
various international organizations. The success of these efforts, however, depends on the volunteer
compliance of member countries submitting annual reports to these organizations, such as UNSCEAR,
IAEA and NEA, and on funding and interest to keep the programmes current. These programmes
would benefit from extending the data-collection schemes to include all relevant technologies, and
investing in research to identify significant parameters to help design streamlined data-collection
schemes that would further decrease the burden on member countries and data-collecting organizations.

224. Uranium mining using the ISL process. Explanations were given in chapter IV of the assumptions
made about radon releases during uranium mining using the ISL process. Existing data are
inconclusive, and radon could be released in varying quantities from the lixiviant used to dissolve the
uranium in the underground ore body, depending on the specifics of the process. Determining more
precisely the amount of radon releases from ISL uranium recovery, and how to better generalize them,
requires additional research and a more complete understanding of the variation in the global processes
and procedures currently in use for mining of uranium using the ISL process.

225. Recycling of metals. In chapter VII on assessing doses from the mining of metals needed in the
construction phase, several assumptions were made in the assessments. Although the assumptions affect
the absolute value of the assessed collective doses, the relative importance when comparing the
different electricity-generating technologies with regards to the construction phase should remain
relevant. However, these assumptions make a difference when comparing to the other components in
the full cycle of electricity generation, such as in table 52. One area of importance that could be studied
further is the effect of recycling metals, which has not been included in the present study because of
insufficient data for large portions of the world. By not explicitly considering recycling, the collective
dose per unit of electricity generated due to the extraction and processing of the metals used in the
construction of a facility is likely to be overestimated by around a factor of two on average, and
possibly a factor of five or more for some regions of the world with developed recycling processes
(assuming the recycled metal is not contaminated with radioactive material). Also, recycling may affect
the different technologies somewhat differently because it is more difficult and costly to recycle
electronics and components (such as in solar panels) than larger bulk metal objects (such as in more
conventional power plant components).

226. Reference year. This annex concentrates on the reference year 2010 for the assessments. Lifetime
commitments would be relevant for comparing the full impact of the different electricity-generating
technologies. Confounders in this type of extrapolation are associated with the changes in the respective
electricity-generating technologies over time, caused by, for example, improving effectiveness of each
technology in both economic and environmental terms and in radiation protection practices.

227. Suess effect. The displacement of 14C in atmospheric CO2 discharges during the coal cycle and
other fossil fuel releases of CO2 (which are depleted in 14C) is known as the Suess effect. Research into
this effect would improve knowledge of the global carbon cycle and could help clarify the role of 14C in
human exposures in the modern world.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 227

X. CONCLUSIONS
228. This annex provides estimates of exposures from the various electricity-generating technologies
that (a) may be used by researchers and policy-makers in their own more comprehensive assessments
for developing energy policy; (b) can be used to help inform the media and the public on these matters
in a balanced perspective; and (c) can highlight possible emerging issues or opportunities for
improvement that may warrant more attention and scrutiny, or future research. The following
summarizes the findings.

229. The Committee has updated its methodology for estimating public exposures due to radioactive
discharges, which is now more flexible for use in evaluating radiation exposures to the public from
diverse electricity-generating technologies. This methodology along with extensive data collection and
analyses has provided the Committee with a sounder basis for comparative studies than was possible
earlier. The Committee has also re-evaluated occupational exposures arising from different electricity-
generating technologies relying on data mainly from dosimetry records of worker exposures. These
evaluations comprised the basis for the current comparative study on radiation exposures of both the
public and of workers from electricity generation. To compare exposures, the Committee has focused
on two metrics. These were (a) the collective doses to defined population groups integrated over
specific time periods resulting from one years electricity generation by each technology in each
geographical region and for the world as a whole, and (b) the relevant collective doses divided by the
amount of electricity generated by each technology. The year 2010 was used as the reference year for
the comparisons.

230. The Committee has conducted this comparative study by investigating sources of radiation
exposure from electricity-generating technologies based on the (a) nuclear fuel cycle, (b) coal cycle, the
combustion of natural gas, oil and biofuels, and (c) geothermal, wind and solar power. Two electricity-
generating technologies (nuclear fuel cycle and coal cycle) were investigated in detail because a more
robust database existed for these technologies. The Committee evaluated the main sources of
radioactive discharges from the life cycle of these electricity-generating technologies. For the nuclear
fuel cycle, these were uranium mining, milling and mill tailings, power plant operation, spent fuel
reprocessing, and decommissioning activities. For the coal cycle, they were the mining for coal, power
plant operation for both a modern coal plant and an older-style coal plant, and coal ash deposits.

231. The Committee estimated that, excluding long-lived globally circulating radionuclides, the
contribution from the coal cycle, assuming discharges from a modern coal plant, was more than half of the
total collective dose to the global public from the discharges due to a single years global electricity
generation, while the nuclear fuel cycle contributed less than a fifth. The contribution from the coal cycle
comes from discharges of natural radionuclides (primarily radon and its radioactive progeny) during coal
mining, combustion of coal at the power plant and from coal ash deposits. Similarly, almost half of the
contribution to public exposures from the nuclear fuel cycle also comes from discharges of natural
radionuclides during uranium mining and milling activities. These values depend on the amount of
electricity generated by each technology; in 2010, the coal cycle produced the largest amount of total
electricity generated, about 40%. Although radon and its progeny are relatively important contributors to
collective doses to the public for both the nuclear fuel cycle and the coal cycle, the associated individual
doses are small compared with doses due to inhalation of naturally occurring radon [U9, W7].
228 UNSCEAR 2016 REPORT

232. The Committee found, however, that the contribution to the collective dose to the public from
each electricity-generating technology was not only because of how much electricity each technology
generated. There were differences due to the collective doses per unit of electricity generated. In normal
operations, the coal cycle gives a higher collective dose per unit of electricity generated than electricity
generation from nuclear power plants, and significantly higher than the other technologies evaluated
with the exception of geothermal energy. Based on the limited information on radon discharges from
geothermal power plants, the collective dose per unit of electricity generated by geothermal energy
could be significant. However, because the use of geothermal technology is not widespread, its
contribution to radiation exposures of the global public is smaller than that of the coal cycle.

233. Previous investigations on electricity generation from the nuclear fuel cycle have examined the
contribution to human exposures made by long-lived radionuclides, such as carbon-14, which are
discharged, circulate globally and continue to contribute to radiation exposures of the public centuries
into the future, albeit as extremely small individual doses. The Committee found that public exposures
from these globally-circulating radionuclides, for one year of discharge and integrated to 100 years,
result in a contribution to exposures from the nuclear fuel cycle that are about the same as the coal
cycle. Over long integration times such as hundreds of years, the contribution from these radionuclides
results in larger collective doses to the global public from the nuclear fuel cycle than the coal cycle.

234. The Committee also assessed the occupational exposures for these technologies. The largest
collective dose to workers per unit of electricity generated resulted from coal mining, because of
exposures to naturally occurring radionuclides. Of all the collective doses evaluated, to both the public
and workers, the exposure of workers from coal mining gave the largest contribution, although it has
fallen over time because of better mining conditions. Regarding the mining of rare earth metals needed
for construction, by far the largest collective dose to workers per unit of electricity generated assessed
in this study came from solar power, followed by wind power. This is because the workers are exposed
to natural radionuclides during mining, and the amount of low-grade ore required to be mined for these
technologies is high.

235. The total collective dose (i.e. to the global public and all exposed workers combined) per unit of
electricity generated by the coal cycle was larger than that generated by the nuclear fuel cycle, even
when considering the long-lived globally-circulating radionuclides integrated out to 500 years. When
considering the amount of electricity generated in the year 2010 by each technology, the coal cycle
resulted in the largest collective dose to the global public and workers combined, followed by the
nuclear fuel cycle. Of the remaining technologies, geothermal energy and combustion of natural gas
were the next largest contributors.

236. Great care should be taken when interpreting and using these results, because this analysis only
gives a perspective on the magnitude and differences of radiation exposures, and cannot be used to
determine whether one form of energy generation is preferable to another. As stated earlier, a number
of factors determine why a certain mix of energy generation technologies may be selected by countries.
Radiation exposure is only one of them.

237. Moreover, when comparing collective doses for the various electricity-generating technologies, it
is important to note that the collective dose from serious accidents, such as those that occurred at the
Chernobyl and Fukushima-Daiichi nuclear power stations, were orders of magnitude larger than the
collective doses to the world population from one years normal operation of all the technologies of
electricity generation that were assessed in the annex. More significantly, the distribution of doses after
an accident is more localized geographically (local populations receive higher doses than the average),
whereas the collective doses from normal operations for electricity generation are often averaged over
populations within each geographical region or the world as a whole.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 229

XI. ACKNOWLEDGEMENTS
The Committee wishes to acknowledge with gratitude the experts directly involved in compiling data,
conducting analyses and quality control checks, and drafting text and conclusions, and the support staff
in their institutes who assisted them. In particular, the Committee would like to thank: the consultant,
L. Hubbard; the expert group, who provided valuable oversight and advice for the work; staff,
T. Anderson and K. Jones, from Public Health England in the United Kingdom, who conducted many
of the analyses; L. Anspaugh, H. Grogan, V. Holahan and J. Simmonds, who provided helpful advice,
drafted text, and reviewed and checked the results. The views expressed in this scientific annex and
related material remain those of the Committee and do not necessarily represent the views of individual
experts, the United Nations or its Member States.

Expert group

Co-chairs: E. Waller (Canada)

Members: T. Anderson (United Kingdom), L. Anspaugh (United States), H. Grogan (United States),
G. Hirth (Australia), V. Holahan (United States), K. Jones (United Kingdom), L. Hubbard (Sweden),
R. Michel (Germany), J. Simmonds (United Kingdom)

Observer: G. Proehl (IAEA)

The Committee would also like to acknowledge with gratitude: E. Rochedo, who conducted some of
the preparatory work to compile data; B. Lauritzen, who compiled some of the data and conducted
some of the analyses on occupational exposures due to mining of metals for construction purposes; and
M. Tzivaki for conducting some specialized quality checks of the calculations.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 231

REFERENCES
A1 ACAA. The American Coal Ash Association. [Internet] Available from (www.ACAA-USA.org)
on 13.01.2016.

A2 ACAA. 2010 Coal Combustion Product (CCP) production and use survey report. American
Coal Ash Association. [Internet] Available from (https://www.acaa-usa.org/Portals/9/Files/
PDFs/2010_CCP_Survey_FINAL_102011.pdf) on 28.02.2016.

A3 AK Steel. Product data sheet. 316/316L stainless steel. UNS S31600 and UNS S31603. AK
Steel Corporation. [Internet] Available from (http://www.aksteel.com/pdf/markets_products/
stainless/austenitic/316_316l_data_sheet.pdf) on 24.03.2016.

A4 Al-Masri, M.S. and R. Shwiekani. Radon gas distribution in natural gas processing facilities
and workplace air environment. J Environ Radioact 99(4): 574-580 (2008).

A5 Al-Masri, M.S. and K. Haddad. NORM emissions from heavy oil and natural gas fired power
plants in Syria. J Environ Radioact 104: 71-74 (2012).

A6 Al-Masri, M.S., K. Haddad, A.W. Doubal et al. Assessment of soil contamination by (210)Po
and (210)Pb around heavy oil and natural gas fired power plants. J Environ Radioact 132:
89-93 (2014).

A7 Al-Saleh, F.S. and G.A. Al-Harshan. Measurements of radiation level in petroleum products
and wastes in Riyadh City Refinery. J Environ Radioact 99(7): 1026-1031 (2008).

A8 Althaus, H.-J., R. Hischier, M. Osses et al. Life cycle inventories of chemicals. Data v2.0.
Ecoinvent Report No. 8. Swiss Centre for Life Cycle Inventories, Dbendorf, CH, 2007.

A9 ANL. Energy and land use. DOE/EV/10154-5. Argonne National Laboratory, 1981.

A10 Anspaugh, L.R. Final report on the investigation of the impact of the release of radon-222, its
daughters, and precursors at The Geysers Geothermal Field and surrounding area. MISC-2818.
Lawrence Livermore National Laboratory, Livermore, California, 1978.

B1 Behlen Building Systems. Facts on building green. Behlen Building Systems. [Internet]
Available from (http://www.behlenbuildingsystems.com/BuildingGreen.htm#RecycledContent)
on 24.11.2015.

B2 Bell, K.G., C. Goodman and W.L. Whitehead. Radioactivity of sedimentary rocks and
associated petroleum. Bull Am Assoc Petrol Geol 24(9): 1529-1547 (1940).

B3 Borromeo, J. Darlington Radiological Environmental Monitoring Report (REMP) Review.

NK38-REP-03481-10003-R000. Ontario Power Generation, 2008.

B4 Britannica. The nuclear fuel cycle. Britannica Nuclear Reactors. [Internet] Available from
(http://www.britannica.com/EBchecked/topic/421763/nuclear-reactor/307282/Emergency-
response?anchor=ref272869) on 02.05.2015.

B5 Brown, A. Pickering Radiological Environmental Monitoring Report (REMP) Review - 2006.

P-REP-03481-00002-R000. Ontario Power Generation, 2007.
232 UNSCEAR 2016 REPORT

B6 Brown, S.H. and R.C. Smith. A model for determining the overall radon release rate and
annual source term for a commercial in-situ leach uranium facility (Chapter 119). in:
International Conference on Radiation Hazards in Mining: Control, Measurement and Medical
Aspects, 4-9 October 1981, Colorado School of Mines, Golden, CO (M. Gomez, ed.) Society
of Mining Engineers of the American Institute of Mining, Metallurgical, and Petroleum
Engineers, New York, 1981.

B7 Brown, S.H. Design improvements and ALARA at U.S. uranium in situ recovery facilities.
pp.649-664 in: ASME 2009 Proceedings of the 12th International Conference on Environmental
Remediation and Radioactive Waste Management, Liverpool, UK, 11-15 October 2009. Paper
no. ICEM2009-16415. The American Society of Mechanical Engineers (ASME), 2009.

B8 Bryan, R.H. and I.T. Dudley. Estimated quantities of materials contained in a 1000-MW(e)
PWR power plant. ORNL-TM-4515. OAK Ridge National Laboratory, Tennessee, 1974.

B9 Bunce, L.A. and F.W. Sattler. Radon-222 in natural gas. Radiol Health Data Rep 7(8): 441-
444 (1966).

C1 Capudean, B. Carbon content, steel classifications and alloy steels. Fabricators &
Manufacturers Association, Intl. [Internet] Available from (http://www.thefabricator.com/
article/metalsmaterials/carbon-content-steel-classifications-and-alloy-steels) on 24.02.2012.

C2 Cheng, C. 2011 Results of Radiological Environmental Monitoring Programs. N-REP-03481-

10010-R000. Ontario Power Generation, 2012.

C3 Classen, M., H.-J. Althaus and S. Blaser. Life cycle inventories of metals. Data v2.1.
Ecoinvent Report No. 10. Swiss Centre for Life Cycle Inventories, Dbendorf, CH, 2009.

C4 Coles, D.G., R.C. Ragaini and J.M. Ondov. Behavior of natural radionuclides in western coal-
fired power plants. Environ Sci Technol 12(4): 442-446 (1978).

C5 Corbett, J.O. The radiation dose from coal burning: a review of pathways and data. Radiat Prot
Dosim 4(1): 5-19 (1983).

D1 D'Amore, F. Radon-222 survey in Larderello geothermal field, Italy (Part 1). Geothermics
4(1): 96-108 (1975).

D2 Dannemand Andersen, P., M. Borup and T. Krogh. Managing long-term environmental

aspects of wind turbines: a prospective case study. Int J Technol Policy Manage 7(4): 339-354
(2007).

D3 de Jong, P., W. van Dijk, E.R. van der Graaf et al. National survey on the natural radioactivity
and 222Rn exhalation rate of building materials in The Netherlands. Health Phys 91(3): 200-
210 (2006).

D4 de Jong, P., W. van Dijk and M. de Rooij. Influence of the porosity on the (222)Rn exhalation
rate of concrete. Health Phys 100(2): 127-137 (2011).

D5 Dones, R., C. Bauer, R. Bolliger et al. Life cycle inventories of energy systems: Results for
current systems in Switzerland and other UCTE countries. Data v2.0. Ecoinvent Report No. 5.
Swiss Centre for Life Cycle Inventories, Dbendorf, CH, 2007.

D6 Drummond, I., P. Boucher, B. Bradford et al. Occurrence of 222Rn and progeny in natural gas
processing plants in western Canada. Health Phys 59(1): 133-137 (1990).
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 233

E1 EC. Assessment of the radiological impact on the population of the European Union from
European Union nuclear sites between 1987 and 1996. Radiation Protection 128 (J.G. Smith et
al., eds.). European Communities, Luxembourg, 2002.

E2 EC. European Commission Radioactive Discharges Database (RADD) for collecting, storing,
exchanging and dissemination of information on radioactive discharges. European
Commission. [Internet] Available from (http://europa.eu/radd/) on 9 and 21 December 2015.

E3 EPRI. Coal ash: Characteristics, management and environmental issues. Technical Update.
Electric Power Research Institute. [Internet] Available from (http://www.epri.com/
abstracts/Pages/ProductAbstract.aspx?ProductId=000000000001019022).

F1 Feuerborn, H.-J. Coal Combustion Products in Europe - an update on production and

utilisation, standardisation and regulation. World of Coal Ash (WOCA) Conference, Denver,
CO, USA, 9-12 May 2011.

F2 FSUE.
2013 , Report on the Environmental Safety of the Federal State Unitary
Enterprise "Production Association Mayak" in 2013. Eco-report-2013-Mayak. Federal State
Unitary Enterprise, Ozersk, 2014. (Russian).

F3 Fthenakis, V. and H.C. Kim. Land use and electricity generation: A life-cycle analysis.
Renewable and Sustainable Energy Reviews 13(67): 1465-1474 (2009).

G1 George, A.C., G. Scandiffio and A.J. Breslin. The distribution of radon and radon daughters in
the geothermal region of Larderello. pp.3-14 in: Proceedings - Second DOE-ENEL Workshop
for Cooperative Research in Geothermal Energy. LBL-11555. Lawrence Berkeley Laboratory,
California, 1980.

G2 Gesell, T.F. Occupational radiation exposure due to 222Rn in natural gas and natural gas
products. Health Phys 29(5): 681-687 (1975).

G3 Globalspec. Carbon steels and alloy steels information. [Internet] Available from
(http://www.globalspec.com/learnmore/materials_chemicals_adhesives/metals_alloys/ferrous_
metals_alloys/carbon_steels_alloy_steels) on 27.02.2012.

G4 Green, D.E. and V. Downing. The report on the pilot study for the measurement of the
occupational exposure of Drax Power Station staff to airborne dust. NER/SSD/M/80/45. UK
Central Electricity Generating Board, 1981.

G5 Green, D.E. and M.M. Tylee. The report on the pilot study for the measurement of the
occupational exposure of Ferrybridge B Power Station staff to airborne dust.
NER/SSD/M/81/180. UK Central Electricity Generating Board, 1981.

H1 Hamlat, M.S., S. Djeffal and H. Kadi. Assessment of radiation exposures from naturally
occurring radioactive materials in the oil and gas industry. Appl Radiat Isot 55(1): 141-146
(2001).

H2 Hedvall, R. and B. Erlandsson. Radioactivity concentrations in non-nuclear industries. J

Environ Radioact 32(12): 19-31 (1996).

H3 Heidrich, C., S. Brown and D. Collier. Naturally occurring radionuclides in Australian Coal
Combustion Products (CCPs). World of Coal Ash (WOCA) Conference, Denver, CO, USA, 9-
12 May 2011.
234 UNSCEAR 2016 REPORT

I1 IAEA. Radiation protection and the management of radioactive waste in the oil and gas
industry. Safety Reports Series No. 34. International Atomic Energy Agency, Vienna, 2003.

I2 IAEA. Extent of environmental contamination by naturally occurring radioactive material

(NORM) and technological options for mitigation. Technical Reports Series No. 419.
International Atomic Energy Agency, Vienna, 2003.

I3 IAEA. Extending the operational life span of nuclear plants. Major International Conference
on Nuclear Plant Life Managements, Shanghai, 15-18 October 2007. International Atomic
Energy Agency. [Internet] Available from (https://www.iaea.org/newscenter/news/extending-
operational-life-span-nuclear-plants).

I4 IAEA. World distribution of uranium deposits (UDEPO) with uranium deposit classification.
IAEA-TECDOC-1629. International Atomic Energy Agency, Vienna, 2009.

I5 IAEA. Measurement and calculation of radon releases from NORM residues. Technical
Reports Series no. 474. International Atomic Energy Agency, Vienna, 2013.

I6 IAEA. PRIS - Power Reactor Information System. International Atomic Energy Agency.
[Internet] Available from (http://www.iaea.org/pris/home.aspx) on 11.10.2013.

I7 IAEA. PRIS - Power Reactor Information System. International Atomic Energy Agency.
[Internet] Available from (http://www.iaea.org/pris/home.aspx) on 18.11.2015.

I8 IAEA. Permanent shutdown reactors. PRIS - Power Reactor Information System. International
Atomic Energy Agency. [Internet] Available from (http://www.iaea.org/PRIS/WorldStatistics/
ShutdownReactorsByCountry.aspx) on 20.07.2015.

I9 IAEA. PRIS - Power Reactor Information System. International Atomic Energy Agency.
[Internet] Available from (https://www.iaea.org/pris/) on 14.07.2015.

I10 IAEA. Naturally occurring radioactive material (NORM VII). Proceedings of an International
Symposium, Beijing, China, 2226 April 2013. International Atomic Energy Agency, Vienna,
2015.

I11 ICRP. Lung cancer risk from radon and progeny and statement on radon. ICRP Publication
115. Annals of the ICRP 40(1). International Commission on Radiological Protection, Elsevier
Ltd., 2010.

I12 IEA. World energy outlook 2004. International Energy Agency, OECD/IEA, 2004.

I13 IEA. Key world energy statistics. International Energy Agency, OECD/IEA, Paris, 2008.

I14 IEA. Coal information 2012. IEA Statistics. International Energy Agency, OECD/IEA, Paris,
2012.

I15 IEA. Electricity information 2012. IEA Statistics. International Energy Agency, OECD/IEA,
Paris, 2012.

I16 IEA. World energy outlook 2012. International Energy Agency, OECD/IEA, Paris, 2012.

I17 IEA. Energy statistics of OECD countries 2013. International Energy Agency, OECD/IEA,
Paris, 2013.

I18 IEA Clean Coal Centre. Life extension of coal-fired power plants. PF 05-13 (http://www.iea-
coal.org.uk/documents/81405/5990/Life-extension-of-coal-fired-power-plants). International
Energy Agency Clean Coal Centre, London, 2005.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 235

I19 InfoMine. Major mining operations around the world. [Internet] Available from
(http://www.infomine.com/minesite/).

I20 ISOE. ISOE database. Information System on Occupational Exposure. [Internet] Available
from (http://www.isoe-network.net/) on 14.03.2016.

J1 Jungbluth, N., M. Stucki and R. Frischknecht. Photovoltaics. Part XII. Data v2.1. ecoinvent
report No. 6. Swiss Centre for Life Cycle Inventories, Dbendorf, CH, 2009.

K1 Kable Intelligence Limited. The top 10 biggest geothermal power plants in the world.
[Internet] Available from (http://www.power-technology.com/features/feature-top-10-biggest-
geothermal-power-plants-in-the-world/) on 12.11.2013.

K2 Kolb, W.A. and M. Wojcik. Enhanced radioactivity due to natural oil and gas production and
related radiological problems. Sci Total Environ 45: 77-84 (1985).

K3 Kruger, P., A. Stoker and A. Umaa. Radon in geothermal reservoir engineering. Geothermics
5(14): 13-19 (1977).

K4 Kvasnicka, J. Radiation protection in the offshore petroleum industry. Volume 4. pp.4-625-

627 in: IRPA9: Proceedings of the International Congress on Radiation Protection
Association, Vienna, 14-19 April 1996.

L1 Liu, F.D., Z.Q. Pan, S.L. Liu et al. The estimation of the number of underground coal miners
and the annual dose to coal miners in China. Health Phys 93(2): 127-132 (2007).

L2 Long, S., S. Sdraulig, B. Tate et al. A survey of naturally occurring radioactive material
associated with mining. Technical Report Series No. 161. Australian Radiation Protection and
Nuclear Safety Agency, 2012.

L3 Lund, J.W. 100 years of geothermal power product. Proceedings, Thirtieth Workshop on
Geothermal Reservoir Engineering, 31 January-2 February 2005, Stanford University,
California. 2005.

M1 Made How. How products are made: copper. Volume 4. [Internet] Available from
(http://www.madehow.com/Volume-4/Copper.html) on 31.03.2011.

M2 Marovi, G., Z. Frani, J. Senar et al. Natural radionuclides in coal and waste material
originating from coal fired power plant. IRPA 12: International Congress of the International
Radiation Protection Association - Strengthening Radiation Protection Worldwide, Buenos
Aires, Argentina, 19-24 October 2008.

M3 Matek, B. 2015 Annual U.S. & global geothermal power production report. Geothermal
Energy Association, Washington, D.C., 2015.

M4 McBride, J.R. Offsite radiological surveillance for Project Gasbuggy June 1967-July 1968.
Radiol Health Data Rep 10(12): 535-546 (1969).

M5 Mora, J.C., B. Robles, J.A. Corbacho et al. Modelling the behaviour of 210Po in high
temperature processes. J Environ Radioact 102(5): 520-526 (2011).

N1 NEI. Nuclear Engineering International Handbook 2007. Nuclear Engineering International,

Kent, UK, 2007.

N2 NREL. Parabolic trough solar field technology. National Renewable Energy Laboratory.
[Internet] Available from (http://www.nrel.gov/csp/troughnet/solar_field.html).
236 UNSCEAR 2016 REPORT

N3 NREL. Life cycle greenhouse gas emissions from solar photovoltaics. NREL/FS-6A20-56487
(http://www.nrel.gov/docs/fy13osti/56487.pdf). National Renewable Energy Laboratory, 2012.

N4 NREL. Wind LCA harmonization. NREL/FS-6A20-57131 (http://www.nrel.gov/

docs/fy13osti/57131.pdf). National Renewable Energy Laboratory, 2013.

O1 O'Donoughue, P.R., G.A. Heath, S.L. Dolan et al. Life cycle greenhouse gas emissions of
electricity generated from conventionally produced natural gas. J Ind Ecol 18(1): 125-144
(2014).

O2 OECD. Nuclear energy data 2007. NEA No. 6248. OECD Nuclear Energy Agency, 2007.

O3 OECD. OECD Factbook 2011-2012. Economic, environmental and social statistics.

Organisation for Economic Co-operation and Development, OECD Publishing, Paris, 2011.

O4 OECD/NEA. Occupational exposures at nuclear power plants. Twentieth annual report of the
ISOE programme, 2010. Information System on Occupational Exposure (ISOE).
NEA/CRPPH/ISOE(2010)5. Jointly sponsored by the OECD Nuclear Energy Agency and the
International Atomic Energy Agency. OECD, 2010.

O5 OECD/NEA. Occupational exposures at nuclear power plants. Twenty-second annual report of

the ISOE Programme, 2012. Information System on Occupational Exposure (ISOE).
NEA/CRPPH/ISOE(2012)8. Jointly sponsored by the OECD Nuclear Energy Agency and the
International Atomic Energy Agency. OECD, 2012.

O6 OECD/NEA. Uranium 2011: Resources, Production and Demand. NEA No. 7059. A Joint
Report by the OECD Nuclear Energy Agency and the International Atomic Energy Agency.
OECD, 2012.

O7 OECD/NEA. ISOE country reports. Information System on Occupational Exposure (ISOE).

Jointly sponsored by the OECD Nuclear Energy Agency and the International Atomic Energy
Agency. OECD, 2013.

O8 OECD/NEA. Uranium 2014: Resources, Production and Demand. NEA No. 7209. A Joint
Report by the OECD Nuclear Energy Agency and the International Atomic Energy Agency.
OECD, 2014.

O9 Organo, C. and D. Fenton. Radiological assessment of NORM industries in Ireland

Radiation doses to workers and members of the public. Radiological Protection Institute of
Ireland, 2008.

O10 Ortegon, K., L.F. Nies and J.W. Sutherland. Preparing for end of service life of wind turbines.
J Cleaner Prod 39: 191-199 (2013).

P1 Pasqualetti, M.J. Geothermal energy and the environment: The global experience. Energy 5(2):
111-165 (1980).

P2 PDEP. Technologically Enhanced Naturally Occurring Radioactive Materials (TENORM)

study report. Pennsylvania Department of Environmental Protection, Harrisburg, PA, 2015.

R1 Razzano, F. and M. Cei. Geothermal power generation in Italy 2010-2014 Update Report.
Proceedings World Geothermal Congress, Melbourne, Australia, 19-25 April 2015.
International Geothermal Association, Bochum, Germany, 2015.

R2 Richard, T.L. Challenges in scaling up biofuels infrastructure. Science 329(5993): 793-796

(2010).
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 237

R3 Rood, A.S., G.J. White and D.T. Kendrick. Measurement of 222Rn flux, 222Rn emanation,
and 226,228Ra concentration from injection well pipe scale. Health Phys 75(2): 187-192
(1998).

R4 Rowan, E.L. and T.F. Kraemer. Radon-222 content of natural gas samples from upper and
middle Devonian sandstone and shale reservoirs in Pennsylvania: Preliminary data. Open-File
Report Series 2012-1159. U.S. Geological Survey, Reston, Virginia, 2012.

R5 Rule, B.M., Z.J. Worth and C.A. Boyle. Comparison of life cycle carbon dioxide emissions
and embodied energy in four renewable electricity generation technologies in New Zealand.
Environ Sci Technol 43(16): 6406-6413 (2009).

S1 Sakoda, A., Y. Ishimori and K. Yamaoka. A comprehensive review of radon emanation

measurements for mineral, rock, soil, mill tailing and fly ash. Appl Radiat Isot 69(10): 1422-
1435 (2011).

S2 Snchez Delgado, M. and T. Lpez Fernndez. Uranium mill tailings piles remediation:
remedial cleanup actions versus long term stewardship. pp.213-223 in: Environmental
Contamination from Uranium Production Facilities and their Remediation. Proceedings of an
International Workshop, Lisbon, 1113 February 2004. International Atomic Energy Agency,
Vienna, 2005.

S3 Shelmet Precision Casting. Common stainless steel alloys. [Internet] Available from
(http://www.shelmetcastings.com/common-stainless-steel-alloys.html) on 22.02.2012.

S4 Smith, K.R., G.M. Crockett, W.B. Oatway et al. Radiological impact on the UK population of
industries which use or produce materials containing enhanced levels of naturally occurring
radionuclides: Part I: Coal-fired electricity generation. NRPB-R327. National Radiological
Protection Board, Chilton, 2001.

S5 Snavely Jr., E.S. Radioouclides in produced water. A literature review. American Petroleum
Institute, Dallas, Texas, 1989.

S6 Solecki, A.T. and D.E. Tchorz-Trzeciakiewicz. Radon exhalation from the Upper Silesian coal
ashes. Geochem J 45: 491-496 (2011).

S7 Spath, P.L., M.K. Mann and D.R. Kerr. Life cycle assessment of coal-fired power production.
NREL/TP-570-25119. National Renewable Energy Laboratory, Colorado, 1999.

S8 Spath, P.L. and M.K. Mann. Life cycle assessment of a natural gas combined-cycle power
generation system. NREL/TP-570-27715. National Renewable Energy Laboratory, Colorado,
2000.

S9 Sullivan, J.L., C.E. Clark, J. Han et al. Life-cycle analysis results of geothermal systems in
comparison to other power systems. ANL/ESD/10-5. Argonne National Laboratory, Illinois,
2010.

T1 Tadmor, J. Atmospheric release of volatilized species of radioelements from coal-fired plants.

Health Phys 50(2): 270-273 (1986).

T2 Tang, Z., S. Ma, J. Ding et al. Current status and prospect of fly ash utilization in China. 2013
World of Coal Ash (WOCA) Conference, Lexington, KY, 22-25 April 2013.
http://www.flyash.info/. 2013.
238 UNSCEAR 2016 REPORT

T3 Tata Steel. The recycled content of steel. Tata Steel Europe Limited [Internet] Available from
(http://www.tatasteeleurope.com/en/company/activities/strip_products_uk/the_environment/po
sitive_environmental_characteristics/recyclability/the_recycled_content_of_steel/).

T4 Tokonami, S. and T. Ishikawa. Radiological aspects of using coal ash (slag) in the building
industry. Jpn J Health Phys 42(3): 227-233 (2007).

U1 UKQAA. UK Quality Ash Association: the power behind PFA. UK Quality Ash Association.
[Internet] Available from (http://www.ukqaa.org.uk/wp-content/uploads/2014/03/6-sided-re-
branded-Mar-2014.pdf) on 01.03.2016.

U2 UN. The energy challenge for achieving the millennium development goals. UN Energy Paper
June 22, 2005. United Nations Energy, 2005.

U3 UNDP. UNDP and energy access for the poor: energizing the millennium development goals.
United Nations Development Programme, 2010.

U4 UNSCEAR. Official Records of the General Assembly, Thirteenth Session, Supplement No.
17 (A/3838). UNSCEAR 1958 Report. United Nations Scientific Committee on the Effects of
Atomic Radiation. United Nations, New York, 1958.

U5 UNSCEAR. Sources and Effects of Ionizing Radiation. UNSCEAR 1977 Report. United
Nations Scientific Committee on the Effects of Atomic Radiation, 1977 Report to the General
Assembly, with annexes. United Nations sales publication E.77.IX.1. United Nations, New
York, 1977.

U6 UNSCEAR. Ionizing Radiation: Sources and Biological Effects. UNSCEAR 1982 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1982 Report to the
General Assembly, with annexes. United Nations sales publication E.82.IX.8. United Nations,
New York, 1982.

U7 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. UNSCEAR 1988 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1988 Report to the
General Assembly, with annexes. United Nations sales publication E.88.IX.7. United Nations,
New York, 1988.

U8 UNSCEAR. Sources and Effects of Ionizing Radiation. UNSCEAR 1993 Report. United
Nations Scientific Committee on the Effects of Atomic Radiation, 1993 Report to the General
Assembly, with scientific annexes. United Nations sales publication E.94.IX.2. United
Nations, New York, 1993.

U9 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.3.
United Nations, New York, 2000.

U10 UNSCEAR. Effects of Ionizing Radiation. Volume II: Scientific Annexes C, D and E.
UNSCEAR 2006 Report. United Nations Scientific Committee on the Effects of Atomic
Radiation. United Nations sales publication E.09.IX.5. United Nations, New York, 2009.

U11 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources: Report to the
General Assembly, Scientific Annexes A and B. UNSCEAR 2008 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation. United Nations sales publication
E.10.XI.3. United Nations, New York, 2010.
 ANNEX B: RADIATION EXPOSURES FROM ELECTRICITY GENERATION 239

U12 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume II: Effects. Scientific Annexes
C, D and E. UNSCEAR 2008 Report. United Nations Scientific Committee on the Effects of
Atomic Radiation. United Nations sales publication E.11.IX.3. United Nations, New York,
2011.

U13 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. Volume I: Report to the
General Assembly and Scientific Annex A. UNSCEAR 2013 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation. United Nations sales publication
E.14.IX.1. United Nations, New York, 2014.

U14 USDE. Energy efficiency & renewable energy. Office of Energy Efficiency & Renewable
Energy. U.S. Department of Energy. [Internet] Available from (http://www.eere.energy.gov/
basics/renewable_energy/crystalline_silicon.html) on 12.03.2015.

U15 USDE. Steel industry profile (30 January 2012). Office of Energy Efficiency & Renewable
Energy. U.S. Department of Energy. [Internet] Available from (http://www1.eere.energy.gov/
manufacturing/resources/steel_profile.html) on 23.02.2012.

U16 USDOL. Industries at a glance. Oil and gas extraction: NAICS 211. US Department of Labor,
Bureau of Labor Statistics. [Internet] Available from (http://www.bls.gov/iag/tgs/iag211.html)
on 23.02.2012.

U17 USEIA. Open data. International Energy Statistics. U.S. Energy Information Administration.
[Internet] Available from (https://www.eia.gov/opendata/) on 12.04.2012.

U18 USEIA. Nuclear explained. US Energy Information Administration. [Internet] Available from
(http://eia.doe.gov/cneaf/nuclear/page/intro.html) on 02.05.2014.

U19 USEIA. Table CT8. Electric power sector consumption estimates, 19602013, United States.
U.S. Energy Information Administration. [Internet] Available from
(eia.gov/state/seds/data.cfm?incfile=/state/seds/sep_use/eu/use_eu_US.html&sid=US).

U20 USEIA. Frequently asked questions: How much coal, natural gas, or petroleum is used to
generate a kilowatthour (kWh) of electricity? U.S. Energy Information Administration.
[Internet] Available from (eia.gov/tools/faqs/faq,cfn?id=667&t=2)

U21 USEPA. Coal ash basics. United States Environmental Protection Agency. [Internet] Available
from (www.epa.gov/coalash/coal-ash-basics) on 09.07.2014.

U22 USEPA. Copper mining and production wastes (8 July 2011). U.S. Environmental Protection
Agency. [Internet] Available from (http://www.epa.gov/radiation/tenorm/copper.html) on
28.02.2012.

U23 USGS. Radioactive elements in coal and fly ash: Abundance, forms, and environmental
significance. Fact Sheet FS-163-97. U.S. Geological Survey, 1997.

U24 USNRC. Occupational radiation exposure at commercial nuclear power reactors and other
facilities 2010. NUREG-0713, Vol. 32. U.S. Nuclear Regulatory Commission, 2012.

U25 USNRC. Capacity factor (net). [Internet] Available from (http://www.nrc.gov/reading-

rm/basic-ref/glossary/capacity-factor-net.html) on 03.03.2016.

V1 van der Heijde, H.B., H. Beens and A.R. de Monchy. The occurrence of radioactive elements
in natural gas. Ecotoxicol Environ Saf 1(1): 49-87 (1977).
240 UNSCEAR 2016 REPORT

V2 van Netten, C., K. Kan, J. Anderson et al. Radon-222 and gamma ray levels associated with
the collection, processing, transmission, and utilization of natural gas. Am Ind Hyg Assoc J
59(9): 622-628 (1998).

W1 Waggitt, P.W. Uranium mine rehabilitation: the story of the South Alligator Valley
intervention. J Environ Radioact 76(1-2): 51-66 (2004).

W2 Wang, C., W. Liu, F. Chen et al. Estimation of radon emissions from coal mining in China.
Paper provided by the Chinese delegation. Official communication to the UNSCEAR
secretariat. 2015.

W3 Wang, C., M. Wang and G. Liu. Radiation exposure to rare earth workers in China. Data
provided by the Chinese delegation. Official communication to the UNSCEAR secretariat.
2015.

W4 WCI. Coal and steel. World Coal Institute, UK, 2009.

W5 White, S.W. and G.L. Kulcinski. Birth to death analysis of the energy payback ratio and CO2
gas emission rates from coal, fission, wind and DT fusion electrical power plants. UWFDM-
1063. University of Wisconsin, Madison, 1998.

W6 Whitehead, N.E., B.J. Barry, R.G. Ditchburn et al. Systematics of radon at the Wairakei
geothermal region, New Zealand. J Environ Radioact 92(1): 16-29 (2007).

W7 Wilkening, M.H., W.E. Clements and D. Stanley. Radon 222 flux measurements in widely
separated regions. pp.717730 in: The Natural Radiation Environment II (J.A.S. Adams et al.,
eds.). CONF-720805-P2. National Technical Information Service, U.S. Department of
Commerce, 1972.

W8 WNA. World uranium mining production. World Nuclear Association. [Internet] Available
from (http://www.world-nuclear.org/information-library/nuclear-fuel-cycle/mining-of-
uranium/world-uranium-mining-production.aspx) on 06.04.2016.

W9 WNA. Nuclear power in France (updated 31 March 2016). World Nuclear Association.
[Internet] Available from (http://www.world-nuclear.org/information-library/country-profiles/
countries-a-f/france.aspx) on 02.08.2016.

W10 WNA. In Situ Leach (ISL) mining of uranium. World Nuclear Association. [Internet]
Available from (http://www.world-nuclear.org/info/Nuclear-Fuel-Cycle/Mining-of-Uranium/
In-Situ-Leach-Mining-of-Uranium) on 31.05.2015.

W11 WNA. Supply of uranium. World Nuclear Association. [Internet] Available from
(http://www.world-nuclear.org/information-library/nuclear-fuel-cycle/uranium-resources/
supply-of-uranium.aspx) on 24.03.2015.

W12 WNA. Processing of used nuclear fuel (updated November 2015). World Nuclear Association.
[Internet] Available from (http://www.world-nuclear.org/information-library/nuclear-fuel-
cycle/fuel-recycling/processing-of-used-nuclear-fuel.aspx) on 14.03.2016.

W13 World Bank. Breakdown of electricity generation by energy source. The World Bank - World
Development Indicators. [Internet] Available from (http://www.tsp-data-portal.org/
Breakdown-of-Electricity-Generation-by-Energy-Source#tspQvChart) on 24.11.2015.

Z1 Zeevaert, T., L. Sweeck and H. Vanmarcke. The radiological impact from airborne routine
discharges of a modern coal-fired power plant. J Environ Radioact 85(1): 1-22 (2006).
 ANNEX C

BIOLOGICAL EFFECTS OF SELECTED INTERNAL

EMITTERSTRITIUM

241
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 243

CONTENTS
I. INTRODUCTION ................................................................................................................................. 245
II. SOURCES AND LEVELS .................................................................................................................... 248
A. Natural sources .......................................................................................................................... 248
B. Artificial sources ........................................................................................................................ 248
III. PHYSICAL, RADIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS .............................. 252
A. Physical characteristics ........................................................................................................... 252
B. Radiological characteristics................................................................................................... 252
C. Biochemical characteristics ................................................................................................... 253
IV. HUMAN EXPOSURE .......................................................................................................................... 255
A. Exposure of the public ............................................................................................................ 255
B. Occupational exposure .......................................................................................................... 258
V. BIOKINETICS AND DOSIMETRY .................................................................................................... 259
A. Information on biokinetics and dosimetry of tritiated compounds ............................ 259
B. Overview of current biokinetic models for tritium ....................................................... 272
C. Intakes of tritium in relation to pregnancy and breast-feeding .............................. 281
D. Uncertainties in dose coefficients for tritium ................................................................. 284
E. Summary of biokinetic and dosimetic models .............................................................. 285
VI. BIOLOGICAL AND HEALTH EFFECTS .......................................................................................... 287
A. Non-radiological effects of tritium in biological systems .......................................... 287
B. Deterministic effects................................................................................................................ 288
C. Stochastic effects of HTO in mammals ............................................................................. 295
D. Effects of tritiated biochemical substrates ...................................................................... 299
VII. RELATIVE BIOLOGICAL EFFECTIVENESS .................................................................................... 304
A. Track structure considerations ............................................................................................ 305
B. RBE literature reviews and experimental studies.......................................................... 307
C. Factors affecting RBE values ................................................................................................. 308
D. Summary of RBE value determinations ............................................................................ 309
VIII. EPIDEMIOLOGICAL STUDIES ......................................................................................................... 316
A. Studies of occupational exposure ...................................................................................... 316
B. Studies of environmental exposure ................................................................................... 320
C. Summary of epidemiological studies................................................................................ 321
244 UNSCEAR 2016 REPORT

IX. RESEARCH NEEDS ............................................................................................................................. 322

X. GENERAL CONCLUSIONS ............................................................................................................... 324
XI. ACKNOWLEDGEMENTS .................................................................................................................. 325
APPENDIX A: TABLES SUMMARIZING STUDIES OF OCCUPATIONAL AND ENVIRONMENTAL
EXPOSURE TO TRITIUM ........................................................................................................................... 327
REFERENCE ................................................................................................................................................. 341
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 245

I. INTRODUCTION
1. The Committee has conducted an independent review of the scientific literature on the
characteristics of tritium, its biokinetics and dosimetry within the human body for various physical and
chemical forms and routes of intake into the body, radiobiological effects of tritium exposure, and
epidemiological data relating to its impact on the health of workers and members of the public.

2. Tritium is a radioactive isotope of hydrogen (symbol 3H, but commonly represented by T).
Chemically, it behaves like other isotopes of hydrogen (protium, 1H, the principal stable isotope, and
deuterium, 2H, the other stable isotope). The word tritium is used here to mean the particular isotope of
hydrogen irrespective of the chemical form in which it occurs.

3. Tritium occurs both naturally, mainly as a result of the interaction of cosmic-ray particles with the
atomic nuclei of air molecules in the upper atmosphere, and as a consequence of the operation of
nuclear reactors and other industries. Tritium in the environment and workplace is encountered
predominantly as tritiated water (HTO) in liquid or vapour form.

4. Tritium emits low-energy beta particles with a short range in body tissues and, therefore, poses a
risk to health as a result of internal exposure only following ingestion in drinking water or food, or
inhalation or absorption through the skin. Unlike external penetrating radiation, such as X-rays and
gamma rays, internal exposure to tritium has the potential to result in heterogeneous dose distribution
within tissues and cells. Other factors that may affect the potential radiotoxicity of tritium include
transmutation and isotopic effects. Transmutation is the term used for the formation of a new element
by radioactive decay, which has the potential to adversely affect metabolic processes. Isotopic effects
apply to low atomic mass elements such as hydrogen, for which tritium atoms with larger mass may
replace the stable protium in cellular processes. Both effects are judged to be minor contributors to
radiotoxicity when compared to the predominant effect of the energy deposition from beta particles
emitted by tritium decay.

5. Five main chemical forms are of interest when considering the biological and health effects of
internal exposure to tritium: HTO, organically bound tritium (OBT), tritiated biochemical substrates
(including DNA precursors), insoluble compounds, and tritiated gases. OBT is the general term used to
describe tritium that is non-exchangeably bound to carbon atoms within organic constituents of cells
and tissues (e.g. proteins, polysaccharides, lipids).

6. Absorbed doses arising from the intake of tritium cannot be measured directly and recourse has to
be made to the use of bioassay (such as the determination of tritium in urine) or to assessments based
on environmental monitoring. Biokinetic models of the behaviour of tritium in the body are used to
determine intake from such measurements and are also used together with dosimetric models to relate
retention of tritium in body tissues to the time-course of dose delivery within tissues. For intake of
tritium as HTO, distribution between organs and tissues and within cells is quite uniform, depending on
their water content, and so the dose is uniformly delivered despite the short range of the low-energy
beta particle emissions.

7. However, some organic substrates containing tritium concentrate in specific organs and tissues,
and even within specific regions within cells. In such cases, the pattern of dose distribution is very
different from that experienced following uniform exposure to external penetrating radiation or
246 UNSCEAR 2016 REPORT

incorporation of HTO, with heterogeneity of dose between organs and tissues, and potentially within
organs and even within cells. For intake of tritiated nucleotides and nucleosides, for example, a small
proportion has been shown to reach cells intact and may then be incorporated into cellular DNA,
resulting in localized energy deposition [D5, N1].

8. There is also some tritium-containing radioactive material with low solubility in aqueous media,
such as tritides of metals (e.g. Ti, Zr, Hf), tritiated luminous compounds, micro fragments of glass and
carbon and beryllium particles contaminated with tritium. Such inhaled particles exhibit long-term
retention in the lungs, leading to prolonged exposure of lung tissue to beta radiation.

9. The International Commission on Radiological Protection (ICRP) has used three main biokinetic
models in the estimation of doses from compounds that contain tritium for protection purposes [I8, I9,
I10, I14, I15, I18]:

(a) A model for tritium absorbed to blood as HTO following either ingestion or inhalation, applied
also to other tritiated compounds, including elemental hydrogen and methane, that partially convert
to HTO after being taken into the body;

(b) A model for tritium absorbed to blood as OBT, mainly following ingestion in food, but also
applied to inhalation of non-specified organic material and to ingestion or inhalation of some
specific tritiated organic compounds;

(c) The generic ICRP models for the human respiratory tract, specifying absorption parameter
values for inhalation of insoluble forms of tritium used in industry, including metal tritides.

10. The existing ICRP biokinetic and dosimetric models for tritium are currently being upgraded on
the basis of recent biokinetic data, especially for recently developed physical and chemical forms of
tritium. This work includes models for tritium as gases, HTO, organic substances and OBT, and
material with low solubility.

11. Electrons with very low energy, including beta particles from 3H, have higher linear energy
transfer (LET) values than electrons generated by the interaction of higher energy photons (e.g. from
external gamma rays). This higher LET may result in greater effectiveness in causing cancer. The
assessment of the effectiveness of different radiation in causing health effects relies on data on their
relative biological effectiveness (RBE). RBE is an empirical quantity that depends on the biological
system, the observed end points, the dose and the experimental conditions. In recent decades, several
tens of experiments have been conducted using mammals (mostly mice) and their cells to determine
RBE for tritium under various experimental conditions and considering a range of biological end
points. However, only a small number of studies were performed to directly measure cancer induction
in mammals.

12. Laboratory studies using animals have demonstrated that tritium, like other sources of radiation,
can interfere with the development of the embryo or fetus, and can induce carcinogenic, heritable and
reproductive effects and cell death. The use of high doses of tritium, for example, in the form of HTO
or tritiated thymidine, has also been shown to induce acute radiation syndrome.

13. The dose and risk from some tritiated biochemical substrates and OBT is greater than that from
HTO due to their longer residence in the body. However, there are few studies looking specifically at
biological effects related to tritiated biochemical substrates and most of them use DNA precursors and
amino acids. There is no appropriate ICRP biokinetic and dosimetric model for use in human risk
assessment and radiation protection for tritiated nuclear acid precursors and there is a practical need for
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 247

the development of such models for intake of tritiated biochemical substrates, including nucleotropic
forms even though the number of workers dealing with such forms of tritium is limited.

14. Most experimental studies on tritium were performed 20 to 30 years ago. While this work was
competently performed at the time, it did not use modern scientific approaches and procedures that are
often more sensitive and can use multiple approaches to test a single question. The application of
modern techniques would be helpful in reinvestigating aspects of tritium dosimetry and effects,
including fetal and embryo studies, and DNA damage analyses.

15. Workers may be subjected to wide-range occupational exposure to tritium in various chemical and
physical forms. Usually, occupational exposure to tritium is low relative to other sources of exposure.
However, historically there have been several cases of occupational exposure of workers (Russian
Federation, Germany), mostly following accidents but also following chronic exposure to considerable
quantities of tritium, resulting in haematological radiation syndrome [M15, O3, S12], including a few
cases of radiation-induced death.

16. The principal source of quantitative information on radiation-induced cancer and other health
effects in humans remains the epidemiological follow-up studies of the Japanese survivors of the
atomic bombings exposed to external radiation [P12, P13, U11]. An important question is the extent to
which these risk estimates are applicable to exposure from internal emitters, including tritium with its
low dose-rate and low-energy beta radiation with heterogeneity of exposure between and within organs
and tissues. Currently, little information on tritium-specific risks can be derived from epidemiological
studies of tritium workers or members of the public potentially exposed to tritium, beyond the
conclusion that tritium-specific risks have not been substantially underestimated.

17. The Committee has agreed to undertake a comprehensive review of the biokinetics, dosimetry and
effects of selected internal emitters. The first radionuclide to be considered is the radioisotope of the
element hydrogen, tritium. The main reasons for this selection are as follows:

The potential for large scale production of tritium in connection with civilian and military fusion
activities, as well as its creation as a by-product from operation of nuclear fission reactors,
especially heavy water reactors;

Exposure of workers and the public to various physical and chemical forms of tritium, including
organic and substrates with low solubility, with a wide range of radiotoxicity that requires
comprehensive scientific analyses;

Professional and public concerns expressed between 2006 and 2010 regarding the radiotoxicity of
tritium, which led to extensive review and data analysis in a number of countries, including
Canada, France and the United Kingdom.
248 UNSCEAR 2016 REPORT

II. SOURCES AND LEVELS

A. Natural sources
18. Tritium was discovered in 1934 by Oliphant, Harteck and Rutherford [O4] and isolated in 1939
by Alvarez and Cornog [A2]; the production of tritium by natural processes was reported by Libby
[L10]. There are three main sources of natural tritium: production in the atmosphere by galactic cosmic
rays, production in the atmosphere by solar flare accelerated particles, and accretion from the sun. This
natural production of tritium is estimated to occur at a rate of about 0.12 to 2.0 tritium atoms per square
centimetre of earth surface per second, with the most probable values being close to 0.2 to 1.0 tritium
atoms per square centimetre per second [J1, N2].

19. Tritium produced by natural processes is rapidly converted into HTO, which then joins the water
cycle. Its concentration in continental surface water and throughout the oceans is about 400 Bq/m3 and
100 Bq/m3, respectively. Humans, on average, ingest about 500 Bq of tritium each year, with a
resulting average annual effective dose of about 0.01 Sv [U11].

B. Artificial sources

1. Nuclear weapon tests

20. In the mid-1950s and early 1960s, tritium was widely dispersed during the above-ground testing
of nuclear weapons. Especially large quantities of tritium in elemental form and as tritium oxide were
released in the environment in a series of hydrogen-bomb tests that started in 1952; their total explosive
fusion yield was 328 Mt. The total amount of tritium released into the atmosphere from the testing of
nuclear weapons from 1945 to 1980 was estimated to be 186,000 PBq [U11]. The quantity of tritium in
the atmosphere from weapon testing peaked in 1963 and has since been decreasing.

21. Tritium is readily recycled in the biosphere and becomes homogenously disseminated in the
hemisphere where it has been released. The International Atomic Energy Agency (IAEA) runs a global
network of 155 stations to measure tritium in precipitation [I1]. Measurements of tritium in drinking
water in the United States in the early 1960s showed concentrations over two orders of magnitude
higher than background levels that decreased with a half-time of about three years (figure I).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 249

Figure I. Environmental tritium in surface water (pCi/L) in the United States in 19511975 [B15]
1 pCi/L = 0.037 Bq/L

2. Production of tritium
22. In countries with developed nuclear technologies, tritium is produced in large quantities for
military and peaceful purposes by means of irradiation with neutrons of lithium enriched with isotope
6
Li at industrial nuclear reactors. Tritium can be released into the environment from operational tritium
production facilities in the form of elemental hydrogen or tritium oxide with high specific activity.

23. Elevated levels of HTO were measured in lakes located in the area of the Mayak facility
(Ozyorsk, Russian Federation) in 1982, 1986 and 20012003 by Chebotina and Nikolin [C12] and in
20092012 by Kazachenok et al. [K5, K6]. In the former measurement series, tritium concentration in
lake water in 20012003 was inversely proportional to the distance from the Mayak facility (figure II).
During the observation period 19822003, the HTO concentration in lake water decreased by a factor
of 2 to 16 while during the same period its mean concentration in major Russian rivers went down by a
factor of about 3, from 8 to 3 Bq/L [S20].
250 UNSCEAR 2016 REPORT

Figure II. Dependence of tritium concentration in lake water (Bq/L) in 20012003 on the distance (km)
from the Mayak facility [C12]

3. Operation of nuclear facilities

24. Tritium is produced in nuclear reactors by ternary fission, one triton per around 1 in 104 fissions
of 235U induced by thermal energy neutrons and by neutron reaction with light elements such as boron,
lithium and hydrogen (deuterium) [N2]. Tritium is produced in much larger quantities in heavy-water-
moderated nuclear reactors through neutron capture by deuterium atoms.

25. Tritium is released into the environment from nuclear reactors, especially heavy water reactors,
and spent fuel reprocessing plants, including waste storage and waste disposal sites. In the future, there
is potential for significant releases during the operation of fusion reactors. Tritium is released
predominantly as HTO or elemental hydrogen, partially converted by environmental biota to OBT.
From 1998 to 2002, the global annual average releases of tritium to the atmosphere and to the aqueous
environment from nuclear facilities were estimated to be 11.7 PBq and 16.0 PBq, respectively. The
resultant average annual collective effective doses from these releases were estimated to be 25 and
10.5 person-Sv, respectively [I2, I3, U11].

26. In the vicinity of nuclear installations, especially near heavy water reactors, tritium activity in
environmental compartments can be above background values. For example, while tritium (HTO)
activity concentrations in air at background locations in Ontario, Canada, range from 0.01 to
0.08 Bq/m3, tritium in the vicinity of CANDU nuclear power plants (NPPs) range from 0.05 to
31 Bq/m3. Fish caught in the vicinity of NPP effluent discharges have HTO activity concentrations up
to 50 Bq/L while in fish from background locations, it was less than 9 Bq/L [C23].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 251

4. Incidental releases from nuclear facilities

27. Large incidental releases of tritium from tritium production facilities have been reported to occur
from Lawrence Livermore Laboratory, United States in 1970 and from Savannah River Plant, United
States in 19741984 shown in table 1 [O2]. The released activity decreased with time from 11 to
18 PBq in early 1970s to 0.3 PBq in 1984. The chemical forms of the released tritium were
predominantly elemental hydrogen (gas) or tritium oxide or their mixture. Monitoring has shown that
elemental tritium was gradually converted in the environment to tritium oxide.

Table 1. Large incidental releases of tritium in the United States [O2]

Lawrence Livermore Laboratory (LLL) and Savannah River Plant (SRP)

Site Year Tritium release (PBq) HTO (%)

LLL 1970 11 <1
SRP 1974 18 <1
SRP 1975 6.7 0.6

SRP 1981 1.2 >99

SRP 1983 2.1 1

SRP 1984 0.3 70

28. Elevated levels of tritium in the environment were also observed following major nuclear
accidentsthe Chernobyl accident in the USSR in 1986 and the Fukushima accident in Japan in 2011.
In May 1986, tritium concentrations in precipitation collected in the Ukraine and the European part of
the Russian Federation had increased by a factor of twothree compared with 1985, as had tritium
concentration in river water [S20].

29. After the Fukushima accident, tritium concentrations in precipitation collected 170700 km
southwest from the Fukushima Daiichi Nuclear Power Station and in plant water collected in its
vicinity were substantially elevated (up to a factor of a few tens) compared with pre-accident levels
[K2, M4]. According to Povinec et al. [P11], the amount of tritium released and deposited over the
north-west Pacific Ocean was in the range of 0.10.5 PBq.

5. Other tritium-bearing facilities and commodities

30. Elemental tritium and tritiated luminous compounds are widely used in the luminizing industry,
e.g. for illuminating watch and compass dials and as permanent warning lights. Metal plates with
incorporated tritium are used in nuclear physics as targets for nuclear reactions, e.g. neutron production.
Other metal plates with incorporated tritium are used as sources of air ionization in industry and
agriculture. Tritiated biochemical substrates are produced at radiopharmaceutical facilities and then
applied for diagnostic health examinations in hospitals, and research activities in medicine and biology.

31. Tritium used in such industrial, health care, research and other applications is partially released
into the working environment, human habitat and natural environment and becomes incorporated in the
bodies of workers and members of general public in various physical and chemical forms.
252 UNSCEAR 2016 REPORT

III. PHYSICAL, RADIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS

A. Physical characteristics
32. Tritium (3H or T) is the heaviest radioactive isotope of hydrogen. The tritium atom has one proton
and two neutrons in its nucleus and one electron. The binding energy of nucleons is 8.4 MeV, and the
diameter of a tritium atom is 1.1 Angstroms. The dissociation energy, T2 to 2T, is 4.59 eV; ionization
energy, T to T+ e, is 13.55 eV.

33. Tritiums physical properties are similar to those of common hydrogen (1H), which dominates in
nature over tritium and the intermediate by mass, stable deuterium (2H or D). Under ambient
conditions, tritium is a colourless highly flammable diatomic gas with the molecular formula T2. It is
possible to make liquid tritium at atmospheric pressure by cooling it to below 25 K (248 C). Liquid
hydrogen can be stored in insulated containers under pressure.

34. Tritium has a high coefficient of diffusion. It readily diffuses through porous substances such as
rubber and can also diffuse through metal. Tritium, like common hydrogen, easily undergoes various
chemical reactions depending on physical and chemical conditions. The prevailing form of tritium in
nature is tritium oxide (T2O) or HTO.

35. Tritium figures prominently in studies of nuclear fusion because of its favourable reaction cross
section and the large amount of energy (17.6 MeV) produced through its reaction with deuterium:

D + T = 4He2 + n + 17.6 MeV

B. Radiological characteristics
36. The nucleus of the tritium atom is unstable and decays with the emission of a beta particle and an
antineutrino to stable 3He. The antineutrino is of no biological significance because it does not interact
with matter:
3
H 3He+ + e + e

37. Tritium has a physical half-life of 12.3 years and, in the pure elemental state, a specific activity of
3.56 1014 Bq/g. Emitted beta particles are very low energy, mean 5.7 keV (90 fJ) and maximum
18.6 keV (300 fJ).

38. In water, the average track length of the beta particle is 0.56 m and the maximum track length is
6 m, which compares with a typical cell nucleus diameter of 615 m, while a cell has a diameter of
10100 m [V1]. Tritium beta particles are completely absorbed by sheets of plastic, glass or metal.
They do not penetrate dead layers of skin. However, following intake of tritium, beta radiation can
irradiate internal organs. Within the body it gives a relatively low absorbed dose per disintegration
compared with other beta-emitting radionuclides, but the ionization density of the electron is greater.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 253

39. Radioactive decay of tritium atom also results in transfer of some recoil energy to a daughter 3He+
positively charged ion. This energy depends on the random dispersion angle between the emitted
electron and antineutrino and comprises 1.0 eV as average and 3.3 eV as maximum [F2, G8]. This
energy is insufficient for either daughter atom self-ionization (required energy of the order 10 keV) or
tissue ionization (about 30 eV). Besides the recoil energy, the daughter 3He+ ion also carries excitation
energy of about 11 eV that can influence the fate of the molecule to which the tritium atom was bound
and result in its chemical transmutation and modification of its chemical properties.

C. Biochemical characteristics

1. Tritiated water
40. Tritium is most commonly found in natural and working environments in the form of HTO, which
has the same chemical properties as ordinary water. Water with a tritium activity of 1 Bq/L contains
less than one tritium atom among 1017 molecules of water. HTO can enter the human body by
inhalation, skin absorption (liquid and vapour) [D3, P9], or ingestion of water or food [B14, I13]. Once
inside the body, HTO diffuses freely and rapidly across cellular membranes, and reaches equilibrium
throughout the total body water pool [H12]. HTO is excreted via urine, faeces, sweat, and breath [N1].
Since HTO quickly reaches equilibrium with the water in the body and is distributed uniformly among
all soft tissues, the concentrations of HTO in sweat, sputum, urine, blood, perspiration, and exhaled
water vapour are considered to be equal [H12].

2. Tritiated gases
41. Tritiated elemental hydrogen (HT or T2) is relatively inert in biological systems and has a very
low uptake into body fluids and tissues [H12]. Humans are mostly exposed to HT by inhalation or skin
contact with contaminated surfaces. A small fraction of inhaled HT is converted to HTO in the human
body. The primary sources of HT are tritium production and processing facilities (such as those
involved in making gaseous tritium light sources), tritium recovery facilities, and nuclear fuel
reprocessing facilities. HT is readily converted to HTO in the environment, with soil microorganisms
playing an important part in this process [A3].

42. Tritiated methane (CH3T) is relatively inert in biological systems. Humans can be exposed to
CH3T by inhalation in workplaces or in the public domain following biochemical degradation of OBT
in the environment. Because of the low solubility of methane in body fluids, the radiological
implications of inhaling CH3T are mostly determined by its oxidation to HTO and biochemical
conversion to OBT in the human body [P5].
254 UNSCEAR 2016 REPORT

3. Organically bound tritium

43. Because tritium atoms are exchangeable with normal hydrogen atoms, a fraction of the tritium
absorbed by plants or animals can become incorporated into organic compounds such as carbohydrates,
fats, proteins, and collagen: this is referred to as OBT 1. Animals, including humans, ingest OBT and
form OBT from HTO within their tissues [D5, K14].

44. A tritium atom in OBT attached to a carbon atom is essentially fixed until the compound is
metabolized (i.e. the tritium is non-exchangeable). However, a tritium atom attached to an oxygen,
sulphur, nitrogen or phosphorus atom is readily exchangeable with hydrogen in water and is not
considered as OBT in this annex [D5, R11, S1] or specifically qualified as exchangeable OBT [K14].
OBT exhibits longer retention times in the body than HTO.

4. Tritiated organic substances

45. A broad spectrum of organic substances labelled with tritium, including biochemical substrates,
are produced and used widely in research and for other purposes. Workers may be exposed by
inhalation or through skin contamination, and also by inadvertent ingestion. In the human body,
labelled biochemical substrates (e.g. amino acids, DNA precursors, glucose, hormones) may be
metabolized with partial loss of tritium label converted to HTO, or be incorporated into biological
macromolecules as OBT [B11, H12]. Foreign organic compounds (such as organic solvents) are usually
rapidly excreted from the body in urine and faeces [B8].

46. Labelled DNA precursors (e.g. 3H-thymidine, 3H-deoxycytidine) belong to a special group that, in
the mammalian body, are partially degraded to HTO and partially incorporated into the DNA of dividing
cells, and thereafter selectively expose the nuclei of proliferating cells to beta radiation [D5, N1].

5. Metal tritides and other low soluble forms of tritium

47. Tritiated compounds with low solubility, which are widely produced and used in industrial or
research facilities, include luminous compounds (as powders), particles of metal tritides that are used as
accelerator targets or as ionization sources in industry and agriculture, and carbon, beryllium and
tungsten dust, micro-fragments of glass contaminated with tritium used in fusion experiments. Airborne
particles with insoluble tritium are inhaled by workers and, depending on their dimensions and
respirability, deposited in the respiratory tract. Following inhalation, such substances can remain in the
pulmonary region of the lungs and expose tissues to both beta radiation and, to a lesser extent, to
bremsstrahlung.

48. Although insoluble particles are largely retained in lung tissues, transported by macrophages to
regional lymph nodes, or escalated from the lungs by mucociliary clearance, some dissolution will occur
and a proportion of their tritium content will be removed and absorbed to blood as HTO. When luminous
powder produced from zinc sulphide granules coated with a thin layer of high activity tritiated polymer is
inhaled, some tritium is detached from the polymer and absorbed to blood as low molecular organic
foreign compounds that are rapidly excreted from the body in urine and faeces [B8, B11, B14].

1
OBT in biological tissues is carbon-bound tritium that was originally formed in living systems through natural environmental or
biological processes from HTO (or HT via HTO). OBT is not exchangeable with hydrogen in water.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 255

IV. HUMAN EXPOSURE

A. Exposure of the public

1. Tritiated water in global water cycle

49. The Committee [U9] has estimated worldwide average annual individual effective doses mainly
from data on ingestion of the globally dispersed HTO created as a result of fallout from above-ground
nuclear weapon testing (figure III) [B25, U9]. The doses received from the inhalation of 3H are
negligible in comparison with those received from ingestion. The derivation of these doses is largely
based on environmental measurements and is described by the Committee in its UNSCEAR 2000
Report [U9], Bouville et al. [B25] and Bennett [B16]. The background concentration of tritium in
humans is calculated from an average of the concentrations in the sources of water ingested, assumed to
be 33% from the atmosphere, 53% from fresh water, 13% from groundwater and 0.7% from ocean
surface water (through fish) [B25, N2].

50. The largest annual doses from tritium in fallout were received by the world population during the
period of intense nuclear weapon testing during the late 1950s and early 1960s, before the Limited Test
Ban Treaty of 1963. The peak global average annual effective dose from tritium in fallout was 7.2 Sv in
1962. Since the majority of atmospheric nuclear weapon tests during that period took place in the
Northern Hemisphere, average doses from tritium were greater in the Northern Hemisphere than in the
Southern Hemisphere. Generally, tritium follows the global water cycle: a large proportion is transferred
to the oceans within a few years of production and a very small fraction is ingested by humans [U9].

Figure III. Worldwide average annual individual effective dose in 1950s1990s from the ingestion
of tritium produced in atmospheric nuclear weapon testing [B25, U9]
Dashed line is for dose from natural tritium
256 UNSCEAR 2016 REPORT

2. Local exposure of the public from nuclear facilities

51. Tritium released from nuclear facilities, especially from those operating with large amounts of
tritium (i.e. tritium production facilities, heavy water reactors or reprocessing plants) may enter the
bodies of people residing in the vicinity in drinking water or via inhalation. In these conditions, human
doses caused by local intake of environmental tritium are usually larger than those caused by global
tritium levels common in neighbouring areas with no facilities releasing tritium.

52. Measurements of tritium (HTO and OBT) in environmental media are carried out routinely in the
vicinity of Canadian nuclear facilities. These measurements allow estimation of doses to members of
the public from all exposure pathways (e.g. inhalation, skin absorption, ingestion of food and drinking
water). In 2006, annual tritium doses to members of the public living in the vicinity of NPPs were less
than 2.5 Sv whereas they were slightly higher in the vicinity of two facilities manufacturing gaseous
tritium light sources (GTLS). The annual tritium doses of respective critical groups for the two GTLS
facilities were 15 and 67 Sv [C23].

53. Kim and Han [K13] studied environmental radiation conditions in 19921993 around the
Wolsong NPP, Republic of Korea, for a CANDU-6 heavy water reactor that had been operational since
1983. Activity concentrations of HTO and OBT were analysed in food samples collected within
1-15 km of the reactor site and HTO was collected from the air in some locations. In water samples
extracted from rice, Chinese cabbage, radish and pumpkin, HTO concentration was in the range of
3-100 Bq/L and that in combustion water obtained from organic parts of vegetables was 4130 Bq/L,
both inversely proportional to the distance from the site. The ratio of tritium concentration per unit
hydrogen mass in OBT to that in free HTO was in the range of 1.02.8, with an average of 1.35. On the
basis of monitoring data, the authors assessed the annual effective dose for adult members of the public
to be in the range of 1.3 Sv in a radius of 01.6 km to 0.15 Sv at 816 km. Although both values are
much lower than the background dose, they are substantially higher than the annual dose from
environmental tritium to the Korean public residing away from NPPs, assessed by Yoon et al. [Y7]
from urine samples of 50 persons to be about 2 nSv.

54. In 2008, Chebotina and Nikolin [C13] measured elevated tritium concentrations in the urine of
45 residents of five towns located in the vicinity of the Mayak facility (figure IV). The average
concentrations, in the range of 100800 Bq/L, correlated inversely with the distance between the town
and the Mayak facility. Those values are much higher than tritium concentrations in potable water
measured in the area in the 2000s, indicating inhalation as a possible intake pathway. The measured
concentrations of tritium in urine correspond to average annual effective doses incurred by the residents
of five towns in 2008 in the range of 314 Sv.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 257

Figure IV. Average concentration of tritium in 2008 in urine of residents of towns located in the
area of the Mayak facility [C13]

3. Organically bound tritium in human tissue

55. Very few authors have directly measured tritium content in human tissue. Bogen and Welford
[B21] for example have summarized results of tritium measurements in the United States environment
carried out in 1960s to early 1970s in non-equilibrium conditions caused by termination of nuclear
weapon tests in 1963 and continued radioactive fallout (decreasing with time) from the stratosphere.
They sampled water vapour from the air, tap water, soil, vegetation, food, animal and human tissue and
they measured HTO in water distilled from the samples and OBT in water from combustion of dried
samples. Both sets of data were presented in terms of HTO activity concentration in water. In all links
of the ecological chain up until 1973, the tritium specific activity in the OBT fraction was higher than
that found in the free water fraction. The ratio of specific activities OBT/HTO decreased by a factor of
about 1.52 from each trophic level: soil 68, vegetation 34, animal 23 and human 1.52. Those
patterns can be interpreted by slower clearance of OBT compared to HTO from various organisms of
the trophic chain and their residues (soil organic matter) in conditions of decreasing HTO concentration
in the environment, as was the case in the 1970s.

56. Ujeno et al. [U3] measured tritium in tissue water distilled from samples of human organs and tissues
(brain, lung, liver, kidney and muscle) collected by staff of Kyoto University during forensic autopsy of eight
dead bodies. The tritium concentration in water from various organs and tissues was similar and not affected
258 UNSCEAR 2016 REPORT

by sex or age. The average HTO concentration in tissue water was 2.50.7 Bq/L, which was similar to that
measured in tap water, rain water and water distilled from local food.

57. Hisamatsu et al. [H13, H14] presented the results of tritium measurements in organs and tissues
collected from 11 human cadavers (10 males, one female, mean ageSD=4616 year), who died
suddenly in 1986 at Akita Prefecture in northern Japan. They measured tritium concentrations in free
water (HTO) distilled from human samples (brain, liver, lung, heart, kidney, blood serum and whole
blood) and in combustion water obtained by combustion of dried samples in oxygen atmosphere. The
mean tritium concentrations in free water from seven diverse organs and tissues were similar, with a
range of 1.51.9 Bq/L and average of 1.6 Bq/L. Mean tritium concentration in combustion water of
various organs/tissues varied in the same range, average 1.7 Bq/L. The ratio of OBT/HTO tritium
concentrations in human tissue varied between 0.95 and 1.3 with an average of 1.1. Tritium
concentrations in local food sampled in 19851987 were also reported. Free water concentration in six
samples of the total human diet varied between 1.4 and 2.2 Bq/L and combustion water varied in the
range of 1.72.2 Bq/L. Their ratio varied between 0.9 and 1.6 with an average value of 1.2. Thus, in
equilibrium conditions of low-level intake of environmental tritium, no differences in tritium
concentrations were revealed between various human organs and tissues, between diet and human
tissue, and between free water tritium and OBT.

4. Measurements of environmental tritium

58. The measurement of environmental tritium in its various forms as gases or vapours (HT, HTO,
organic molecules), liquids (HTO or OBT in solution) and solids (OBT, hydrides) is a key step for dose
assessment and evaluation of health and environmental risks. Sampling, storage and treatment are
important points in the analytical procedure for tritium. The final form of tritium for analysis is usually
water, and low concentrations of tritium in water (few Bq/L) are currently measured either by gas
proportional counter or by liquid scintillation counter. They can also be determined indirectly using a
sensitive mass spectrometer, measuring the amount of the decay product, helium-3, formed in a water
sample in a closed vessel during a given period [W8].

59. Reference water that is virtually tritium-free is used as calibration blanks for the analytical system
and a recent comparison of these water sources gave results ranging from 0.004 to 0.17 Bq/L [F9].
Analytical procedures have been developed to measure OBT [B2] and respective standards are under
development. Recent studies of the speciation of tritium as OBT are investigating variations in the
hydrogen content of different forms and identifying compounds solubilized in the samples during labile
exchange [B1].

B. Occupational exposure
60. In working environments, tritium is present in various physical and chemical forms depending on
the production processes. Working environments may contain tritium in a variety of different chemical
forms, including HTO, elemental hydrogen, organic solvents, airborne particles of metal tritides (e.g.
Ti, Zr, Er), tritium contaminated glass and dust particles, luminous compounds, and labelled
biochemical substrates (e.g. amino acids, DNA precursors, glucose, hormones) [B11, H16, I2, I3, J1].

61. Occupational exposure to tritium is usually low relative to other sources of exposure. For example,
in 2006 average annual tritium doses for NPP workers in Canada ranged between 0.07 and 0.26 mSv,
which represented between 14 and 29% of the total effective dose. In the same year, workers employed at
two Canadian GTLS manufacturing facilities had annual tritium doses of 0.19 and 0.3 mSv [C23].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 259

V. BIOKINETICS AND DOSIMETRY

62. Biokinetic models describe the time-dependent deposition and translocation of radionuclides in
the body and the rates at which they are removed from the body. Biokinetic models are used to
calculate the number of nuclear transformations of radionuclides in each source organ during a
specified period following an intake. Dosimetric models are then used to calculate the absorbed doses
to specific organs and tissues (referred to as target organs) per nuclear transformation of radionuclides
in each source organ (i.e. each site of radionuclide deposition or transit in the body).

63. For protection purposes, ICRP calculates values of committed effective dose as a doubly weighted
sum of organ and tissue dosesfirst, adjustment to take account of the relative effectiveness of
different radiation types in causing stochastic effects using radiation weighting factors (wR), and
second, adjustment for differences between organs and tissues in their contribution to total detriment
from stochastic effects using tissue weighting factors (wT) [I23].

64. Because the range of the beta radiation emitted by tritium is short, all the radiation energy is
generally assumed to be absorbed in the tissues and organs in which the tritium decays. Therefore,
organ or tissue dose from tritium radiation is entirely determined by a relevant biokinetic model and not
by radiation transport in the body. The biokinetic model for tritium depends on the type of tritiated
compound taken into the body, as this dictates its deposition, translocation, retention and excretion.

65. In animal studies, the absorbed dose in tissue resulting from an acute administration of tritium can
be calculated using information on the initial activity concentration of tritium in the tissue and the rate
of removal of tritium from that tissue arising from biological processes and radioactive decay. This
sections present the basis for the biokinetic models used for the intake of tritiated compounds by
inhalation, ingestion or absorption through the skin. Examples of dose coefficients (committed effective
dose per unit activity taken into the body) for tritium calculated with the various biokinetic models are
provided.

A. Information on biokinetics and dosimetry of tritiated compounds

1. Tritiated water

(a) Early biokinetics of HTO in mammals

66. Tritiated water can enter the human body via ingestion of food and drink andmostly for
occupational exposureby inhalation of HTO vapour or direct absorption through skin exposed to
water or water vapour. Following ingestion, absorption from the alimentary tract into the bloodstream is
complete within a time range from a few minutes to some tens of minutes. Following inhalation, almost
the entire amount of inhaled HTO vapour is absorbed very rapidly from the respiratory tract into the
bloodstream [B7, P9]; absorption through skin provides an additional common route of entry into the
bloodstream [D3, O6, P9].

67. Following uptake to blood from the alimentary or respiratory tracts or through the skin, HTO is
transported by the circulatory system to all the body organs and tissues and diluted uniformly in body
260 UNSCEAR 2016 REPORT

water. This process takes from a few hours to some tens of hours. A small fraction of tritium from HTO
(0.5 to 4%) exchanges very rapidly with hydrogen in organic molecules as OH, NH and SH bonds
throughout body tissues. Another small fraction (from less than 1 to 3% in humans) is gradually converted
to OBT as a result of biochemical processes, i.e. CH bonds in organic molecules [B8, H12, P9].

68. The absorption of HTO through the skin from either the vapour or liquid phase has been
investigated by several authors. DeLong et al. [D3] exposed mice, rats and human adult volunteers to
HTO vapour in air. Animals were sacrificed following exposure. Urine and blood samples were
collected from human subjects for 48 hours after the end of exposure. Absorption rates of HTO,
calculated from measurements of tritium in blood and total body water, suggested that a delay occurred
in the distribution of the absorbed HTO. The absorption rate of HTO through the skin was the same
whether the skin was covered with a cloth (cotton) or uncovered. It was also proportional to the water
vapour pressure. This suggested a single diffusion mechanism for percutaneous absorption. However,
the absorption rate for the vapour phase was larger than could be accounted for by diffusion due to
vapour pressure alone, perhaps as a result of capillary action. The absorption rate increased with
increasing skin temperature. DeLong et al. [D3] and Pinson and Langham [P9] concluded similarly that
the quantity of HTO entering the body through the total skin, when exposed to an atmosphere
containing a given activity per unit volume, would be about equal to that entering through the lungs.
Osborne [O6] exposed volunteers to HTO in air and measured the tritium activity in urine, showing a
correlation between skin absorption rate and skin temperature.

(b) Long term biokinetics of HTO in mammals

69. Several studies have examined the biological half-time of HTO in a total of about 400 adults by
measuring tritium activity concentrations in urine. Butler and Leroy [B28] found this parameter to vary
with the amount of water ingested (decreasing with increasing water intake rate), with the ambient
temperature (decreasing with increasing ambient temperature) and with age (decreasing with increasing
age in adults). Their study, based on 310 cases of HTO intake, showed that the biological half-time of
HTO varied from about 4 to 18 days, with a mean of 9.5 days. During the warmer months, the average
half-time was lower; the difference being attributed to increased water intake. Other studies, based on
fewer cases, showed similar results [H12]: from 6 days for 8 cases [R16], to 12 days for 5 cases [B7].
In a few cases of accidental intake of large amounts of HTO, the excretion rate was accelerated with
diuretics and increased intake of fluids (e.g. [S2, T14]). The ICRP model uses a biological half-time of
10 days for HTO and 40 days for OBT formed from HTO in the body of adults [I8, I9, I10, I14, I17].

70. A number of studies have reported evidence for the presence of a second exponential component
of tritium activity concentration in human urine associated with the formation of OBT from HTO and
its subsequent removal. The biological half-time for OBT removal mostly ranged from 23 to 104 days
(mean 59 days) and its contribution to total excretion in 17 study subjects varied between 0.01 and
0.7% (mean 0.2%) [B7, H9, L4, S2, S17, T14]. A value of 40 days, based on carbon turnover in the
body, was adopted by ICRP [I13]. This was derived from the ratio of reference values for the body
content of carbon (16 kg) and daily carbon intake (0.3 kg) for adults [I7].

71. Some studies have reported an even longer component for the removal of tritium from the body
(e.g. [M15, M16, S2]). However, the parameters of this component as derived from data obtained on
five subjects occupationally exposed to either HTO or tritiated luminous compounds are very uncertain.
Such a component would contribute in only a minor way to tissue doses, as it represents less than 1% of
total OBT. The biological half-times of tritium following acute intake of HTO by adults reported by
various authors are shown in table 2.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 261

Table 2. Biological half-times of tritium in humans following acute HTO intake

Biological half-time (days)

Number
Study Compartment 1 Compartment 2 Compartment 3
of cases
(body water) (organically bound) (organically bound)
Pinson and Langham [P9] 9 11.3
Foy and Schnieden [F10] 10 511
(mean: 7.5)
Wylie et al. [W9] 7 6.412.1
(mean: 8.5)
Butler and Leroy [B28] 310 418
(mean: 9.5)
Osborne [O6] 30 6.414.4
(mean: 10.5)
Snyder et al. [S17] 1 8.7 34
Sanders and Reinig [S2] 1 6.1a
23 344
Minder [M15] 1 1030 139 to 230

Lambert et al. [L4] 1 9.1b 36

Moghissi et al. [M16] c

3 21 and 26 280, 550140,
350190
Henry [H9] 1 7.5 63

Balonov et al. [B7] 5 1113 39 to 76

(mean: 11.90.3) (mean: 517)
Trivedi et al. [T14] 8 6.212.8 d
58104
(mean: 8.4) (mean: 7418)
a
Oral diuretic administered from day 335 post-intake.
b
HT/HTO acute intake.
c
Data for three tritium luminous dial painters collected 610 months after termination of employment.
d
During the initial period when the exposed individuals increased fluid intake one month post-intake, the biological half-time
varied from 5.0 to 8.1 days with a mean of 6.3 days.

72. The partitioning of HTO and OBT after intake of HTO was examined by Takeda and Kasida [T1]
as part of a study of the biokinetics of HTO in rats. These investigators found that initially, the ratio of
tissue-bound tritium to total tritium was about 3% in the kidney and 15% in other tissues.
Measurements of tritium in human tissue are generally unavailable, but information can be derived by
biophysical modelling using long-term measurements of tritium in urine or, in some cases, blood
samples. The published data from the six studies [B7, H9, L4, S2, S17, T14] are available for such
analysis (table 2). Additional information can be derived from tritium measurements for organic
components of urine (urea) or blood (proteins, dehydrated cells). However, the number of reliable
measurements is limited [L4, R16, T16] and their representativeness with regard to tritium content in
organs and tissues is questionable.

73. The long-term retention of tritium in the human body as shown in figure V was modelled by
several authors as linear recurrent two-compartment models with transfer of tritium from HTO to OBT
(synthesis of biomolecules with OBT) and subsequent catabolic loss (degradation of biomolecules with
OBT) with all tritium excretion as HTO [B7, S17, T14]. Using this approach, the peak OBT activity
262 UNSCEAR 2016 REPORT

was calculated by Balonov and Chipiga [B6] from the data of the six studies referred to above [B7, H9,
L4, S2, S17, T14]. The peak OBT activity was estimated to be reached in 1638 days (mean 265 days)
after a single HTO intake and accounted for 0.11% (mean 0.40.2%) of the initial intake of tritium.
Balonov and Chipiga estimated the contribution of OBT to total soft tissue doses after intake of HTO
using 17 available data sets and obtained values of 1.84.6% (mean 3.00.9%) [B6]. It is notable that
the contribution of OBT to the total dose was estimated to be similar in nine subjects with relatively
low previous occupational tritium intake [H9, L4, S2, S17] or in five volunteers with no previous
tritium intake [B7] (3.00.9%) and in eight workers at a Canadian heavy water reactor (3.00.8%),
indicating that the contribution of previous chronic HTO intake by the workers was insignificant.

Figure V. Schematic two-compartment recurrent model of tritium biokinetics following HTO intake
in the body [B7, S17]
I(t) is HTO intake rate (Bq/day); W(t) is tritium activity in the body as HTO (Bq); B(t) is tritium activity in the body as OBT (Bq);
U(t) is tritium activity excreted from the body as HTO (Bq): , and are transfer rate constants (day-1)

74. Trivedi et al. [T14] assessed the contribution of OBT to effective dose following acute intake of
HTO by calculations based implicitly on a simpler linear two-compartment model with no transfer of
tritium from HTO to OBT. Instead, it was assumed that some fraction of HTO taken in was instantly
converted to OBT that was gradually degraded to and excreted as HTO (figure VI). The contribution of
OBT to the effective dose assessed with this model was 39% (mean 5.32.1%) for 15 subjects [B7,
H9, I14, S17, T14] excluding data from Rudran [R16] obtained on workers occupationally exposed to
tritiated luminous compounds. In another study by Trivedi et al. [T16] the tritium concentrations in
urine and blood samples both as HTO and OBT for six workers chronically exposed to low levels of
tritium were measured. The activity concentration of OBT per gram of hydrogen in OBT from urine
and blood samples (their ratio was 0.950.25) was assumed to be equal to that in body tissues. By
means of a simple equilibrium model, they calculated the contribution of OBT to the total dose to be
equal to 4.79.9% (mean 6.93.1%) for the six workers. This is in good agreement with the previous
results from 15 subjects with acute intake of HTO.

75. The estimate of Trivedi et al. for the contribution of OBT to dose following acute intake of HTO
(figure VI) is 1.8 times greater than that obtained by Balonov and Chipiga using a more physiologically
realistic model (figure V); it does not account for the gradual bioaccumulation of OBT and assumes its
instant formation from HTO. A similar conservative approach is used by the ICRP model for HTO (see
also figure VIII) which is applied for radiation protection purposes [I14]. According to this model, 97%
of HTO absorbed to blood is distributed in body water (T1/2=10 days in adults) and 3% is instantly
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 263

converted to OBT (T1/2=40 days in adults); the contribution of OBT to the effective dose in the ICRP
model is about 9% under assumption of uniform OBT distribution in organs and tissues.

Figure VI. Schematic two-compartment model of tritium biokinetics following HTO intake in the
body [T14]
I(t) is HTO intake rate (Bq/day); x is fraction of tritium instantly converted to OBT (dimensionless); W(t) is tritium activity in the
body as HTO (Bq); B(t) is tritium activity in the body as OBT (Bq); U(t) is tritium activity excreted from the body as HTO (Bq):
and are transfer rate constants (day1)

76. Other authors [H12, J3, N1, S2, T9] have used the limited available human data to develop a
three-compartment model of tritium retention following intake of HTO. The results showed
considerable variation, both in the observed biological half-times and in the proportions of tritium
entering OBT compartments. The model parameters suggested by Taylor [T9] for adults are shown in
table 3. The resulting committed effective dose per unit intake of HTO by adults, on the basis of this
model, is 1.7 10-11 Sv/Bq. The current ICRP dose coefficient for HTO is 1.8 10-11 Sv/Bq [I15].

Table 3. Parameter values for HTO model proposed by Taylor [T9]

Model component Distribution (%) Biological half-time (days)

HTO 99 10
OBTa 0.98 40
OBT b
0.02 350
a
Short-term OBT compartment.
b
Long-term OBT compartment.
264 UNSCEAR 2016 REPORT

2. Inhalation and skin absorption of elemental tritium gas

77. Tritiated elemental hydrogen (HT) is only slightly soluble in body fluids and has a much lower
uptake into biological systems than HTO. After inhalation, most of the HT is exhaled, but a small
fraction is dissolved in body fluids and is then oxidized to HTO by anaerobic bacteria in the
gastrointestinal tract, the only known biological site of HT oxidation [I6].

78. The ICRP based its assessment of the effective dose resulting from the inhalation of HT primarily
on exposure of the lungs, as opposed to exposure of the skin. Oxidation to HTO in vivo was not
considered [I9]. The absorption of HT through the skin appears to be negligible and it does not convert
to HTO on contact with the skin [H12].

79. The studies of Pinson and Langham [P9] and Peterman et al. [P3, P4] found that exposure to HT
resulted in excretion of HTO in urine, and that the HTO, formed from the oxidation of HT, was retained
in the body and excreted with the usual biological half-time of about 10 days. About 0.01% of the HT
inhaled by human volunteers was converted to HTO in the body [P3, P4, P9].

80. Using their biokinetic models for HT and HTO and data from studies with human volunteers,
Peterman et al. [P3] concluded that the effective dose from inhaled HT was dominated by two roughly
equal contributions: the effective dose resulting from exposure of the lungs to HT in inhaled air, and the
effective dose resulting from exposure to HTO caused by the oxidation of HT. The current dose
coefficient for inhalation of HT given by the ICRP Publication 68 [I15] is based on the work of
Peterman et al. [P4] and its human respiratory tract model.

81. Most of the energy of the tritium beta particles is not deposited in the target cells of the
respiratory tract due to their short range. The average depth of the nuclei of these cells range from about
10 to 50 m in the extrathoracic, bronchial and bronchiolar regions of the respiratory tract [I16]. In the
current ICRP model of the human respiratory tract [I16], the tritium beta particles are assumed to
deliver a dose only in the alveolar-interstitial region after inhalation of HT. Trivedi and Gentner [T15]
noted that the nuclei of target cells within the alveolar-interstitial region, at depths of less than 10 m,
are assumed to receive some dose.

3. Contact of skin with tritium-gas-contaminated surfaces

82. Eakins et al. [E1] applied tritium-gas-contaminated metal surfaces to the forearms of four
volunteers. The estimated average body content of HTO and OBT were about 0.5 and 0.3%,
respectively, of the applied activity of tritium gas. The results did not depend on the type of material
tested and the initial body content of organic tritium was less than that of HTO for each volunteer.
Urinary excretion of tritium, initially primarily OBT, reached a peak about 24 hours after exposure. Up
to 50% of the OBT was excreted via urine with a biological half-time of about onetwo days; the rest of
the OBT was excreted with a biological half-time of 0.10.2 days. Onethree weeks after exposure, the
tritium was excreted predominantly in the form of HTO with a biological half-time of about 14 days.
The effective dose resulting from intake of tritium from contact with a contaminated surface was
estimated to be about 9 1012 Sv/Bq [J4].

83. Similarly, Trivedi [T13] used hairless rats and showed that when HT-contaminated stainless steel
was brought in contact with intact skin, tritium was fixed as OBT and HTO in the skin and
demonstrated biphasic excretion of OBT and HTO.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 265

4. Ingestion of organically bound tritium

84. While HTO diffuses freely in body tissues and enters into equilibrium with body fluids in a
matter of minutes or hours [H12, P9], several factors determine the distribution in the body of OBT
ingested through the diet. These include the biochemical composition of OBT as a mixture of
tritiated carbohydrates, fats and proteins, the oxidation rates of the dietary constituents, absorption
from the alimentary tract, and the synthesis and retention of organic forms and the HTO and OBT
excretion rates [D5].

85. Experiments indicate that about 320 times more tritium is bound to organic compounds in more
metabolically active tissue after ingestion of food containing OBT than that after ingestion of the same
activity of tritium in the form of HTO [P7, R11, T3, T6]. This ratio depends on the animals studied
(rabbits, rats, mice), the kind of tritium-labelled food (e.g. alfalfa, wheat, meat, shrimp) and the
duration of tritium intake (single, few weeks, three generations). Takeda and Kasida [T1] found that,
following the intake of HTO by animals, 15% of the tritium was incorporated into organic constituents
of rat tissue. Therefore, several tens of per cent of the tritium ingested as OBT would be expected to be
incorporated into organic molecules in mammal tissues. The assumption made in the ICRP
Publication 60 [I13] that 50% of tritium is incorporated into OBT in tissues after an intake of OBT
stems from this reasoning.

86. On the basis of all available data, several authors [I12, K15, M1, P6, R11, R12] suggestedas a
rounded valuethat approximately nine times more OBT may be present after intake of OBT than after
intake of HTO. This corresponds to a range of 945% for the proportion of tritium reaching blood that
was retained as OBT, the remainder being converted to HTO (the wide range resulting from the
differing metabolic roles of different OBT molecules, e.g. as an energy source or a structural
component). Further, it was also suggest that the use of a value of 50% would be suitable for general
radiological protection applications.

87. While the biokinetic parameters used by ICRP are supported by a range of data, there have been
suggestions that the contribution to dose from OBT may be greater for intake of either HTO or OBT
than predicted by ICRP models. However, the conclusions of Takeda [T4], Komatsu et al. [K21] and
Rodgers [R12] are reasonably consistent with the ICRP conclusion that the contribution to dose from
OBT after intake of HTO will be small (<10%) and that the overall dose from intake of OBT will be
greater than from HTO by about a factor of two (the data in table 4 show the ratio in ICRP dose
coefficients to be about 2.5).

88. Some experimental data suggest that after chronic intake of HTO, equilibrium tissue concentrations
of HTO and OBT are similar [C25]. Etnier et al. [E3] used a four-compartment model of hydrogen
metabolism to show theoretically that OBT in food can increase the cumulative total body dose by a factor
1.74.5 times the free body water dose alone. This model is regarded as providing a reliable representation
of tritium biokinetics. The predictions in the model were demonstrated by Takeda et al. [T3] using rats fed
with tritiated wheat and HTO. After chronic ingestion of tritiated wheat (22 days), the amount of OBT in
the tissue of rats exposed to tritiated wheat was about 611 times higher than when exposed to HTO.

89. Hunt et al. [H19] measured the retention time of tritium in volunteers who had eaten fish that
contained OBT or HTO as a consequence of discharges from the General Electric Healthcare Ltd. plant
in the United Kingdom. The excreta from five volunteers were screened for a period of up to 150 days
after intake. The results suggested biological half-times ranging from four to eleven days for the
retention of the total amount of tritium, with no evidence for a statistically significant contribution from
a component with a longer half-time.
266 UNSCEAR 2016 REPORT

90. Animal studies have shown a non-uniform distribution of OBT in soft tissue, and also a non-
uniform distribution of tritium in the various organic compounds in the body [D5, I13]. However, with
the exception of the special case of DNA precursors, discussed later, any non-uniformity of distribution
within cells appears to be small.

91. Richardson and Dunford [R9] conducted a literature review of the studies of exposure to OBT. They
proposed two biokinetic models governed by the overall metabolic reactions of the principal nutrients:
carbohydrates, fats, and proteins. The parameters for two models of differing complexitycalled the
HCNO-S and HCNO-C modelswere evaluated on the basis of biochemical reactions. The simpler
model has a single compartment representing the principal nutrients. The more complex model includes
compartments representing the longer-term retention of carbohydrates as glycogen, fats as adipose tissue,
and proteins in bone and soft tissue. The differences in the water and organic contents of tissues and
organs and the different biokinetics for the different organic components were considered [R8, R9].

92. Melintescu et al. [M10] and Galeriu and Melintescu [G1] developed other physiologically based
multicompartment models where OBT exchange rates were associated with energy metabolism of
major groups of human organs and tissues. Model predictions for HTO intake were successfully tested
against available human data, e.g. from Trivedi et al. [T14]. Final model solutions were presented as
dose coefficients for both HTO and OBT for all ICRP age groups separately for males and females and
sex-average values.

93. Table 4 shows values of committed effective dose per unit intake for adults obtained from
Richardson and Dunford [R8], an earlier paper of Richardson et al. [R6], other physiologically based
models of Melintescu et al. [M10] and Galeriu and Melintescu [G1] and the corresponding values
calculated using the ICRP models for HTO and OBT. The dose coefficients from the various studies are
similar to the ICRP values [I17]. Richardson and Dunford [R8] also provided dose coefficients for tritiated
nutrients, with the lowest value for tritiated carbohydrates equal to 3.3 1011 Sv/Bq, which is 21% lower
than the ICRP value for OBT, and the highest value for tritiated protein being 8.4 1011 Sv/Bq, which is
twice the ICRP value for OBT [I17, R8].

Table 4. Adult effective dose coefficients from physiologically-based biokinetic models and ICRP
models for ingestion of HTO and organically bound tritium

Biokinetic model Dose coefficient (1011 Sv/Bq)

Ingestion of HTO
ICRP [I14] 1.8
Richardson et al. [R6] 1.8 (males); 2.2 (females)
Melintescu et al. [M10] 1.7 (males); 2.4 (females)

Galeriu and Melintescu [G1] 2.0

Ingestion of OBT
ICRP [I14] 4.2
Richardson et al. [R6] 4.2 (males); 6.1 (females)

Richardson and Dunford [R8] and ICRP [I7] 5.07.4

Melintescu et al. [M10] 3.9 (males); 5.7 (females)
Galeriu and Melintescu [G1] 4.9
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 267

94. Physiological models of OBT biokinetics proposed by Richardson et al. [R6] and Melintescu et al.
[M10] provide sex- and age-differentiated biokinetics and dose coefficients. This analysis was based on
estimates of daily intake rates of carbon in females (228 g carbon) and males (303 g carbon) and
calculated biological half-times of carbon of 51 days and 40 days for female and male adults,
respectively. Table 4 presents the dose coefficients from physiologically-based OBT biokinetic models
by sex. The dose coefficient is higher in females by about 2040% for HTO and up to 50% for OBT.
Further, the authors that the modelling of age dependence of the dose coefficients both for HTO and
OBT is in good agreement with ICRP data [I17, M10].

5. Intake of tritiated biochemical substrates

95. Tritiated biochemical substrates, such as glucose, amino acids, hormones, and DNA and RNA
precursors, are produced and widely used in biomedical research, leading to possible exposure by
inhalation, skin contamination and possible inadvertent ingestion. A general metabolic feature is that
such biochemical substrates may be directly incorporated into organic molecules in body tissue if
absorbed to blood and transported intact to sites of active metabolism within cells. The extent of
incorporation of tritium into specific forms of OBT is determined by such factors as the chemical
compound containing tritium, its isomeric form, the position of the label in the molecule, and the
amount of carrier [B11, B12, F2, L1, T2, T4, T8, T10]. Catabolism of labelled compounds will result in
tritium being partially oxidized and entering the body water as HTO or catabolized and excreted as low
molecular weight organic substances.

(a) Tritiated glucose and amino acids

96. Studies by Takeda [T2, T4] and Balonov et al. [B8, B11, B12], have shown that the bound fraction
of tritium administered intraperitoneally to rats as biochemical substrates varied from 35% (tritiated
D,L-alanine, glucose) to 5080% (tritiated L-tyrosine, L-lysine). The bound fraction of tritium from
tritiated biochemical substrates was substantially larger than that of HTO. The low retention of tritium as
OBT after administration of tritiated glucose is consistent with its rapid catabolism while the high
retention of tritiated lysine is consistent with its incorporation into protein as an essential amino acid.

97. Retention of tritium administered as the amino acids, glycine, leucine and methionine, was
intermediate between that of glucose and lysine. D-isomers of 3H-amino acids were assimilated
considerably less than L-isomers. Thus, the level of binding of 3H2-L-leucine-2,3 in different tissues,
except kidneys, was 22.5 times higher than that of the D-isomer, and 1.52 times higher than that of
racemic mixture. Similar results were obtained with 3H-lysine isomers [B8, B12].

98. Figure VII shows the retention curves for OBT in rat spleen after intraperitoneal injection of HTO
and some biochemical substrates [B8, B12]. Tritium was rapidly excreted after administration of
3
H-glucose with a half-time of 34.5 days. Retention functions for the majority of 3H-amino acids
showed two components: one with a half-time from 0.7 to 2 days and the second with a half-time from
7 to 16 days, presumably reflecting metabolism of two groups of functionally different proteins.
Retention times were shown to be tissue specific. In actively proliferating tissuesbone marrow, small
intestine, testisthe observed kinetics of OBT is influenced by the processes of cell differentiation and
movement of labelled cells out of the organ. Due to high demand for L-lysine in tissues and its
reutilization, the kinetics of its excretion from tissues is slowed twothree-fold relative to other amino
acids [B8, B12, T2, T4].
268 UNSCEAR 2016 REPORT

Figure VII. Specific activity of bound tritium in rat spleens normalized to intraperitoneally injected
activity of tritium (per g of body weight) after injection of HTO, D-glucose-6-3H, glycine-2-3H2 and
L-lysine-6-3H [B8, B12]

99. Table 5 presents estimated values of absorbed dose (D) in four organs and tissues and the
contribution of OBT to dose for HTO and three biochemical substrates derived from the same studies of
intraperitoneal administration to rats [B8, B12]. Injection of 3H-glucose did not create a dose
significantly higher than the dose from an equal amount of HTO, and differed only in the increased
contribution to dose from OBT (723% against 2.54%). The dose following administration of
3
H-glycine was 540% greater than for HTO, and contribution of OBT to the dose reached about
20-40%. Notably, for administration of L-lysine-3H, the dose was twoeight times that for HTO, and
the contribution from OBT was >90%.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 269

Table 5. Absorbed dose, D, in rat organs and tissues (mGy) and OBT contribution to dose (%) after
intraperitoneal injection of 37 kBq/g of HTO and tritiated biochemical substrates [B8, B12]

HTO D-glucose-6-3H Glycine-2-3H2 L-lysine-6-3H

Organ, tissue
D (mGy) OBT (%) D (mGy) OBT (%) D (mGy) OBT (%) D (mGy) OBT (%)
Bone marrow 21 4 21 19 26 38 92 97
Small intestine 22 23 59 95
Testis 20 3 19 7 21 24 40 92
Muscle 20 2.5 28 43 150 98

100. Absorption coefficients for ingestion of and skin contamination by tritiated biochemical substrates
were obtained in experiments in which preparations were administered to rats. The absorption
coefficient was defined as the average ratio of OBT levels in tissue after ingestion or skin application to
those after intraperitoneal injection. While 3H-glucose and 3H-amino acids were completely absorbed
by the gastrointestinal tract [B12], absorption of 3H-thymidine was considerably lower (1020%)
because of its degradation to 3H-thymine in the gastrointestinal tract. In contrast, 3H-deoxycytidine was
absorbed almost completely (60100%). In rats fed HTO, tritiated amino acids, glucosamine, or
tritiated DNA or RNA precursors for 22 days, the greatest concentrations of OBT were found after
exposure to amino acids, with intermediate concentrations found after exposure to DNA and RNA
precursors [T4]. Studies using rat everted gut sacs showed only about 2% of tritiated thymidine crossed
the intestinal epithelium [L2].

101. During six hours after the application of preparations on rat skin, 14% of tritiated substrates were
absorbed in blood, and 0.010.5% of tritium activity was measured in the skin layer at the place of
application after its decontamination [B8, B12].

(b) Tritiated nucleic acid precursors

102. The DNA precursors, deoxythymidine and deoxycytidine labelled with tritium, have most
commonly been used in studies of cell kinetics. For work involving RNA, tritiated uridine and adenine
are the precursors that have been used [H16]. After tritiated thymidine has been orally administered to
humans or animals, about 2% is incorporated into DNA during the synthesis stage of the cell cycle
[L2], and the remainder appears as HTO. Tritiated thymidine is available for only a short time after
intake and primarily for uptake by rapidly cycling cells such as those of the bone marrow or gut.
However, prolonged administration of 3HTdR throughout gestation results in labelling of slower
cycling cells [K11].

103. According to Feinendegen et al. [F5], nuclei of 530% of proliferating cells in mammals are
labelled by tritium. As the average range of tritium beta radiation is considerably less than the
dimensions of the nuclei of mammal cells, and distribution of DNA-bound tritium is extremely
inhomogeneous both on the scale of organs and tissues and also within cells, the concept of the average
organ or tissue dose in the case of incorporation of 3H-nucleosides requires care in its interpretation. An
alternative is to estimate dose to radiosensitive nuclei in rare cases of tritiated DNA-precursor
incorporation by workers.
270 UNSCEAR 2016 REPORT

104. Especially in embryo,3H-deoxynucleosides are actively included in intensively proliferating cell

systems of bone marrow and small intestine, and considerably less in tissues with small frequency of
mitoses (e.g. muscle, liver). A high concentration of 3H-CdR in bone marrow of rats is noteworthy
[B12], contrasting with lower values obtained in mice [F5]. The dose in cell nuclei of bone marrow
labelled by 3H-CdR estimated in studies by Balonov et al. [B8, B12] according to the methodology
suggested by Feinendegen and Cronkite [F4, F5] is larger by two orders of magnitude than the average
tissue dose from an equal amount of injected HTO.

105. Taylor [T10] reviewed the biokinetics of 11 xenobiotic tritiated organic compounds and estimated
that the clearance half-time in humans was less than 40 days in all cases. Some organic compounds may
be incorporated directly into structural components and retained for longer periods.

6. Metal tritides and other forms with low solubility

106. Tritiated compounds with low solubility include luminous compounds (powder and paint), tritides
of metals (e.g. Ti, Zr, Hf), microfragments of glass and carbon and beryllium particles contaminated
with tritium. Such compounds are produced and widely used in industry (painting wrist watches and
compasses with luminous paint, ionization sources) and research (e.g. accelerator targets, tritium carrier
in fusion physics). Tritiated compounds with low solubility are considered as sources of internal
exposure in workplaces.

(a) Metal tritides

107. When tritides of metals (e.g. Ti, Zr, Hf) are used for industrial or research purposes, the tritium
chemically bound in the crystal matrix is gradually desorbed from the metal surface in the form of HTO
or HT. Due to external exposure of the device surface to radiation fluxes and heat, metal tritides can
also be released into the work environment in particulate form. The main pathway of radiation exposure
of workers dealing with devices containing metal tritides is the inhalation of HTO vapour and airborne
particles of metal tritides.

108. Balonov et al. [B8, B11] reported that, following short-term inhalation by rats, titanium tritide
(TiT) showed slow lung clearance during one month after exposure. Similar experiments of Cheng et
al. confirmed slow clearance of a fraction of TiT (1 m count median diameter (CMD)) after
intratracheal instillation of TiT suspension into rats [C18, Z8]. Similar results were obtained in
experiments on rats with hafnium tritide (1 m CMD) [Z7] and zirconium tritide (0.3 m CMD) [Z9].
The size of the slow cleared fractions and uptake rates of tritium dissolved from articles depend on the
methods of TiT particle production and particle size.

109. In a series of in vitro experiments with powders of titanium, hafnium and zirconium tritides aimed
at simulating time-dependent absorption functions of tritium in the respiratory tract of rats and humans,
the dissolution of tritium in synthetic serum ultrafiltrate was studied during 30200 days by Cheng et
al. [C17, C19, Z9]. The dynamics of tritium release from particles was well described with two
exponentials, one with a half-time in the range from one day (TiT) to about 50 days (HfT and ZrT) and
the second with a half-time in the range from one month (TiT) to one year (ZrT) and several hundred
years (HfT). The long-term dissolution half-time from coarse powder (~100 m) was larger than from
fine powder (~1 m) [C17].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 271

110. In vitro and in vivo experiments were complemented with dosimetric considerations of beta
radiation self-absorption in particulate material [C19], which became important for particle sizes of
0.1 m and more. From the available data, it was concluded that airborne ZrT and HfT should be
considered in human internal dosimetry as material of slow solubility (ICRP Type S), and TiT as
material of medium solubility (ICRP Type M).

(b) Components of nuclear fusion reactors

111. Tritium is present in nuclear fusion facilities in various physical and chemical forms, some of
which have radiological properties different from HTO and HT. In the 1980s, features of tritiated
microballoon glass fragments that could be potentially inhaled by workers were studied both in vitro
and in vivo [C27, C28]. The median diameter of glass fragments was initially estimated to be about
4 m but, accounting for fragment shape, the mass median diameter was specified as 20 m.
Experiments in vitro have shown that 98% of tritium is released from glass with a half-time of 39 days
and 2% with longer half-time of 23280 days. After intratracheal instillation of tritiated glass fragments
in rats, 93% of the tritium activity was removed from glass with half-time of 60.5 days and 7% with a
longer half-time of 433 days, consistent with the results of the in vitro study. The resultant dose to the
lung was three orders of magnitude greater than the dose to other tissues, and approximately 40 times
greater than the dose incurred in the lung from the inhalation of a similar quantity of HTO. However,
because of the low inhalability of tritiated glass fragments of such sizes, this ratio would be much lower
in a direct comparison of inhalation doses.

112. Since 1999, when the Joint European Torus (JET) fusion tokamak started its operation with
tritium, tritiated dust and flakes were observed predominantly during maintenance operations. Those
carbon, beryllium and tungsten particles were formed due to the interaction of plasma with the carbon-
based first internal wall of the fusion reactor. The activity median aerodynamic diameter (AMAD) of
the dust particles was assessed to be ~4 m with high specific activities of up to 3 GBq/g and AMAD
of flakes as about 100 m [D4]. After inhalation, the former can deposit in different parts of the
respiratory tract and expose its tissue to beta radiation. The in vitro dissolution tests with tritiated dust
have shown that 15% of tritium activity was dissolved in lung serum simulant within one minute and a
further 120% of tritium were dissolved over the next 100 days [H15]. Slow excretion of tritium
deposited in lungs may complicate its individual monitoring [R13]. From the available data, it can be
concluded that carbon and beryllium dust particles from fusion reactors can be classified as material of
medium or slow solubility (ICRP Type M or S).

(c) Luminous compounds

113. The basis for self-luminous tritium paint is fine zinc sulphide powder (~10 m) coated with a thin
layer (0.010.1 m) of tritiated polymer with high specific activities. In order to reach higher light
intensities, tritiated polymers (polystyrene, silicon rubber) with specific activities up to 20 TBq/g have
been used [B8, I3]. During the paint luminizing process, workers may be exposed to T2, HTO, and
vapours of tritiated organic solvents via inhalation and skin contact. Particles deposited in the lungs will
irradiate tissues by emission of beta particles and bremsstrahlung. HTO and unknown organic
compounds have been detected in urine [B8, R15], formed by degradation of polymers by radiolysis,
oxidation and isotopic exchange.

114. In the past, some workers were exposed to high internal doses that were fatal in a few cases [M15,
S12]. Lambert and Vennart have shown that radiological control in the workplace usually necessitates
continuous biological monitoring of workers for tritium in urine at relatively short intervals [L5].
272 UNSCEAR 2016 REPORT

115. In vitro experiments have shown that 0.55% of tritium from tritiated polystyrene-based Soviet
luminous powder (PS-A) was gradually released in buffer mixture as HTO and low molecular organic
foreign compounds [B8, B11]. In vitro studies of the dissolution of commercial luminous powder made
from tritium-labelled polystyrene in bovine serum over five days [R15] showed that an average of 12%
dissolved on the first day, and about 2% of the remaining activity on subsequent days.

116. Similar uptake (0.55% depending on compound age) of both HTO and low molecular organic
compounds was observed in experiments with PS-A luminous powder orally administered to rats and in
a similar study with three volunteers [B8, B11]. Substantial fractions of the tritium compounds were
rapidly excreted in urine and faeces of rats, a proportion was retained for some days in the liver/kidney,
and the remainder was catabolized to HTO. The half-time of organic tritium in human urine was about
one day and the remaining tritium was excreted as HTO with a mean half-time of about 16 days.

117. Experiments on cats showed that absorption of tritium from luminous paints depended on the
plastic substrate involved, with values of 0.007 for polystyrene, about 0.03 for silicone rubber and
0.8 for polyester [H12, W6]. Balonov et al. [B8, B11] also reported that following intratracheal
instillation of PS-A luminous powder into rats, the lung specific activity showed essentially no decrease
within five months, demonstrating its very low solubility.

118. In summary, from the available experimental and human data, it can be concluded that the
specific characteristics of tritiated luminous compounds differ substantially from those of HTO. Special
attention should be paid to individual bioassay monitoring.

B. Overview of current biokinetic models for tritium

119. Tritium may enter the body by inhalation, absorption through skin, and ingestion. The first two
are the more frequent routes of intake in the workplace, while the latter mostly contributes to exposure
of members of the public. Furthermore, skin contact with tritium-contaminated surfaces, such as metal
and glass, has been shown to result in the formation of OBT in the body [E1]. This has been shown to
be a route of tritium intake in the workplace [H12].

120. The fate of tritium once taken into the body is determined mostly by its chemical form. One can
expect to find HTO in most workplaces and environmental media where tritium is present. In general,
consideration of the biokinetics of OBT refers to the non-exchangeable component resulting from
tritium atoms bonding directly to carbon in organic molecules and exhibiting retention times relating to
carbon turnover; the exchangeable component of tritium in organic molecules, in the form of hydroxyl
and sulphydryl groups, for example, has the same metabolism and distribution in the body as HTO.

121. Tritiated compounds may also exist as airborne particles, e.g. metal tritides or luminous powder. The
retention and clearance of these particles from the respiratory tract depend on several factors, such as
particle size and chemical composition. For dose assessment purposes, tritium absorbed into blood after
tritiated particles have been inhaled is usually treated as being in the form of HTO [C18, C19, C20, I17].

122. Tritiated compounds are categorized by ICRP for radiation protection purposes according to the
metabolic model that best describes their dynamics after intake and subsequent uptake. Three primary
metabolic models are currently used by ICRP either separately or in combination to calculate
committed effective dose for an intake of tritiated compounds [I8, I9, I10, I12, I14, I17]:
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 273

(a) A model for tritium absorbed to blood as HTO following either ingestion or inhalation, applied
also to other tritiated compounds that partially convert to HTO after being taken into the body;

(b) A model for tritium absorbed to blood following intake of OBT, mainly by ingestion in food,
but also applied to inhalation of non-specific organic molecules, and to ingestion or inhalation of
some specific tritiated organic compounds;

(c) The generic ICRP model for the human respiratory tract, specifying absorption parameter
values for inhalation of poorly soluble forms of tritium.

123. All three models and respective sets of dose coefficients are widely used by the Committee in its
main reports, by the IAEA in its international standards (e.g. [I4]) and also in documents of WHO
[W7], FAO and the CODEX Alimentarius Commission [C24]. The ICRP modelsand especially dose
coefficientsare included in numerous national regulations on worker and public protection against
internal exposure with different forms of tritium.

1. Dosimetry of tritium using model for HTO

(a) Biokinetic model for HTO

124. The ICRP biokinetic model for systemic HTO [I12, I14, I17], illustrated in figure VIII, is used to
calculate the committed effective dose from:

(a) Intakes of (HTO);

(b) HTO formed following inhalation of elemental hydrogen (HT, T2);

(c) HTO formed following inhalation of tritiated hydrocarbon vapours and gases (e.g. CH3T).

Use of this model involves assumptions about the conversion of the other forms of tritium to HTO.

125. For modelling HTO intake, an instantaneous translocation to blood is assumed for both inhalation
and ingestion. The ICRP model further assumes that HTO is transferred from blood with a biological
half-time of six hours and distributed uniformly throughout the body. It is also assumed that 97%
remains as HTO, while 3% is instantly converted to OBT. In adults, HTO is assumed to be retained
with a biological half-time of 10 days, and OBT with the biological half-time of carbon, calculated as
40 days (figure VIII). ICRP values for the partitioning between HTO and OBT after acute intake of
HTO in various age groups, and the corresponding biological half-times, are given in table 6 [I12, I14,
I17]. Age dependence of biological half-time for HTO was derived by ICRP from available human
observations and physiological data [I7].

126. The calculation of committed effective dose per unit intake (dose coefficient) for adults
resulting from the intake of HTO, as given by ICRP [I14, I17], is based on the model presented in
figure VI. The current value for intake of HTO by adults both by inhalation and ingestion computed
by ICRP is 1.8 10-11 Sv/Bq [I14, I15, I17].

127. The contribution of the OBT fraction to the committed effective dose in the HTO model has been
shown to be about 10% [I8, I9, I10, J3]. This is adequate for estimates of dose from both acute and
prolonged intake [T16].
274 UNSCEAR 2016 REPORT

Table 6. ICRP parameter values for distribution and retention of tritium after acute HTO intake
[I12, I14, I17]

Initial distribution (%) Biological half-time (days)

Age
HTO component OBT component HTO component OBT component
3 months 97 3 3.0 8
1 year 97 3 3.5 15
5 years 97 3 4.6 19

10 years 97 3 5.7 26
15 years 97 3 7.9 32
Adult 97 3 10.0 40

Figure VIII. ICRP model for biokinetics of HTO [I12, I14, I17]

(b) Absorption of HTO through skin

128. Immersion in a vapour or liquid containing HTO also results in its absorption through the skin
[D3, O6, P9]. In the context of occupational exposure to HTO vapour, ICRP [I8, I9, I10] referred to the
work of Osborne [O6], Hill and Johnson [H12], and the review of Myers and Johnson [M26] and
concluded that about 1% of the HTO activity per cubic metre of air may be assumed to be absorbed
through the skin in a minute. On this basis, the amount absorbed through skin contributes about one
third of the total HTO intake for a given HTO concentration in air when the exposed individual is active
during exposure (i.e. breathing in more air than when at rest). The work of Osborne [O6] was based on
direct measurements of the absorption rate of HTO vapour through the skin of the whole body.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 275

(c) Inhalation of tritium gas

129. Following inhalation of HT, a small fraction (about 0.01%) is dissolved in body fluids and
oxidized to HTO [P4]. The latter is the predominant contributor to the committed effective dose. HT is
not significantly absorbed through the skin and does not readily convert to HTO on the skin. Irradiation
of the lungs by inhaled HT does not significantly increase the committed effective dose [I16] because
of the short range of the tritium beta particles in lung tissue. The dose coefficient for inhalation of HT is
therefore about 0.01% of the dose coefficient used for inhalation of HTO.

(d) Inhalation of tritiated hydrocarbons

130. Tritiated methane is the only tritiated hydrocarbon for which ICRP recommends a dose coefficient
for inhalation based on the HTO model. Tritiated methane is known to be formed as a result of
microbial degradation within tritiated waste. About 1% of the inhaled tritiated methane is assumed to be
converted to HTO [P5]. The current ICRP dose coefficient for tritiated methane is therefore 1% of that
for HTO [I17]. The ICRP approach is considered to give a conservative dose coefficient [P5].

131. Carlisle et al. reported in a more recent study on rats that the fraction of tritiated methane retained
in the body as HTO and OBT after acute inhalation to be about 0.06 to 0.13%, which is lower than that
assumed by the current ICRP model [C6]. However, the observed conversion of tritiated methane to
OBT in rats was greater than that estimated by the ICRP model in all human tissue examined [I17], in
particular in the liver, where the conversion was observed to be 22 times greater than that of HTO
estimated by ICRP. The authors further suggested that some of tritium taken in as tritiated methane is
converted directly to OBT. They concluded that the committed dose to some organs is one third to one
tenth of that estimated by ICRP and that the ICRP value of effective dose may be conservative.

(e) Summary of dose coefficients based on HTO model

132. Table 7 presents the current ICRP committed effective dose coefficients for tritiated compounds
based on the biokinetic model for HTO and table 8 illustrates the effect of age on the computed
committed effective dose per unit intake for the inhalation of HTO [I17].

Table 7. ICRP effective dose coefficients based on HTO model for various tritiated compounds,
modes of intake and age groups [I14, I17]

Dose coefficient (Sv/Bq)

Tritiated compound Mode of intake
Infants (1 year old) Adults
HTO Inhalation 4.8 1011
1.8 1011
HTO Ingestion 4.8 1011 1.8 1011
HT Inhalation 4.8 1015 1.8 1015
CTH3 Inhalation 4.8 1013 1.8 1013
276 UNSCEAR 2016 REPORT

Table 8. ICRP effective dose coefficients for HTO inhalation by various age groups [I17]

Age Dose coefficient (Sv/Bq) Ratio of dose coefficient to that of an adult

3 months 6.4 1011
3.6
1 year 4.8 1011
2.7
5 years 3.1 1011
1.7

10 years 2.3 1011 1.3

15 years 1.8 1011 1.0
Adult 1.8 1011
1.0

2. Dosimetry of OBT and tritiated biochemical substrates using model for OBT
133. Figure IX illustrates the ICRP model [I14] used to calculate committed effective dose for the
ingestion of OBT. It is assumed that OBT, once taken into the body, is translocated to blood completely
and instantaneously, and is transferred from blood to tissue with a biological half-time of six hours,
with 50% retained in tissue as OBT and 50% transformed into HTO. The uptake and retention of
tritium in various tissues depends on the metabolic activity of the individual tissues and the constituent
chemical forms of OBT. However, the ICRP model assumes uniform distribution of doses from OBT to
all soft tissue in the body while a greater metabolic activity leads to greater uptake and more rapid loss.

Figure IX. ICRP model for biokinetics of OBT [I14]

134. The ICRP model for the ingestion of OBT [I12, I14] is intended to represent the biokinetics of the
average dietary content of the different chemical forms of OBT. The model was developed in the
absence of information about the exact proportions of the various chemical components of OBT in the
human diet and the turnover of these components [I12, I14]. On the basis of the turnover of hydrogen in
Reference Man, less than 10% of tritium taken in daily as OBT in diet is assumed to be excreted daily
in the form of OBT (mostly in the form of urea, with about 3% in faeces) while the rest is assumed to
be excreted as HTO [R8].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 277

135. Table 9 gives the assumed partitioning between HTO and OBT and the corresponding biological
half-times after dietary intake of OBT in various age groups [I14]. The ICRP committed effective dose
coefficients for OBT are shown in table 10 [I17]. Table 11 shows the effect of age on the committed
effective dose per unit intake for the ingestion of OBT.

136. The ICRP model for OBT and relevant dose coefficients can also be used for prospective dose
assessment in cases of intake of tritiated biochemical substrates if more specific biokinetic information
is not available. For most biochemical compounds, this approach will lead to some overestimation of
effective dose [T10]. Considering a range of biochemical compounds for which biokinetic data are
available from animal experiments, only the essential amino acid, L-lysine-3H, resulted in estimated
tissue doses greater than those for HTO (by a factor of 28) [B11, B12, T2].

137. The ICRP model for OBT and the corresponding dose coefficients are not applicable for dose
assessments in cases of occupational intake of tritiated nucleic acid precursors. After tritiated DNA-
precursors are taken in by humans or administered to animals, some fraction of tritium is incorporated
into DNA during the synthesis stage of the cell cycle [L3], and the remainder appears as HTO or
metabolized biochemical substrates. As distribution of DNA-bound tritium and its radiation energy is
extremely inhomogeneous in organs, tissues and inside cells, the concept of average organ or tissue
dose in this case requires careful consideration. An alternative approach is to calculate dose taking the
localization of tritium in cell nuclei into account [F2, F3, F4, F5, N1].

138. The National Council on Radiation Protection and Measurements (NCRP) [N1] examined the
absorbed dose resulting from the ingestion of 3HTdR based mainly on theoretical considerations. For
acute intake, it concluded that the absorbed dose to stem cells and bone marrow (per unit intake)
resulting from the ingestion of tritiated thymidine is greater by an order of magnitude than that resulting
from the ingestion of HTO. This conclusion may need to be revised when additional data on stem cells
and 3HTdR distribution and incorporation rates become available. It was noted that biokinetics of
tritiated nucleic acid precursors is strongly dependant on mammal species [B11, B12]. Without
appropriate human biokinetic data, the modelling uncertainty remains unspecified.

Table 9. Partitioning between HTO and OBT after dietary intake of OBT [I14]

Initial distribution (%) Biological half-time (days)

Age
HTO component OBT component HTO component OBT component
3 months 50 50 3.0 8
1 year 50 50 3.5 15

5 years 50 50 4.6 19
10 years 50 50 5.7 26
15 years 50 50 7.9 32

Adult 50 50 10.0 40
278 UNSCEAR 2016 REPORT

Table 10. ICRP effective dose coefficients based on OBT model for various modes of intake and age
groups [I17]

Dose coefficient (Sv/Bq)

Tritiated compound Mode of intake
Infants (1 year old) Adults
OBT Inhalation 1.1 1010 4.1 1011
OBT Ingestion 1.2 1010 4.2 1011

Table 11. ICRP effective dose coefficients for OBT ingestion for various age groups [I17]

Age Dose coefficient (Sv/Bq) Ratio of dose coefficient to that of an adult

3 months 1.2 1010
2.9

1 year 1.2 1010 2.9

5 years 7.3 1011 1.7

10 years 5.7 1011

1.4
15 years 4.2 1011
1.0
Adult 4.2 1011 1.0

3. Revised ICRP dosimetry models for HTO and OBT

139. ICRP has developed revised models with improved physiological realism focussing on
occupational intakes of radionuclides. Intakes by members of the public will be considered by ICRP at
a later stage.

140. The HTO systemic model includes compartments representing blood, extravascular body water
that exchanges rapidly with blood, and two components of retention of tritium converted in vivo to
OBT. The revised ICRP model structure is shown in figure X. The transfer coefficient from blood to
excreta is set to yield an initial removal half-time from the body of 10 days. The transfer coefficients
from compartments OBT-1 and OBT-2 back to extravascular HTO correspond to half-times of 40 days
and one year, respectively; the net retention half-times in these compartments are slightly longer than
40 days and one year due to recycling of activity. Excretion pathways from blood are not shown in
figure X but the following division is assumed on the basis of reference data for water balance ICRP
Publication 89 [I19]: urine, 55%; faeces, 4%; exhalation, 12%; and loss through skin (sweat plus
insensible loss), 29%.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 279

Figure X. Revised ICRP systemic model for HTO

Reproduced with the permission of ICRP

141. The revised model for systemic tritium applied to intake of OBT is a modification of the model for
OBT applied in ICRP Publication 56 and is shown in figure XI. In relation to occupational exposures, this
model is assumed to apply to biogenic organic compounds of tritium for which specific information is
not available. It is assumed that 50% of tritium initially entering blood transfers immediately to
compartment OBT-1 and 50% is converted immediately to HTO within the blood compartment. Tritium
entering OBT-1 or blood subsequently follows the HTO model defined in figure X.

142. The revised model for HTO predicts that OBT would represent about 56% of total body tritium
following chronic exposure to HTO. The OBT model with the default initial division of activity between
OBT-1 (50%) and blood (50%) predicts that OBT would represent about 6570% of total-body tritium in a
worker who is chronically exposed to a biogenic form of tritium. The adult dose coefficient calculated using
the revised model for HTO is 1.9 1011 Sv/Bq compared with the current value of 1.8 1011 Sv/Bq.
Values for intakes of biogenic organic compounds are given as 3.5 1011 Sv/Bq for inhalation (Type F) and
5.1 1011 Sv/Bq for ingestion (complete intestinal absorption, fA=1). The value for ingestion compares with
the current value for OBT of 4.2 1011 Sv/Bq.

143. The ICRP will also publish revised dose coefficients for inhaled tritiated methane and particulate
forms for which the HTO systemic model is applied. The revised value for inhalation of tritiated methane
by adults is 5.8 1014 Sv/Bq compared with the current value of 1.8 1013 Sv/Bq, reflecting the
assumption of 0.3% deposition and absorption to blood as HTO compared with 1% in the current model.
Revised values for inhaled particulate forms are 1.3 1011 Sv/Bq for Type F and 2.4 1011 Sv/Bq for
Type M, compared with current values of 6.2 1012 Sv/Bq and 4.5 1011 Sv/Bq, respectively. The
value for Type S materials is unchanged at 2.6 1010 Sv/Bq.
280 UNSCEAR 2016 REPORT

Figure XI. Revised ICRP systemic model for OBT

Reproduced with the permission of ICRP

4. Dosimetry of tritiated compounds using model for low solubility

particulate tritium
144. Dose coefficients for inhaled particulate tritium were introduced by the ICRP Publication 71 [I17]
in 1995 following publication of a number of scientific papers in which those forms of tritium were
identified and partially characterized [B8, B11, H12, I3, R15, T8, W6]. The papers described conditions
of occupational exposure in research and industry during contact of workers with neutron generators
and particle accelerators and with luminous powder.

145. Calculation of dose coefficients for particulate tritium was based on the generic ICRP model of
the respiratory tract [I16], categorizing tritium aerosols according to the absorption types specified in
the model, i.e. fast (F), medium (M) and slow (S). It was assumed that: (a) beta radiation of tritium
located in the alveolar part of the respiratory tract was fully absorbed in lung tissue, and (b) tritium
separated from the particles deposited in the respiratory tract, due to both fast and slow processes,
behaved as HTO. Therefore, the ICRP generic lung model was combined with the specific ICRP model
for HTO (see figure VIII). The clearance of tritium from the respiratory tract following deposition of
tritium particles included the escalation of particles through the bronchial tree to the alimentary tract
and dissolution and absorption to blood as HTO.

146. Data on the metabolism of inhaled particles of metal tritides are scarce [B8, B11, C17, C18, C19,
C20]. In the absence of specific information, ICRP has considered these particles as Type M [I17],
meaning that their rate of absorption from the respiratory tract to blood is moderate. Richardson and
Hong [R7] conducted dosimetric modelling of inhaled tritiated particles and reported that, while the
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 281

dose to the alveole can be up to two orders of magnitude higher than that from the same activity of
inhaled HTO, taking account of self-absorption of beta radiation in particulate material can reduce dose
by up to an order of magnitude, depending on particle size, in the range of 0.110 m.

147. Following inhalation of tritiated organic compounds such as luminous ones, uptake of HTO is
accompanied by uptake of low molecular compounds of tritium originating from degradation of
tritiated polymer. The contribution of the tritiated low molecular compounds to the effective dose is
generally low because of their rapid excretion in urine and faeces. However, these elevated
concentrations of organic compounds of tritium in excreta should be recognized when interpreting the
results of individual monitoring.

148. The available dose coefficients for particulate tritium are applicable mostly for occupational
radiation protection of workers in research and industry dealing with neutron generators, particle
accelerators, and air ionizers and the manufacture of luminous products. However, for the substantially
lower probability of exposure of members of the public, ICRP has provided committed effective dose
coefficients for different age groups (table 12).

Table 12. ICRP effective dose coefficients (Sv/Bq) for inhalation of particulate tritium of various types [I17]

Age Type F Type M Type S

3 months 2.6 1011
3.4 1010
1.2 109
1 year 2.0 1011 2.7 1010 1.0 109
5 years 1.1 1011 1.4 1010 6.3 1010

10 years 8.2 1012 8.2 1011 3.8 1010

15 years 5.9 1012 5.3 1011 2.8 1010
Adult 6.2 1012 4.5 1011 2.6 1010

149. Dose coefficients for Type F aerosols are lower than those for HTO despite the assumption of
rapid dissolution and absorption of deposited particles. This is because a proportion of inhaled particles
is subsequently exhaled while deposition of inhaled HTO vapour is assumed to be complete. The main
contribution to effective dose from inhalation of Type M and S aerosols is from lung dose. The dose
coefficients presented in table 12 can be applied for internal dose assessment in cases of inhalation of
various airborne tritium particles, e.g. metal and graphite tritides and iron hydroxide used in tritium
facilities, luminous powder and other particulate forms.

C. Intakes of tritium in relation to pregnancy and breast-feeding

1. Pregnancy and tritium intake

150. A few studies have investigated the transfer of tritiated compounds to the embryo and fetus and
their distribution and retention in fetal tissues. In prenatal dosimetry, the term embryo refers to the
developing human offspring up to the end of the eighth week of pregnancy, from the initial stages of
growth up to the end of organogenesis. At this time, the embryo weighs less than about 10 g.
282 UNSCEAR 2016 REPORT

151. Harrison et al. [H6] measured the transfer, distribution and retention of tritium in rats and guinea
pigs following the administration of HTO, tritiated glucose and tritiated food (i.e. liver and cress).
Transfer and retention of tritium in fetal tissue were similar for HTO and tritiated food. However, the
retention of tritium in both maternal and fetal tissue for tritiated glucose was lower. The ratio of the
concentrations of tritium in the fetus (CF) to that in the mother (CM) at the end of organogenesis in rats
ranged from 0.4 to 1.0. At the end of pregnancy, the ratio ranged from 1.3 to 1.5 in the case of rats and
from 1.1 to 1.3 in the case of guinea pigs.

152. Takeda et al. [T5] exposed female rats to single oral administrations of HTO, tritiated thymidine
or tritiated lysine. No significant differences were observed between the tritium concentrations in the
fetus and in maternal tissue. Tritium concentrations in the fetus were greater following the ingestion of
tritiated lysine than following the ingestion of HTO or tritiated thymidine. Tritiated lysine also resulted
in greater prenatal and neonatal absorbed doses than tritiated thymidine and HTO by factors of 1.5
(when tritium was administered on the thirteenth day of gestation) and 6 (when tritium was
administered on the first day of nursing).

153. Pietrzak-Flis et al. [P7] studied tritium incorporation in rats chronically exposed to tritiated food
(activity concentration, 48.1 kBq/g) or HTO (activity concentration, 37.0 kBq/mL) for three successive
generations (from three weeks before mating of the parents to the delivery of the F3 generation). The
analysis of tissue at various ages showed that the absorbed dose rates were higher in rats exposed to
HTO. However, the amount of tritium incorporated into the organic fraction of tissue was several times
higher after exposure to tritiated food.

154. Kowalska [K22, K23] reported further results from the same study showing that the amount of
tritium incorporated into the amino acids of rat-brain proteins and the main rat-brain phospholipids and
gangliosides was higher after the ingestion of tritiated food than after the ingestion of HTO. The highest
tritium concentrations in rat-brain proteins followed in utero exposure while the highest tritium
concentrations in phospholipids and gangliosides were found in 21-day-old rats exposed during
pregnancy and lactation.

155. The ICRP Publication 88 [I18] has provided biokinetic and dosimetric models and dose
coefficients for the embryo, fetus and newborn as a result of intake of radionuclides by the mother. The
term fetus refers to the developing human offspring after the eighth week of pregnancy [I18]. The
equivalent dose to the embryo is assumed to be the same as that to the uterus wall and proportional to
the concentration of HTO in maternal body water. For the calculation of the equivalent dose to the
fetus, the HTO concentration in fetal body water was assumed to be equal to that in the mother. The
ICRP uses a simple approach to the calculation of fetal doses for the majority of elements and their
radioisotopes, including tritium, considering data collected from studies of animals and humans [I18,
I21]. These fetal doses from tritium are calculated on the basis of relative concentrations of tritium
averaged over the whole body of the fetus (CF) and that of the mother (CM). The CF:CM ratios are taken
mainly from studies presenting data obtained at short times post-intake.

156. HTO rapidly crosses the placenta after it is inhaled or ingested by the mother. The resulting
equivalent dose to the fetus depends on the water content of the fetus over the course of its
development. As gestation progresses, the water content decreases relative to the amounts of protein,
fat and mineralsfrom about 95% of the total body mass at the sixth week of pregnancy to about 70%
at birth. For comparison, the percentage of total body water in a non-pregnant woman is about 50%.
The total water content of the mothers body is known to increase during pregnancy. This is mainly due
to the increase of about 50% in the blood volume in order to carry additional nutrients and other
substances needed for the fetus. Calculations have shown that the biological half-time of HTO in the
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 283

mother varies from about 10 days at the start of pregnancy to about 12 days at term. This variation does
not significantly affect the ICRP fetal dose coefficients for HTO [I18].

157. On the basis of an average body water content of 80% for the fetus and 50% for the mother, ICRP
used a CF:CM ratio of 1.6 in the derivation of the dose coefficients for HTO. This ratio is applied to both
the HTO and OBT components for intake of OBT and is assumed to remain constant throughout
pregnancy. Tritium is assumed to be uniformly distributed throughout all tissues of the fetus. Harrison
et al. [H7] showed that the CF:CM ratio ranged from about 1.4 to 1.8, on the basis of data for the relative
water content of the mother to that of the fetus.

158. Following birth, the ICRP biokinetic model for the three-month-old infant [I18] is used to
calculate the committed effective dose per unit intake by the newborn child. Table 13 presents the
committed effective dose coefficients for the unborn and the three-month-old child. The ICRP [I18] has
published a range of dose coefficients for in utero exposure of the child following maternal intake by
inhalation or ingestion, before or during pregnancy, considering acute and chronic exposure. Account is
taken of the retention of activity in body tissue at birth and dose delivered post-natally. The purpose of
the ICRP fetal dose coefficients is to allow comparison with doses to other age groups in order to
ensure that protection does not neglect doses received in utero [C29, I22].

Table 13. ICRP prenatal and infant effective dose coefficients [I18]

Tritiated compound Prenatal dose coefficient (Sv/Bq) Dose coefficients for

and mode of intake Acute maternal intake a
Prolonged maternal intake b 3-month-old child (Sv/Bq)

HTO inhalation 3.6 10 11

3.1 1011 6.4 1011
HTO ingestion 3.6 1011 3.1 1011 6.4 1011

OBT ingestion 7.6 1011 6.3 1011 1.2 1010

a
Values are for acute intake by the mother at the end of the tenth week of pregnancy. Acute intake occurring at other times yields
lower dose coefficients.
b
Prolonged intake by the mother, beginning at the start of pregnancy and continuing for the duration of the pregnancy.

2. Intake of tritium from maternal milk

159. The ICRP [I21] has developed dose coefficients for newborns for intake of tritium from maternal
milk. The approach involved the estimation of the activity transferred to milk as a function of maternal
intake for various intake scenarios (acute or prolonged intake regimes, and for various maternal intake
times relative to birth). Nursing was assumed to continue for six months after birth, and the ingestion
dose coefficients for infants were applied [I14]. Maternal intake (by ingestion and by inhalation) during
pregnancy and during lactation was considered.

160. The HTO and OBT models were modified by ICRP [I21] to account for the transfer of tritium to
milk fed to infants (up to one year). The rate of OBT transferred to milk was taken to be that of carbon.
The ICRP developed dose coefficients on the basis of the assumption of six feeds per day and an
average daily milk intake by the infant of 0.8 L. Table 14 shows selected committed effective dose
coefficients for nursing infants resulting from maternal intake of tritium.
284 UNSCEAR 2016 REPORT

Table 14. ICRP effective dose coefficients for nursing infant [I21]

Dose coefficient (Sv/Bq)

Tritiated compound and mode of intake
Acute maternal intakea Prolonged maternal intakeb
HTO inhalation 2.2 1011 2.0 1011
HTO ingestion 2.2 1011 2.0 1011

OBT ingestion 3.5 1011 3.0 1011

a
Values are based on acute maternal intake at one week after birth. Acute intake occurring at other times yields lower dose coefficients.
b
Values are based on prolonged intake during the lactation period (up to six months after birth). These dose coefficients are
greater than those for prolonged intake during pregnancy.

D. Uncertainties in dose coefficients for tritium

161. Dose coefficients applied for assessment of internal dose for purposes of human radiation
protection are defined by ICRP as regulatory parameters without any uncertainty [I23]. They are
presented in ICRP publications for reference persons [I7, I19] as values depending on the radionuclide,
its physical and chemical form, exposure pathway (inhalation or ingestion) and a persons age [I12, I14,
I15, I17, I18, I21]. Uncertainties of ICRP dose coefficients for some radionuclides, including tritium as
HTO, are systematically considered in a specific NCRP publication [N4] and by Leggett et al. [L8].
The need for individual values of metabolic parameters arises mostly in cases of emergency intake of
large radionuclide activities when probabilities of radiation-induced acute health effects and risk of
stochastic effects require consideration on an individual basis in the context of possible decorporation
or other medical treatment.

162. For the most common form of tritium, HTO, dose coefficients for adults are determined on the
basis of numerous human observations and are, therefore, associated with low uncertainty. Greater
uncertainties apply to dose coefficients for children exposed to HTO from the environment. Leggett et
al. [L8] considered the sources, quality and completeness of data underlying the biokinetic models for
tritium (as HTO) and concluded that the dose coefficient for HTO is known within a factor of two for
an adult and between a factor two and three for a five-year-old child.

163. Exchange model parameters for HT inhalation are also derived from human observations, which
can generate moderate uncertainty of corresponding dose coefficients. More uncertain are dose
coefficients for OBT based on animal experiments and modelling and for tritium aerosols of various
types based on both in vitro and in vivo animal experiments [P4, P9].

164. Harrison et al. [H7] reviewed the assumptions used by ICRP in the derivation of dose coefficients
for tritium along with their associated uncertainties. The assumptions considered were those related to
absorption to blood, the biological half-time of HTO and OBT in adults and children, the transfer to the
fetus, the heterogeneity of the distribution in tissues and cells, and the RBE value of the beta particles
emitted by tritium.

165. Harrison et al. [H7] estimated uncertainties using ranges on the central values for the
incorporation of tritium into OBT in body tissue of 0.010.1 and 0.150.75 of the tritium activity
reaching blood after intake as HTO and OBT, respectively. Biological half-times in adults were taken
to vary from about 5 to 20 days for HTO and from about 20 to 200 days for OBT, these ranges being
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 285

the 2.5 and 97.5 percentiles of the log-normal distributions. They also considered CF:CM ratios ranging
from 1.4 to 1.8 based on the relative content of water in the mother and the fetus. The 2.5 and 97.5
percentiles of the log-normal distribution for the OBT transfer and distribution in fetal tissue were taken
to be 1.2 and 2. A range of 12.5 was used for the RBE value of tritium beta particles compared to
gamma rays. This analysis gave median (50%) values for the dose coefficient of 2.3 1011 Sv/Bq for
HTO and 5.6 1011 Sv/Bq for OBT for ingestion or inhalation by adults, excluding consideration of
RBE value. Table 15 shows probability distributions of the committed effective dose coefficients for
adults and for the fetus after ingestion by the mother during pregnancy, including the range of RBE
values. Dose coefficients for other age groups were estimated to vary by a factor of two to three for
HTO and OBT.

Table 15. Probability distributions of effective dose coefficients from ingestion of HTO or OBT by
adults and for fetus after ingestion by mother during pregnancy [H7]
Intakes during pregnancy assumed to take place at 10 weeks after conception

Distributions of effective dose coefficients (1011 Sv/Bq)

Age Form
5 percentile 50 percentile 95 percentile
Adult HTO 2.1 3.9 6.6

OBT 3.9 8.7 20

Fetus HTO 3.7 7.6 14

OBT 6.9 17 40

166. Hamby [H2], using Monte Carlo sampling, calculated that the dose coefficient for intake of HTO
by adults varied by a factor of 15 from its highest to lowest modelled values, with a median value of
2.2 1011 Sv/Bq and a geometric standard deviation of 1.6. This range was most sensitive to the
biological half-time, the linear energy transfer (LET), and the reference radiation (i.e. whether X- or
gamma radiation). When the quality factor was set to unity, the geometric mean of the dose coefficient
was 1.3 1011 Sv/Bq with a geometric standard deviation of 1.4.

167. Melintescu et al. [M10] also assessed the effect of uncertainty on the RBE value of the
tritium beta radiation, the retention of HTO and OBT and the presence of tritium in the DNA
hydration shell. For the adult male, setting the radiation weighting factor to unity (based on the
RBE value) resulted in dose coefficients ranging from 1 to 2.9 10 11 Sv/Bq for the intake of HTO
and from 5 to 6.7 10 11 Sv/Bq for the ingestion of OBT. Accounting for the variability in the
radiation weighting factor, the resulting dose coefficient range was from 1.5 to 5.1 10 11 Sv/Bq
and 5 to 11.3 10 11 Sv/Bq for HTO and OBT, respectively. Similarly, the dose coefficient for a
one-year-old was found to vary from 2.9 to 6.6 10 11 Sv/Bq and 14 to 18 10 11 Sv/Bq for HTO
and OBT, respectively, when the radiation weighting factor was set to unity.

E. Summary of biokinetic and dosimetic models

168. HTO behaviour following intake into the mammalian body is well understood with regard to both
early distribution in the body followed by excretion as part of water exchange (half-time of adult
humans 418 days) and simultaneous conversion of a small fraction (of the order of 1%) of tritium into
OBT with subsequent longer term excretion (half-time of adult humans 23104 days). The average
contribution of OBT to effective dose of adults derived from 17 human observations was 3.0%. The
286 UNSCEAR 2016 REPORT

data on longer term processes with half-times of more than 100 days are highly uncertain and need
further study. The ICRP biokinetic model for HTO for all age groups is moderately conservative, with
an OBT contribution to dose equal to 9% for adults. The biokinetics of HT following inhalation is also
well understood. However, the current ICRP model considers only HTO formed in vivo and does not
take account of an additional component of dose to cells of the alveolar-interstitial region of the
respiratory tract.

169. The ingestion of food containing OBT by members of the public results in the degradation of a
substantial fraction of the OBT to HTO in the gastrointestinal tract and in body tissue by catabolic
reactions, and uptake and conversion of the rest of the OBT into tissue macromolecules. Both the
fractions and subsequent excretion rates of OBT from tissue depend on food origin and composition.
The ICRP model for OBT ingestion by humans is generally consistent with experimental results.
However, it does not consider the non-uniform deposition of OBT between various organs and tissues.
Physiologically-based OBT models, such as the ones proposed by Richardson and Dunford [R8] and
Galeriu and Melintescu [G1], consider non-uniform deposition. Dose coefficients calculated using these
models are generally similar to ICRP values.

170. Tritiated biochemical substrates, such as glucose, amino acids, hormones, DNA and RNA
precursors, may be directly incorporated into organic molecules in body tissues if absorbed to blood
and transported to sites of active metabolism within cells. The extent of incorporation of tritium into
specific forms of OBT is determined by such factors as the chemical compound containing tritium,
its isomeric form, position of the label in the molecule, and the amount of carrier. OBT formed from
tritiated precursors of biological macromolecules is retained in tissue longer than HTO. Catabolism
of labelled compounds will result in tritium being partially oxidized and entering the body water as
HTO or catabolized and excreted as low molecular weight organic substances. Following intake of
tritiated precursors of biological macromolecules, the internal dose to mammal tissue is usually up to
ten times larger than the dose from intake of an equal amount of HTO, and the contribution of OBT
to dose may dominate.

171. Labelled DNA precursors (e.g. 3H-thymidine, 3H-desoxycitidine) entering the mammalian body
by various routes are partially degraded to HTO and partially incorporated into the DNA of dividing
cells, and thereafter selectively expose nuclei of proliferating cells to beta radiation. 3H-
deoxynucleosides are preferentially incorporated into the proliferating cell systems of embryo and
fetus, bone marrow and small intestine at any age and, to a substantially lesser extent, in tissue with
lower frequencies of mitoses (e.g. muscle, liver). For both acute and prolonged intake of tritiated DNA
precursors, the absorbed dose to nuclei of proliferating cells may be larger by one to two orders of
magnitude than the dose from intake of equal amounts of HTO. As the average range of tritium beta
radiation is considerably less than the dimensions of the nuclei of mammal cells, the use of average
organ or tissue dose in the case of incorporation of 3H-nucleosides requires careful consideration. The
ICRP dose coefficients for OBT should not be directly applied to intake of tritiated DNA precursors.

172. Tritium particles of low solubility (e.g. metal tritides, luminous powder) inhaled by workers in
occupational conditions partially deposit in the respiratory tract and may be retained in the lungs for
long periods. Material-specific labile fractions of tritium segregate from particles as HTO or as organic
molecules that are absorbed to blood and excreted. Effective doses from inhalation of low soluble
tritium particles can be assessed by means of the ICRP human respiratory tract model combined with
the HTO model. Tritium as HTO or biochemical substrates is easily transported through the placenta to
the embryo and fetus and secreted in maternal milk. In conditions of chronic tritium intake by the
mother, committed internal doses incurred by the embryo and fetus and by the suckling infant correlate
with the dose of the mother and are not substantially different in magnitude.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 287

VI. BIOLOGICAL AND HEALTH EFFECTS

A. Non-radiological effects of tritium in biological systems

1. Transmutation
173. Transmutation is the conversion of one element into another through radioactive decay. When
tritium undergoes decay, it becomes helium-3 (3He), a stable, inert gas. Helium is chemically very
different from hydrogen and, therefore, this could make a significant contribution to the effect of
tritium when organically combined. If a tritium atom is bound to a DNA molecule when it decays, most
of the kinetic energy will accompany the beta radiation as it is ejected from the nucleus, but some
energy will provide a kick-back to the 3He atom as recoil energy. Kacena [K1] determined that the
recoil energy was too small (up to 3 eV) to cause ionization of the DNA molecule on its own. However,
the resultant 3He atom would break free from the DNA molecule. In complex molecules, the effect of
this conversion to a positively charged carbonium ion would be difficult to distinguish because of the
proximity to the deposition of energy and associated events from the beta emission.

174. Myers and Johnson [M26] and Gracheva and Korolev [G8] performed comprehensive reviews of
transmutation effects. They noted that the degree of damage caused by transmutation of tritium into
helium could theoretically vary significantly, depending on the position of the tritium atom in specific
DNA nucleotides. The studies covered several test systems in the S13 virus, in two strains of the
bacterium E. coli, in Drosophila Melanogaster (fruit fly) and in cultured mammalian cells. On the basis
of the position of the tritium in the nucleic acid, varying degrees of damage were observed in
experiments. The most pronounced mutagenic effect was detected with cytosine-5-3H, which in the case
of tritium decay converted to uracil. In some simple biological systems (virus, E. coli), that
transmutation resulted in an elevated mutation rate that was 3400 times larger than the mutation rate
caused by beta radiation only [G8]. The reviewers argued, however, that the increase in the mutation
rate in mammals (resulting from transmutation) would not likely exceed 5% of the normal rate and that
this was too small to be detected.

175. Carsten [C11] discussed the possibility that such effects would be manifest in humans after
ingesting HTO or OBT as food. He suggested that the risk was small enough to pose no significant
hazard, primarily because only 2% of the hydrogen atoms in DNA were located at the 5-position of the
cytosine ring and damage would be minimal. Feinendegen and Bond [F3] reached the same
conclusionthat the effects of intracellular tritium are overwhelmingly due to beta irradiation of the
nucleus and transmutation effects do not produce a measurably increased effect under most
conditions. The United Kingdom Advisory Group on Ionising Radiation (AGIR) [H16] came to a
similar conclusion. If DNA damage in mammals did occur from transmutation, it is unlikely that it
could be distinguished from radiation-induced damage and thus would be already accounted for in
measured RBE values.

2. Isotopic effects
176. While chemically similarly to hydrogen, tritium has slightly different physical properties due to
its increased mass. Diabat and Strack [D5] noted that synthetic reaction rates decreased as atomic mass
288 UNSCEAR 2016 REPORT

increased, causing a significant isotopic effect of OBT depletion with tritium compared with HTO from
which biological macromolecules are synthesized. In contrast, once fixed to carbon, C-T bonds are cleft
more slowly than C-H bonds. In equilibrium conditions of OBT synthesis in plants grown in a medium
with HTO, the isotopic ratio of tritium specific activity in hydrogen of OBT in bulk organic matter to
that in water was in the range of 0.60.8 [D5].

177. Intracellular discrimination between tritium and hydrogen atoms due to isotopic differences (in
particular a mass ratio of 3) have been considered in the past (see e.g. [C11, M2]) and concluded to be
of little consequence for the risk of tritium. However, contrary to the traditional view of OBT in
biomatter, experiments that used denaturing agents have suggested that a proportion of tritium may be
designated as buried tritium, a tightly bound HTO [B13]. In such a fraction, the buried tritium in
biomacromolecules, such as proteins, is in positions where the exchange rates are substantially reduced
as a consequence of the three-dimensional structure that arises upon the folding of these
biomacromolecules. The hydrogen bridges between the molecules of water are stronger than between
organic configurations, resulting in accumulation of tritons both inside the biopolymers and within their
primary hydration shields. There is an enrichment of tritium in the newly identified buried hydrogen
bonds compared to the free water in the cell. In most biomolecules, the enrichment may be 1.4-fold but
in DNA, where the hydration shell consists of 11 molecules per nucleotide and is not readily permeable
to ions, the enrichment in the water trapped in the core may be twofold. While this will certainly result
in slightly more beta tracks originating from HTO within and around the DNA, it remains true that the
vast majority of beta tracks encountered by the DNA will have originated from HTO outside the DNA
since that is where most of the HTO is situated. The effect on radiation dose to the DNA will therefore
be small but may increase the RBE value in experimental determinations.

B. Deterministic effects

1. Lethality
178. Brues et al. [B26] investigated the lethality of tritium by giving mice single injections of
0.126-8.4 GBq of HTO. The dose that killed 50% of the population within 30 days (LD50/30) was about
9 Gy, which corresponds to an initial activity concentration of tritium of 37 MBq/g of body weight
(BW). Furchner [F13] reported an LD50/30 of 8 Gy, for HTO given in a single intraperitoneal injection
of 33 MBq/g to CF1 female mice. Yamamoto et al. [Y2] (see also [Y6]) reported an LD50/30 of about
8 Gy after a single injection of 0.56 GBq in C57BL/6N female mice and about 13 Gy with an injection
of 0.93 GBq in female (C57BL/6N C3H/He) F1 mice. Overall, these data suggest a LD50/30 dose of
the order of 10 Gy.

179. Yamamoto et al. [Y2] also investigated the lethality of a continuous oral administration of HTO
as drinking water in (C57BL/6N and C3H/He) F1 female mice. The tritium concentration reached a
plateau in organs and blood after about seven days. Haematopoietic death typically occurred after two
weeks following ingestion of HTO with activity concentrations from 0.15 to 0.6 TBq/L in drinking
water. The lowest absorbed dose to cause death was estimated to be about 11 Gy from the continuous
ingestion of 0.15 TBq/L of HTO.

180. The dose from tritium necessary to cause death in mice appears to be similar to that from acute
external irradiation by X-rays or gamma rays, with tritium LD50/30 values of about 69 Gy,
corresponding to an acute intake of the order of 1 GBq of HTO. In the literature, there are some
published historical cases of radiation sickness of workers caused by tritium [M15, O3, S12], including
two lethal cases.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 289

181. In 1963, a worker in a radiological laboratory in the former Soviet Union had an intake of about
350 GBq of HTO [O3] due to violation of safety rules. Monitoring of tritium in urine started 25 days
later, when the worker had visited a doctor because of deteriorating health. Tritium concentration in
urine declined with a half-time of seven days and the committed equivalent dose in soft tissues was
assessed to be about 12 Sv. Treatment started immediately after visiting the doctor. Pronounced
radiation sickness symptoms were observed during 1.52 months after the accident. After about
1.5 months, regeneration of the haemopoietic system began but substantial recovery of leukocyte and
neutrophil concentration in blood took more than three months. The patient returned to work after
six months. Medical surveillance over 2.5 years showed recovery of working capacity and the absence
of substantial adverse effects in internal organs.

182. In the late 1950s and early 1960s, several workers at two European facilities for production of
tritiated luminous compounds used large quantities of tritium (410 TBq per annum) [M15, S12]. In the
course of chemical operations, each worker incorporated substantial amounts of tritium. Measurements
of tritium in urine started in the 1960s and were used for dose assessment. In case A-1, tritium
concentrations in urine varied in the range of 240 MBq/L. From these data, the equivalent dose in soft
tissue was assessed as 35 Sv over the four years preceding death [S12]. An assistant to A-1 (A-2)
received a dose that was about half of that received by A1 and stopped working with tritium after she
showed moderate anaemia. Workers A-3 and A-4, who succeeded A-1 and A-2, worked under
improved conditions and incurred lower doses without health effects. In case B-2, the dose incurred
over three years before death was assessed from tritium measurements in urine in the range from a few
sieverts up to 20 Sv. In both cases A-1 and B-2 there was substantial contribution to dose from OBT, as
indicated by measurements made on autopsy samples, and both workers died with clinical signs of an
aplastic pan-myelocytopenia complicated by pulmonary or other symptoms [S12].

2. Effects of HTO on animal embryo and fetus

183. Straume and Carsten [S23] reviewed the current literature on exposure to tritium during fetal
development. During particular periods of development, some fetal cells will be dividing rapidly and
differentiating to form tissues and organs, while other cell types may be showing very little or no cell
proliferation. Tritium that is incorporated into OBT of low proliferation cells could result in larger
integrated doses to these cells, since the tritium would not be diluted by further cell proliferation.
However, uptake will be dominated by those cells undergoing rapid division. Commerford et al. [C26]
reported that the dose from tritium incorporated from HTO into macromolecules, such as DNA and
histones, of rapidly dividing cells was small compared with that from the same activity of tritium in the
form of HTO.

184. In a study by Laskey et al. [L7], rats were continuously fed HTO with activity concentrations of
0.37370 kBq/mL from conception of the first generation until delivery of the second. The
corresponding dose rates were 0.0330 mGy/d. Exposure to HTO with an activity concentration of
370 kBq/mL resulted in a 30% weight reduction of the testes in the first generation of adult males, but
there was no impairment of overall growth or reproductive ability. In the second generation of
newborns, decreases in the weight of the brain, overall body weight and litter size were found. An
increase in resorption for rats exposed to HTO with an activity concentration at 370 kBq/mL was also
found. However, Laskey et al. [L7] did not observe any effects on litter size or resorption at or below an
activity concentration of 37 kBq/mL (which corresponds to a dose rate of about 3 mGy/d).

185. Bursian et al. [B27] assessed the effects of continuous exposure of rats to HTO from conception
to birth. The activity concentrations used were 0, 37, 370 and 3,700 kBq/mL. In utero exposure to doses
290 UNSCEAR 2016 REPORT

as low as 0.66 Gy (corresponding to the highest concentration used) produced measurable and
persistent decreases in brain weight and increases in norepinephrine concentrations at 21 and 45 days
after birth. No differences from the controls were observed in the rate of turnover or the concentrations
of dopamine, acetylcholinesterase or monoamine oxidase.

186. Jones et al. [J8] gave pregnant squirrel monkeys water with tritium levels ranging upwards from
2 kBq/mL throughout gestation. No effects on body weight, body dimensions, organ weights,
haematological patterns, or the histology of organs or tissues, with the exception of ovaries, were
observed in newborn progeny. However, the number of primary oocytes in female progeny decreased
markedly with increasing levels of HTO in maternal drinking water.

187. Yamada et al. [Y1] studied the effect of prolonged in vitro exposure to HTO and 60Co gamma
radiation on pre-implantation mouse development after mating female C57BL/C3H F1 and male ICR
mice. With the development to blastocyst as the end point, the LD50 was 4.4, 8.5 and 15.8 MBq/mL
corresponding to beta radiation dose rate of about 10, 20 and 40 mGy/h, respectively, for pronuclear,
early two-cell, and late two-cell embryos, respectively. Compared to 60Co gamma radiation, the RBE
value of tritium beta radiation was in the range of 1.01.7.

188. The effects of tritium on the morphogenesis and development of rats and mice have been studied by
several authors. Wang and Zhou [W3] reported modifications in the cognitive function of young rats born
to mothers that had been injected with HTO on the thirteenth day of pregnancy that resulted in 0.1 and
0.3 Gy of in utero exposure. Gao et al. [G2] showed decreases in cognitive behaviour and a significant
decrease in hippocampal pyramidal cells in the brains CA1 area with intraperitoneal injections of HTO
on the thirteenth day of gestation that resulted in 0.09 and 0.27 Gy of in utero exposure. Sun et al. [S25]
reported a decrease in the brain weight of mice exposed on the thirteenth day of gestation to a dose of
0.4 Gy (from the injection of HTO of 964 kBq/g of body weight); both the thickness of the somatosensory
cortex and pyramidal cell density were significantly decreased at this dose. Jain and Bhatia [J2] also
observed pathological changes in the cerebellum of mice following exposure of the mothers to HTO.
An initial injection of HTO was given at 17 days post-conception followed by additional intake of HTO
of 111 and 11.1 kBq/mL. The observed damage was dose dependent.

189. Zamenhof and van Marthens [Z3] studied how five generations of rats were affected by pre- and
post-natal exposure to HTO. Female rats were given water containing HTO with an activity
concentration of 111 kBq/mL beginning in adolescence and continuing throughout pregnancy. This
exposure to tritium did not produce any signs of radiation-induced disease in the mothers. The courses
and outcomes of pregnancy were also normal, but 60% of the newborn rats exhibited haematomas,
oedemas and subdural haemorrhages. None of these effects lasted beyond 30 days. Zamenhof and van
Marthens [Z4] subsequently found that there were decreases in the weight and DNA and protein
content of the brain in later generations of rats continuously drinking water containing HTO with an
activity concentration of 111 kBq/mL (estimated average daily intake for a 60 g ratthe weight of a
30-day-oldwas 2.1 MBq, corresponding to an absorbed dose rate of about 1.4 to 8 mGy/d). The
effects were generally most evident in all generations (F2F5) except the F1 generation. In addition, all
but the F1 generation showed some recovery over time.

190. In summary, radiation-induced effects on the embryo or fetus (principally neurological or

reproductive tissue impairment) have been demonstrated in laboratory animals exposed to tritium. Such
effects start occurring at chronic intake of activity concentrations of HTO of about 50100 kBq/mL of
body water and are consistent with similar effects from external photon irradiation. Embryofetal
effects in animal studies are observed at doses of about 0.40.6 Gy from chronic intake of HTO.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 291

3. Immunological effects of HTO

191. The Committee has conducted a number of reviews of the effects of exposure to ionizing radiation
on the immune system, the most recent being given in its UNSCEAR 2006 Report, annex D [U12]. The
major conclusions from that review were the following:

High doses of radiation produce immunosuppression mainly through destruction of cells.

Lymphocytes are very radiosensitive, and their destruction is currently used as an early
indicator of the level of an accidental acute exposure. Radiation-induced changes in immune
parameters seem to be more dependent on total dose than on dose rate. Persisting effects on
immune system have been observed after exposure to ionizing radiation.

At low doses and dose rates, the effects of ionizing radiation on the immune system may be
suppressive or stimulatory. The long-term effect of low radiation doses on the immune
function in relation to human health needs to be further evaluated.

192. However, very few studies specifically reference exposure to tritium. Some reports have come
from studies of workers exposed occupationally and also from experiments carried out in the in vitro or
in vivo systems.

193. Tuschl et al. [T17] investigated some immunological parameters in ten NPP workers exposed
during a four-week period to external radiation (total effective doses ranged from 1.4 to 9.8 mSv) and
tritium inhalation (committed effective doses ranged from 1.2 to 2.8 mSv). Twenty-five days after the
beginning of the exposure, only the CD4+/CD8+ T cell ratios were markedly elevated compared to the
ratios detected in non-exposed controls and the effect was mainly due to an increase in the absolute
numbers of CD4+ T helper cells. In five exposed subjects who agreed to give blood samples five
months after the first sampling, the CD4+/CD8+ T cell ratios were still elevated.

194. Milaci [M14] analysed blood samples from 53 workers exposed to tritium. The workers were
separated into three groups depending on the duration of occupational exposure: 05, 615 and
16-30 years. The groups were compared with each other and with a control group. Effective doses were
not provided, but the average tritium activity concentration in urine of the exposed subjects was
1.9 kBq/L. Although total leukocyte counts did not differ from the control group, lymphocyte and
eosinophil counts were higher in the workers exposed to tritium and varied with the duration of the
exposure, being the lowest in the 615 years group. Chromosomal aberrations were detected in 49% of
workers exposed for 10.5 years and with a significantly higher average tritium concentration in urine
(3.5 kBq/L). However, the average tritium activity concentration in the exposed subjects with no
aberrations was 0.35 kBq/L. Alkaline phosphotase and myeloperoxidase activities in granulocytes were
significantly lower in all exposed workers. The author interpreted the increase in lymphocyte counts as
a stimulation of the immune system by tritium and in eosinophil counts as a compensatory reaction of
the bone marrow in response to impaired function of granulocytes. The workers had no clinical
manifestations of immunity disorders.

195. In 1980, Kirillova and Luzanov [K16] compared prolonged exposure of CBA mice to HTO and
external gamma radiation (from 137Cs) delivered in equal daily radiation doses (cumulative dose
4.1 Gy). They observed a decrease in the immune response. The antibody and rosette-forming capacity
of spleen cells was lower in the group exposed to HTO than in the group exposed to gamma radiation.
On the basis of the total immune response observed over the entire treatment period, they concluded
that HTO was 1.27 times more efficient in decreasing the immune response than gamma radiation.
They suggested that the observed diminution of the immune response was due mainly to damage to the
lymphoid tissue and also to disruption of haematopoiesis. Later on, Kirillova [K19] conducted a similar
292 UNSCEAR 2016 REPORT

experiment, but used rats instead of mice. After a prolonged intake of HTO, the decrease in normal
killer cells (NK) 2 activity in rats was less and the recovery was more rapid than observed in mice for
the same total dose (about 8 Gy). The authors linked this observation to the higher metabolic rate in
mice compared to rats.

196. In another study of changes in the concentration of NK cells as an end point, Kirillova [K17]
treated mice to HTO at an activity concentration of 0.37 MBq/g of body mass over six months and
compared them with a non-treated group of mice. The exposure to HTO resulted in a decrease in the
activity of NK cells by 3545% in the treated mice when compared with the reference group (p <0.05)
at three months after administration of HTO had ceased. Six months later, the NK cell function had
recovered and was even higher by a factor of 1.8 in the mice that had been treated with HTO than in the
control group.

197. Kirillova et al. [K18] also studied the quantitative and qualitative degree of recovery of the
immunological system in mice after prolonged exposure (throughout life; around 725 days for the
controls and 650 days for the treated mice) to HTO with an activity concentration of 0.37 MBq/g of
body weight giving a total dose of 8.7 Gy. The authors observed a depopulation of cells in bone
marrow, spleen and thymus, which persisted until the end of the lives of the animals. They suggested
that the disruption in immunological response was caused early in the treatment by a reduction in the
numbers of lymphocytes and by a lower activity of B cells and tritium helpers and, later after radiation
exposure, by the impairment of tritium helpers function. There was a direct relationship between the
immunodeficiency and the dose rate and the total absorbed dose of beta radiation.

198. Smirnov et al. [S14] concluded that prolonged exposure of CBA mice to daily intake of HTO of
0.19, 0.37 and 0.74 MBq/g of body weight up to doses ranging from 0.2 to 1 Gy, resulted in an
impairment of humoral immunity at various stages of the immunopoesis at all dose levels. They also
showed a direct relationship between the depopulation of colony forming units (CFU) (early precursors
of T and B lymphocytes) and dose rate. They concluded that the production of antibody forming cells
was a function of the absorbed dose and that a prolonged exposure to HTO resulted in impaired
humoral immunity at various stages of immunopoesis. They also demonstrated that the mechanisms
generating the immunological response were highly sensitive to the tritium beta radiation.

199. As demonstrated more recently by Umata et al. [U4], a single intraperitoneal injection of HTO to
C57BL/6N mice to give a total whole-body dose of 3 Gy significantly increased the number of variants
of the T-cell receptor expressed on splenocytes. The frequency of apoptotic cells of the spleen 12 hours
after HTO injection increased to 5.0%.

200. In summary, experimental studies carried out in animals indicate that prolonged exposure of
animals to HTO associated with relatively high radiation doses (in the range of 18.7 Gy) resulted in
some diminution in the immune response in animals which, depending on the dose, could be reversed.
In contrast, the few studies carried out in workers occupationally exposed to low doses of tritium
demonstrate that such exposure can stimulate immune functions, but also increase the frequency of
chromosome aberrations in lymphocytes. However, it is not clear whether such effects, even at high
doses, had any significant health consequences. More research on the effects of acute and protracted
low-level exposure to tritium on the immune system is warranted.

2
A cytotoxic lymphocyte.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 293

4. Germ cell effects of HTO

(a) Effects in females

201. Because tritium distributes itself throughout the body, it can be taken up by developing oocytes
and incorporated into DNA. Tritium incorporated into human oocyte DNA could theoretically irradiate
the oocytes over 30 or more years. Because oocytes do not divide until fertilized, there is little turnover
of the DNA molecules, which implies that the biological half-time of the tritium embedded in oocyte
DNA could approach the radioactive half-life of tritium of 12.3 years. Forell et al. [F8] pointed out that
if a biological process were to gradually exchange all the components of DNA moleculesincluding
that of tritium-labelled DNAit would take 50 years to replace 25% of a cells genome (DNA). The
author estimated that in resting lymphocytes 2,000 bases per hour were turned over. Therefore, most of
the tritium incorporated into an oocyte will remain there for its whole life. This issue was also discussed
by the United Kingdom Health Protection Agency [H16] in relation to the discharges from a
radiopharmaceutical plant. They concluded that tritium could be incorporated during pregnancy into the
DNA of fetal oocytes and remain there until fertilization decades later.

202. Straume and Carsten [S23], in reviewing the effects of radiation exposure on oocytes, reported
that most of the information on radiosensitivity in humans came from autopsies of women who had
been exposed to substantial doses of external radiation [L13], and from fertility histories of women who
had undergone radiotherapy or had been exposed as a consequence of the atomic bombings in Japan
[B5, L14, U5]. In all cases, the data were for short-term external exposure of adult women.

203. In female mice, data suggest that premature oocytes are more radiosensitive than mature oocytes
[B3]. This is in contrast to results in human females that demonstrate that from 3 days after birth, all
oocytes are equally radiosensitive. In women, exposure of 2.56 Gy to X- or gamma radiation will lead
to permanent sterility [I11, I20]. There is no temporary sterility in human females as there is in mice.

204. Dobson and Kwan [D6, D7] continuously exposed non-inbred Swiss-Webster mice to HTO and
60
Co gamma rays during the early period of oocyte development from conception until 14 days after
birth. Oocyte survival following tritium exposure decreased exponentially with dose rate with no
threshold; the LD50 level was 0.074 MBq/mL of body water corresponding to 4.4 mGy/day. Exposure to
60
Co gamma rays was shown to be less effective at the same dose and did not follow an exponential
relationship. As a result, the RBE value of tritium beta radiation, compared with gamma radiation,
increased with decreasing dose and dose rate from about 1.6 at gamma-ray dose of 0.4 Gy to about 3 at
lower doses. In studies of both mice and rats, Satow et al. [S4, S5] studied the effectiveness of exposure
to tritium in killing immature oocytes. Statistically significant oocyte resorption was found at activity
concentrations of HTO of 0.34 MBq/g of body weight and more, corresponding to a total dose of
77 mGy and more.

205. A number of animal studies have investigated the continuous administration of HTO or tritiated
thymidine throughout pregnancy and analysis of the subsequent effect on oocytes in the offspring. This
method of continuous administration in rats was pioneered by Fliedner et al. [F7] and Schreml et al.
[S7] and subsequently adapted by Lambert and Phipps [L6] using mice. Using this method, Haas et al.
[H1] studied the effect of HTO infusion in utero on post-natal oocyte development in rats. During the
first 21 days of life post-conception, a dose dependent reduction in oocyte numbers was observed. At
birth, 54 MBq infused had reduced the numbers by 50% whereas 215 MBq produced total aplasia of
oocytes. They suggested that tritiated thymidine was about ten times more effective than HTO for this
effect. In a subsequent paper [S8], these authors concluded that this factor was about 3.7 in relation to
radiation dose to the oocyte cell nucleus.
294 UNSCEAR 2016 REPORT

206. Lambert and Phipps [L6] exposed pregnant SAS/4 mice to HTO in drinking water and, by
constant infusion, to tritiated thymidine throughout pregnancy. A number of parameters were studied in
the offspring including oocyte survival at 14 days of life. They concluded that for this parameter
tritiated thymidine was about twothree times as effective as HTO in causing oocyte lethality.

207. Pietrzak-Flis and Wasilewska-Gomulka [P8] also studied the effect of constant intake of HTO or
tritiated food on oocyte survival in Wistar rats from birth and sampled at ages 21 and 71 days. They
found that tritiated food was more effective at reducing oocyte numbers.

208. In summary, the female reproductive system shows great discrepancies between mice and
humans. For humans, the reproductive system is radiosensitive to a dose of 2.56 Gy given in a single
fraction. Exposure to doses within that range will cause permanent ovarian failure due to the killing of
oocytes, and will be accompanied by features associated with menopause. It is anticipated that such a
dose of tritium in the ovaries would have a comparable effect.

(b) Effects in males

209. Unlike oocytes, spermatogonia are continuously produced from stem cells throughout adult life.
Like all tissues that are rapidly replaced, there are certain germ-cell stages that are highly sensitive to
cell killing by ionizing radiation. Experiments in mice conducted by Oakberg in 1955 and 1959 showed
that the most sensitive cells are the type A and B spermatogonia, which can be reduced by 50% with
doses of only about 0.3 Gy of acute X-rays (reported in [S23]). The spermatid and spermatozoa stages
are much less sensitive than the spermatogonia stage (Oakberg and Clark, 1964 as reported in [S23]).
Lambert [L3] found a 27% reduction in spermatogonia of mice injected with tritiated thymidine at an
activity concentration of 0.19 MBq/g of body mass and with HTO at an activity concentration of
2.2 MBq/g of body mass (dose to the cell nucleus of 84 mGy and 49 mGy, respectively).

210. Carr and Nolan [C9] studied the reduction of testis mass in mice following single injections of
tritiated thymidine (0.0370.74 MBq/g of body mass) or HTO (0.371.48 MBq/g of body mass) and
60
Co gamma rays (delivered to match the dose-rate vs. time curve in the 1.48 MBq/g HTO group). The
radiobiological effect was investigated at times from one hour to 24 weeks after injection.
Measurements of the testicular retention of tritium were also made at these times. There was a
progressive loss in mass, up to 30% after 45 weeks, followed by an irregular recovery, which was
more delayed in the case of the animals injected with tritiated thymidine. Time-integrated fractional
testis mass loss was a linear function of injected HTO activity. The RBE value of HTO compared with
60
Co gamma rays was 1.43 at gamma-radiation dose of 0.6 Gy. As only one gamma-radiation dose level
was used, RBE dependence on dose could not be assessed.

211. Balonov et al. studied reduction of testis mass in mice following single intraperitoneal injection of
HTO (0.412.6 MBq/g of body mass) or exposure to 137Cs gamma radiation delivered during ten days
with exponential reduction of dose rate with a half-time of 2.5 days similar to HTO excretion rate
[B10]. The range of tritium beta-radiation dose was 0.123.4 Gy and that of gamma-radiation dose was
0.253.7 Gy. Statistically significant relative testis mass reduction was observed in the dose range from
0.25 Gy of both tritium beta radiation and gamma radiation. RBE values in terms of relative testis mass
reduction increased from 1.9 at mass reduction of 40% (dose of tritium beta radiation 0.8 Gy) to 2.2 at
mass reduction of 10% and corresponding lower dose of about 0.2 Gy. In summary, it has been shown
by studies using both external radiation and tritium administration that certain stages of spermatogonia
development are particularly sensitive to radiation. Most studies concluded that a ratio of effects
(mostly lethal) compared to gamma radiation was in the range of 1.42.2.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 295

C. Stochastic effects of HTO in mammals

1. Carcinogenicity
212. Many laboratory studies on animals have demonstrated that exposure to tritium, both as HTO and
tritiated compounds, can induce cancer although the carcinogenic effect of exposure to tritium has not
been studied as extensively as that of gamma radiation and X-rays.

213. Cahill et al. conducted two studies involving the administration of HTO to pregnant Sprague-
Dawley rats to term in a range resulting in whole-body doses during gestation of 0.0666.6 Gy [C1,
C2]. In their first study, increased incidence of mammary fibroadenomas was detected in dams exposed
at 3.3 Gy and 6.6 Gy, but not at lower doses. In the second study, offspring surviving beyond 30 days
were observed throughout their life and neoplasia recorded. Intrauterine exposure to doses of up to
0.66 Gy had no significant effects on overall cancer incidence rate or onset of mammary
fibroadenomas. In addition, females exposed in utero to 3.3 or 6.6 Gy had lower incidence rates of
mammary fibroadenomas and at 6.6 Gy females had a lower incidence of overall neoplasia compared to
the control unexposed rats. These females, however, were sterile and had reduced mean life spans.

214. Seyama et al. [S13] reported on a series of studies involving acute intraperitoneal injections of
relatively high levels of HTO to 78-week-old female (C57BL/6N C3H/He) F1 mice at the activities
resulting in internal whole-body doses from 2.0 to 10.5 Gy. The animals were observed for up to
750 days and cumulative neoplasia was compared to the effects of chronic irradiation by either gamma
rays or fission neutrons. The effect (incidence of cancer) seems to have nearly saturated at the lowest
dose level; that is, the total incidence of tumours was similar at 500 days and later in all exposed groups
(about 80%) in contrast to the control group (less than 5%). The authors also studied induction of
thymic lymphoma in mice that received 7.9 or 10.5 Gy from a single intraperitoneal injection with
those that received equal doses in four subsequent injections with weekly intervals. In the latter case,
the latent period was much shorter and lifetime lymphoma incidence was significantly higher than after
a single injection.

215. Yamamoto et al. [Y3] reported a study involving continuous oral administration of five levels of
HTO to female (C57BL/6N C3H/He) F1 mice from 10 weeks of age resulting in a dose rate to soft
tissues of 0.010.24 Gy/day. Lifetime tumour incidence approached the maximum level (83%) already
at the lowest dose rate (0.01 Gy/day) while spontaneous incidence was 54%. Exposure to larger dose
rates accelerated development of most studied tumours, which resulted in substantial life shortening.
The main cause of death of mice exposed to higher dose rates (0.0960.24 Gy/day) was thymic
lymphoma. At lower dose rates, non-thymic lymphomas and solid tumours were also observed. Both
shortening of the latent period and life shortening due to development of tumours significantly
increased with dose rate of tritium beta radiation and the increases were significantly larger than those
after irradiation of mice with X-rays or gamma radiation from a 60Co source. In a later similar study
[Y4], the same authors exposed mice to three lower levels of HTO in drinking water from ten weeks of
age resulting in a dose rate to soft tissues of 0.00020.0036 Gy/day. The life span was discernibly
shortened in the group with a larger dose rate (0.0036 Gy/day) due to shortening of latent periods for
tumour development. In the groups of mice with lower dose rates, both this effect and incidence of
thymic lymphoma were missing.

216. A study by Johnson et al. [J5] estimated the lifetime incidence of myeloid leukaemia in seven
groups of about 750 CBA/H mice each; radiation exposure was approximately 0, 1, 2 and 3 Gy both for
HTO and for X-rays. The lifetime incidence of leukaemia in these mice increased from 0.13% in the
296 UNSCEAR 2016 REPORT

control group to 68% in groups exposed to higher radiation doses. The results were fitted to various
equations relating leukaemia incidence to radiation dose, using both the raw data and data corrected for
cumulative mouse-days at risk. The calculated RBE values for tritium beta rays compared to X-rays
ranged from 1.00.5 to 1.30.3. A best estimate of the RBE value for this experiment was about 1.20.3.

217. Gragtmans et al. [G9] estimated RBE for tritium radiation in reference to 200 kVp X-rays, using
acceleration of breast tumour appearance in the female Sprague-Dawley rat as the end point. Chronic
X-ray doses of 0.32.0 Gy were delivered over ten days. Intraperitoneal injections of HTO ranging in
concentration from 45 to 370 MBq/100 g of body weight were administered, followed by four
additional injections at two-day intervals and half of the initial concentration. RBE estimations were
based on various criteria including the tumour incidence per Gy at 450 days post-irradiation and the
time required to induce tumours in 50% of the animals at risk. The results suggest that tritium beta rays
are about 1.11.3 times more effective than chronic 200 kVp X-rays for acceleration of the appearance
of rat mammary tumours. However, the uncertainties involved in these calculations are such that the
effects of tritium beta rays could not be reliably distinguished from those of chronic 200 kVp X-rays.

218. Revina et al. [R1] described a study in rats which were administered HTO intragastrically five
times a week for six months. The effects (leukaemia and other cancer) on these animals were compared
with those on a group chronically exposed to gamma radiation delivered in daily doses comparable to
the tritium dose rate. The main problems with this study are the high doses and the fact that there was
only one dose point in the tritium and gamma exposed groups. An estimate of RBE of about 2.5 was
made in this study.

219. Intraperitoneal injection of HTO to male N5 mice several times over 30 days (so that the exposure
period comprised all stages of spermatogenesis) to give a total dose of 1.5 Gy resulted in a statistically
significant increase of leukaemia incidence among their young (less than 210-day-old, but not yet
one-year-old) offspring [D1]. It appeared that the overall leukaemia incidence in the offspring of the
HTO-exposed fathers was significantly dependent on the maturation stage of the sperm-forming cells
during the HTO exposure. However, in a paper by Balonov et al. [B12] which summarizes Russian
studies on tritium carcinogenicity. Mice and rats were given HTO in drinking water at a dose range of
0.2425.3 Gy. Although most malignancies were increased, the absence of a positive doseresponse
and occasionally a negative response makes interpretation of these data difficult.

220. Yin et al. [Y5] used 12-day-old male and female pups of C3H/HeN mice that were given a single
intraperitoneal injection of HTO at the activities of 0.23, 0.92 and 3.70 MBq/mouse and then observed
for 14 months for the development of tumours. In the males, a significantly increased incidence of liver
neoplasms was detected whereas in the females, only an insignificantly elevated incidence of ovarian
cancer was observed in mice exposed to the highest concentration of HTO.

221. Studies in which administration of tritiated material was compared with the effects of similarly
protracted X- or gamma-radiation exposure are the most reliable sources of RBE data. This is important
in the absence of such data for human exposure. Four of the more comprehensive studies [G9, J5, R1,
S13] were critically examined in the AGIR report [H16] and the conclusion was that, due to
deficiencies in the experimental design and statistical analysis, the findings from these studies should
be treated with caution. However, taken together, they indicate an RBE value with a central estimate in
the range 0.82.5 with an upper 97.5 percentile value of no more than about 3.

222. Straume [S22] undertook a literature review of the risks, including cancer induction, from
exposure to tritium. Because information was not available for humans, cancer-risk estimates for
tritium were derived from experimental animal (mostly mouse) studies. Straume calculated a skewed
risk distribution (with a fiftieth percentile risk per unit dose of 81 106 mGy1 with a 90% confidence
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 297

range of 38185 106 mGy1) using Monte Carlo methods and distributions of dose-rate effectiveness
and multiplying by best estimate RBEs for tritium (based a central value of the RBE of 23 ranging up
to 4.5). This was comparable to radiation risk estimates in the Committees UNSCEAR 1988 Report
[U6], in the ICRP Publication 60 [I13] and in the BEIR V Report of the Committee on the Biological
Effects of Ionizing Radiation [N6].

2. Heritable effects
223. In its UNSCEAR 2001 Report [U10], the Committee noted that no radiation-induced hereditary
diseases had (to that date) been demonstrated in humans. Nonetheless, since such effects are seen in
plants and animals, the Committee provides an approach for estimating such risks [U10].

224. Russell et al. [R17] studied the incidence of seven specific locus mutations in (101 C3H) F1
wild-type male mice using about 40,000 offspring of males exposed to HTO with weighted mean doses
of 4.3 Gy to post-spermatogonial germ cells and 6.2 Gy to spermatogonia. The observed mutation
spectrum was similar to those following previous exposure to external X- or gamma radiation. The
radiosensitivity of post-meiotic cells was similar to that observed for acute exposure to X-rays. For
spermatogonia, comparison was made with earlier experiments with low-dose rate gamma radiation
showing that the mutation rate was twice as high in the case of tritium exposure.

225. Pomerantseva et al. [P10] studied reciprocal translocations (RTs) in mouse stem spermatogonia
induced by HTO and 137Cs gamma radiation. HTO was administered to males by a single
intraperitoneal injection and excreted with a half-time of about 2.5 days. In order to adjust exposure
conditions, the dose rate of gamma radiation was reduced exponentially with the same half-time. Mean
doses of testis cells were 0.5, 1.0, 1.9 and 3.4 Gy of tritium beta radiation and 1.0, 1.9 and 3.7 Gy of
gamma radiation [B9]. In the post-sterile period, threefive months after exposure commenced, ten
twelve males from each experimental group and fourfive males from control groups were sacrificed.
From each male, 100200 spermatocytes were analysed at the stage of diakinesis-metaphase of the first
meiotic division, and the number of multivalents in the form of rings and chains recorded. The groups
of animals exposed to beta and gamma radiation at a dose of 2 Gy were also studied over twoeight
months and reciprocal translocation frequency did not significantly change during this period. This
observation indicates that elimination of cells with translocation from populations of spermatogonia
exposed continuously at low dose rates is insignificant. The increase of reciprocal translocation
frequency with increasing dose was observed over the entire dose range up to 34 Gy. The RBE of
tritium in this study was estimated to be 1.8; dose dependence of RBE was not observed.

226. The ICRP Publication 103 [I23] has stated that there continues to be no direct evidence that
exposure of parents to radiation leads to excess heritable disease in offspring. Nevertheless, on the
basis of the results of animal experiments and, citing the Committees UNSCEAR 2001 Report [U10],
ICRP estimated, for protection purposes, a nominal genetic risk of about 0.2% Gy1 for up to the
second generation (grandchildren). For low-LET radiation, the ICRP value for the probability of severe
heritable effects is 0.5% per Gy for the reproductive population, estimated on the basis of mouse data.

227. Straume and Carsten [S23] noted that the heritable effects observed for other low-LET radiation
were also present following exposure to HTO. By grouping the RBE studies with genetic end points
(such as chromosome aberrations and mutations in mice), they determined that the RBE values ranged
from 1 to 3, with the higher values associated with low doses and low-dose rates, largely owing to the
curvilinear response for the reference radiation.
298 UNSCEAR 2016 REPORT

3. Germ cell effects

228. This subsection presents effects of tritium exposure of mammalian germ cells in one or several
generations that result in progeny death. These effects are interpreted as stochastic, presumably caused
by damage to genetic material of a single cell, but not transferred to the next generations because of the
lethal nature of the radiation-induced mutation.

229. Carsten and Commerford studied dominant lethal mutations (DLMs) in Hale-Stoner-Brookhaven
strain mice resulting from chronic HTO ingestion [C10]. In a two-generation study, mice were
maintained on drinking water with 0.11 MBq/mL HTO. Radiation doses to germ cells of the second
generation animals tested for DLMs were 0.28 Gy in females and 0.38 Gy in males. Mice were mated
at the age of eight weeks, pairing exposed males and females and exposed and non-exposed.
Statistically significant increases in DLM frequency were detected in all the exposed groups compared
with the control one. There was no effect of tritium exposure on breeding efficiency.

230. Mewissen et al. [M12, M13] studied cumulative genetic effects following exposure of male
C57BL/6M mice to tritium for six and ten generations. At each generation, weaned male breeders aged
35 days either received a single injection of tritiated thymidine (0.037 MBq/g of body weight) or were
exposed for five weeks to HTO (0.37 MBq/mL). The dose to male sperm over a 35-day period of
exposure in each generation was estimated at 0.037 Gy from HTO. At 10 weeks of age, all breeders
were sibling-mated. At the fifth generation, the average litter size was 6.56 and 6.72 vs. 6.92,
respectively, in the sibling line receiving tritiated thymidine, exposed to HTO and in the control. The
observed variations are significant at the 0.01% level by chi square test. Also, the average weight of
mice at weaning consistently decreased through successive generations in the sibling line exposed to
tritiated thymidine and to a lesser extent in those exposed to HTO, whereas individual weight remained
fairly constant in control mice. F1 and F2 offspring from the ninth generation were studied for litter size
and infant mortality. The litter size had decreased and infant mortality increased in experimental
groups. DLM frequency (pre-implantation death) had increased in both experimental groups.

231. Balonov and Kudritskaia studied the frequency of DLMs in germ cells of male randomly bred
mice induced by HTO and 137Cs gamma radiation [B9]. HTO was administered to males by a single
intraperitoneal injection and excreted with a half-time of about 2.5 days. In order to adjust exposure
conditions, the dose rate of gamma radiation was reduced exponentially with the same half-time. DLM
frequency was estimated from the results of four weekly matings of each male with 24 intact females
beginning from the tenth day after the exposure launch. At 1718 days after the beginning of mating,
the females were dissected and the number of yellow bodies in the ovaries, places of implantation and
dead embryos were counted. It was concluded that HTO beta radiation was more effective than 137Cs
gamma radiation (figure XII). RBE values tended to increase with dose and effect reduction: from 1.6
at DHTO = 2 Gy to 2.2 at DHTO = 0.5 Gy. Linear extrapolation of RBE dependence on dose to zero gives
RBEmax = 2.6 [B9, B12].

232. In summary, the studies presented here demonstrated that internal exposure of mammal germ cells
to HTO can induce DLMs of the progeny at a wide range of radiation doses. RBE estimation showed
tritium beta radiation to be more effective than gamma rays for this biological end point by factors of
1.6 to 2.6 [B9, B12].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 299

Figure XII. Frequency of induced post-implantation death of randomly bred mouse embryos depending
on dose of 137Cs gamma radiation (lower curve) and of HTO beta radiation (upper curve) [B9, B12]

D. Effects of tritiated biochemical substrates

233. The extent of cellular injury caused by a tritiated biochemical substrate depends largely on where
it is incorporated into a cell and the duration of exposure. Tritiated DNA precursors, such as tritiated
thymidine, are theoretically more efficient in causing cellular injury because they form part of the basic
building block of a DNA strand. This effect was already reported in 1958 by Painter et al. [P1]. On the
other hand, compounds containing tritium that are not close to the DNA in the cell, such as fats or some
amino acids included in a non-nuclear protein, should pose a lesser risk.

234. Whereas radiobiological studies with HTO aimed mostly at specification of RBE values for either
deterministic or stochastic effects in mammals, similar studies with tritiated biochemical substrates
aimed to reveal their effects compared with those from HTO. In practical terms, these studies are used
for specification of dosimetric models for OBT and tritiated biochemical substrates. It is understood
that direct studies of OBT-related biological effects are hardly feasible either in the environment or in
experiments on mammals because of low tritium concentration in OBT. Therefore, some tritiated
biochemical substrates are used as an experimental surrogate for OBT in food, which is potentially a
factor of public exposure.

235. Studies on the effects of nucleotropic forms of tritium such as tritiated DNA-precursors are of
special interest because they promote better understanding of radiobiological mechanisms of internal
300 UNSCEAR 2016 REPORT

exposure. In practical terms, their results will be used as scientific bases for future models for the
protection of workers dealing with those forms of tritium. However, the number of workers potentially
thus exposed is rather limited. Thus, exposure to and incorporation of tritiated thymidine is essentially
largely of scientific interest.

1. Studies in vivo
236. In experimental studies on mammals, studies have focused on biological effects resulting from
tritiated thymidine, the first synthesized tritium-labelled nucleoside, because of its pronounced
intranuclear localization in contrast to uniformly distributed HTO. Effects of tritiated amino acids with
varied intracellular distribution and of uniformly distributed 3H-glucose have been studied to a lesser
extent. Deterministic radiation-induced effects were the focus of these studies, except for one series of
experiments (table 16) studying stochastic effects in germ cells of male mice, i.e. induction of heritable
reciprocal translocations in spermatogonia and non-heritable DLMs.

Table 16. Time-integrated frequency of excess post-implantation embryo death (DLM, % weekdays),
frequency of reciprocal translocation (RT) in spermatogonia and relative testis mass reduction (RTMR, %)
in one month after injection of HTO and tritiated biochemical substrates in male mice [B9, B10, B12]

Tritium Intake (MBq/g Testis cell dose DLM RT frequency RTMR

compound of body weight) (Gy) (%week) (%) (%)
HTO 3.3 1.0 38 0.440.20 472
Glucose 3.3 0.9 32 0.460.16 464
Glycine 3.1 1.3 15 0.360.11 443

D,L-Lysine 3.1 1.5 63 0.380.15 514

L-Lysine 1.9 2.2 130 0.870.22 643

0.04 10 0.210.07
Thymidine 0.4 35 0.560.20 294

1.1 35 a
0.440.17

Deoxycytidine 0.4 20 0.120.06 144

1.1 56 0.260.11 353

Control 0 0.0250.014 0
a
Low fertility.

237. Lambert [L3] found that the number of resting primary spermatocytes per tubule in rat
spermatogonia was halved in 72 hours by exposure to HTO and tritiated thymidine concentrations of
2.2 and 0.55 MBq/g of body mass, respectively, given as a single injection. He pointed out that these
values should be viewed with caution owing to uncertainties in several factors, such as the time of death
of the spermatogonia and, therefore, the dose from tritium that induced it.

238. Baker and McLaren studied effects of tritiated thymidine on the developing oocytes of randomly
bred Q strain mice [B4]. Following seven intraperitoneal injections of pregnant mice with 3H-thymidine
(0.15, 1.5 and 15 MBq per injection), tritium label was detected in the ovaries of their progeny. The
total number of oocytes in the ovaries was reduced in all the exposed groups proportionally to injected
tritium activity. The primordial oocytes were more affected than multilayered follicles. Other functions
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 301

(e.g. body weight, fertility, ovarian weight) were affected only in the group with the highest injected
activity. It was concluded that mouse oocytes are highly sensitive to beta radiation from incorporated
3
H-thymidine during embryonic life.

239. In the experiments of Carr and Nolan [C9] on reduction of testis mass of the mouse following
single injections of tritiated thymidine (0.0370.74 MBq/g of body mass) or HTO (or exposed to 60Co
gamma rays), a significant effect on the testis mass was seen after the injection of tritiated thymidine at
0.037 MBq/g of body mass, which delivered an estimated average absorbed dose to the testis of about
0.035 Gy during 16 weeks. The authors assessed that tritium from tritiated thymidine fixed in the
testis was about twice as effective as the more labile and uniformly distributed tritium from HTO and
that, in terms of injected amount tritiated thymidine, it is unlikely to be more than five times as
effective as HTO even at very low injected amounts.

240. Balonov et al. [B12], in a series of experiments on randomly bred male mice, studied both
reduction of testis mass and frequency of DLMs in germ cells induced by 3H-glucose, two amino acids
(3H-glycine and 3H-lysine) and two nucleosides (3H-thymidine and 3H-deoxycytidine) following
intraperitoneal injections. Tritiated glucose and amino acids were administered in single injections and
nucleosides in six portions during three days. In the post-sterile period, threefive months after
exposure commenced, tentwelve males from each experimental group and fourfive males from
control groups were sacrificed and reciprocal translocations in mouse stem spermatogonia counted. The
effects were assessed in comparison with HTO [B9, B10]. For injected tritiated nucleosides, the
concept of tissue or cell dose was not applied. The main results of experiments are presented in
table 16. As the time dependence of DLMs was different following administration of HTO and various
tritiated substrates, especially the nucleotropic forms, the presented parameter is time-integrated excess
post-implantation embryo death.

241. The administration of all the tested tritiated substrates resulted in the production of three
radiobiological effects that are qualitatively similar to those observed for HTO. As far as the time-
integrated DLM frequency is concerned, the order of effectiveness of the various chemical forms, per
unit of injected activity per gram of body weight, is as follows: 3H-glycine HTO and 3H-glucose
D,L-lysine-3H 3H-deoxycytidine L-lysine-3H 3H-thymidine. DLM frequencies induced by labelled
lysine, thymidine, and deoxycytidine are fiveeight times higher than that from an equal HTO activity.
Per unit of testis dose, 3H-glycine is three times less effective than HTO, 3H-glucose and 3H-D,L-lysine
do not differ from HTO, and 3H-L-lysine is more effective by a factor of 1.5. The difference in genetic
efficiency will not only reflect different dynamics of retention of bound tritium in the testes but also its
different location with regard to the cell nucleus. The nuclear location of 3H-nucleosides and 3H-lysine
is well established, and the suggestion of a predominantly extranuclear location of 3H-glycine in germ
cells is consistent with observations. In contrast to DLM, reciprocal translocation frequencies in stem
spermatogonia reflect late mutagenic effects of tritium. The deterministic effect of testis mass reduction
at one month after tritium injection generally agrees with assessed testis doses [B9, B10, B12].

242. In summary, experiments on rodents with administration of biochemical substrates labelled with
tritium confirmed the theoretical consideration that some of them are more efficient with regard to
induction of both deterministic (cell death) and stochastic (mutation) effects compared with
administration of equal activities of HTO. Per unit of intake, 3H-thymidine is 510 times more efficient
than HTO. The experiments also demonstrated elevated efficiency of other nucleotropic forms of
tritium, L-lysine-3H and 3H-desoxicytidine. For 3H-labelled nucleosides, the application of concepts of
tissue and cell dose is hampered by lack of data on their distribution in mammalian cells.
302 UNSCEAR 2016 REPORT

2. Studies in vitro
243. In 1973, Snow [S16] showed that tritiated thymidine at activity concentrations between 0.37 and
3.7 kBq/mL (0.01 and 0.1 Ci/mL) significantly reduced the number of cells in mouse blastocysts
cultivated in vitro after exposure starting during the two-cell stage. Activity concentrations exceeding
3.7 kBq/mL were lethal to the two-cell embryos. Several other authors have published consistent
findings [H11, K10, M23, O5, S21].

244. By incubating embryos in solutions of tritiated thymidine, Streffer et al. [S24] found that an
activity concentration of 18.5 kBq/mL almost completely inhibited the development of mouse
blastocytes and was about 1,000 times more effective than similar activity concentrations of HTO.
Similarly, they found that activity concentrations of 1.85 MBq/mL of HTO and 1.85 kBq/mL of
tritiated thymidine reduced the yield of fully developed blastocytes to 5060%.

245. The extreme ratios of toxicity reported refer to concentrations of tritium in the culture media
rather than dose to cells. In fact, Streffer et al. [S24], when culturing blastocysts, estimated that tritiated
thymidine at a concentration of 1.85 kBq/mL resulted in a dose rate to DNA of 70 mGy/h and its effect
was similar to that of 1.85 MBq/mL of HTO, which gave a (uniform) dose rate of 60 mGy/h. The
tritiated thymidine in these in vitro cultures was available for the duration of the experiment, thus
resulting in the labelling of all DNA synthesis in the developing embryo. This situation is different
from the situation that might arise in vivo. This was referred to by Furuno-Fukushi [F14].

246. Particularly interesting are those papers that compared the effects of exposure to different tritiated
compounds. Clerici et al. [C22] compared the toxicity of tritiated thymidine with that of the toxicity of
four tritiated amino acids; arginine, lysine, histidine and aspartic acid with regard to growth and
development of two-cell mouse embryos exposed in vitro. Surprisingly, arginine was the most lethal of
all the tritiated compounds, requiring 1.1 kBq/mL of medium to kill 50% of the embryos. In
comparison, tritiated thymidine had an LD50 of about 3.0 kBq/mL. The LD50 of the other amino acids
was 2.2 kBq/mL for lysine, 4.8 kBq/mL for histidine and 14.8 kBq/mL for aspartic acid. The LD50 for
tritiated tryptophan (1 kBq/mL) was almost the same as that for tritiated arginine [K12]. The authors
commented that this was surprising because tryptophan was not excessively incorporated into histones.
However, two non-histone chromosomal proteins with high amounts of tryptophan were identified.

247. Similarly, in vitro experiments performed by Mller et al. [M24] on preimplantation mouse
embryos have shown that, given the heterogeneous distribution and specific incorporation into DNA,
tritiated thymidine is 1,0005,000 times more effective than HTO in inducing harmful effects at the
same level of applied activity concentration. Mller et al. also found that tritiated arginine, a histone
precursor, was more damaging than tritiated thymidine for a number of in vitro end points in the mouse
embryo, including blastocyst formation, hatching of blastocysts, trophoblast outgrowth, inner cell mass
formation, number of cells per embryo and micronucleus formation. The mice embryos were incubated
in vitro in solutions with activity concentrations of 0.37 kBq/mL and 0.93 kBq/mL of both tritiated
arginine and tritiated thymidine. The authors postulated that the greater radiotoxicity of the tritiated
arginine was due to faster uptake and possibly because histone synthesis is not restricted to the S phase
of the cell cycle while thymidine would be incorporated only during DNA synthesis. The latter
assumption was confirmed some years later in cell-cycle specific experiments [M25]. These studies
emphasize the importance of understanding the intracellular distribution of tritium-labelled
biochemicals, with effects relating more directly to nuclear dose than to averaged cell or tissue dose or
to levels of administered activity.

248. Furuno-Fukushi et al. [F14] treated lymphocytic leukaemia cells of mice for 50 hours with
various tritiated compounds: thymidine, lysine, arginine, leucine, and aspartic acid. Cell doses for HTO
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 303

and tritiated amino acids or nucleus doses for 3H-thymidine ranged from approximately 0.1 to 8 Gy (as
interpreted from graphical data). Cell survival decreased exponentially with increased substrate activity
concentration in culture medium for all compounds with the effects being greatest for thymidine,
followed by arginine, lysine, leucine and aspartic acid whereas cell-mutation frequencies increased
linearly. The concentrations for detectable cell killing and mutagenesis were about 37 kBq/mL for
tritiated thymidine, 37370 kBq/mL for tritiated amino acids and 18.5185 MBq/mL for HTO. When
activities in cells were measured and converted to dose in cells for tritiated amino acids and dose in
nuclei for 3H-thymidine, the response both in terms of cell survival and mutation frequency per unit
dose was estimated equal for all amino acids; however, elevated by a factor of 23 for 3H-thymidine.

249. Wang et al. [W4] examined the effects of tritiated biochemical substrates on cultured embryonic
mid-brain cells of mice using the following tritiated compounds: thymidine, uridine, arginine and
glutamic acid. The cells were exposed to different concentrations of these compounds over a 20-hour
period. Assays of cell proliferation and differentiation and DNA and protein content were conducted.
Contrary to the studies by Mller et al. [M24] and Clerici et al. [C22], Wang and Zhou [W3, W4] found
that both tritiated thymidine and tritiated uridine were more radiotoxic than tritiated arginine and
tritiated glutamic acid. This was probably due to the different biological end points studied. Table 17
provides a summary of the effects of exposure to the tritiated compounds measured as the tritium
activity concentration (ID50) in the culture medium and corresponding dose for cells or nuclei (the latter
for 3H-TdR) necessary to inhibit the cellular processes (proliferation and differentiation) by 50%.
Tritiated thymidine behaved very differently from the other three tritiated compounds, with a much
steeper doseresponse curve. This is evident in table 17 where the ID50 for tritiated thymidine is much
lower. However, corresponding absorbed doses for cells or nuclei do not vary much whether the
tritium-labelled substance is distributed uniformly or concentrated in the nucleus as is 3H-thymidine.

Table 17. Inhibitory effect of beta radiation from tritiated biochemical substrates on cellular
proliferation and differentiation [W4]

Proliferation Differentiation
OBT ID50 Absorbed dose ID50 Absorbed dose
(kBq/mL) (Gy) (kBq/mL) (Gy)
3
H-Thymidine 29 0.58 21 0.42
3
H-Arginine 193 0.85 163 0.66
3
H-Uridine 193 0.60 141 0.43
3
H-Glutamic acid 525 0.95 438 0.77

250. In summary, incubating mammalian cells and embryos in media containing biochemical
substrates labelled with tritium may result in various biological effects, such as death of cells and
embryos, cell mutations and inhibited proliferation and differentiation. The radiobiological efficiency
of these outcomes, assessed per unit labelled substrate concentration in culture media, varies by a factor
of up to 1,000 compared with HTO and by a factor of up to some tens between substrates. These
differences reflect the active involvement of labelled biochemical substrates in biochemical processes
that lead to incorporation of tritium into cell organelles and subsequent exposure of cells and nuclei
(such as in the case of labelled DNA precursors) with tritium beta radiation. However, the
radiobiological efficiency of the different substrates is comparable when assessed per unit of cell or
nuclei radiation dose.
304 UNSCEAR 2016 REPORT

3. Biophysical models
251. In order to study the impact of tritiated biochemical substrates or OBT, Chen [C14] performed
microdosimetric simulations to compare differences in energy deposition between uniform distribution
of tritium within a cell (as expected with HTO) and a non-uniform distribution based on the assumption
that all OBT was bound uniformly within biologically critical sites of dimensions from 10 nm to 2 m.
The dose mean lineal energies within these critical targets were calculated to be a factor of 1.7 higher
for OBT bound to the critical site compared to HTO over a wide range of target dimensions. This effect
results from a localized increase in dose to the critical target due to a non-uniform distribution of
energy deposition within the cell. However, the extent of any increase would depend on the extent to
which OBT preferentially localizes within critical targets.

252. Alloni et al. [A1] simulated radiobiological effects of tritium concentration, depending on its
chemical form, either in the cytoplasm or in the nucleus of the target cell. The biophysical track-
structure code PARTRAC was used to calculate nuclear doses, DNA damage yields and fragmentation
patterns for different localization of tritium in human interphase fibroblasts. For tritium distributed
selectively in the cytoplasm but excluded from the cell nucleus, the dose in the nucleus is 15% of the
average dose in the cell. In the low- and medium-dose regions investigated in the paper, numbers of
double-strand breaks (DSBs) are proportional to the nuclear dose, with about 50 DSB/Gy. These results
illustrate the potential for over- or underestimating the risk associated with tritium intake when its
distribution at subcellular levels is not appropriately considered.

253. In summary, while many studies have examined how tritiated biochemical substrates are
partitioned within the body and within the cell, studies specifically looking at health effects due to
exposure to those substances are limited. Those that are available indicate that most organic compounds
have about the same effectiveness as HTO, since they are distributed throughout cells and do not lead to
preferential irradiation of the nucleus. Incorporation of some tritiated amino acids and tritiated
nucleosides (e.g. thymidine), however, can lead to the accumulation of tritium in the nucleus with
longer retention times and a proportionately larger dose per unit intake.

VII. RELATIVE BIOLOGICAL EFFECTIVENESS

254. The RBE is the ratio of the absorbed dose of a reference radiation needed to cause a specific
biological response divided by the absorbed dose of the radiation of interest that causes the same
response. RBE values are experimentally observed values and differ for particular radiation types
according to the biological system and end point under consideration, dose, dose rate, and the reference
radiation. RBE values are the basis for but have to be distinguished from the concept of radiation
weighting factors (wR) used by ICRP in the calculation of equivalent and effective doses, in which
simplifications are made that are considered appropriate by ICRP for protection purposes. In
radiological protection, the RBE for stochastic effects at low doses (RBEM) is of particular interest.

255. RBE data and biophysical considerations (see below) indicate that lower energy electrons (such as
those released by tritium) or photons are biologically more effective than higher energy gamma rays for
a range of deterministic and stochastic end points. Although ICRP [I13, I23] recognized that there was
evidence for a significant variation in RBE values for low-LET radiation (e.g. increasing RBE with
decreasing photon energy), it was argued that a more detailed distinction was not warranted for the
purposes of radiological protection. Thus, a value for wR of 1 was chosen for practical reasons to apply
to all electrons and photons, including beta particles from tritium [C30].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 305

A. Track structure considerations

256. Tritium decay results in the production of a very low-energy beta particle (average energy
5.7 keV) of short range (average track length in water 0.56 m) and, as a result, the average ionization
density (and LET) produced by the emitted beta particle is significantly higher than that produced by
higher energy electrons or photons, such as 60Co gamma rays (see table 18). Lower energy photons or
electrons similar to those produced by tritium decay also show a significant shift in microdosimetric
energy deposition patterns towards higher lineal energy (y) compared to higher energy photon or
electron fields. The spectra of energy deposition in low-pressure proportional counters over a range of
simulated tissue site sizes for tritium, 250 kVp X-rays and 60Co gamma rays were measured by Ellett
and Braby [E2]. The results were then interpreted using the earlier site model of the Kellerer-Rossi
theory of dual radiation action (DRA) (e.g. [K7]) to estimate the RBE value for limiting low doses. The
DRA model simply assumes that the biological effect is proportional to the square of the energy
deposited in some small volume, often taken to be about 1 m in diameter. They reported theoretical
RBE values for tritium of 3.75 compared to 60Co gamma rays and 1.5 compared to 250 kVp X-rays
(half-value layer 1.8 mm Cu), assuming a critical site size of 1 m.

Table 18. Track average LET, L , in water for various radiations based on a cut-off energy, , of
100 eV [I24]

Radiation L (keV/m)
60
Co gamma rays 0.22
200 kV X-rays 1.7
3
H beta rays 4.7
50 kV X-rays 6.3

257. On the nanometre scale also, analysis of the energy deposition patterns of tritium beta particles
has shown tritium to be more effective in producing larger sized clusters of ionization which can be
enfolded within a 2.3 nm diameter sphere compared with photons with energies above 100 keV [M17].
This represents ionization events on the dimensional scale of DNA.

258. A joint task group of ICRP and the International Commission of Radiation Units and
Measurements suggested a relationship between what the group called a quality factor, Q(y), and lineal
energy, y, defined as the energy imparted in a 1 m diameter spherical tissue volume divided by its
mean chord length [I25]. The relationship was based in part on general observations and theoretical
considerations, with special consideration given to the experimental data on chromosome aberrations in
human lymphocytes. The value of Q(y) obtained for tritium beta particles was approximately 2
compared to orthovoltage X-rays. This is supported by theoretical calculations performed by Bigildeev
et al. [B19] on the basis of similar microdosimetric quantities on the micrometre scale.

259. Morstin et al. [M20] calculated the lineal energy spectra of tritium beta particles for spherical sites
with diameters from 1 nm to 10 m and showed that the mean values varied by more than an order of
magnitude over this range. Then, when they applied the assumptions of the site model of DRA theory,
they found that the predicted RBE of low doses of tritium beta particles relative to 250 kVp X-rays rose
from a value of <1.1, for assumed 10 nm sensitive sites, to a peak of ~1.5 for 1 m and then decreased
to ~0.6 for 10 m. The corresponding predicted RBE value of tritium beta radiation relative 60Co
gamma rays was ~1.5 for 10 nm sites, ~2.9 for 1 m and ~1.6 for 10 m. Morstin et al. [M20] pointed
out limitations of the DRA approach and they also considered the possibility of two different pathways
306 UNSCEAR 2016 REPORT

of radiation damage related to two different target sizes. They produced bidimensional correlated
distributions of lineal energy for spherical sites of 10 nm and 20 nm diameter (to represent DNA
double-strand break formation) within a gross sensitive volume of 1 m diameter. By then assuming
arbitrarily (by somewhat questionable analogy to the DRA theory) a squared dependence of RBE on
the product of the lineal energies for the large and small sites, they obtained estimated theoretical RBE
values for tritium compared to 250 kVp X-rays of 1.6 for the 10 nm sites, and 1.8 for the 20 nm sites,
within the 1 m gross sensitive volumes.

260. In a recent theoretical study, Chen [C16] performed microdosimetric simulations to compare dose
mean lineal energies for HTO and OBT with that for 60Co gamma rays in the same size range from
10 nm to 2 m of spherical radiosensitive sites. Compared with 60Co gamma rays, the estimated RBE
value varied from 1.3 to 3.5 for HTO and for 2.3 to 5.6 for OBT.

261. Most of the above theoretical calculations of RBE values are based on the assumptions of uniform
interaction between pairs of elementary biological sub-lesions within sensitive sites of approximately
1 m. However, a number of experimental investigations have indicated that the biological
effectiveness of radiation at low doses is determined predominantly by patterns of energy deposition
over much smaller distances down to nanometre dimensions and, therefore, micrometre sized simulated
volumes will not typically provide an adequate description of these patterns [G3, G5, G12, K8].

262. The effectiveness of low-energy electrons, similar to those produced by tritium, can be studied
using ultrasoft X-rays (0.15 keV) that interact in the cell to produce low-energy electrons. Data from a
range of laboratories around the world, with few exceptions, show ultrasoft X-rays to have increased
effectiveness for a wide range of biological end points compared to equal doses of conventional X-rays
or gamma rays [G6, G7, H10], with RBE values typically increasing with decreasing ultrasoft X-ray
energy down to CK X-rays (0.28 keV; producing a single photoelectron with a range less than 7 nm).
RBEs greater than unity were also found for TiK and CuK X-rays with energies (4.5 keV and 8.0 keV,
respectively) similar to the average energy of the emitted beta particle from tritium.

263. Hill [H11] reviewed several studies that looked at in vitro end points, such as dicentric aberrations
in human chromosomes, micronuclei induction, and mutations over a range of photon energies from
ultrasoft X-rays to 60Co gamma rays. He observed a pronounced trend of an increase in RBE values
with decreasing photon energy for several biological end points, particularly for the induction of
dicentrics in human lymphocytes. He noted that, because of differences in cell types and biological end
points, the extent to which lower photon energy caused RBE values to rise was still uncertain. Recent
data by Frankenberg et al. [F11] report an RBEM (maximum RBE value for very low doses) of about 4
for soft (mammography) X-rays compared to 200 kVp X-rays.

264. The percentage of absorbed dose deposited by low-energy electrons (0.15 keV) in tissue is ~33%
for 60Co gamma rays, ~49% for 220 kV X-rays and rising to ~78% for tritium beta particles [N5]. These
are similar to the low-energy electrons produced by ultrasoft X-rays. It has been inferred that these low-
energy secondary electron track ends produced by low-LET radiation are the predominant cause of
DSB induction, cell inactivation and other cellular effects, with isolated sparse ionizations and
excitations apparently having little biological effect [B24, G6]. The contribution to absorbed dose of
these low-energy electrons is large for tritium beta particle compared to orthovoltage X-rays or gamma
rays. Therefore, the ultrasoft X-ray data would predict an increase in biological effectiveness as a result
of increased clustering of ionization events on the nanometre (DNA) scale leading to an increase in the
number of DSB per unit absorbed dose, along with a slight increase in complexity of the breaks due to
additional associated damage within a few base pairs.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 307

265. Using Monte Carlo simulation, Moiseenko et al. [M18] modelled DSBs and single-strand breaks
(SSBs) in cells exposed to tritium beta particles and low-energy photons. They found that a direct
energy deposition of 10 eV could result in an SSB. They further studied base damage associated with
DSB and were able to differentiate between simple DSBs and complex DSBs. They later developed a
Monte Carlo model to calculate yields of DSBs in DNA after irradiation with 137Cs gamma radiation,
orthovoltage X-rays (typically 150300 kVp) and tritium beta particles [M19]. The RBE values for
DSB production for tritium beta radiation (with 137Cs gamma radiation as the reference) was 1.2 for the
total DSB yield and 1.3 for complex DSBs. They explained that low-energy X-rays and tritium beta
particles tended to deposit energy in a more clustered fashion than 137Cs gamma rays. They concluded
that tritium beta particles were more efficient in producing DSBs in DNA compared to 137Cs gamma
rays and that their relative effectiveness was even greater for the production of complex DSBs.

266. In summary, track structure considerations suggest that the low-energy beta particles produced by
tritium decay are more biologically effective than hard X-rays and gamma rays per unit absorbed dose,
at least in producing DSBs in DNA. This is a result of the average ionization density along the track of
the tritium beta particle being significantly higher than produced by much higher energy photons.
Theoretical calculations based on microdosimetric considerations suggest an RBE value of
approximately 2, relative to 60Co gamma rays.

B. RBE literature reviews and experimental studies

267. In studies aimed at deriving values of RBE for tritium beta radiation, there is the potential for the
values obtained to depend on the total dose and rate at which dose from both tritium and the reference
radiation are delivered. In the case of 3H, the dose is expected to be protracted in time since the dose
rate is dictated by the rate of tritium loss from the body and, to lesser extent, by its radioactive
disintegration. Experiments using X- or gamma rays, on the other hand, often deliver dose in a single
acute exposure because this is more convenient in practice. It is generally accepted that the same dose
delivered in a protracted manner can have a lower effect than an acute dose would, due to the greater
opportunity for DNA repair in the protracted case [I13, N3, N7, U7, U8].

268. Most of the lower values for the RBE of tritium reported in the literature are from studies that
used higher doses and dose rates of the reference radiation. This trend can be explained by the high
dose rate of the reference radiation reducing the apparent relative effectiveness of the tritium doses. A
review of tritium RBE studies was carried out by Ujeno [U2] illustrated this phenomenon. Those
studies, which included external reference radiation, showed a tendency for an inverse relationship
between dose rate and RBE value. The author concluded that use of a RBE value of 1 would be
reasonable for assessing the dose from a very large intake of tritium but that a figure larger than 1
would be more appropriate for environmental and occupational exposure situations. Therefore, the
differences in RBE values can be viewed assuming that the response at high acute doses is less
dependent upon or is independent of photon/electron energy.

269. In numerous studies of tritium radiobiological effects conducted since the early 1960s, the focus
has been on the RBE values for the tritium beta radiation for assessing stochastic health risks (primarily
cancer induction and heritable effects) of lower doses in mammals to be used in radiation protection of
humans and environment. Those studies were carefully reviewed by several groups of authors over the
past two decades [H16, K20, L12, O1, P2, S23].
308 UNSCEAR 2016 REPORT

270. Straume and Carsten [S23] provided a comprehensive review of the literature on the
carcinogenic, heritable, developmental and reproductive effects associated with tritium exposure.
They identified 33 published studies on the RBE of the tritium beta radiation using HTO: 12 studies
used X-rays (200500 kVp) as the reference radiation; 21 studies used gamma rays from 137Cs or
60
Co as the reference radiation. Combining these studies, they calculated an arithmetic mean of 1.8
and 2.3 for the range of RBE values when X-rays and gamma rays, respectively, were used as the
reference radiation.

271. The AGIR report [H16], discussed RBE values of tritium beta radiation and also the relationship
between RBE value and the dose and dose-rate effectiveness factor (DDREF), the use of different
reference radiation, and some experimental studies. They noted that the RBE values for tritium beta
radiation using HTO from a wide variety of cellular and genetic studies were generally found to be in
the range of 12 when X-rays were the reference, and 23 when gamma rays were the reference.

272. Little and Lambert [L12] subsequently conducted a comprehensive analysis of several peer-
reviewed studies with the intent of determining maximum low-dose RBE values, denoted RBEM. The
biological end points of these tritium studies included carcinogenesis, chromosomal aberration, and cell
death. The studies were divided into in vivo and in vitro experiments, and subdivided further according to
whether the reference exposure was prolonged or acute. To be classified as a prolonged exposure, the
external irradiation dose rate had to be comparable to that for exposure to tritium. The authors concluded
that the overall aggregated results implied RBE values with a central estimate in the range of 1.22.5 and
a 97.5 upper percentile of no more than 3.0. The six studies with prolonged gamma radiation as reference
exposure and end points apart from cell survival yielded aggregate RBE values of 2.19 (95% CI: 2.04,
2.33) versus 1.17 (95% CI: 0.96, 1.39) when chronic X-rays were used as reference radiation.

273. A review of RBE values by Kocher et al. [K20] was conducted to support assessments of cancer
risk from known exposure to ionizing radiation and estimation of the probability of causation. A
defining characteristic of analysis of data on RBEs by these authors was that estimates of radiation
effectiveness factors were expressed as subjective probability distributions to represent uncertainty
arising from uncertainties in the underlying estimates of RBE values.

C. Factors affecting RBE values

274. The reference radiation types used in RBE studies were 150250 kVp X-rays, and gamma
radiation from 60Co (1,173 and 1,332 keV) or 137Cs (662 keV). However, studies such as those by Bond
et al. [B23], Sasaki [S3] and Schmid [S6] reported in the ICRP Publication 92 [I20], demonstrated that
orthovoltage X-rays are typically about twice as effective at low doses as are high energy gamma rays
and that this difference is consistent with biophysical calculations [E2, K9].

275. Straume and Carsten [S23] discussed this issue, indicating that RBE can vary significantly with
changes in dose rate and radiation quality. Furthermore, as noted by AGIR [H16], the doseresponse
curve for acute doses of low-LET reference radiation is often curvilinear in relation to the response to
radiation with higher LET. This curvilinearity of response to reference radiation in the low-dose range
means that the RBE between the two types of radiation is maximum at lower doses. The maximal RBE
value is referred to as the RBEM.

276. Cell mutations and chromosome aberrations represent chromosome damage that could lead to
cancer development. However, as Little and Lambert [L12] noted, chromosome damage may be only
one of many steps in carcinogenesis. Hill [H11] also pointed out that many studies used the induction of
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 309

dicentric aberrations as an end point because it was a reliable and repeatable method for comparing
biological response, recognizing that this was a short-term effect not directly relevant to quantitative
assessment of long-term stochastic effects.

277. The yield of DSBs is considered to be strongly dependent on biological systems and cellular
environments and has been used for RBE determinations. After reviewing data on DSB yields for low-
energy electrons and high-energy protons of comparable microdosimetric characteristics to those of
tritium in the dimensions relevant to DSBs, Chen [C15] estimated that average yields of 2.7, 0.93, 2.4
and 1.6 1011 DSBs/GyDa were reasonable estimates of DSB yields for tritium in plasmid DNAs,
yeast cells, Chinese hamster V79 cells and human fibroblasts, respectively. If a biological system is not
specified, the DSB yield from tritium exposure can be estimated as a simple average over
experimentally determined yields as 2.30.7 1011 DSBs/GyDa in various biological systems.

D. Summary of RBE value determinations

278. In summary, about 50 different experimental estimates of the RBE values for the tritium beta
radiation in animals or animal cells have been reported that ranged from 1.0 to 5.0 (centred around
2-2.5) and 0.4-8.0 (centred around 1.52) with gamma rays and orthovoltage X-rays as reference
radiation, respectively. There is tendency for RBE values to increase with decreasing doses. RBE
values derived from stochastic effect studies are generally higher (centred around 2.53 compared with
prolonged gamma radiation) than those obtained from studies of deterministic effects (cell killing in
vivo and in vitro). Considerable variation from one experiment to another exists. Only three
experimental studies directly addressed carcinogenic effects in mammals. In addition to experimental
uncertainty and the choice of a reference radiation, a number of other factors contribute to this
variability, including differences in radiosensitivity of tissues, organs and organisms; differences in the
biological end points; variation in dose and dose rate; and choice of in vitro or in vivo test systems.

279. Tables 19 to 22 provide summaries of the studies of the RBE of tritium beta radiation in
mammals, the studies being grouped by the experimental condition (in vivo studies in tables 19 and 20,
and in vitro studies, including murine and human cells, in tables 21 and 22) and according to the type of
radiation used as reference (X-rays in tables 19 and 21, and gamma rays in tables 20 and 22). In all the
listed studies, tritium was used in HTO form. Tables 19 and 22 present studies in chronological order
with both stochastic and deterministic end points; these are separated in summary table 23. When two
or more end points or various exposure conditions were used in the same study, they are presented in
tables as separate studies.

280. Using the data from some of these studies, Little and Lambert [L12] recalculated RBE values for
the tritium beta radiation and their results are also provided in the tables 19 to 22 and used in summary
table 23.
 310 UNSCEAR 2016 REPORT
Table 19. In vivo studies using X-rays as reference radiation

Reference RBE values for the tritium beta radiation

Study reference Biological end point Exposure conditions Dose range (Gy)
radiation (95% CI where indicated)
Lambert [L3] Spermatogonial survival in mice Single injection of HTO and 200 kVp X-rays 0.05 tritium beta 2.3
prolonged X-rays 0.11 X-rays

Gragtmans et al. [G9] Cumulative incidence of Five HTO injections every two days 200 kVp X-rays 0.463.85 tritium beta
mammary tumours in S-D rats and acute and prolonged X-rays 0.571.78 X-rays (acute) 0.68
0.292.00 X-rays (prolonged) 1.11.3

Gragtmans et al. [G9] Cumulative percentage of Five HTO injections every two days 200 kVp X-rays 0.463.85 tritium beta
mammary tumours in S-D rats and acute and prolonged X-rays 0.571.78 X-rays (acute) 0.83
0.292.00 X-rays (prolonged) n.a.

Chopra and Heddle [C21] Chromosome aberrations in Single injection of HTO and 250 kVp X-rays 2.06.0 tritium beta 1.14 (0.81.5)
peripheral blood in female mice prolonged X-rays 1.56 X-rays

Chopra and Heddle [C21] Chromosome aberrations in Single injection of HTO and 250 kVp X-rays 1.54.5 tritium beta 1.21 (0.81.9)
spermatogonia in mice prolonged X-rays 1.54.5 X-rays

Johnson et al. [J5] Myeloid leukaemia in mice Single injection of HTO and 200/150 kVp 0.853.04 tritium beta 1.24 (0.631.85)
prolonged X-rays X-rays 1.062.64 X-rays

Kozlowski et al. [K24] Chromosome aberrations in Continuous intake of HTO and 250 kVp X-rays 0.6 tritium beta 12
bone marrow in mice acute X-rays 0.5 X-rays 0.43 (0.200.81)a
a
Recalculated value by Little and Lambert [L12].
Table 20. In vivo studies using gamma rays (prolonged irradiation) as reference radiation

Reference Dose range RBE values for the tritium beta radiation
Study reference Biological end point Exposure conditions
radiation (Gy) (95% CI where indicated)
Furchner [F13] Mortality in mice Single injection of HTO and Cobalt-60 gamma 5.316.5 tritium beta 1.7
prolonged gamma (tritium simulator) 12.316.5 gamma

Dobson and Kwan [D6, D7] Oocyte survival in mice Continuous intake of HTO Cobalt-60 gamma 0.070.88 tritium beta 2.8
and prolonged gamma (tritium simulator) 0.221.25 gamma

Carr and Nolan [C9] Testes weight loss in mice Single injection of HTO and Cobalt-60 gamma 0.140.58 tritium beta 1.43 (1.061.80)
prolonged gamma (tritium simulator) 0.58 gamma

Russell et al. [R17] Seven specific locus mutations in F1 Single injection of HTO and Caesium-137 gamma 69 tritium beta 2.2
mice from spermatogonia exposure prolonged gamma 0.4 Gy/h

Balonov et al. [B10] Testes weight loss in mice Single injection of HTO and Caesium-137 gamma 0.123.4 1.82.2
prolonged gamma (tritium simulator) 0.253.7

Balonov et al. [B9, B12] Dominant lethal mutations in male Single injection of HTO and Caesium-137 gamma 0.53.4 tritium beta 1.62.2
mice prolonged gamma (tritium simulator) 1.03.7 gamma RBEM = 2.6

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

Pomerantseva et al. [P10] Reciprocal translocations in mice Single injection of HTO and Caesium-137 gamma 0.53.4 tritium beta 1.8
and Balonov et al. [B12] spermatogonia prolonged gamma (tritium simulator) 1.03.7 gamma

Zhou et al. [Z5] Dominant lethal mutations in female Single injection of HTO and Cobalt-60 gamma 0.040.91 tritium beta 2.5
mice prolonged gamma (tritium simulator) 0.532.7 gamma 2.94 (2.004.28)a

Ijiri [I26] Apoptosis of small intestinal cells in Single injection of HTO and Caesium-137 gamma 0.00.29 tritium beta 2.1 (1.72.5)
mice prolonged gamma (tritium simulator) 0.02.9 gamma 1.6 (1.22.0)a

Ijiri [I26] Apoptosis of descending colon cells Single injection of HTO and Caesium-137 gamma 0.00.2 tritium beta 1.8 (1.42.2)
in mice prolonged gamma (tritium simulator) 0.00.4 gamma 1.4 (1.21.6)a

Satow et al. [S5] Oocyte killing in mice Single injection of HTO and Caesium-137 gamma, 0.040.25 tritium beta 1.13.5
prolonged gamma (tritium simulator) 0.040.25 gamma

Satow et al. [S5] Teratogenic effects on rat embryos Single injection of HTO and Caesium-137 gamma 2.06.0 tritium beta 2.6
prolonged gamma (tritium simulator) 1.756.8 gamma 1.01 (0.571.78)a

Zhou et al. [Z6] Dominant lethal mutations-oocytes Single injection of HTO and Cobalt-60 gamma 0.20.6 tritium beta 2.83.4
prolonged gamma (tritium simulator) 0.72.7 gamma

Zhou et al. [Z6] Dominant lethal mutations- Single injection of HTO and Cobalt-60 gamma 0.20.6 tritium beta 1.63.9
spermatocytes prolonged gamma (tritium simulator) 0.72.1 gamma

Zhou et al. [Z6] Dominant skeletal mutations- Single injection of HTO and Cobalt-60 gamma 0.20.6 tritium beta 3.53.9
spermatogonia prolonged gamma (tritium simulator) 0.72.9 gamma

311
 312 UNSCEAR 2016 REPORT
Reference Dose range RBE values for the tritium beta radiation
Study reference Biological end point Exposure conditions
radiation (Gy) (95% CI where indicated)
Zhou et al. [Z6] Oocyte survival Single injection of HTO and Cobalt-60 gamma 0.20.6 tritium beta 1.42.0
prolonged gamma (tritium simulator) 0.72.9 gamma

Zhou et al. [Z6] Spermatogonial survival Single injection of HTO and Cobalt-60 gamma 0.20.6 tritium beta 2.12.8
prolonged gamma (tritium simulator) 0.72.9 gamma

Zhou et al. [Z6] Chromosome aberrations- Continuous intake of HTO Cobalt-60 gamma 0.20.6 tritium beta 2.93.8
spermatogonia and prolonged gamma (tritium simulator) 0.72.9 gamma

Zhou et al. [Z6] Primary oocyte survival Continuous intake of HTO Cobalt-60 gamma 0.20.6 tritium beta 1.5
and prolonged gamma (tritium simulator) 0.72.9 gamma

Zhou et al. [Z6] Spermatogonial survival Continuous intake of HTO Cobalt-60 gamma 0.20.6 tritium beta 2.32.5
and prolonged gamma (tritium simulator) 0.72.9 gamma

Seyama et al. [S13] Cancer in mice Single injection of HTO and Caesium-137 gamma 2.010.5 tritium beta 2.5b
prolonged gamma (tritium simulator) and gamma
a
Recalculated value by Little and Lambert [L12].
b
Calculated at 500 days.
Table 21. In vitro studies using X-rays as reference radiation

Dose range RBE values for the tritium beta radiation

Study reference Biological end point Exposure conditions Reference radiation
(Gy) (95% CI where indicated)
Bocian et al. [B20] Chromosome aberrations in HTO (2 h or 53 h) 180 kVp X-rays 0.282.45 tritium beta 1.17 (1.131.21)
human lymphocytes and acute X-rays 0.53.0 X-rays 1.91 (0.64, 3.18)b

Prosser et al. [P14] Chromosome aberrations in HTO (30 min or 24 h) 250 kVp X-rays 0.24.0 tritium beta RBEM=1.13 (0.951.31)
human lymphocytes and acute X-ray 0.14.1 X-rays

Vulpis [V3] Chromosome aberrations in HTO (20 min to 2.5 h) 250 kVp X-rays 0.257.0 tritium beta 2.6 at 0.25 Gy
human lymphocytes and acute X-rays 0.059.0 X-rays 1.10 at 7 Gy
8.0 (0.215.8)a

Little [L11] Transformation in mouse cells HTO (5168 h) 220 kVp X-rays 0.255.0 tritium beta <12a
and acute X-rays 0.54.0 X-rays

Kamiguchi et al. [K3, K4] Chromosome-type aberrations HTO (~80) and acute X-rays 220 kVp X-rays 0.142.06 tritium beta 1.08 max dose
in human sperm 0.253.74 tritium beta 1.96 min dose
0.231.82 X-rays 1.39 (1.261.54)a

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

Kamiguchi et al. [K4] Chromatid-type aberrations in HTO (~80) and acute X-rays 220 kVp X-rays 0.142.06 tritium beta 1.65 max dose
human sperm 0.253.74 tritium beta 3.0 min dose
0.231.82 X-rays 2.17 (1.732.73)a

Kamiguchi et al. [K4] Chromosome breakage HTO (~80) and acute X-rays 220 kVp X-rays 0.142.06 tritium beta 1.14 max dose
aberrations in human sperm 0.253.74 tritium beta 2.07 min dose
0.231.82 X-ray 1.47 (1.331.62)a

Kamiguchi et al. [K4] Chromosome-exchange HTO (~80) and acute X-rays 220 kVp X-rays 0.142.06 tritium beta 1.54 min dose
aberrations in human sperm 0.253.74 tritium beta 2.81 min dose
0.231.82 X-rays 1.96 (1.492.62)a
a
Recalculated by Little and Lambert [L12].
b
Recalculated by Prosser et al. [P14].

313
 314 UNSCEAR 2016 REPORT
Table 22. In vitro studies using gamma rays (prolonged irradiation) as reference radiation

Dose range RBE values for the tritium beta radiation

Study reference Biological end point Exposure conditions Reference radiation
(Gy) (95% CI where indicated)
Ueno et al. [U1] Cell survival in mouse cells HTO (20 h) and Cobalt-60 gamma 1.38.0 tritium beta 1.5
prolonged gamma 2.09.0 gamma 1.31.6a

Ueno et al. [U1] Micronuclei in mouse cells HTO (20 h) and Cobalt-60 gamma 1.38.0 tritium beta 2.0
prolonged gamma 2.09.0 gamma 1.82.3a

Ueno et al. [U1] Mutation induction in mouse cells HTO (20 h) and Cobalt-60 gamma 1.54.7 tritium beta 1.8
prolonged gamma 1.54.7 gamma

Yamada et al. [Y1] Mouse pronuclear embryo cell survival Prolonged HTO and Cobalt-60 gamma 0.0090.07 Gy/h tritium beta 1.09
prolonged gamma 0.020.12 Gy/h gammaa (0.501.68)

Yamada et al. [Y1] Mouse early 2-cell embryo survival Prolonged HTO and Cobalt-60 gamma 0.0090.10 Gy/h tritium beta 1.70
prolonged gamma 0.020.12 Gy/h gammaa (1.212.20)

Yamada et al. [Y1] Mouse late 2-cell embryo survival Prolonged HTO and Cobalt-60 gamma 0.0090.19 Gy/h tritium beta 1.25
prolonged gamma 0.020.30 Gy/h gammaa (0.881.62)

Matsuda et al. [M3] Chromosome aberrations in mouse HTO (2 h) and Cobalt-60 gamma 0.090.34 tritium beta 2.0
zygotes prolonged gamma 0.050.30 gamma 1.62 (1.302.07)a

Tanaka et al. [T7] Dicentric chromosome aberrations in Prolonged HTO and Cobalt-60 and 0.142.10 tritium beta 2.12.3
human lymphocytes prolonged gamma Caesium-137 gamma 0.054.0 gamma 2.39 (2.202.59)a

Tanaka et al. [T7] Chromosome aberrations in human Prolonged HTO and Cobalt-60 and 0.142.10 tritium beta n.a.
lymphocytes: centric rings prolonged gamma Caesium-137 gamma 0.054.0 gamma 3.14 (2.563.86)a

Tanaka et al. [T7] Chromosome aberrations in human Prolonged HTO and Cobalt-60 and 0.142.10 tritium beta 2.22.7
lymphocytes: dicentrics and centric rings prolonged gamma Caesium-137 gamma 0.054.0 gamma 2.52 (2.332.72)a

Tanaka et al. [T7] Chromosome aberrations total in human Prolonged HTO and Cobalt-60 and 0.131.11 tritium beta 1.13
bone marrow cells prolonged gamma Caesium-137 gamma 0.252.0 gamma 1.30 (0.961.76)a

Tanaka et al. [T7] Chromosome aberrations in human Prolonged HTO and Cobalt-60 and 0.131.11 tritium beta 3.1
bone marrow cells (chromatids) prolonged gamma Caesium-137 gamma 0.252.0 gamma 4.96 (3.736.59)a
a
Recalculated by Little and Lambert [L12].
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 315

281. The RBE values derived from 48 experiments, on mammals in vivo (28) and mammalian cells in
vitro (20), is presented in table 23. The RBE values are presented in three study groups: for all studies,
for studies of stochastic effects (carcinogenic, genetic, cytogenetic, cell transformations), and for
carcinogenic effect in mammals separately; the latter two compared with effects of prolonged exposure
to photon radiation (gamma and X-rays). The latter two RBE sets of values are more relevant to effects
of low radiation doses on humans. In line with the work of Little and Lambert [L12], the data presented
in table 23 are combined mostly from maximum low-dose RBE values (RBEM if available).

Table 23. Summary of tritium radiation RBE values from experimental studies using different end
points and different reference radiation

Studies of stochastic effectsa Studies of carcinogenic effect

All studies with prolonged reference in mammals with prolonged
Reference exposure reference exposure
Studies
radiation Number RBE value Number RBE value RBE value
Number of
of (mean/median of (mean/median (mean/median
studies
studies and range) studies and range) and range)
Prolonged 2.5 / 2.5 3.0 / 3.0
21 9 1 2.5
gamma (1.03.9) (1.83.9)
In vivo
1.1 / 1.2 1.2 / 1.2 1.3 / 1.3
X-rays 7 4 2
(0.42.3) (1.11.3) (1.21.3)

Prolonged 2.1 / 1.8 2.5 / 2.4

12 8
gamma (1.15.0) (1.35.0)
In vitro
Acute 2.4 / 1.7
8
X-rays (1.18.0)
a
Including carcinogenic effects.

282. In broad terms, the RBE values from all studies ranged from 1.0 to 5.0 (centred around 22.5) and
0.48.0 (centred around 1.52) with gamma rays and orthovoltage X-rays as reference radiation,
respectively. Studies show a general tendency of RBE values to increase with lower doses. There is
some tendency for RBE values derived from studies of stochastic effects (centred around 2.53 when
compared with prolonged gamma radiation) to be generally higher than those obtained from studies of
deterministic effects such as cell killing in vivo and in vitro (not presented separately in table 23).

283. Only three of the studies identified have used cancer incidence as an end point [G9, J5, S13].
These studies include accelerated cancer incidence, i.e. cancer occurring at an early age, rather than an
increase in the overall incidence of disease, as the end point. The small number of studies and
ambiguous end points limit the opportunity to come to a clear conclusion regarding RBE values or their
range for the carcinogenic effect of tritium in mammals.
316 UNSCEAR 2016 REPORT

VIII. EPIDEMIOLOGICAL STUDIES

284. Epidemiological studies of groups of people exposed (or potentially exposed) to tritium fall into
two broad categories: those exposed at work and those exposed in the environment. Workplace
exposure generally provides a better opportunity for assessing the tritium-specific risks to health
following doses received by particular tissues or organs from internally deposited tritium because
monitoring for occupational exposure to tritium will usually have been conducted at the facility where
exposure occurred (or potentially occurred) and doses may be estimated from these monitoring results.
Further, data from monitoring for other sources of occupational exposure to ionizing radiation are likely
to be available if assessments for exposure to tritium have been performed at a facility. Such dose
monitoring data are required when tritium-specific risk is estimated to distinguish it from others
radiation risk sources. Epidemiological studies of tritium workers have generally been unsatisfactory
regarding the use of tritium-specific doses monitoring data. Moreover, studies of occupational exposure
involve mainly adult men and do not include children who may be more sensitive to tritium-induced
adverse health effects.

285. Members of the public are exposed not only to natural sources of tritium, but also to
anthropogenic sources such as tritium produced in nuclear weapon explosions, particularly the fallout
from the atmospheric thermonuclear weapon tests of the early 1960s. Public exposure to tritium also
occurs as a result of releases from nuclear power and nuclear weapon facilities, or from luminizing,
radiochemical and other plants, and from devices containing tritium, such as wristwatches with tritium-
based luminous paint or emergency exit signs. Studies of environmental exposure to tritium have the
advantage that they usually involve exposed (or potentially exposed) individuals other than just adults
who are fit enough to be at work. However, a substantial drawback of such environmental studies is
that bioassay monitoring for exposure to tritium is unlikely to have been conducted, which greatly
reduces the reliability of environmental studies to assess risks specific to tritium exposure. However,
there are some instances where such monitoring has been reported for environmental studies.

286. Any analysis of risk in terms of tritium exposure will need to take account of other sources of
radiation exposure, such as penetrating radiation from external sources and from intake of other
radionuclides, to appropriately distinguish tritium-specific risks from those arising from other sources
of radiation exposure. It is important to ensure, if there is a positive correlation between tritium dose
and other radiation doses, that any risk from these other radiation doses are taken into consideration.

A. Studies of occupational exposure

287. Given the variety of sources of exposure to tritium, epidemiological studies of the risks to health
from tritium exposure would seem to be attractive, especially in an occupational setting since those
workers potentially exposed to tritium are likely to have been monitored for such potential exposure
through the analysis of urine samples. The results of this monitoring should have been recorded and, if
these records still exist, the data could be made available for scientific use through the production of
tritium-specific doses to organs/tissues. Unfortunately, although a number of epidemiological studies of
tritium workers have been conducted in various countries, few of these studies have made direct use of
tritium monitoring data or have used tritium-specific doses derived from urinalysis data. Studies of
exposure to tritium in the workplace are considered below in four broad groupings of studies.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 317

288. All workers at installations with nuclear reactors or reprocessing plants will have been exposed to
tritium to some extent because tritium is produced (at a low frequency) in ternary nuclear fission.
However, studies of workers exposed to tritium have concentrated on those workers who are likely to
have received non-trivial doses from tritium because of certain features of operations at the sites, such
as the presence of heavy-water-moderated reactors or tritium production or processing facilities. The
epidemiological studies considered in this section focus on studies of workers at such nuclear sites
rather than workers exposed to very low levels of tritium because of work at other sites, including light-
water-moderated or gas cooled, graphite moderated reactors.

1. Studies of workers at installations where tritium is present

289. The weakest of the epidemiological studies of occupational exposure to tritium are those studies
that consider workers at sites where exposure to tritium occurs, but make no distinction between
workers exposed (or potentially exposed) to tritium and other workers at the site not so exposed. Both
the workers monitored for potential exposure to tritium and at least some of the other workers at a site
are likely to have been exposed to other sources of radiation.

290. For example, Cragle et al. [C31] studied mortality among almost 10,000 white male workers
employed at the Savannah River Site, United States during 19521980, and they noted that around
5,000 workers would have been exposed to tritium with 800 of these having received a dose of at least
0.5 mSv from tritium. The workers were found to have standardized mortality ratios (SMRs) that were
generally less than 1.0 when compared to the population of the United States (which is probably a
reflection of the healthy worker effect), but no distinction was made in the analysis between workers
monitored for exposure to tritium and other workers. So, although the results of this study are broadly
reassuring as far as workers who have been employed at the Savannah River Site, United States are
concerned, they are of limited informative value when assessing the risk arising from exposure to
tritium, except that the risk of tritium exposure cannot have been grossly underestimated or this would
be apparent in the overall results of the study.

291. In another study, McGeoghegan and Binks [M5] examined mortality and cancer incidence among
workers at the Capenhurst site in the United Kingdom, a nuclear installation that has handled tritium in
relation to nuclear weapon production. This study did not specifically identify those workers monitored
for potential exposure to tritium. McGeoghegan and Binks [M5] reported that radiation workers at
Capenhurst had a significantly low SMR for all causes of death and a significantly low standardized
registration ratio (SRR) for all incident cancer, but workers at this site were exposed to external
radiation and to radionuclides other than tritium and it is not possible in this study to disentangle risk
posed by tritium from that posed by other sources of radiation. The findings of studies of workforces at
establishments where tritium is present to some extent in non-trivial quantities, but which do not
distinguish tritium workers from other radiation workers, are summarized in appendix A, table A1.

2. Studies of workers monitored for potential exposure to tritium

292. The next set of studies embraces those workers who have been monitored for potential exposure to
tritium, but for whom tritium-specific doses are not available or, if available, have not been used. These
studies typically identify those workers at an installation who have been monitored for potential exposure
to tritium and then calculate SMRs for that particular group. For example, in the study of workers of the
United Kingdom Atomic Energy Authority (UKAEA), Beral et al. [B17] identified the subset of workers
who had been monitored for potential exposure to tritium and calculated separate SMRs.
318 UNSCEAR 2016 REPORT

293. In some studies, the SMRs for the tritium workers are compared with the SMRs for other workers
at the installation to generate rate ratios (RRs), which have the benefit of addressing (at least, to some
extent) the healthy worker effect that may be present if the analysis is limited to the calculation of
SMRs alone and the reference population is the general population of a country or region. For example,
in a follow-up study of UKAEA workers, Fraser et al. [F12] not only calculated SMRs for tritium
workers but also compared these SMRs with equivalent SMRs for radiation workers at the UKAEA
who had not been monitored for potential exposure to tritium, to generate RRs. However, the absence
of tritium-specific doses in these studies means that quantitative tritium-specific risk estimates cannot
be generated, although the calculation of SMRs and RRs for tritium workers does permit the
identification of possible large effects arising from exposure to tritium, as happened when Beral et al.
and Fraser et al. [B17, F12] found significantly raised prostate cancer SMRs and RRs for UKAEA
workers who had been monitored for potential exposure to tritium. The findings of this group of studies
of workers monitored for potential exposure to tritium are presented in appendix A, table A2.

3. Studies of workers using occupational dose estimates

294. Some studies of workplace exposure to tritium have used occupational dose estimates, but have
not directly used estimates of tritium-specific doses, if available. Typically, these studies identify
workers at an installation who have been monitored for potential exposure to tritium, calculate SMRs
and possibly also RRs, and then conduct a doseresponse analysis in terms of recorded doses of
penetrating radiation from external sources rather than tritium-specific doses. For example, in a nested
case-control study of prostate cancer risk among UKAEA workers, Rooney et al. [R14] found that for
workers who had been monitored for potential exposure to tritium (or to one of four other radionuclides
frequently found in the same workplace environment as tritium), the relative risk of prostate cancer
significantly increased with the recorded dose of external radiation, whereas it did not for other
workers. For those workers either monitored for potential exposure to tritium or not so monitored but
assessed to have the potential for exposure to tritium, the relative risk significantly increased with the
assessed level of potential exposure. However, although the significantly increased relative risk was
confined to those workers monitored for potential exposure to tritium (rather than those assessed to be
potentially exposed, but not monitored) no use was made of the tritium monitoring data to derive
tritium-specific doses for analysis [R14]. The absence of tritium-specific doses in this study
substantially limits the interpretation of the associations found in terms of tritium-specific risk.

295. Sometimes tritium doses have been derived from monitoring data but are then included with
external doses since it is usually argued that tritium produces whole-body doses that are essentially
equivalent to (and generally smaller than) doses received from external sources of penetrating gamma
rays. For example, in the study of workers of the United Kingdom Atomic Weapons Establishment
(AWE), Beral et al. [B18] added the recorded whole-body dose from tritium based upon monitoring
data to the recorded whole-body dose from external sources. They found a significantly increasing
trend of RR for prostate cancer mortality with increasing whole-body dose (driven by one death with a
cumulative external dose >100 mSv), but did not conduct an analysis in which the tritium-specific dose
was separated out from the external dose. Any inference concerning a tritium-specific risk obtained
from such studies of a positive doseresponse for tritium should rely on an assumption of a positive
correlation between external doses and tritium doses.

296. Zablotska et al. [Z1] studied mortality in Canadian nuclear industry workers, and added recorded
tritium doses derived from urinalysis results to recorded doses from external sources. Again, this did
not permit any tritium-specific risk to be identified from the published results because any findings of
analyses based on tritium-specific doses were not presented. However, of interest are the results for the
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 319

ERR/Sv for leukaemia (excluding CLL) and for all solid cancer, with external doses combined with
tritium doses, in comparison with the ERR/Sv estimates using external doses not combined with tritium
doses (i.e. for external doses alone): for the combined doses, the ERR/Sv estimates are 18.9 (95% CI:
<2.08, 138) and 2.80 (95% CI: 0.038, 7.13), respectively, and for external dose alone, the ERR/Sv
estimates are 16.3 (CI not given) and 2.67 (CI not given), respectively. Therefore, the ERR/Sv
estimates show a small increase when the tritium doses are included with external doses but,
unfortunately, it is not possible from the published results to derive ERR/Sv estimates for tritium doses
alone. The findings of studies that have used records of occupational doses of radiation, but have not
used tritium-specific doses for the analysis, are presented in appendix A, table A3.

4. Studies of workers using tritium-specific dose estimates

297. Few studies of workers have tritium-specific dose estimates derived from occupational exposure
records and use these doses in a tritium-specific risk analysis. If tritium-specific doses are used in an
analysis that appropriately adjusts for any effect of doses received from external sources of radiation
(and for any doses received from any other internally deposited radionuclides), these tritium-specific
doses should enable estimates to be made of tritium-specific risks, although due account must be taken
of the precision of these estimates since the number of workers included in such an analysis may be
small, leading to limited statistical power.

298. Zablotska et al. [Z2] conducted a study of Canadian nuclear industry workers and used tritium-
specific doses in addition to external doses. However, in most analyses, tritium doses were combined
with external doses, so the tritium-specific risk was assessed for only one analysis: the ERR/Sv for all
solid cancer was reported as 4.71 (95% CI: <5.92, 8.58). The wide confidence interval for this
estimate is indicative of the limited power of studies of tritium exposure in just one country.

299. Hamra et al. [H4] studied the tritium-specific risk of leukaemia among workers at the SRS and
supplemented tritium-specific doses derived directly from occupational monitoring records with annual
tritium doses reconstructed from external dose records using tritium monitoring results combined with a
job exposure matrix. The authors, in a Bayesian analysis, reported that for leukaemia excluding CLL,
the ERR/10 mGy estimate was 0.281 (90% credibility interval: 1.136, 0.548) while if the constraint
was imposed that the ERR/10 mGy for the dose from tritium beta particles was greater than that for the
dose from penetrating gamma rays, then the ERR/10 mGy estimate became 0.334 (90% credibility
interval: 0.049, 0.817). The sensitivity of the results to this constraint and the width of the credibility
intervals are notable.

300. Studies of adverse health effects in the offspring of workers exposed to tritium in the
preconceptional period, which have made use of tritium monitoring results to calculate tritium-specific
doses, have been conducted in Canada [G10, M9] and the United Kingdom [H17, H18]. No association
between adverse health effects in offspring and the preconceptional dose derived from tritium
monitoring data were found. However, the UK-study illustrated a difficulty in the interpretation of
findings based upon the assessed potential for historical exposure to tritium rather than the use of
monitoring of dose records. In this regard, a highly significant association between the risk of
leukaemia and non-Hodgkins lymphoma in the offspring of Sellafield workers was assessed to
potential for exposure to tritium of fathers in the preconceptional period and was not confirmed by
tritium dose monitoring data. This calls into question the reliability of assessed potential for exposure to
tritium during historical operations when monitoring of those workers most likely to have received
doses resulted in different conclusions. The findings of these studies that have explicitly used tritium-
specific doses are summarized in appendix A, table A4.
320 UNSCEAR 2016 REPORT

B. Studies of environmental exposure

301. The situation is much more uncertain in studies of members of the public potentially exposed to
tritium because direct measurements of tritium body burdens in non-occupationally exposed people are
rare and not undertaken as part of an epidemiological study. Therefore, any individual assessments of
likely environmental exposure to tritium for use in epidemiological studies have to rely upon modelling
results, such as estimates of intake and consequent doses to people based upon measurements of tritium
in environmental media; but the derived tritium-specific doses are generally very small, leading to
extremely low statistical power to detect any tritium-specific effects. Some studies rely only on
measures of proximity to a source of tritium, such as a nuclear facility, and this inevitably leads to
results that have an uncertain interpretation because the relationship between linear distance from the
facility and tritium dose has not been established, and the relationship between distance and level of
exposure could be very complex and far from being directly proportional if, say, the wind rose is
significantly anisotropic. Further, such studies must take other relevant exposure into account,
including other sources of radiation. As a consequence, studies of public exposure to tritium have to be
examined carefully and their findings need to be interpreted accordingly.

302. In only one epidemiological study of environmental exposure to tritium has individually assessed
exposure to tritium been used, and that is in a historical cohort study of cancer incidence during
1986-2005 among residents of two areas near the Pickering heavy-water-moderated CANDU reactor
site in Ontario, Canada [W5]. In this study, exposure to tritium at residences occupied in 1985 was
estimated from an atmospheric dispersion model that used discharge and meteorological data, and the
exposure data were used in an analysis of cancer risk in relation to tritium exposure. Assessed annual
individual effective doses from tritium were very low (maximum for an adult, 2.36 Sv), and the
limited number of cases of cancer available for study presented problems for some of the analyses, so
the lack of detection of an effect of tritium exposure upon cancer risk must be viewed in this light.
Nonetheless, this study does illustrate what may be done to address tritium-specific risk in a study of
environmental exposure, but it also demonstrates the problems of achieving reasonable statistical power
in a practicably sized study of such exposure.

303. With only one epidemiological study of environmental exposure to tritium that takes account of
assessed tritium-specific doses to individuals, very little can be reliably inferred from studies of health
effects in the vicinities of installations producing, processing or storing tritium. Nuclear installations
that include nuclear reactors or reprocessing plants inevitably discharge tritium to some limited extent
because, for example, ternary fission product tritium will be produced/processed during the operation at
such installations, although such discharges are likely to lead to very small doses. Studies around
installations that include heavy-water-moderated reactors or tritium production or processing plants are
likely to address higher doses from tritium discharges, although even then tritium-specific doses will, in
general, be low; few of these studies have been conducted, and they will be briefly considered in
appendix A, table A5.

304. One possibility that can be eliminated by environmental studies is that the risk of childhood
leukaemia from exposure to tritium has been grossly underestimated, and that this is the cause of
excesses of childhood leukaemia incidence that have been reported around certain nuclear installations,
as has been suggested by Fairlie [F1]. Substantial quantities of tritium were released into the
environment by atmospheric nuclear weapon testing during the late 1950s and early 1960s, particularly
by thermonuclear weapon testing in the early 1960s [U8]. If there has been a serious underestimation of
the tritium-specific risk of childhood leukaemia then it would be apparent in the rates of childhood
leukaemia incidence following this period of intense nuclear weapon testing, particularly in the
Northern Hemisphere where most of the testing took place and tritium-specific doses were highest.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 321

Examination of childhood leukaemia incidence rates from around the world has not revealed any
evidence for an increase of childhood leukaemia risk that might be attributed to tritium fallout, or that
the risk of childhood leukaemia has been greater in the Northern Hemisphere than in the Southern
Hemisphere after this period of intense nuclear weapon testing [W1, W2]. It cannot be claimed that the
results of this study show that the risk of childhood leukaemia arising from exposure to fallout from
nuclear weapon testing is less than predicted by standard radiation-induced leukaemia risk models (or
that there is no risk) because the statistical power is insufficient for this purpose, but the study can
exclude a risk that is much greater than predicted, as has been claimed.

C. Summary of epidemiological studies

305. The great majority of epidemiological studies of tritium workers have not used estimates of
tritium-specific doses in their analyses, which limits the inferences that may be made about tritium-
specific risk using the findings of these studies. There is no indication, however, from studies of tritium
workers that tritium-specific risks have been seriously underestimated. Some results from a few studies
that have used tritium-specific doses are available, which represents progress in the epidemiological
approach to tritium-specific risks, but the conclusions that may be drawn about tritium-specific risks to
health from these few studies are limited.

306. For scientific purposes, it should not be assumed that the effect per unit absorbed dose for tritium
is the same as that for an external source of penetrating gamma rays (i.e. the assumption of an RBE
value for tritium beta particles of unity), since this question is a component of the epidemiological
study. A separate analysis of risk is required using tissue-specific absorbed doses received from tritium
rather than equivalent or effective doses. However, tritium-specific absorbed doses to tissues have only
rarely been used.

307. Owing to the limited numbers of tritium workers in particular countries, and limited exposure of
most of these workers, it is unlikely that epidemiological studies of individual nuclear facilities, or
indeed individual countries, will have sufficient statistical power to have a reasonable prospect of
detecting the risks predicted by standard models, or of risks not far removed from those predicted.
Consequently, international collaboration is required to provide a study large enough to properly
investigate tritium risk. Some studies of tritium workers have already been conducted in Canada, the
United Kingdom and the United States. Other candidate countries for studies of tritium workers include
France and the Russian Federation, and it is possible that other countries, such as China and India,
might be able to contribute worker data. It is clear that a coordinated effort is needed if a serious
epidemiological evaluation of tritium risk is to be made.

308. It is unlikely that epidemiological studies of environmental exposure to tritium will produce
meaningful tritium-specific risk estimates because such exposure is, in general, unlikely to produce
tritium-specific tissue/organ doses that are not low or very low. Further, measures of tritium exposure
of members of the public based upon monitoring data are rare, and indirect assessments of tritium-
specific doses, particularly those based upon linear distance from a point of discharge, are likely to be
associated with considerable uncertainty. Those measurements of the presence of tritium in human
tissue that have been made do not indicate that assessments of the amounts of tritium entering members
of the general public and being retained have been seriously underestimated.

309. The absence of a discernible impact upon global childhood leukaemia incidence rates of tritium
released into the environment by atmospheric nuclear weapon testing demonstrates that the risk of
childhood leukaemia posed by tritium has not been grossly underestimated, as has been proposed by
some commentators.
322 UNSCEAR 2016 REPORT

IX. RESEARCH NEEDS

(a) Heterogeneity, concept of mean dose

310. The biodistribution of organic forms of tritium is heterogeneous within tissues and cells. The
issue of the relevance of the mean organ dose concept as a risk indicator therefore arises. There is no
appropriate dosimetric model for use in human risk assessment and radiation protection for tritiated
nuclear acid precursors. The development of an appropriate microdosimetric approach to better
understand the distribution of dose from various organic forms of tritium within a cell and within
tissues and organs is of importance. The uptake and long-term retention of heterogeneously distributed
organic forms of tritium in tissue and cells is particularly of concern in assessing doses to germ cells,
the embryo, the fetus and the infant. Another physico-chemical form of tritium that is a factor of
human, mostly occupational, exposure and deserves further study of biodistribution in body organs and
tissues and biokinetics is tritium dust and flakes formed in nuclear fusion reactors on carbon, beryllium
or tungsten basis.

(b) RBE studies

311. Up-to-date methods should be used to gather further knowledge on the RBE of tritium beta
particles, especially as OBT, focusing not only on aspects relating to carcinogenesis but also on non-
cancer effects. In particular, research is required for the various stages of in utero and early childhood
exposure. Research aimed at addressing real or practical situations should be prioritizedfor instance
long-term intake of HTO/OBT as food. Data regarding the potential induction of heritable/
transgenerational effects should be critically assessed. New approaches should be investigated, in the
light of the latest advances in biology.

(c) Mechanistic studies

312. Mechanistic studies should emphasize cellular damage and notably the types and frequencies of
DNA damage caused by tritium beta radiation. Of particular interest is the complexity of DNA damage
(that might affect efficiency and fidelity of repair) induced by either DNA synthesis precursors (tritiated
thymidine) or tritium-labelled amino acids in chromatin binding proteins, the triggering of (DNA)
damage signalling pathways and activation of protective processes (e.g. repair, cell cycle arrest,
apoptosis, differentiation) in the context of toxicity, and genome instability. Data are also lacking on the
metabolism and biological/mechanistic effects associated with organic tritium (tritiated biochemical
substances) in situations of chronic environmental exposure of the public.

(d) Environmental considerations

313. Given the still incomplete knowledge on tritium accumulation and behaviour in sediment,
targeted multidisciplinary studies with rigorous protocols need to be used to provide experimental
verification of the hypotheses regarding the possible influence of the activity of microorganisms in
aquatic sediments when organic tritium is remobilized in aquatic animals. In general, the scientific data
regarding the conversion of HTO into organic tritium along the food chain should be enhanced.
Reliable quantitative estimates are required.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 323

314. Some information on the levels of tritium in the environment is available, but it would be of value
to enhance this database. Measurements in targeted environmental media, including those giving a
historical record of exposure, such as tree-rings, particularly in the vicinity of tritium-handling
facilities, would be desirable to provide additional confidence in the current understanding of tritium
behaviour in the environment and its transfer to food.

315. Some measurements of tritium in humans exposed in the environment are available, from
urinalysis and measurements in tissue, for example, at autopsy. It would be desirable to expand this
rather sparse database by gathering data from urine sampling programmes, particularly in residents near
tritium-handling facilities, and from measurements made on various tissues as the opportunities arise.

(e) Epidemiological issues

316. Epidemiological studies are currently very limited in the robustness of their conclusions due to
insufficient statistical power and lack of information on tritium-specific doses. It should be noted that
recorded tritium doses for workers in presently available reports are of the order of ten milligray, which
implies that very large cohorts would be needed to demonstrate a statistically significant increase in
cancer risk at conventionally predicted increases in risk. However, there may be facilities not currently
included in epidemiological studies at which workers were exposed to relatively high levels of tritium
in the early years of tritium production or processing, and it will be important to include such facilities
in any international collaboration. Such studies should use available tritium monitoring data, and
exposure information may be improved by dose reconstruction methods, such as the job exposure
matrix methodology, which has already been used at one tritium production/processing facility. A
coordinated international approach based on standardized dosimetric assessments would be required in
order to make progress in this field. Such international collaboration is the only realistic prospect of
obtaining tritium-specific risk estimates from occupational epidemiological studies.

317. Since tritium will continue to find large-scale use, especially if commercial fusion reactors come
into operation, it would be advisable to seriously consider international collaborative projects to
investigate risks posed by tritium to workers that make full use of relevant occupational data, and that
these studies should be based on a common protocol for the determination of tritium-specific doses
from occupational monitoring data.

318. Unless circumstances can be identified in which relatively large numbers of members of the public
have been exposed to relatively high levels of tritium, it is difficult to see how tritium-specific risk
estimates can be obtained from epidemiological studies. Should a biomarker of tritium-specific
exposure be identified and developed, this could provide opportunities for studies but, even then, the
very low exposure from industrial tritium releases would be difficult to distinguish from the
background presence of tritium.

319. A more realistic approach to assessing exposure and inferring risks to members of the public is
careful monitoring for the presence of tritium in the environment in the vicinity of facilities that
discharge tritium, and the possible monitoring of tritium in selected members of the public in these
areas through, for example, urinalysis. However, monitoring programmes conducted in areas around
facilities handling tritium must also record data on the general levels and variability of tritium away
from such facilities in order that results can be appropriately interpreted.
324 UNSCEAR 2016 REPORT

X. GENERAL CONCLUSIONS
320. Tritium is a radioactive isotope of hydrogen and thus behaves chemically like hydrogen. Humans
are exposed internally to beta radiation emitted by tritium either in occupational settings or as members
of the public. Workers are generally exposed to higher levels of tritium, as HTO, HT, metal tritides and
dust, luminous compounds, tritiated biochemical substrates and some other anthropogenic chemical
forms. The general public is exposed to environmental HTO and OBT in food.

321. Absorbed doses arising from the intake of tritium cannot be measured directly and recourse has to
be made to the use of biokinetic and dosimetric models such as those of ICRP for dose assessment
based on environmental measurements and of bioassay methods (such as the tritium measurement in
urine) combined with models for retrospective determination of doses from individual measurements.

322. The ICRP has developed models for estimating the dose from the intake of HTO, or other tritiated
compounds that partially convert to HTO after being taken into the body, for inhalation of tritiated
gases and low soluble particulate tritium, and for OBT. The current ICRP biokinetic models for tritium
intake by workers and members of the public are reasonably consistent with experimental results, and
improved models are under development. There is a practical need for the development of biokinetic
models for intake of tritiated biochemical substrates, including nucleotropic forms.

323. Doses, and hence risk, from some tritiated biochemical substrates and OBT are greater than those
from HTO, due to their longer residence in the body and potentially also as a result of their localization
within cells, specifically their proximity to DNA. Direct studies of OBT-related biological effects are
not generally feasible because of low tritium concentrations in OBT. However, there are a few studies
looking specifically at biological effects related to tritiated biochemical substrates, most of them using
DNA precursors and amino acids. There is no appropriate dosimetric model for use in human risk
assessment and radiation protection for tritiated DNA precursors. It should be noted, however, that the
number of workers dealing with these forms of tritium is rather limited.

324. In laboratory studies of mammals, tritium has been shown to induce both stochastic and
deterministic biological effects, consistent with the effects induced by other types of ionizing radiation,
and consistent with its generally uniform distribution throughout body tissue, particularly as HTO. The
severity of deterministic effects increases with increasing tissue dose above thresholds, as observed
with other radiation. Exposure to tritium can also induce stochastic effects, such as cancer or heritable
effects, in laboratory mice and rats. However, to date, there is no epidemiological evidence of
stochastic health effects being induced by tritium exposure in humans.

325. A review of the RBE studies of tritium beta radiation indicates a range of values from about unity
to several-fold higher compared to gamma rays and orthovoltage X-rays depending on many factors
such as biological end point, test system, dose and dose rate and choice of reference radiation. RBE
values derived from about 50 in vivo and in vitro experiments on mammals, for different end points,
ranged from 1.0 to 5.0 (centred around 22.5) and from 0.4 to 8.0 (centred around 1.52) with regard to
gamma rays and orthovoltage X-rays, respectively. Studies also showed a general tendency of RBE
values to increase with lower doses. However, the Committee emphasizes that the ability to draw
specific conclusions for carcinogenic effect in mammals is limited because of the lack of pertinent data.

326. A number of epidemiological studies have been conducted of workers or members of the public
potentially exposed to tritium. Unfortunately, the majority of these studies do not use the results of
tritium monitoring to calculate tritium-specific doses for use in the analyses. This makes it very difficult
to reliably interpret the findings of these studies in terms of tritium-specific risk as distinct from risk
from other types of exposure, principally external sources of penetrating radiation and other internal
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 325

emitters. These limitations apply particularly to studies of members of the public, but also extend to
studies of tritium workers, because any occupational tritium monitoring data that might be available
have not been fully utilized. Consequently, at present, little information of substance on tritium-specific
risk can be derived from epidemiological studies of tritium workers or members of the public
potentially exposed to tritium beyond the conclusion that tritium-specific risks have not been seriously
underestimated.

327. Epidemiological studies of occupational exposure to tritium offer the best prospect of
investigating tritium-specific risk to health, but certain requirements have to be fulfilled if this prospect
is to be realized. First, tritium-specific doses derived from tritium monitoring and other occupational
data need to be calculated from existing records or reconstructed. Second, exposure to other sources of
exposure to radiationsuch as external sources of penetrating gamma radiation and intake of other
radionuclidesneeds to be taken into account so that analysis in terms of tritium-specific doses may be
adjusted for the presence of other exposure. Third, since the numbers of tritium workers and the tritium-
specific doses they receive are limited, international collaboration is required to achieve reasonable
statistical power in studying tritium workers to meaningfully investigate tritium-specific risk, and such
international collaboration should use a common protocol for the determination of tritium-specific
doses. Fourth, the success of an international collaborative study will depend on the numbers of tritium
workers available and the doses they have received, so the chance of achieving meaningful results will
depend on the participation of as many countries as possible in such a study.

328. As far as the effects of tritium exposure of the public are concerned, there is effectively no
information on tritium-specific risk that can be obtained from presently available epidemiological
studies. It is unlikely that epidemiological studies of members of the public potentially exposed to
tritium will produce results that are interpretable in terms of tritium exposure with any acceptable
degree of reliability. This is because environmental tritium exposure is generally very low, and any
effect of such exposure against the background of other risk factors will provide a very small signal of
tritium risk against this large background noise.

329. Suggestions that reports of excesses of childhood leukaemia incidence near certain nuclear facilities
could be due to releases of tritium from these installations because of a serious underestimation of the risk
of childhood leukaemia from exposure to tritium are implausible. Large quantities of tritium were released
into the environment by atmospheric nuclear weapon testing in the early 1960s and there is no evidence
from childhood leukaemia registration rates following exposure to tritium fallout of any major
underestimation of the risk of childhood leukaemia from exposure to tritium.

XI. ACKNOWLEDGEMENTS
The Committee wishes to acknowledge with gratitude the help of the experts, J. Harrison (United
Kingdom), B. Lambert (United Kingdom) and R. Wakeford (United Kingdom), who were directly
involved in conducting the evaluation, and J. Chen (Canada) for critically reviewing the manuscript. In
particular, the Committee would like to thank M. Balonov (consultant) who mainly elaborated this
scientific annex based on a first draft prepared by K. Bundy (Canada), D. Chambers (Canada), R. Lane
(Canada) and B. Thriault (Canada). The views expressed in the scientific annex remain those of the
Committee and do not necessarily represent the views of individual experts, the United Nations or its
Member States.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 327

APPENDIX A: TABLES SUMMARIZING STUDIES OF

OCCUPATIONAL AND ENVIRONMENTAL EXPOSURE TO TRITIUM

Tables A1 to A4 present four groups of studies that provide increasing levels of information on tritium-
specific risk.

Table A5 presents studies of environmental exposure to tritium released from heavy-water-moderated

nuclear reactors or tritium production/processing plants.
Table A1. Studies of workforces that include workers (potentially) exposed to tritium though not explicitly identified

Occupational exposure to tritium at installations with heavy-water-moderated nuclear reactors and/or tritium production/processing plants will generally be greater than exposure at installations where there are
very low ambient levels of tritium resulting from, for example, the release of tritium generated in ternary nuclear fission. Consequently, studies of workers exposed to very low levels of tritium at, for example,
nuclear power stations with only light-water-moderated reactors are not included in this table

Study references Summary of study Summary of findings relating to tritium Relevance for this report
Cragle et al., and Cragle Historical cohort study of mortality among white About 5,000 workers exposed to tritium, and about Not possible to derive tritium-specific risk, because
and Watkins [C31, C32] male workers at SRS, USA, first employed 800 workers had recorded dose >0.5 mSv from tritium workers not analysed separately, and
19521975. Total 9,860 workers; 1,722 had died tritium. SMRs for all causes and all cancers were less tritium-specific doses not used
before 1987. SMRs, stratified by employee pay code, than 1.0, which was the case for most causes of
calculated using general population of US white mortality, with no SMR significantly above 1.0.
males as reference. Mortality trends, particularly for Marginally significant positive trend of leukaemia
various cancers, with recorded external doses. mortality with external dose
Annual tritium whole-body doses available, but not
used. Various subgroups examined, but not tritium
workers

Richardson et al., and Historical cohort study of mortality before 2003 SMRs significantly <1.0 for all causes and all cancers. Not possible to derive tritium-specific risk, because

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

Richardson and Wing (5,098 deaths) among 18,883 workers at SRS before Using nested casecontrol approach, marginally tritium workers not analysed separately, and
[R2, R3, R4] 1987. Cumulative recorded whole-body doses from significant positive trend of leukaemia (excluding tritium-specific doses not used
external exposure combined with tritium-specific CLL) mortality with cumulative whole-body dose;
doses from urinalysis. Tritium-specific doses alone strongest for myeloid leukaemia. Adjusting
not used, and workers monitored for potential indirectly for smoking, some evidence for positive
tritium exposure not considered separately trend of lung cancer mortality with cumulative
whole-body dose

McGeoghegan and Binks Historical cohort study of mortality (19461995) and Most SMRs and SRRs <1.0; several significant Tritium workers not analysed separately and tritium-
[M5] cancer incidence (19711991) among 12,540 association between bladder cancer incidence and specific doses not used, so not possible to derive
workers at Capenhurst, UK. Tritium had been cumulative external radiation dose when dose tritium-specific risk
processed before 1988. Annual external whole- lagged 20 years
body radiation doses were included in the analysis.
Unclear how many workers exposed to tritium.
Tritium workers not analysed separately and
internal doses, such as tritium, were not considered

McGeoghegan and Binks Historical cohort study of mortality (19551995) and Most SMRs and SRRs <1.0; several significant trends Tritium workers not analysed separately and tritium-
[M6] cancer incidence (19711995) among 2,628 workers with cumulative external radiation dose found for specific doses not used, so not possible to derive
at Chapelcross, UK. Tritium production from 1980. prostate cancer; no case monitored for tritium and tritium-specific risk
Annual external whole-body radiation doses used. only 2 at Chapelcross after tritium production
Unclear how many workers exposed to tritium. started
Tritium workers not analysed separately and
internal doses not considered

329
 330 UNSCEAR 2016 REPORT
Study references Summary of study Summary of findings relating to tritium Relevance for this report
McGeoghegan et al., Historical cohort study of mortality (19471992) and SMRs for all causes and all cancers <1.0 for radiation Tritium workers not analysed separately and tritium-
Douglas et al., Omar et cancer incidence (19711986) among 14,319 workers. Significant trends of mortality and specific doses not used, so not possible to derive
al., and Smith and workers at Sellafield, UK. Tritium released from incidence for leukaemia (excluding CLL) with tritium-specific risk
Douglas [D8, M7, O5, reprocessing from 1952, and tritium production increasing cumulative external dose
S15] during 19551962. Doses from external radiation
and Pu included in the analyses, but not tritium-
specific doses. Tritium workers not identified or
analysed separately

Carpenter et al. [C7] Historical cohort study of mortality during 19461988 SMRs for radiation workers and other workers Although tritium workers included, not considered
(13,505 deaths) among 75,006 workers at Sellafield, significantly <1.0 for all causes and all cancers. The separately and tritium-specific doses not analysed
UKAEA and AWE; 40,761 monitored for external mortality from leukaemia (excluding CLL) increased alone, so not possible to derive tritium-specific risk
radiation (6,900 deaths). Tritium doses at Sellafield, significantly with increasing cumulative whole-
AWE and Harwell generally included with whole- body dose. No analysis performed for tritium-
body external dose, but excluded for Winfrith and specific doses and tritium workers not considered
Dounreay. Tritium workers not considered as a as a separate group
separate group

Johnson et al. [J7] Historical cohort study of mortality before 1997 SMRs for all causes and all cancers significantly <1.0 Neither tritium workers nor tritium-specific doses
(6,516 deaths) among 22,543 workers at AWE, UK for all workers, all radiation workers, and all workers considered separately, so not possible to derive
during 19511982. Tritium-specific doses combined monitored for internal emitters. For all internal dose tritium-specific risk
with external radiation doses; no analysis of tritium- workers, the SMR for kidney cancer was significantly
specific doses alone. Tritium workers not raised. Significant positive trends with cumulative
considered separately from other internal dose external (plus tritium) dose for mortality from
workers multiple myeloma, bladder cancer and lung cancer
(doses lagged 11 years). No mention of prostate
cancer [B18]

Muirhead et al. [M22] Historical cohort study of mortality (26,731 deaths) SMRs significantly <1.0 for all causes and all cancers. Tritium workers included but not analysed separately
and cancer incidence (11,165 cases) before 2002 Significantly raised SMR for pleural cancer. and tritium-specific doses not used, so not possible
among 174,541 radiation workers included in the Leukaemia (excluding CLL) and all other cancers to derive tritium-specific risk
UK National Registry for Radiation Workers (NRRW). significantly increased with increasing external
Analyses for recorded external doses. Internal doses dose. Exclusion of internal dose workers increased
(including tritium) not included; internal dose slopes for other cancers, but alternatively adjusting
workers identified and either excluded or presence for internal monitoring had little impact
adjusted for in subsidiary analysis. Tritium workers
not identified and not separately analysed

Gribbin et al. [G11] Historical cohort study of cancer mortality during SMRs significantly <1.0 for all causes and all cancers. No tritium-specific doses and no separate analysis of
19561985 (227 deaths) among 8,977 male workers Marginally non-significant positive trend of tritium workers, so not possible to derive tritium-
employed by Atomic Energy of Canada Limited mortality from leukaemia excluding CLL and specific risk
(AECL) during 1956 to June 1980. Analyses for cumulative external dose
recorded external doses. Tritium-specific doses not
considered. Tritium workers not analysed separately
 Study references Summary of study Summary of findings relating to tritium Relevance for this report
Sont et al. [S19] Historical cohort study of cancer incidence during Collective cumulative dose of 1,144.5 man Sv from Tritium-specific doses combined with external doses
19691988 among 191,333 Canadian workers external gamma radiation compared with that of and not considered separately, and tritium workers
monitored for exposure to radiation during 1951 122.6 man Sv from tritium. For nuclear workers, not considered as a separate group, so not possible
1988, using the Canadian National Dose Registry mean cumulative tritium dose received from tritium to derive tritium-specific risk. Caution needs to be
(NDR). Tritium doses estimated from routine of 5.56 mSv compares with 26.43 mSv received from exercised because of problems with the NDR data
urinalysis and added to whole-body external external exposure; for analysis these two (see also [Z2])
radiation doses. Tritium-specific doses not analysed components were combined. SIRs for all cancers
separately and tritium workers not considered as a significantly <1.0 for males and females. ERR/Sv
separate group whole-body dose for all cancers significantly raised

Metz-Flamant et al. [M11] Historical cohort study of mortality during 19682004 Significant trend with external dose for myeloid Tritium workers not treated as a separate group and
(6,310 deaths) among 59,021 nuclear workers in leukaemia tritium-specific doses not used, so not possible to
France during 19501994. Analyses of recorded derive tritium-specific risk
external photon doses. Tritium-specific doses (and
other internal doses) not included. Tritium workers
not analysed separately

Azizova et al., Hunter et Historical cohort study of mortality and disease Study principally concerned with risks from external Tritium-specific doses not used and tritium workers
al., Moseeva et al., and incidence among workers of the Mayak complex in radiation and from Pu; doses from this exposure not identified, so not possible to derive tritium-
Sokolnikov et al. the Russian Federation. Tritium-specific doses not were high in the early years of operations at the specific risk

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

[A7, H20, M21, S18] used and tritium workers not considered separately Mayak complex

Cardis et al., Fix et al., and Historical cohort study of mortality (15,825 deaths), in ERR/Sv cumulative dose for all cancers excluding Neither tritium worker nor tritium-specific doses
IARC [C3, F6, I5] particular cancer mortality (3,976 deaths), among leukaemia was 0.07 (90% CI: 0.39, 0.30) while considered separately so not possible to derive
95,673 workers at Sellafield, UKAEA and AWE in the ERR/Sv for all leukaemia excluding CLL was 2.18 tritium-specific risk
UK, Hanford, Rocky Flats and Oak Ridge National (90% CI: 0.13, 5.7)
Laboratory (ORNL) in the USA, and AECL in Canada
(the IARC 3-country study). For workers at all sites
(except Winfrith and Dounreay) tritium-specific
doses included, but combined with recorded
external doses. Separate analysis with tritium-
specific doses not done and tritium workers not
considered separately

331
 332 UNSCEAR 2016 REPORT
Study references Summary of study Summary of findings relating to tritium Relevance for this report
Cardis et al., Thierry-Chef Historical cohort study of mortality among 407,391 Significant association between cumulative (lagged) No information on tritium-specific doses provided
et al., and Vrijheid et al. nuclear industry workers from 15 countries (the radiation dose and all-cause mortality and no separate analysis of tritium workers
[C4, C5, T11, V2] IARC 15-country study). Total of 18,993 deaths (ERR/Sv=0.42), mainly due to a dose-related conducted, so not possible to derive tritium-specific
included, 5,233 from cancer. Photon doses obtained increase in all cancer mortality (ERR/Sv=0.97). risk. 15-country study has to be treated with caution
from records at individual installations. Tritium- Among 31 specific types of malignancies, significant because of problems with the Canadian worker data
specific doses included with whole-body doses, and association for lung cancer (ERR/Sv=1.86) and (see also [Z2])
no separate analysis conducted. Tritium workers not borderline non-significant associations for multiple
considered separately myeloma (ERR/Sv=6.15) and ill-defined and
secondary cancers (ERR/Sv=1.96). Stratification on
duration of employment had a large effect on
ERR/Sv estimates, reflecting a strong healthy
worker survivor effect in the contributing cohorts

Hamra et al., Leuraud et Historical cohort study of mortality (66,632 deaths) ERR/Gy for all cancers excluding leukaemia 0.39 (90% No analysis conducted for tritium-specific doses and
al., Richardson et al., and among 308,297 radiation workers from nuclear CI: 0.12, 0.67), all leukaemia excluding CLL 2.96 (90% tritium workers not considered as a separate group,
Thierry-Chef et al. installations in USA, UK and France. Doses used CI: 1.17, 5.21). Intended that tritium-specific doses so not possible to derive tritium-specific risk
[H5, L9, R5, T12] derived from recorded external radiation doses. excluded from doses used in analysis; seems likely
Tritium-specific doses not included in the study that for some installations and for some years
(although in some tritium doses included with tritium doses included in external dose records
external doses). Tritium workers not analysed
separately

Daniels et al., and Nested case-control (1:4 matching) study of ERR per 100 mGy (total low-LET radiation dose to Tritium-specific doses alone not used, and tritium
Schubauer-Berigan et al. leukaemia mortality before 2006 (369 deaths) and RBM) 0.09 (95% CI: 0.17, 0.65) for leukaemia workers were not considered as a separate group,
[D2, S9, S10] radiation exposure among 105,245 workers at six excluding CLL (ERR per 100 mGy for exposure 614 so not possible to derive tritium-specific risk
nuclear sites in the USA. 66 (17.9%) cases and 227 years prior to diagnosis 1.9 (95% CI: <0, 8.0)).
(15.4%) controls were exposed to tritium. Internal Tritium-specific doses to RBM available (mean
doses to red bone marrow (RBM) from urinalysis 0.2 mGy; maximum, 85.1 mGy) but not used alone
results were included with recorded external
photon doses. No separate analysis using tritium-
specific doses and tritium workers not treated as a
separate group
Table A2. Studies of workers monitored for (potential) exposure to tritium, identified and investigated as such, but tritium-specific doses not available or not used

Study reference Summary of study Summary of findings relating to tritium Relevance for this report
Beral et al. [B17] Historical cohort study of mortality (3,373 deaths Tritium doses not quantified. All causes SMR for tritium Absence of tritium-specific doses does not permit
during 19461979) in 39,546 workers at UKAEA. workers significantly low at 0.59, and all cancers SMR non- a reliable conclusion on the potential role of
Exposure to tritium occurred at UKAEA sites. Of significantly low at 0.77. On the basis of 6 deaths, prostate tritium in the raised prostate cancer SMR for the
20,382 workers exposed to external radiation, cancer SMR significantly raised, at 8.89; all 6 deaths with group of tritium workers
1,418 also potentially exposed to tritium, and cumulative external doses 50 mSv (SMR=12.77)
tritium workers were considered separately.
Only doses from external radiation used, not
tritium doses (apart from Harwell from 1977
when tritium included in whole-body
exposure)

Carpenter et al. [C8] Historical cohort study of mortality (4,149 For tritium workers, SMRs for all causes and all cancers Absence of tritium-specific doses does not permit
deaths) during 19461988 among 40,761 significantly <1.0, and compatible with other radiation a reliable conclusion about tritium-specific risk
external radiation workers at Sellafield, UKAEA workers. SMR and RR for prostate cancer non-significantly
and AWE. 4,111 workers monitored for tritium raised; testicular cancer the only cancer with significantly
analysed separately, but tritium-specific doses elevated RR buccal cavity and pharynx the only cancer with
not used, only a flag indicating tritium significantly reduced RR. Little evidence that RRs varied with
monitoring period since, or age at, or calendar year of first monitoring or

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

with age at first monitoring. For UKAEA or AWE workers, data
available on duration of monitoring; among tritium-
monitored workers, for prostate cancer, significant variation
with number of years monitored (highest in workers
monitored in 24 years). Significantly raised RR for lung
cancer among tritium workers with external dose <10 mSv,
but not 10 mSv. Non-significantly raised RR (=1.39) for
prostate cancer among tritium workers with external doses
l0 mSv

333
 334 UNSCEAR 2016 REPORT
Study reference Summary of study Summary of findings relating to tritium Relevance for this report
Gillies and Haylock, and Historical cohort study of mortality (19,613 SMRs for all causes and all cancers significantly <1.0 for both Tritium workers considered separately, and all
McGeoghegan et al. deaths) and cancer incidence (10,411 cases) radiation and non-radiation workers, but RRs <1.0, cancer risk (both mortality and incidence)
[G4, M8] before 2006 among 64,956 workers employed significantly for all causes of death. All causes and all cancers significantly greater vs. external only radiation
during 19462002 by British Nuclear Fuels plc RRs for internal radiation workers vs. other radiation workers workers. However, tritium-specific doses not
(BNFL) at Sellafield, Springfields, Capenhurst did not differ significantly from 1.0. Mortality from both used so not possible to draw reliable conclusions
and Chapelcross. 42,431 external radiation cancer and non-cancer increased significantly with increasing about tritium-specific risk
workers, and 22,675 workers also monitored for cumulative external dose; for cancer trend significantly less
tritium, Pu and U. 1,757 workers were for internal radiation workers than external only radiation
monitored for tritium, 1,062 for tritium only. workers, and same pattern for cancer incidence. For tritium
Analyses with external doses only, not internal workers, SMR for all causes significantly <1.0, but SMR for all
doses (including tritium), but workers flagged cancers non-significantly >1.0 (contrasting with SMR of 1.0
for tritium monitoring. In some analyses, for Pu workers and SMR significantly <1.0 for U workers).
tritium workers considered separately Tritium worker SMRs significantly >1.0 for pleural and female
breast cancer. RR comparing all cancer SMR for tritium
workers vs. external only radiation workers significantly >1.0.
Mortality RR for female breast cancer significantly >1.0
(2 deaths among tritium workers) and also for all smoking-
related cancer. All cancer SIR for tritium workers non-
significantly >1.0 (and non-significantly greater than SIRs for
Pu and U workers). SIRs for pleural, testicular and non-
melanoma skin cancer significantly >1.0. Incidence RR for all
cancers significantly >1.0 for tritium workers vs. external only
radiation workers, and incidence RR significantly raised for
non-smoking-related solid cancer and for solid cancer
excluding lung, liver and bone. Tritium workers stated to
have significantly increasing incidence of digestive cancer
with increasing cumulative external dose

Boice et al. [B22] Historical cohort study of mortality during 1944 Mean tritium dose for workers with positive monitoring result Tritium workers did not have unusual patterns of
2009 (3,681 deaths) among 7,270 workers at 8 mSv and maximum 195.5 mSv. (Mean external dose 26 mSv mortality. Tritium-specific doses incorporated into
the Mound nuclear facility, USA, during 1944 and maximum 939.1 mSv.) For workers with non-zero tritium total organ/tissue doses in analysis, so that
1979. 4,509 workers monitored for external dose, SMRs for all causes and all cancers significantly <1.0. tritium-specific risk cannot be derived
radiation. Tritium processed at Mound during For specific cancers, SMRs were generally <1.0, including
19541997; 4,134 workers monitored for prostate cancer and (significantly) lung cancer. For dose
tritium (1,125 with positive urinalysis result). response analysis, with tritium doses included in total
Tritium doses estimated on assumption that organ/tissue doses, significant positive trend for oesophageal
intake was HTO, and added to the doses from cancer and negative trend for liver cancer
other sources. Tritium-specific doses alone not
used in analysis
Table A3. Studies of workers monitored for (potential) exposure to tritium, identified and investigated as tritium workers, and occupational dose records used, but
tritium-specific doses not available or not explicitly used

Study reference Summary of study Summary of findings relating to tritium Relevance for this report

Beral et al. [B18] Historical cohort study of mortality during 19511982 <2% of workers had a tritium dose >10 mSv. SMRs for Very few prostate cancer deaths for tritium workers,
(3,115 deaths) among 22,552 workers at AWE, UK, tritium workers vs. other radiation workers not and no analysis with tritium-specific doses alone, so
before 1983. 9,389 workers monitored for external unusual: RRs of 1.02 for all cancers and 0.97 for other reliable conclusions cannot be drawn about tritium-
radiation; 1,562 also monitored for tritium. Whole- causes (did not vary notably with different dose- specific risk
body doses from tritium added to recorded external lagging periods). Tritium workers had a significant
doses. Tritium workers considered as a separate trend of prostate cancer with cumulative whole-
group for some analyses, but tritium-specific doses body dose compared with workers not monitored
not used for tritium exposure (on the basis of 3 deaths; trend
driven by 1 death with cumulative whole-body dose
100 mSv). For tritium workers, SMR for prostate
cancer non-significantly elevated at 2.50 with
RR 1.27 vs other radiation workers

Fraser et al. [F12] Historical cohort study of mortality during 19461986 SMRs for tritium workers compared with SMRs for Tritium-specific doses not quantified, so no reliable
(5,509 deaths) and cancer morbidity during 1971 other radiation workers significantly raised RR only conclusion on tritium-specific risk of prostate cancer
1984 (1,594 cases) in 39,718 UKAEA workers, for prostate cancer, with SMR in tritium workers 2.82 (or other cancer) can be reached. Cannot assume

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

including 1,702 workers monitored for tritium, (7 deaths) (see also [B17]). Significant association of that external dose gives an acceptable measure of
considered separately. Only recorded external doses prostate cancer mortality with cumulative external tritium dose
used (except tritium doses at Harwell from 1977 radiation dose for tritium workers all 7 deaths with
included in whole-body doses) cumulative external doses 100 mSv (SMR 5.31).
High case fatality for most cancer led to cancer
morbidity results similar to mortality

Rooney et al. [R14] Nested casecontrol study (1:3 matching) of 136 Risk of prostate cancer significantly increased in men Exposure to tritium could not be separated from
UKAEA workers diagnosed with prostate cancer monitored for tritium, with RR 14.26 (95% CI: 3.09, exposure to four other radionuclides. Absence of
between 1946 and 1986, and 404 matched controls. 133.16). Significantly raised RR for men working tritium-specific doses prevents a reliable estimate of
65% of study subjects had been monitored for >10 years with heavy-water-moderated reactors. the tritium-specific risk of prostate cancer
external radiation (a matching criterion). Monitored Significant trend of RR with increasing external dose
and assessed exposure to tritium (and a number of for those men likely to have been exposed to tritium
other radionuclides) included in analysis. Although or one of four other radionuclides (Cr-51, Fe-59,
assessed level of tritium exposure used in some Co60 or Zn-65). Assessed potential for tritium
analyses, tritium-specific doses derived from exposure, and likely level of exposure, gave
monitoring data were not used significant trend of risk with assessed degree of
exposure. When only men assessed as potentially
tritium exposed, but not monitored for exposure,
included in analysis, no significant increase in
prostate cancer RR. Owing to multiple, not possible
to disentangle the independent effects of tritium,
Cr-51, Fe-59, Co-60 and Zn-65

335
 336 UNSCEAR 2016 REPORT
Study reference Summary of study Summary of findings relating to tritium Relevance for this report

Atkinson et al. [A5, A6] Historical cohort study of mortality before 1998 SMRs for all causes and all cancers for non-radiation Absence of tritium-specific doses, so little can be
(10,249 deaths) among 51,367 UKAEA workers workers, radiation workers and internal dose derived about tritium-specific risk
before April 1996; 26,395 monitored for external workers all significantly <1.0. For tritium workers, no
radiation. Some tritium-specific doses included with SMR significantly raised, but prostate cancer SMR for
recorded external doses, although not for Dounreay tritium workers significantly higher than for other
and Winfrith. Tritium workers considered separately radiation workers. Previously reported significant
in some analyses. Tritium-specific doses not used positive trend of prostate cancer risk with
cumulative external dose for tritium workers [F12].
No longer significant and no trend with external
dose for 19801997

Ashmore et al. [A4] Historical cohort study of mortality during 19511987 For all causes of death, when tritium doses included ERR/Sv for all causes of death did not change
(5,426 deaths) among 206,620 Canadian radiation with whole-body external doses, ERR per 10 mSv, substantially when tritium doses included with
workers during 19511983, using Canadian NDR. 2.6 (90% CI: 1.6, 3.6), with tritium doses excluded (i.e. external doses. Separate results for tritium-specific
Tritium doses from urinalysis added to whole-body external doses only) and, ERR per 10 mSv, 2.5 (90% doses alone not presented, so little can be
external radiation doses. Tritium-specific doses and CI: 1.5, 3.5). Risk estimates for tritium doses alone concluded about tritium-specific risk. Caution needs
tritium workers not considered separately not given. The group of workers monitored for to be exercised because of problems with the NDR
potential exposure to tritium not considered data used (see [Z2])
separately

Zablotska et al. [Z1] Historical cohort study of mortality during 19571994 Mean cumulative dose 13.5 mSv; among workers with ERR/Sv for all solid cancers and all leukaemia did not
(1,599 deaths) among 45,468 Canadian nuclear non-zero doses, mean 19.7 mSv. SMRs for all causes materially change when tritium doses excluded
industry workers (from AECL, Ontario Hydro, Hydro of death and all deaths from cancer significantly from whole-body doses, but tritium-specific doses
Qubec and New Brunswick Power) during 1957 <1.0. ERR/Sv total cumulative dose marginally non- alone not used, so not possible to draw reliable
1994, using Canadian NDR tritium doses from significantly positive for all solid cancers combined conclusions about tritium-specific risk. Caution
urinalysis and added to recorded external radiation and marginally significantly positive for all needs to be exercised because of problems with the
doses. Tritium-specific doses and tritium workers not leukaemia excluding CLL; ERR/Sv show small NDR data used (see also [Z2])
considered separately increases (from 2.67 to 2.80, and from 16.3 to 18.9,
respectively) when tritium doses included with
external doses

Schubauer-Berigan et al. Historical cohort study of mortality (41,508 deaths) SMRs for all causes and all cancers significantly <1.0. Absence of tritium-specific doses alone, and tritium
[S11] among 119,195 radiation workers at five nuclear ERR per 10 mSv whole-body dose 0.14% (95% CI: workers considered separately, so little can be
sites in the USA. Tritium-specific doses from 0.17%, 0.48%) for all cancers combined and 2.0% derived about tritium-specific risk
urinalysis added to recorded external doses but (95% CI: 0.71%, 3.5%) for all haematopoietic and
separate analysis with tritium-specific doses alone lymphatic cancer. When neutron and tritium doses
not done separate analysis for all cancers and all together excluded from the whole-body dose, ERR
haematopoietic and lymphatic cancers with both per 10 mSv became 0.18% (95% CI: 0.14%, 0.53%)
neutron and tritium doses excluded from and 2.0% (95% CI: 0.73%, 3.7%), respectively
cumulative doses. Tritium workers not considered
separately
Table A4. Studies of workers monitored for (potential) exposure to tritium, identified and investigated as tritium workers, and tritium-specific doses available and used in
analyses so that tritium-specific risk may be examined explicitly

Study reference Summary of study Summary of findings relating to tritium Relevance for this report

Hazelton et al. [H8] Historical cohort study using a biologically- Of collective dose from external gamma radiation and Analyses performed with tritium-specific doses alone,
based analysis of lung cancer incidence during tritium combined, tritium contributes 9.0%. 95,430 males, although most results for gamma and tritium doses
19691988 (400 cases, 322 in men) among 60,677 with non-zero doses (mean cumulative gamma combined. Modelling predicts ~2 lung cancer cases
191,042 Canadian radiation workers during plus tritium dose, 18.2 mSv), 9,013 with non-zero tritium attributable to tritium exposure. Caution needs to
19511988, making use of Canadian NDR. doses (2 253 with tritium doses >14.95 mSv). Significant be exercised because of problems with the NDR
Tritium-specific doses used in analysis, mainly doseresponse for men with gamma-ray and tritium data used (see [Z2])
in combination with whole-body gamma-ray doses combined, and with gamma-ray doses alone, but
doses for tritium doses alone doseresponse only marginally
significant. Allowing RBE for the tritium absorbed dose to
vary did not improve fit significantly. When dataset
restricted to 69,826 men not flagged for neutron
exposure, doseresponse for tritium doses alone not
significant. For 95,603 women (44,238 with non-zero
radiation doses, mean cumulative gamma plus tritium
dose 3.8 mSv), doseresponse non-significant, but
consistent with doseresponse for men. For men,

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

modelling predicts ~31 cases attributable to gamma
radiation and ~2 cases to tritium exposure

Zablotska et al. [Z2] Review by Canadian Nuclear Safety Commission 42,228 workers first monitored since 1965 had no Analysis of mortality from all solid cancers, individual
of employment and dose records for Canadian significant ERR/Sv for solid cancer (-1.20) or leukaemia solid cancers and leukaemia showed risk due solely
nuclear workers led to corrections and (14.4, p=0.28). This contrasted significantly with ERR/Sv to whole-body gamma doses and that the addition
improvements to records. Mortality (489 cancer for solid cancer in 3,088 AECL workers first monitored of tritium doses did not improve fit of the model.
deaths during 19561994) in revised historical during 19561964: ERR/Sv=7.87 (p <0.01) (but no dose- Study provides estimates of tritium-specific risk, but
cohort of 45,468 nuclear workers (originally related risk of leukaemia). Very likely that dose uncertainties associated with estimates are large
studied by [Z1]) reanalysed. Particular attention information in Canadian NDR incomplete for early AECL due to generally low tritium doses received by a
paid to accuracy of records for AECL workers workers. Mean cumulative tritium dose in revised cohort limited number of workers
before 1965. Tritium doses from urinalysis. 3.02 mSv, 14% of mean cumulative total dose;
Summary doses for whole-body external 809 workers had tritium doses >50 mSv. When in revised
(gamma) and internal (mainly tritium) radiation cohort tritium doses added to risk model for solid cancer
used for risk analysis. Workers with neutron or with gamma doses as second linear term, fit of model did
high internal exposure excluded. Tritium not improve; in model with two separate dose terms, for
workers not considered separately gamma component ERR/Sv was 2.56 (95% CI: 0.11, 6.79),
while for tritium component ERR/Sv was 4.71 (95% CI:
<5.92, 8.58). For individual cancer types, adding tritium
doses to risk model did not improve fit, and risk due
solely to gamma doses

337
 338 UNSCEAR 2016 REPORT
Study reference Summary of study Summary of findings relating to tritium Relevance for this report

Hamra et al. [H3, H4] Reconstruction of tritium-specific annual doses From 75,523 dose records for 19541978 the proportion of Illustrates what may be done using occupational
for workers at SRS, USA, using tritium urinalysis the whole-body dose from tritium was calculated for records to derive tritium-specific doses for use in
data, recorded external doses, and a job- various jobs, areas and time periods, and tritium doses epidemiological analysis. Tritium-specific leukaemia
exposure matrix approach [H3]. Reconstructed assigned for 43,590 person-years. Under a strong risk is uncertain because of limited data, and the
tritium-specific doses used in a study of assumption that the leukaemia risk per 10 mGy from dependence of leukaemia ERR/Sv upon the strong
tritium-specific leukaemia risk [H4], using tritium is always greater than that from external gamma assumption that the risk per unit absorbed dose
Bayesian approach informed by experimental radiation, authors derived ERR per 10 mGy of 0.298 (90% from tritium is always greater than that for gamma
studies of tritium credibility interval: 0.027, 0.702) for leukaemia and 0.344 radiation means that the findings must be treated
(90% credibility interval: 0.049, 0.817) for leukaemia with caution
excluding CLL. Risks obtained without the restriction that
the leukaemogenic effect of tritium is always greater than
that of external gamma radiation leads to values of 0.141
(90% credibility interval: 0.323, 0.649) and 0.281 (90%
credibility interval: 1.136, 0.548), respectively

HSE [H17, H18] Case-control study investigating exposure of Highly significant association between risk of LNHL in Paternal preconceptional tritium doses derived from
fathers working at Sellafield before the offspring and assessed potential for paternal exposure to monitoring data do not indicate raised risk of LNHL
conception of their children and an increased tritium in preconceptional period. Association not in offspring. Risk indicated by assessed potential for
risk of leukaemia and non-Hodgkins reproduced when doses derived from measured tritium exposure must be treated with caution given
lymphoma (LNHL) in these children. Exposure exposure to tritium based on contemporary monitoring difficulties of retrospective assessment of potential
considered included tritium, both assessed data used for historical tritium exposure
potential for exposure and doses derived from
contemporary tritium monitoring data

McLaughlin et al. [M9] Matched case-control study (1:8 matching) of No association between childhood leukaemia risk and No indication of measured paternal preconceptional
childhood leukaemia (112 cases during 1950 paternal preconceptional radiation dose found for either tritium exposure increasing the risk of childhood
1988) and paternal preconceptional exposure recorded external whole-body dose or tritium dose. For leukaemia in offspring
of workers in Ontario, Canada. Analyses tritium exposure, no father of an affected child was
included external and tritium doses from recorded as exposed while 14 control fathers had
occupational records preconceptional exposure

Green et al. [G10] Matched case-control study (1:1 matching) of No association between risk of congenital anomalies and Tritium doses received prior to conception did not
congenital anomalies in offspring of workers of the tritium dose received during the preconceptional increase the risk of congenital anomalies in the child
Ontario Hydro. Preconceptional tritium doses period
based on monitoring data
Table A5. Studies of environmental exposure to tritium released from heavy-water-moderated nuclear reactors or tritium production/processing plants

Study reference Summary of study Summary of findings related to tritium Relevance for this report

Wanigaratne et al. Historical cohort study of cancer incidence 19862005 More than half of Pickering and all north Oshawa This study used assessed tritium-specific exposure
[W5] using data from the Ontario Cancer Registry, among residents experienced modelled average tritium residents at the same address for at least six
people living near Pickering heavy water CANDU concentration levels <2.9 Bq/m2, representing annual years for analysis of cancer incidence. However,
nuclear reactor site in Ontario in 1985 and in effective dose of 0.47 Sv for average adult. The all doses based on residential histories not
comparison area (north Oshawa) further from site. cancer SRR significantly less than 1.0 for both cohorts. reconstructed. Tritium-specific dose estimates
Atmospheric tritium concentrations estimated for Effect of emigration from Ontario not assessable. Only very small (maximum annual effective dose,
each 1985 residential location using atmospheric for female childhood cancer in Pickering, SRR 2.36 Sv), and number of cancer cases (available
dispersion model with tritium discharge and significantly raised. For Pickering residents living at the for just one installation) limited, so power of
meteorological data. Model predictions compared same address in 1985 as in 1979 (non-movers), study to detect any risk was low. Nonetheless,
with monitoring results. In addition to all cancers assessed tritium exposure not associated with risk of study attempts to address tritium-specific cancer
combined, leukaemia and cancer of the lung, thyroid, lung cancer or female breast cancer (other cancer case risk, not done in any other environmental
female breast and of childhood examined numbers too small for this analysis) exposure study

Grosche et al. [G13] Comparison of childhood leukaemia incidence around If releases of tritium from Krmmel site responsible for Absence of human monitoring data and assessed
Krmmel nuclear power station, Germany and SRS, excess childhood leukaemia cases in immediate vicinity, tritium-specific doses from this study limits
USA. Study conducted because of suggestion that then the much larger releases from SRS would possible conclusions on tritium-specific risk.

ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM

tritium discharges from Krmmel (a boiling (light) detectably increase childhood leukaemia risk in Nonetheless, absence of a detectable increased
water reactor) responsible for marked excess of neighbourhood. Around SRS, however, statistically non- risk of childhood leukaemia around SRS does not
childhood leukaemia cases in the vicinity of the site. In significant deficit of childhood leukaemia incidence support serious underestimation of the tritium-
contrast, SRS produced/processed relatively large found. Therefore, theory of childhood leukaemia excess specific risk of childhood leukaemia
quantities of tritium for weapons around Krmmel due to tritium exposure not supported
by this study

Cragle et al., Cragle Geographical study of childhood leukaemia incidence For children born within 25 km of the nuclear power In the absence of assessed tritium-specific doses
and Watkins, and (19641986) and mortality (19501987) within 25 km stations, (marginally) non-significant excess of to individuals, little may be concluded from this
McLaughlin et al. of nuclear installations in Ontario, Canada, in childhood leukaemia mortality and a non-significant study about the tritium-specific risk of childhood
[C31, C32, M9] particular, the heavy-water-moderated CANDU reactor excess of childhood leukaemia incidence. Non- leukaemia
power stations significant excess of childhood leukaemia mortality
among those resident within 25 km of the nuclear
power stations

Johnson and Rouleau Geographical study of birth abnormalities, perinatal and No unusually high mortality or abnormality rates found in Lack of assessed individual exposure to, or doses
[J6] infant mortality during 19711988 within 25 km of study area. Only association between tritium release from, tritium means no substantial conclusions
Pickering heavy-water-moderated CANDU reactor site levels and birth abnormalities was for CNS on tritium-specific risk may be drawn from this
in Ontario, Canada. Health outcome data analysed for abnormalities, but not reproduced using ground study
airborne and waterborne tritium emissions from the monitoring data. Although some evidence of elevated
site, and using ground-monitored airborne tritium risk of Downs syndrome around Pickering, no
concentration data consistent associations with tritium release levels and
ground monitoring data found

339
 340 UNSCEAR 2016 REPORT
Study reference Summary of study Summary of findings related to tritium Relevance for this report

Richter and Stockwell Cancer mortality (during 19801991) among residents No increase in environmental tritium levels as a result of Lacking measured increases of exposure to tritium
[R10] of Lamar County, Mississippi, near the Salmon the nuclear tests detected. Observed cancer mortality resulting from the nuclear tests, this study
(underground) nuclear test site, following two nuclear rates for Lamar County no different from those provides no information on tritium-specific risk.
tests in 1964 and 1966. Residents worried that tritium expected for all Mississippi. No association between Conducted as a public reassurance exercise
released due to these two explosions detectably cancer mortality rate and distance from detonations
increased cancer risk in vicinity
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 341

REFERENCE
A1 Alloni, D., C. Cutaia, L. Mariotti et al. Modeling dose deposition and DNA damage due to
low-energy beta (-) emitters. Radiat Res 182(3): 322-330 (2014).
A2 Alvarez, L.W. and R. Cornog. Helium and hydrogen of mass 3. Phys Rev 56(6): 613 (1939).
A3 Amano, H., M. Atarashi, H. Noguchi et al. Formation of organically bound tritium in plants
during the 1994 chronic HT release experiment at Chalk River. Fusion Sci Technol 28(3P1):
803-808 (1995).
A4 Ashmore, J.P., D. Krewski, J.M. Zielinski et al. First analysis of mortality and occupational
radiation exposure based on the National Dose Registry of Canada. Am J Epidemiol 148(6):
564-574 (1998).
A5 Atkinson, W.D., D.V. Law, K.J. Bromley et al. Mortality of employees of the United Kingdom
Atomic Energy Authority, 1946-97. Occup Environ Med 61(7): 577-585 (2004).
A6 Atkinson, W.D., D.V. Law and K.J. Bromley. A decline in mortality from prostate cancer in
the UK Atomic Energy Authority workforce. J Radiol Prot 27(4): 437-445 (2007).
A7 Azizova, T.V., E.S. Grigorieva, N. Hunter et al. Risk of mortality from circulatory diseases in
Mayak workers cohort following occupational radiation exposure. J Radiol Prot 35(3):
517-538 (2015).
B1 Bacchetta, A. Analyse et speciation du tritium dans des matrices environnementales. Thesis,
Universite Pierre et Marie Curie, Paris VI (2014). (French).
B2 Baglan, N., S.B. Kim, C. Cossonnet et al. Organically bound tritium (OBT) behaviour and
analysis: outcomes of the seminar held in Balaruc-les-Bains in May 2012. Radioprotection
48(1): 127-144 (2013).
B3 Baker, T.G. Comparative aspects of the effects of radiation during oogenesis. Mutat Res 11(1):
9-22 (1971).
B4 Baker, T.G. and A. McLaren. The effect of tritiated thymidine on the developing oocytes of
mice. J Reprod Fertil 34(1): 121-130 (1973).
B5 Baker, T.G. and P. Neal. Action of ionizing radiation on the mammalian ovary. in: The Ovary.
Volume III: Regulation of Oogenesis and Steroidogenesis (L. Zuckerman and B.J. Weir, eds.).
Academic Press, Inc., New York, 1977.
B6 Balonov, M. and L. Chipiga. Dose assessment for intake of tritiated water in humans: role of
tritium incorporation in organic matter. Radiat Hyg 9(4): 1-8 (2016). (Russian).
B7 Balonov, M.I., E.I. Dolgirev and I.A. Likhtarev. Exchange kinetics and dosimetry of tritium
oxide in man for different routes of administration. Health Phys 27(4): 367-375 (1974).
B8 Balonov, M.I. Tritium dosimetry and standardization [Dozimetriya i normirovanie tritiya].
Ehnergoatomizdat 16: 152 (1983). (Russian).
B9 Balonov, M.I. and O. Kudritskaia. Mutagenic action of tritium on the germ cells of male mice.
I. The induction of dominant lethal mutations by tritium oxide and an assessment of its relative
biological effectiveness. Genetika 20(2): 224-232 (1984). (Russian).
B10 Balonov, M.I., O. Kudritskaia and G. Bruk. Relative biological effectiveness of tritium oxide
by the criterion of death of the germ cells in mice. Radiobiologiia 24(1): 114-117 (1984).
(Russian).
B11 Balonov, M.I., I.A. Likhtarev and Y. Moskalev. The metabolism of 3H compounds and limits
for intakes by workers. Health Phys 47(5): 761-773 (1984).
342 UNSCEAR 2016 REPORT

B12 Balonov, M.I., K.N. Muksinova and G.S. Mushkacheva. Tritium radiobiological effects in
mammals: review of experiments of the last decade in Russia. Health Phys 65(6): 713-726
(1993).
B13 Baumgartner, F. and W. Donhaerl. Non-exchangeable organically bound tritium (OBT): its
real nature. Anal Bioanal Chem 379(2): 204-209 (2004).
B14 Belloni, P., G.F. Clemente, S. di Pietro et al. Tritium levels in blood and urine samples of the
members of the Italian general population and some exposed subjects. Radiat Prot Dosim 4(2):
109-113 (1983).
B15 Bennett, B.G. Environmental tritium and the dose to man. Proceedings of the Third
International Congress of IRPA. IRPA, Washington D.C., 1973.
B16 Bennett, B.G. Worldwide dispersion and deposition of radionuclides produced in atmospheric
tests. Health Phys 82(5): 644-655 (2002).
B17 Beral, V., H. Inskip, P. Fraser et al. Mortality of employees of the United Kingdom Atomic
Energy Authority, 1946-1979. Br Med J (Clin Res Ed) 291(6493): 440-447 (1985).
B18 Beral, V., P. Fraser, L. Carpenter et al. Mortality of employees of the Atomic Weapons
Establishment, 1951-82. Br Med J 297(6651): 757-770 (1988).
B19 Bigildeev, E.A., V. Michalik and L. Wilhelmova. Theoretical estimation of quality factor for
tritium. Health Phys 63(4): 462-463 (1992).
B20 Bocian, E., B. Ziemb-Zak, O. Rosiek et al. Chromosome aberrations in human lymphocytes
exposed to tritiated water in vitro. Curr Top Radiat Res Q 12(1-4): 168-181 (1978).
B21 Bogen, D.C. and G.A. Welford. "Fallout tritium" distribution in the environment. Health Phys
30(2): 203-208 (1976).
B22 Boice, J.D., Jr., S.S. Cohen, M.T. Mumma et al. Mortality among mound workers exposed to
polonium-210 and other sources of radiation, 1944-1979. Radiat Res 181(2): 208-228 (2014).
B23 Bond, V.P., C.B. Meinhold and H.H. Rossi. Low-dose RBE and Q for x-ray compared to
gamma-ray radiations. Health Phys 34(5): 433-438 (1978).
B24 Botchway, S.W., D.L. Stevens, M.A. Hill et al. Induction and rejoining of DNA double-strand
breaks in Chinese hamster V79-4 cells irradiated with characteristic aluminum K and copper L
ultrasoft X rays. Radiat Res 148(4): 317-324 (1997).
B25 Bouville, A., S.L. Simon, C.W. Miller et al. Estimates of doses from global fallout. Health
Phys 82(5): 690-705 (2002).
B26 Brues, A.M., A.N. Stroud and L. Rietz. Toxicity of tritium oxide to mice. Proc Soc Exp Biol
Med 79(1): 174-176 (1952).
B27 Bursian, S.J., D.F. Cahill, J.W. Laskey et al. Some aspects of brain neurochemistry after
intrauterine exposure to tritium. Int J Radiat Biol Relat Stud Phys Chem Med 27(5): 455-461
(1975).
B28 Butler, H.L. and J.H. Leroy. Observation of biological half-life of tritium. Health Phys 11(4):
283-285 (1965).
C1 Cahill, D.F., J.F. Wright, J.H. Godbold et al. Neoplastic and life-span effects of chronic
exposure to tritium. I. Effects on adult rats exposed during pregnancy. J Natl Cancer Inst
55(2): 371-374 (1975).
C2 Cahill, D.F., J.F. Wright, J.H. Godbold et al. Neoplastic and life-span effects of chronic
exposure to tritium. II. Rats exposed in utero. J Natl Cancer Inst 55(5): 1165-1169 (1975).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 343

C3 Cardis, E., E.S. Gilbert, L. Carpenter et al. Effects of low doses and low dose rates of external
ionizing radiation: cancer mortality among nuclear industry workers in three countries. Radiat
Res 142(2): 117-132 (1995).
C4 Cardis, E., M. Vrijheid, M. Blettner et al. Risk of cancer after low doses of ionising radiation:
retrospective cohort study in 15 countries. Br Med J 331(7508): 77 (2005).
C5 Cardis, E., M. Vrijheid, M. Blettner et al. The 15-Country Collaborative Study of Cancer Risk
among Radiation Workers in the Nuclear Industry: estimates of radiation-related cancer risks.
Radiat Res 167(4): 396-416 (2007).
C6 Carlisle, S.M., P.A. Burchart, C. McCauley et al. Biokinetics of inhaled radioactive methane in
rats: a pilot study. Appl Radiat Isot 62(6): 847-860 (2005).
C7 Carpenter, L., C. Higgins, A. Douglas et al. Combined analysis of mortality in three United
Kingdom nuclear industry workforces, 1946-1988. Radiat Res 138(2): 224-238 (1994).
C8 Carpenter, L.M., C.D. Higgins, A.J. Douglas et al. Cancer mortality in relation to monitoring
for radionuclide exposure in three UK nuclear industry workforces. Br J Cancer 78(9):
1224-1232 (1998).
C9 Carr, T.E. and J. Nolan. Testis mass loss in the mouse induced by tritiated thymidine, tritiated
water, and 60Co gamma irradiation. Health Phys 36(2): 135-145 (1979).
C10 Carsten, A.L. and S.L. Commerford. Dominant lethal mutations in mice resulting from chronic
tritiated water (HTO) ingestion. Radiat Res 66(3): 609-614 (1976).
C11 Carsten, A.L. Tritium in the environment: isotopic effects and transmutation. pp.419-458 in:
Advances in Radiation Biology (J.T. Lett and H. Adler, eds.). Academic Press, New York,
1979.
C12 Chebotina, M.Y. and O. Nikolin. Tritium water bodies in the area of PA Mayak. Radiation
Safety Issues 1: 79-84 (2005). (Russian).
C13 Chebotina, M.Y. and O.A. Nikolin. The current tritium concentrations in human urine in the
area of nuclear fuel cycle facilities. Dokl Biol Sci 447(6): 390-391 (2012).
C14 Chen, J. Radiation quality of tritium. Radiat Prot Dosim 122(1-4): 546-548 (2006).
C15 Chen, J. Estimated yield of double-strand breaks from internal exposure to tritium. Radiat
Environ Biophys 51(3): 295-302 (2012).
C16 Chen, J. Radiation quality of tritium: a comparison with 60Co gamma rays. Radiat Prot Dosim
156(3): 372-375 (2013).
C17 Cheng, Y.S., A.R. Dahl and H.N. Jow. Dissolution of metal tritides in a simulated lung fluid.
Health Phys 73(4): 633-638 (1997).
C18 Cheng, Y.S., M.B. Snipes, Y. Wang et al. Biokinetics and dosimetry of titanium tritide
particles in the lung. Health Phys 76(2): 120-128 (1999).
C19 Cheng, Y.S., Y. Zhou, Y.S. Wang et al. Dose estimate of inhaled hafnium tritide using the
ICRP 66 lung model. Health Phys 82(6): 817-824 (2002).
C20 Cheng, Y.S., M.B. Snipes, R.F. Kropf et al. Radiation dosimetry of metal tritides. Health Phys
68(6) Suppl.: S53 (1995).
C21 Chopra, C. and J.A. Heddle. Cytogenetic measurements of the relative biological effectiveness
of tritium. INFO-0287. Atomic Energy Control Board, Ottawa, Canada, 1988.
C22 Clerici, L., M.J. Carroll, M. Merlini et al. The toxicity of tritium: the effects of tritiated amino-
acids on preimplanted mouse embryos. Int J Radiat Biol Relat Stud Phys Chem Med 45(3):
245-250 (1984).
344 UNSCEAR 2016 REPORT

C23 CNSC. Tritium releases and dose consequences in Canada in 2006. Part of the tritium studies
project. INFO-0793. Canadian Nuclear Safety Commission, Ottawa, Ontario, 2009.
C24 CODEX Alimentarius Commission. Codex general standard for contaminants and toxins in
food and feed. CODEX STAN 193-1995. 2013.
C25 Commerford, S.L., A.L. Carsten and E.P. Cronkite. The distribution of tritium in the glycogen,
hemoglobin, and chromatin of mice receiving tritium in their drinking water. Radiat Res 72(2):
333-342 (1977).
C26 Commerford, S.L., A.L. Carsten and E.P. Cronkite. The turnover of tritium in cell nuclei,
chromatin, DNA, and histone. Radiat Res 92(3): 521-529 (1982).
C27 Cool, D.A. and H.D. Maillie. Dissolution of tritiated glass microballoon fragments:
implications for inhalation exposure. Health Phys 45(3): 791-794 (1983).
C28 Cool, D.A. and H.D. Maillie. Tritium distribution and excretion following intratracheal
instillation of glass microballoon fragments in rats. Health Phys 46(3): 599-606 (1984).
C29 Cooper, J.R., M.R. Bailey, F.A. Fry et al. Guidance on the application of dose coefficients for
the embryo and fetus from intakes of radionuclides by the mother. Doc NRPB 16(2): 2005.
C30 Cox, R., H.G. Menzel and J. Preston. Internal dosimetry and tritium--the ICRP position. J
Radiol Prot 28(2): 131-135 (2008).
C31 Cragle, D.L., R.W. McLain, J.R. Qualters et al. Mortality among workers at a nuclear fuels
production facility. Am J Ind Med 14(4): 379-401 (1988).
C32 Cragle, D.L. and J.P. Watkins. Mortality among workers at the Savannah river nuclear fuels
production facility. American Statistical Association, VA, 1999.
D1 Daher, A., M. Varin, Y. Lamontagne et al. Effect of pre-conceptional external or internal
irradiation of N5 male mice and the risk of leukemia in their offspring. Carcinogenesis 19(9):
1553-1558 (1998).
D2 Daniels, R.D., S. Bertke, K.M. Waters et al. Risk of leukaemia mortality from exposure to
ionising radiation in US nuclear workers: a pooled case-control study. Occup Environ Med
70(1): 41-48 (2013).
D3 Delong, C.W., R.C. Thompson and H.A. Kornberg. Percutaneous absorption of tritium oxide.
Am J Roentgenol Radium Ther Nucl Med 71(6): 1038-1045 (1954).
D4 Di Pace, L., E. Letellier, H. Maubert et al. Biological hazard issues from potential releases of
tritiated dust from ITER. Fusion Eng Des 83(10-12): 1729-1732 (2008).
D5 Diabate, S. and S. Strack. Organically bound tritium. Health Phys 65(6): 698-712 (1993).
D6 Dobson, R.L. and T.C. Kwan. The RBE of tritium radiation measured in mouse oocytes:
increase at low exposure levels. Radiat Res 66(3): 615-625 (1976).
D7 Dobson, R.L. and T.C. Kwan. The tritium RBE at low-level exposure--variation with dose,
dose rate, and exposure duration. Curr Top Radiat Res Q 12(1-4): 44-62 (1978).
D8 Douglas, A.J., R.Z. Omar and P.G. Smith. Cancer mortality and morbidity among workers at
the Sellafield plant of British Nuclear Fuels. Br J Cancer 70(6): 1232-1243 (1994).
E1 Eakins, J.D., W.P. Hutchinson and A.E. Lally. The radiological hazard from tritium sorbed on
metal surfaces. Health Phys 28(3): 213-224 (1975).
E2 Ellett, W.H. and L.A. Braby. The microdosimetry of 250 kVp and 65 kVp x rays, 60 Co
gamma rays, and tritium beta particles. Radiat Res 51(2): 229-243 (1972).
E3 Etnier, E.L., C.C. Travis and D.M. Hetrick. Metabolism of organically bound tritium in man.
Radiat Res 100(3): 487-502 (1984).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 345

F1 Fairlie, I. Hypothesis to explain childhood cancer near nuclear power plants. Int J Occup
Environ Health 16(3): 341-350 (2010).
F2 Feinendegen, L.E. Tritium-Labeled Molecules in Biology and Medicine. Academic Press,
New York, 1967.
F3 Feinendegen, L.E. and V.P. Bond. Transmutation versus beta irradiation in the pathological
effects of tritium decay. pp.221-231 in: Tritium (A.A. Moghissi and M.W. Carter, eds.).
Messenger Graphics, Phoenix, 1973.
F4 Feinendegen, L.E. and E.P. Cronkite. Effect of microdistribution of radionuclides on
recommended limits in radiation protection, a model. Curr Top Radiat Res Q 12(1-4): 83-99
(1978).
F5 Feinendegen, L.E., E.P. Cronkite and V.P. Bond. Radiation problems in fusion energy
production. Radiat Environ Biophys 18(3): 157-183 (1980).
F6 Fix, J.J., L. Salmon, G. Cowper et al. A retrospective evaluation of the dosimetry employed in
an international combined epidemiological study. Radiat Prot Dosim 74(1-2): 3953 (1997).
F7 Fliedner, T.M., R.J. Haas, H. Stehle et al. Complete labeling of all cell nuclei in newborn rats
with H3-thymidine. A tool for the evaluation of rapidly and slowly proliferating cell systems.
Lab Invest 18(3): 249-259 (1968).
F8 Forell, B., L.S. Myers, Jr. and A. Norman. DNA repair synthesis in minimally stressed human
lymphocytes. Int J Radiat Biol Relat Stud Phys Chem Med 41(5): 535-545 (1982).
F9 Fourr, E., P. Jean-Baptiste, A. Dapoigny et al. Reference waters in French laboratories
involved in tritium monitoring: how tritium-free are they? Radioprotection 49(2): 143-145
(2014).
F10 Foy, J.M. and H. Schnieden. Estimation of total body water (virtual tritium space) in the rat,
cat, rabbit, guinea-pig and man, and of the biological half-life of tritium in man. J Physiol 154:
169-176 (1960).
F11 Frankenberg, D., K. Kelnhofer, K. Bar et al. Enhanced neoplastic transformation by
mammography X rays relative to 200 kVp X rays: indication for a strong dependence on
photon energy of the RBE(M) for various end points. Radiat Res 157(1): 99-105 (2002).
F12 Fraser, P., L. Carpenter, N. Maconochie et al. Cancer mortality and morbidity in employees of
the United Kingdom Atomic Energy Authority, 1946-86. Br J Cancer 67(3): 615-624 (1993).
F13 Furchner, J.E. Relative biological effectiveness of tritium beta-particles and Co60 gamma-rays
measured by lethality in CF1 mice. Radiat Res 6(4): 483-490 (1957).
F14 Furuno-Fukushi, I., A.M. Ueno and H. Matsudaira. Cell killing and mutation to 6-thioguanine
resistance after exposure to tritiated amino acids and tritiated thymidine in cultured
mammalian cells (L5178Y). Radiat Res 110(3): 428-438 (1987).
G1 Galeriu, D. and A. Melintescu. Retention of tritium in reference persons: a metabolic model.
Derivation of parameters and application of the model to the general public and to workers. J
Radiol Prot 30(3): 445-468 (2010).
G2 Gao, W.M., B. Wang and X.Y. Zhou. Effects of prenatal low-dose beta radiation from tritiated
water on learning and memory in rats and their possible mechanisms. Radiat Res 152(3): 265-
272 (1999).
G3 Geard, C.R., R.D. Colvett and N. Rohrig. On the mechanics of chromosomal aberrations: a
study with single and multiple spatially-associated protons. Mutat Res 69(1): 81-99 (1980).
G4 Gillies, M. and R. Haylock. The cancer mortality and incidence experience of workers at
British Nuclear Fuels plc, 1946-2005. J Radiol Prot 34(3): 595-623 (2014).
346 UNSCEAR 2016 REPORT

G5 Goodhead, D.T. An assessment of the role of microdosimetry in radiobiology. Radiat Res

91(1): 45-76 (1982).
G6 Goodhead, D.T. and H. Nikjoo. Current status of ultrasoft X rays and track structure analysis
as tools for testing and developing biophysical models of radiation action. Radiat Prot Dosim
31(1-4): 343-350 (1990).
G7 Goodhead, D.T. Soft x-ray radiobiology and synchrotron radiation. pp. 683-705 in:
Synchrotron Radiation in Biosciences (B. Chance et al., eds.). Oxford University Press,
Oxford, 1994.
G8 Gracheva, L.M. and V.G. Korolev. Genetic Effects of Radionuclide Decay in Cells.
Atomizdat, Moscow, 1977. (Russian).
G9 Gragtmans, N.J., D.K. Myers, J.R. Johnson et al. Occurrence of mammary tumors in rats after
exposure to tritium beta rays and 200-kVp X rays. Radiat Res 99(3): 636-650 (1984).
G10 Green, L.M., L. Dodds, A.B. Miller et al. Risk of congenital anomalies in children of parents
occupationally exposed to low level ionising radiation. Occup Environ Med 54(9): 629-635
(1997).
G11 Gribbin, M.A., J.L. Weeks and G.R. Howe. Cancer mortality (1956-1985) among male
employees of Atomic Energy of Canada Limited with respect to occupational exposure to
external low-linear-energy-transfer ionizing radiation. Radiat Res 133(3): 375-380 (1993).
G12 Griffin, C.S., M.A. Hill, D.G. Papworth et al. Effectiveness of 0.28 keV carbon K ultrasoft
X-rays at producing simple and complex chromosome exchanges in human fibroblasts in vitro
detected using FISH. Int J Radiat Biol 73(6): 591-598 (1998).
G13 Grosche, B., D. Lackland, L. Mohr et al. Leukaemia in the vicinity of two tritium-releasing
nuclear facilities: a comparison of the Kruemmel Site, Germany, and the Savannah River Site,
South Carolina, USA. J Radiol Prot 19(3): 243-252 (1999).
H1 Haas, R.J., W. Schreml, T.M. Fliedner et al. The effect of tritiated water on the development of
the rat oocyte after maternal infusion during pregnancy. Int J Radiat Biol Relat Stud Phys
Chem Med 23(6): 603-609 (1973).
H2 Hamby, D.M. Uncertainty of the tritium dose conversion factor. Health Phys 77(3): 291-297
(1999).
H3 Hamra, G., L.A. Nylander-French and D. Richardson. Dose reconstruction for an occupational
cohort at the Savannah River nuclear facility: evaluation of a hybrid method. Radiat Prot
Dosim 131(2): 188-197 (2008).
H4 Hamra, G., D. Richardson, R. Maclehose et al. Integrating informative priors from
experimental research with Bayesian methods: an example from radiation epidemiology.
Epidemiology 24(1): 90-95 (2013).
H5 Hamra, G.B., D.B. Richardson, E. Cardis et al. Cohort Profile: The International Nuclear
Workers Study (INWORKS). Int J Epidemiol 45(3): 693-699 (2016).
H6 Harrison, J.D., V.M. Levack and R. Kozlowski. Biokinetics, dosimetry and effects of tritium
in the embryo and fetus. NRPB-M962. National Radiological Protection Board, Didcot, 1998.
H7 Harrison, J.D., A. Khursheed and B.E. Lambert. Uncertainties in dose coefficients for intakes
of tritiated water and organically bound forms of tritium by members of the public. Radiat Prot
Dosim 98(3): 299-311 (2002).
H8 Hazelton, W.D., S.H. Moolgavkar, S.B. Curtis et al. Biologically based analysis of lung cancer
incidence in a large Canadian occupational cohort with low-dose ionizing radiation exposure,
and comparison with Japanese atomic bomb survivors. J Toxicol Environ Health A 69(11):
1013-1038 (2006).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 347

H9 Henry, P. Etude d'une contamination accidentelle par l'eau tritiee. IAEA-SM-150/24.

Symposium on Assessment of Radioactive Organ and Body Burdens, Stockholm, Sweden, 22-
26 November 1971. pp.641-657. International Atomic Energy Agency, Vienna 1972. (French).
H10 Hill, M.A., D.L. Stevens, K.M. Stuart Townsend et al. Comments on the recently reported low
biological effectiveness of ultrasoft X rays. Radiat Res 155(3): 503-510 (2001).
H11 Hill, M.A. The variation in biological effectiveness of X-rays and gamma rays with energy.
Radiat Prot Dosim 112(4): 471-481 (2004).
H12 Hill, R.L. and J.R. Johnson. Metabolism and dosimetry of tritium. Health Phys 65(6): 628-647
(1993).
H13 Hisamatsu, S., Y. Takizawa, M. Itoh et al. Fallout 3H in human tissue at Akita, Japan. Health
Phys 57(4): 559-563 (1989).
H14 Hisamatsu, S., T. Ohmura, Y. Takizawa et al. Tritium level in Japanese diet and human tissue.
J Radioanal Nucl Chem 156(1): 89-102 (1992).
H15 Hodgson, S.A., J.E. Scott and A. Hodgson. In vitro dissolution of tritium-loaded particles from
the JET fusion machine. Radiat Prot Dosim 127(1-4): 55-59 (2007).
H16 HPA. Review of risks from tritium. RCE-4. Health Protection Agency, Chilton, 2007.
H17 HSE. HSE investigation of leukaemia and other cancers in the children of male workers at
Sellafield. Health and Safety Executive, Sudbury, UK, 1993.
H18 HSE. HSE investigation of leukaemia and other cancers in the children of male workers at
Sellafield: Review of results published in October 1993. Health and Safety Executive,
Sudbury, UK, 1994.
H19 Hunt, J., T. Bailey and A. Reese. The human body retention time of environmental organically
bound tritium. J Radiol Prot 29(1): 23-36 (2009).
H20 Hunter, N., I.S. Kuznetsova, E.V. Labutina et al. Solid cancer incidence other than lung, liver
and bone in Mayak workers: 1948-2004. Br J Cancer 109(7): 1989-1996 (2013).
I1 IAEA. Environmental Isotope Data No. 1: World survey of isotope concentration in
precipitation (1953-1963). Technical Reports Series No. 96. International Atomic Energy
Agency, Vienna, 1969.
I2 IAEA. Management of tritium at nuclear facilities. Technical Reports Series No. 234.
International Atomic Energy Agency, Vienna, 1984.
I3 IAEA. Safe handling of tritium. Review of data and experience. Technical Reports Series No.
324. International Atomic Energy Agency, Vienna, 1991.
I4 IAEA. Radiation protection and safety of radiation sources: International basic safety
standards. General safety requirements Part 3. IAEA Safety Standards Series No. GSR Part 3.
International Atomic Energy Agency, Vienna, 2014.
I5 IARC. Direct estimates of cancer mortality due to low doses of ionising radiation: an
international study. IARC Study Group on Cancer Risk among Nuclear Industry Workers.
Lancet 344(8929): 1039-1043 (1994).
I6 Ichimasa, Y., M. Ichimasa, T. Shiba et al. Fixation of tritium gas by rats. Radiat Prot Dosim
16(1-2): 127-130 (1986).
I7 ICRP. Report on the task group on reference man. ICRP Publication 23. International
Commission on Radiological Protection, Pergamon Press, Oxford, 1975.
I8 ICRP. Limits for intakes of radionuclides by workers. ICRP Publication 30 (Part 1). Annals of
the ICRP 2(3-4). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1979.
348 UNSCEAR 2016 REPORT

I9 ICRP. Limits for intakes of radionuclides by workers. ICRP Publication 30 (Part 2). Annals of
the ICRP 4(3-4). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1980.
I10 ICRP. Limits for intakes of radionuclides by workers. ICRP Publication 30 (Part 3). Annals of
the ICRP 6(2-3). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1981.
I11 ICRP. Nonstochastic effects of ionizing radiation. ICRP Publication 41. Annals of the ICRP
14(3). International Commission on Radiological Protection, Pergamon Press, Oxford, 1984.
I12 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 1.
ICRP Publication 56. Annals of the ICRP 20(2). International Commission on Radiological
Protection, Pergamon Press, Oxford, 1990.
I13 ICRP. 1990 Recommendations of the International Commission on Radiological Protection.
ICRP Publication 60. Annals of the ICRP 21(1-3). International Commission on Radiological
Protection, Pergamon Press, Oxford, 1991.
I14 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 2.
Ingestion dose coefficients. ICRP Publication 67. Annals of the ICRP 23(3-4). International
Commission on Radiological Protection, Pergamon Press, Oxford, 1993.
I15 ICRP. Dose coefficients for intakes of radionuclides by workers. ICRP Publication 68. Annals
of the ICRP 24(4). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1994.
I16 ICRP. Human respiratory tract model for radiological protection. ICRP Publication 66. Annals
of the ICRP 24(1-3). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1994.
I17 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 4.
Inhalation dose coefficients. ICRP Publication 71. Annals of the ICRP 25(3-4). International
Commission on Radiological Protection, Pergamon Press, Oxford, 1995.
I18 ICRP. Doses to the embryo and fetus from intakes of radionuclides by the mother. ICRP
Publication 88. Annals of the ICRP 31(1-3). International Commission on Radiological
Protection, Pergamon Press, Oxford, 2001.
I19 ICRP. Basic anatomical and physiological data for use in radiological protection: reference
values. ICRP Publication 89. Annals of the ICRP 32(3-4). International Commission on
Radiological Protection, Pergamon Press, Oxford, 2002.
I20 ICRP. Relative biological effectiveness (RBE), quality factor (Q), and radiation weighting
factor (wR). ICRP Publication 92. Annals of the ICRP 33(4). International Commission on
Radiological Protection, Pergamon Press, Oxford, 2003.
I21 ICRP. Doses to infants from ingestion of radionuclides in mothers' milk. ICRP Publication 95.
Annals of the ICRP 34(3-4). International Commission on Radiological Protection, Elsevier
Ltd., 2004.
I22 ICRP. Assessing dose of the representative person for the purpose of the radiation protection
of the public. ICRP Publication 101a. Annals of the ICRP 36(3). International Commission on
Radiological Protection, Elsevier, 2006.
I23 ICRP. The 2007 Recommendations of the International Commission on Radiological
Protection. ICRP Publication 103. Annals of the ICRP 37(2-4). International Commission on
Radiological Protection, Elsevier Ltd., 2007.
I24 ICRU. Linear energy transfer. ICRU Report 16. International Commission on Radiation Units
and Measurements, Washington, D.C., 1970.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 349

I25 ICRU. The quality factor in radiation protection. ICRU Report 40. International Commission
on Radiation Units and Measurements, Bethesda, 1986.
I26 Ijiri, K. Cell death (apoptosis) in mouse intestine after continuous irradiation with gamma rays
and with beta rays from tritiated water. Radiat Res 118(1): 180-191 (1989).
J1 Jacobs, D.G. Sources of tritium and its behavior upon release to the environment. U.S. Atomic
Energy Commission, Division of Technical Information, Tennessee, 1968.
J2 Jain, N. and A.L. Bhatia. Radiobiological effects of low doses of tritiated water on developing
mouse cerebellum from 17th day post-coitum. Indian J Exp Biol 34(9): 891-894 (1996).
J3 Johnson, J.R. The estimation of the effective dose equivalent from tritiated water exposures
using tritium concentrations in urine. Radiat Prot Dosim 2(4): 245-247 (1982).
J4 Johnson, J.R. and D.W. Dunford. Dosimetric models of 3H from skin absorption following
contact with T2-contaminated surfaces. Health Phys 48(1): 110-113 (1985).
J5 Johnson, J.R., D.K. Myers, J.S. Jackson et al. Relative biological effectiveness of tritium for
induction of myeloid leukemia in CBA/H mice. Radiat Res 144(1): 82-89 (1995).
J6 Johnson, K.C. and J. Rouleau. Tritium releases from the pickering nuclear generating station
and birth defects and infant mortality in nearby communities 1971-1988. INFO-0401. Atomic
Energy Control Board, Ottawa, Canada, 1991.
J7 Johnson, P., W.D. Atkinson and J.L. Nicholls. Updated analysis of mortality in workers at UK
atomic weapons establishments. Proceedings of the SRP Sixth International Symposium:
Achievements & Challenges: Advancing Radiation Protection into the 21st Century,
Southport. Society for Radiological Protection, London, 1999.
J8 Jones, D.C., J.S. Krebs, D.P. Sasmore et al. Evaluation of neonatal squirrel monkeys receiving
tritiated water throughout gestation. Radiat Res 83(3): 592-606 (1980).
K1 Kacena, V. Chemical effects of decay of incorporated radioisotopes. pp.190-210 in: Biological
Effects of Transmutation and Decay of Incorporated Radioisotopes, Proceedings of a Panel,
Vienna, 9-13 October 1967. International Atomic Energy Agency, Vienna, 1968.
K2 Kakiuchi, H., N. Akata, H. Hasegawa et al. Concentration of (3)H in plants around Fukushima
Dai-ichi Nuclear Power Station. Sci Rep 2: 947 (2012).
K3 Kamiguchi, Y., H. Tateno and K. Mikamo. Dose-response relationship for the induction of
structural chromosome aberrations in human spermatozoa after in vitro exposure to tritium
beta-rays. Mutat Res 228(2): 125-131 (1990).
K4 Kamiguchi, Y., H. Tateno and K. Mikamo. Types of structural chromosome aberrations and their
incidences in human spermatozoa X-irradiated in vitro. Mutat Res 228(2): 133-140 (1990).
K5 Kazachenok, N., I. Popova, V. Melnikov et al. Pattern of 3H distribution in surface water
bodies and drinking water supply sources in the area of PA Mayak influence. ANRI 3: 43-51
(2013). (Russian).
K6 Kazachenok, N., V. Kostyuchenko, I. Popova et al. Contemporary levels of radioactive
contamination of environmental objects at EURT and other territories in the area of PA Mayak
influence. Radiation Safety Issues 1: 33-48 (2014). (Russian).
K7 Kellerer, A.M. and H.H. Rossi. RBE and the primary mechanism of radiation action. Radiat
Res 47(1): 15-34 (1971).
K8 Kellerer, A.M., Y.M. Lam and H.H. Rossi. Biophysical studies with spatially correlated ions.
4. Analysis of cell survival data for diatomic deuterium. Radiat Res 83(3): 511-528 (1980).
K9 Kellerer, A.M. Electron spectra and the RBE of X rays. Radiat Res 158(1): 13-22 (2002).
350 UNSCEAR 2016 REPORT

K10 Kelly, S.J. and J. Rossant. The effect of short-term labelling in (3H) thymidine on the viability
of mouse blastomeres: alone and in combination with unlabelled blastomeres. J Embryol Exp
Morphol 35(1): 95-106 (1976).
K11 Kember, N.F. and B.E. Lambert. Slowly cycling cells in growing bone. Cell Tissue Kinet
14(3): 327-330 (1981).
K12 Killen, H.M. and J. Carroll. The effects of tritium on embryo development: the embryotoxic
effects of [3H]tryptophan. Int J Radiat Biol 56(2): 139-149 (1989).
K13 Kim, C.K. and M.J. Han. Dose assessment and behavior of tritium in environmental samples
around Wolsong nuclear power plant. Appl Radiat Isot 50(4): 783-791 (1999).
K14 Kim, S.B., N. Baglan and P.A. Davis. Current understanding of organically bound tritium
(OBT) in the environment. J Environ Radioact 126: 83-91 (2013).
K15 Kirchmann, R., S. Bonotto, S.D. Soman et al. Transfer and incorporation of tritium in aquatic
organisms. pp.187-203 in: Behaviour of Tritium in the Environment. Proceedings series.
International Atomic Energy Agency, Vienna, 1979.
K16 Kirillova, E.N. and V.M. Luzanov. Mouse immune response to prolonged tritium oxide intake.
Radiobiologiia 20(4): 560-565 (1980). (Russian).
K17 Kirillova, E.N. Normal killer function in CBA mice as affected by long-term intake of tritium
oxide. Radiobiologiia 25(6): 792-795 (1985). (Russian).
K18 Kirillova, E.N., V.M. Manko and K.N. Muksinova. Recovery of humoral immunity parameters
in mice under a long-term action of tritium oxide. Immunologiya 2: 38-41 (1986). (Russian).
K19 Kirillova, E.N. The immunity indices of rats after the long-term action of tritium oxide or
gamma irradiation. Radiobiologiia 30(2): 175-178 (1990). (Russian).
K20 Kocher, D.C., A.I. Apostoaei and F.O. Hoffman. Radiation effectiveness factors for use in
calculating probability of causation of radiogenic cancers. Health Phys 89(1): 3-32 (2005).
K21 Komatsu, K., Y. Okumura and K. Sakamoto. Radiation dose to mouse liver cells from
ingestion of tritiated food or water. Health Phys 58(5): 625-629 (1990).
K22 Kowalska, M. Incorporation of tritiated water (HTO) and organically bound tritium (OBT)
into phospholipids and gangliosides of rat brain. J Radiat Res 26(4): 385-394 (1985).
K23 Kowalska, M. Incorporation of tritiated water (HTO) or organically bound tritium (OBT) into
amino acids of rat brain proteins. J Radiat Res 26(1): 99-108 (1985).
K24 Kozlowski, R., S.D. Bouffler, J.W. Haines et al. In utero haemopoietic sensitivity to alpha,
beta or X-irradiation in CBA/H mice. Int J Radiat Biol 77(7): 805-815 (2001).
L1 Lambert, B.E. and R.J. Clifton. Radiation doses resulting from the administration of tritiated
folic acid and tritiated water to the rat. Br J Radiol 40(469): 56-61 (1967).
L2 Lambert, B.E. and R.J. Clifton. Radiation doses resulting from the ingestion of tritiated
thymidine by the rat. Health Phys 15(1): 3-9 (1968).
L3 Lambert, B.E. Cytological damage produced in the mouse testes by tritiated thymidine,
tritiated water and x-rays. Health Phys 17(4): 547-557 (1969).
L4 Lambert, B.E., H.B. Sharpe and K.B. Dawson. An accidential intake of tritiated water. Am Ind
Hyg Assoc J 32(10): 682-686 (1971).
L5 Lambert, B.E. and J. Vennart. Radiation doses received by workers using tritium in industry.
Health Phys 22(1): 23-30 (1972).
L6 Lambert, B.E. and M.L. Phipps. Some effects of irradiation of mice in utero with tritiated
compounds. Curr Top Radiat Res Q 12(1-4): 197-211 (1978).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 351

L7 Laskey, J.W., J.L. Parrish and D.F. Cahill. Some effects of lifetime parental exposure to low
levels of tritium on the F2 generation. Radiat Res 56(1): 171-179 (1973).
L8 Leggett, R.W., A. Bouville and K.F. Eckerman. Reliability of the ICRPs systemic biokinetic
models. Radiat Prot Dosim 79(1-4): 335-342 (1998).
L9 Leuraud, K., D.B. Richardson, E. Cardis et al. Ionising radiation and risk of death from
leukaemia and lymphoma in radiation-monitored workers (INWORKS): an international
cohort study. Lancet Haematol 2(7): e276-e281 (2015).
L10 Libby, W.F. Atmospheric helium three and radiocarbon from cosmic radiation. Phys Rev
69(11-12): 671-672 (1946).
L11 Little, J.B. Induction of neoplastic transformation by low-dose-rate exposure to tritiated water.
Radiat Res 107(2): 225-233 (1986).
L12 Little, M.P. and B.E. Lambert. Systematic review of experimental studies on the relative
biological effectiveness of tritium. Radiat Environ Biophys 47(1): 71-93 (2008).
L13 Lushbaugh, C.C. and R.C. Ricks. Some cytokinetic and histopathologic considerations of
irradiated male and female gonadal tissues. Front Radiat Ther Oncol 6: 228-248 (1972).
L14 Lushbaugh, C.C. and G.W. Casarett. The effects of gonadal irradiation in clinical radiation
therapy: a review. Cancer 37(2 Suppl): 1111-1125 (1976).
M1 Martin, J.R. and J.J. Koranda. Biological half-life studies of tritium in chronically exposed
kangaroo rats. Radiat Res 50(2): 426-440 (1972).
M2 Mathur-De Vre, R. and J. Binet. Molecular aspects of tritiated water and natural water in
radiation biology. Prog Biophys Mol Biol 43(2): 161-193 (1984).
M3 Matsuda, Y., T. Yamada and I. Tobari. Chromosome aberrations induced by tritiated water or
60
Co gamma-rays at early pronuclear stage in mouse eggs. Mutat Res 160(2): 87-93 (1986).
M4 Matsumoto, T., T. Maruoka, G. Shimoda et al. Tritium in Japanese precipitation following the
March 2011 Fukushima Daiichi Nuclear Plant accident. Sci Total Environ 445-446: 365-370
(2013).
M5 McGeoghegan, D. and K. Binks. The mortality and cancer morbidity experience of workers at
the Capenhurst uranium enrichment facility 1946-95. J Radiol Prot 20(4): 381-401 (2000).
M6 McGeoghegan, D. and K. Binks. The mortality and cancer morbidity experience of employees
at the Chapelcross plant of British Nuclear Fuels plc, 1955-95. J Radiol Prot 21(3): 221-250
(2001).
M7 McGeoghegan, D., M. Gillies, A.E. Riddell et al. Mortality and cancer morbidity experience of
female workers at the British Nuclear Fuels Sellafield plant, 1946-1998. Am J Ind Med 44(6):
653-663 (2003).
M8 McGeoghegan, D., K. Binks, M. Gillies et al. The non-cancer mortality experience of male
workers at British Nuclear Fuels plc, 1946-2005. Int J Epidemiol 37(3): 506-518 (2008).
M9 McLaughlin, J.R., W.D. King, T.W. Anderson et al. Paternal radiation exposure and leukaemia
in offspring: the Ontario case-control study. Br Med J 307(6910): 959-966 (1993).
M10 Melintescu, A., D. Galeriu and H. Takeda. Reassessment of tritium dose coefficients for the
general public. Radiat Prot Dosim 127(1-4): 153-157 (2007).
M11 Metz-Flamant, C., O. Laurent, E. Samson et al. Mortality associated with chronic external
radiation exposure in the French combined cohort of nuclear workers. Occup Environ Med
70(9): 630-638 (2013).
M12 Mewissen, D.J., A.S. Ugarte and J.H. Rust. Genetic effects from exposure of male mice to
tritium for six generations. Radiat Res 70: 629-640 (1977).
352 UNSCEAR 2016 REPORT

M13 Mewissen, D.J. and A.S. Ugarte. Cumulative genetic effects from exposure of male mice to
tritium for ten generations. pp.215-229 in: Proceedings of International Symposium on
Biological Implications of Radionuclides Released from Nuclear Industries, Vienna, 26-30
March 1979. International Atomic Energy Agency, Vienna, 1979.
M14 Milacic, S. Changes in leukocytes caused by tritium contamination. Health Phys 86(5):
457-459 (2004).
M15 Minder, W. Internal contamination with tritium. Strahlentherapie 137(6): 700-704 (1969).
(German).
M16 Moghissi, A.A., M.W. Carter and E.W. Bretthauer. Further studies on the long-term evaluation
of the biological half-life of tritium. Health Phys 23(6): 805-806 (1972).
M17 Moiseenko, V.V., A.J. Walker and W.V. Prestwich. Energy deposition pattern from tritium
and different energy photons--a comparative study. Health Phys 73(2): 388-392 (1997).
M18 Moiseenko, V.V., R.N. Hamm, A.J. Waker et al. Calculation of radiation-induced DNA
damage from photons and tritium beta-particles. Part I: Model formulation and basic results.
Radiat Environ Biophys 40(1): 23-31 (2001).
M19 Moiseenko, V.V., A.J. Waker, R.N. Hamm et al. Calculation of radiation-induced DNA
damage from photons and tritium beta-particles. Part II: Tritium RBE and damage complexity.
Radiat Environ Biophys 40(1): 33-38 (2001).
M20 Morstin, K., M. Kopec, P. Olko et al. Microdosimetry of tritium. Health Phys 65(6): 648-656
(1993).
M21 Moseeva, M.B., T.V. Azizova, E.S. Grigoryeva et al. Risks of circulatory diseases among
Mayak PA workers with radiation doses estimated using the improved Mayak Worker
Dosimetry System 2008. Radiat Environ Biophys 53(2): 469-477 (2014).
M22 Muirhead, C.R., J.A. O'Hagan, R.G. Haylock et al. Mortality and cancer incidence following
occupational radiation exposure: third analysis of the National Registry for Radiation Workers.
Br J Cancer 100(1): 206-212 (2009).
M23 Muller, W.U. and A. Spindle. Induction of sister chromatid exchange in preimplantation
mouse embryos in vitro by 3H-thymidine or ultraviolet light in combination with caffeine.
Teratog Carcinog Mutagen 6(2): 107-114 (1986).
M24 Muller, W.U., C. Streffer, M. Molls et al. Radiotoxicities of [3H]thymidine and of
[3H]arginine compared in mouse embryos in vitro. Radiat Res 110(2): 192-198 (1987).
M25 Muller, W.U., N. Heckeley and C. Streffer. Effects of cell cycle specific exposure to 3H-
thymidine or 3H-arginine on development and cell proliferation of mouse embryos. Radiat
Environ Biophys 35(4): 267-271 (1996).
M26 Myers, D.K. and J.R. Johnson. Toxicity and dosimetry of tritium: a review. INFO-0377.
Atomic Energy Control Board, Advisory Committee on Radiological Protection, Ottawa,
Canada, 1991.
N1 NCRP. Tritium and other radionuclide labeled organic compounds incorporated in genetic
material. NCRP Report No. 63. National Council on Radiation Protection and Measurements,
Bethesda, 1979.
N2 NCRP. Tritium in the environment. NCRP Report No. 62. National Council on Radiation
Protection and Measurements, Bethesda, 1979.
N3 NCRP. Influence of dose and its distribution in time on dose-response relationships for low-
LET radiations. NCRP Report No. 64. National Council on Radiation Protection and
Measurements, Bethesda, 1980.
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 353

N4 NCRP. Uncertainties in internal radiation dose assessment. NCRP Report No. 164. National
Council on Radiation Protection and Measurements, Bethesda, 2009.
N5 Nikjoo, H. and D.T. Goodhead. Track structure analysis illustrating the prominent role of low-
energy electrons in radiobiological effects of low-LET radiations. Phys Med Biol 36(2): 229-
238 (1991).
N6 NRC. Health effects of exposure to low levels of ionizing radiation. Committee on the
Biological Effects of Ionizing Radiations, BEIR V. National Research Council, National
Academy Press, Washington D.C., 1990.
N7 NRC. Health risks from exposure to low levels of ionizing radiation. Committee to Assess
Health Risks from Exposure to Low Levels of Ionizing Radiation, BEIR VII - Phase 2.
National Research Council, The National Academies Press, Washington D.C., 2006.
O1 Okada, S., K. Sakai and N. Nakamura. Relative biological effectiveness of tritiated water on
cultured mammalian cells at molecular and cellular level. Radiat Prot Dosim 16(1-2): 137-140
(1986).
O2 Okada, S. and N. Momoshima. Overview of tritium: characteristics, sources, and problems.
Health Phys 65(6): 595-609 (1993).
O3 Okladnikova, N.D., V.F. Khoryakov and E.V. Odintsova. The case of radiation sickness
caused by exposure to tritium. Med Radiol 6: 595-609 (1969). (Russian).
O4 Oliphant, M., P. Harteck and L. Rutherford. Transmutation effects observed with heavy
hydrogen. Proc R Soc London Ser A: 692-703 (1934).
O5 Omar, R.Z., J.A. Barber and P.G. Smith. Cancer mortality and morbidity among plutonium
workers at the Sellafield plant of British Nuclear Fuels. Br J Cancer 79(7-8): 1288-1301 (1999).
O6 Osborne, R.V. Absorption of tritiated water vapour by people. Health Phys 12(11): 1527-1537
(1966).
P1 Painter, R.B., R.M. Drew and W.L. Hughes. Inhibition of HeLa growth by intranuclear
tritium. Science 127(3308): 1244-1245 (1958).
P2 Paquet, F. and H. Metivier. Are the risks from tritium exposures being underestimated?
J Radiol Prot 29(2): 175-181 (2009).
P3 Peterman, B.F., J.R. Johnson, R.G. Dunford et al. Internal dosimetry of tritiated hydrogen gas.
CFFTP-G-84034. Canadian Fusion Fuels Technology Project, Ottawa, 1985.
P4 Peterman, B.F., J.R. Johnson and R.G.C. McElroy. HT/HTO conversion in mammals. Int J
Appl Radiat Isot 36(7): 600 (1985).
P5 Phipps, A.W., G.M. Kendall, T.P. Fell et al. Doses from radioactive methane. Radiat Prot
Dosim 30(3): 191-195 (1990).
P6 Pietrzak-Flis, Z., I. Radwan and L. Indeka. Tritium in rabbits after ingestion of freeze-dried
tritiated food and tritiated water. Radiat Res 76(2): 420-428 (1978).
P7 Pietrzak-Flis, Z., I. Radwan, Z. Major et al. Tritium incorporation in rats chronically exposed
to tritiated food or tritiated water for three successive generations. J Radiat Res 22(4): 434-442
(1981).
P8 Pietrzak-Flis, Z. and M. Wasilewska-Gomulka. Effect of lifetime intake of organically bound
tritium and tritiated water on the oocytes of rats. Radiat Environ Biophys 23(1): 61-68 (1984).
P9 Pinson, E.A. and W.H. Langham. Physiology and toxicology of tritium in man. J Appl Physiol
10(1): 108-126 (1957).
354 UNSCEAR 2016 REPORT

P10 Pomerantseva, M.D., M.I. Balonov, L.K. Ramaiia et al. Mutagenic action of tritium on the
germ cells of male mice. II. The genetic damages in stem spermatogonia induced by tritium
oxide and gamma radiation. Genetika 20(5): 782-787 (1984). (Russian).
P11 Povinec, P.P., M. Aoyama, D. Biddulph et al. Cesium, iodine and tritium in NW Pacific waters a
comparison of the Fukushima impact with global fallout. Biogeosciences 10(8): 5481-5496 (2013).
P12 Preston, D.L., Y. Shimizu, D.A. Pierce et al. Studies of mortality of atomic bomb survivors.
Report 13: Solid cancer and noncancer disease mortality: 1950-1997. Radiat Res 160(4):
381-407 (2003).
P13 Preston, D.L., E. Ron, S. Tokuoka et al. Solid cancer incidence in atomic bomb survivors:
1958-1998. Radiat Res 168(1): 1-64 (2007).
P14 Prosser, J.S., D.C. Lloyd, A.A. Edwards et al. The introduction of chromosome aberrations in
human lymphocytes by exposure to tritiated water in vitro. Radiat Prot Dosim 4(1): 21-26 (1983).
R1 Revina, V.S., V.S. Voronin, V.K. Lemberg et al. Comparative evaluation of the carcinogenic
effects of chronic exposure to tritium oxide and external gamma-radiation. Radiobiologiia
24(5): 697-700 (1984). (Russian).
R2 Richardson, D.B. and S. Wing. Leukemia mortality among workers at the Savannah River
Site. Am J Epidemiol 166(9): 1015-1022 (2007).
R3 Richardson, D.B., S. Wing and S. Wolf. Mortality among workers at the Savannah River Site.
Am J Ind Med 50(12): 881-891 (2007).
R4 Richardson, D.B. and S. Wing. Evidence of confounding by smoking of associations between
radiation and lung cancer mortality among workers at the Savannah River Site. Am J Ind Med
54(6): 421-427 (2011).
R5 Richardson, D.B., E. Cardis, R.D. Daniels et al. Risk of cancer from occupational exposure to
ionising radiation: retrospective cohort study of workers in France, the United Kingdom, and
the United States (INWORKS). Br Med J 351: h5359 (2015).
R6 Richardson, R.B., D.W. Dunford and S.R. Peterson. Influence of gender differences in the
carbon pool on dose factors for intakes of tritium and 14C-labeled compounds. Health Phys
81(3): 302-312 (2001).
R7 Richardson, R.B. and A. Hong. Dose to lung from inhaled tritiated particles. Health Phys
81(3): 313-324 (2001).
R8 Richardson, R.B. and D.W. Dunford. A biochemical-based model for the dosimetry of dietary
organically bound tritium--Part 2: Dosimetric evaluation. Health Phys 85(5): 539-552 (2003).
R9 Richardson, R.B. and D.W. Dunford. A biochemical-based model for the dosimetry of dietary
organically bound tritium--Part 1: Physiological criteria. Health Phys 85(5): 523-538 (2003).
R10 Richter, B.S. and H.G. Stockwell. Descriptive study of deaths from cancer associated with
residential proximity to the site of underground nuclear detonations. Arch Environ Health
53(2): 109-113 (1998).
R11 Rochalska, M. and Z. Szot. The incorporation of organically-bound tritium of food into some
organs of the rat. Int J Radiat Biol Relat Stud Phys Chem Med 31(4): 391-395 (1977).
R12 Rodgers, D.W. Tritium dynamics in mice exposed to tritiated water and diet. Health Phys
63(3): 331-337 (1992).
R13 Rodriguez-Rodrigo, L., J. Elbez-Uzan and C. Alejaldre. Tritium and workers in fusion
devices-lessons learnt. J Radiol Prot 29(3): 351-360 (2009).
R14 Rooney, C., V. Beral, N. Maconochie et al. Case-control study of prostatic cancer in employees
of the United Kingdom Atomic Energy Authority. Br Med J 307(6916): 1391-1397 (1993).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 355

R15 Rudran, K. Radiation doses to lungs and whole body from use of tritium in luminous paint
industry. Radiat Prot Dosim 25(2): 117-125 (1988).
R16 Rudran, K. Significance of in vivo organic binding of tritium following intake of tritiated
water Radiat Prot Dosim 25(1): 5-13 (1988).
R17 Russell, W.L., R.R. Cumming, E.M. Kelly et al. Induction of specific locus mutations in the
mouse by tritiated water. pp.489-497 in: Proceedings of the International Symposium on the
Behaviour of Tritium in the Environment. International Atomic Energy Agency, Vienna, 1979.
S1 Saito, M., M.R. Ishida and C.C. Travis. Dose-modification factor for accumulated dose to cell
nucleus due to protein-bound 3H. Health Phys 56(6): 869-874 (1989).
S2 Sanders Jr., S.M. and W.C. Reinig. Assessment of tritium in man. pp.534-542 in: Diagnosis
and Treatment of Deposited Radionuclides (H.A. Kornberg and W.D. Norwood, eds.).
Excerpta Medica Foundation, Amsterdam, 1968.
S3 Sasaki, M.S. Primary damage and fixation of chromosomal DNA as probed by monochromatic
soft x-rays and low-energy neutrons. pp.369-384 in: The Early Effects of Radiation on DNA
(E.M. Fielden and P. O'Neill, eds.). Springer-Verlag, Berlin, 1991.
S4 Satow, Y., H. Hori and J.Y. Lee. Teratogenic effect of fission neutron and tritium water on rat
embryo. J UOEH 11 Suppl: 416-431 (1989).
S5 Satow, Y., H. Hori, J.Y. Lee et al. Effect of tritiated water on female germ cells: mouse oocyte
killing and RBE. Int J Radiat Biol 56(3): 293-299 (1989).
S6 Schmid, E. Is there reliable experimental evidence for a low-dose RBE of about 4 for
mammography X rays relative to 200 kV X rays? Radiat Res 158(6): 778-781 (2002).
S7 Schreml, W., R.J. Haas, F. Planas-Bohne et al. Distribution and dosimetry of tritium in
newborn rats after in utero exposure to 3H-TdR. Radiat Res 58(2): 239-252 (1974).
S8 Schreml, W. and T.M. Fliedner. Distribution of tritiated compounds (tritiated thymidine and
tritiated water) in the mother-fetus system and its consequences for the radiotoxic effect of
tritium. Curr Top Radiat Res Q 12(1-4): 255-277 (1978).
S9 Schubauer-Berigan, M.K., R.D. Daniels, D.A. Fleming et al. Chronic lymphocytic leukaemia
and radiation: findings among workers at five US nuclear facilities and a review of the recent
literature. Br J Haematol 139(5): 799-808 (2007).
S10 Schubauer-Berigan, M.K., R.D. Daniels, D.A. Fleming et al. Risk of chronic myeloid and
acute leukemia mortality after exposure to ionizing radiation among workers at four U.S.
nuclear weapons facilities and a nuclear naval shipyard. Radiat Res 167(2): 222-232 (2007).
S11 Schubauer-Berigan, M.K., R.D. Daniels, S.J. Bertke et al. Cancer mortality through 2005
among a pooled cohort of U.S. nuclear workers exposed to external ionizing radiation. Radiat
Res 183(6): 620-631 (2015).
S12 Seelentag, W. Two cases of tritium fatality. pp.267-280 in: Tritium (A.A. Moghissi and M.W.
Carter, eds.). Messenger Graphics, Phoenix, 1973.
S13 Seyama, T., O. Yamamoto, A. Kinomura et al. Carcinogenic effects of tritiated water (HTO) in
mice: in comparison to those of neutrons and gamma-rays. J Radiat Res 32 Suppl 2: 132-142
(1991).
S14 Smirnov, D.G., E.N. Kirillova and K.N. Muksinova. The early changes in humoral immunity
under the prolonged action of tritium oxide with different dose rates. Radiobiologiia 30(1):
129-133 (1990). (Russian).
S15 Smith, P.G. and A.J. Douglas. Mortality of workers at the Sellafield plant of British Nuclear
Fuels. Br Med J (Clin Res Ed) 293(6551): 845-854 (1986).
356 UNSCEAR 2016 REPORT

S16 Snow, M.H. Abnormal development of pre-implantation mouse embryos grown in vitro with
(3 H) thymidine. J Embryol Exp Morphol 29(3): 601-615 (1973).
S17 Snyder, W.S., B.R. Fish, S.R. Bernard et al. Urinary excretion of tritium following exposure of
man to HTO--a two exponential model. Phys Med Biol 13(4): 547-559 (1968).
S18 Sokolnikov, M., D. Preston, E. Gilbert et al. Radiation effects on mortality from solid cancers
other than lung, liver, and bone cancer in the Mayak worker cohort: 1948-2008. PLoS One
10(2): e0117784 (2015).
S19 Sont, W.N., J.M. Zielinski, J.P. Ashmore et al. First analysis of cancer incidence and
occupational radiation exposure based on the National Dose Registry of Canada. Am J
Epidemiol 153(4): 309-318 (2001).
S20 Soyfer, V.N., V. Goryachev, S. Vakulovsky et al. Tritium Studies of Russia's Natural Waters.
Geos, Russia, 2007. (Russian).
S21 Spindle, A., K. Wu and R.A. Pedersen. Sensitivity of early mouse embryos to [3H]thymidine.
Exp Cell Res 142(2): 397-405 (1982).
S22 Straume, T. Tritium risk assessment. Health Phys 65(6): 673-682 (1993).
S23 Straume, T. and A.L. Carsten. Tritium radiobiology and relative biological effectiveness.
Health Phys 65(6): 657-672 (1993).
S24 Streffer, C., D. van Beuningen and S. Elias. Comparative effects of tritiated water and
thymidine on the preimplanted mouse embryo in vitro. Curr Top Radiat Res Q 12(1-4): 182-
193 (1978).
S25 Sun, X.Z., M. Inouye, H. Yamamura et al. Effects of prenatal treatment with tritiated water on
the developing brain in mouse. Int J Radiat Biol 71(3): 309-313 (1997).
T1 Takeda, H. and Y. Kasida. Biological behavior of tritium after administration of tritiated water
in the rat. J Radiat Res 20(2): 174-185 (1979).
T2 Takeda, H. Comparative metabolism of tritium in rat after single ingestion of some tritiated
organic compounds versus tritiated water. J Radiat Res 23(3): 345-357 (1982).
T3 Takeda, H., K. Arai and T. Iwakura. Comparison of tritium metabolism in rat following single
or continuous ingestion of tritium labeled wheat versus tritiated water. J Radiat Res 26(1):
130-139 (1985).
T4 Takeda, H. Incorporation and distribution of tritium in rats after chronic exposure to various
tritiated compounds. Int J Radiat Biol 59(3): 843-853 (1991).
T5 Takeda, H., Y. Nishimura and J. Inaba. Transfer of tritium to prenatal and neonatal rats from
their mothers exposed to tritiated compounds. Radiat Prot Dosim 53(1-4): 281-284 (1994).
T6 Takeda, H., H.M. Lu, K. Miyamoto et al. Comparative biokinetics of tritium in rats during
continuous ingestion of tritiated water and tritium-labeled food. Int J Radiat Biol 77(3): 375-
381 (2001).
T7 Tanaka, K., S. Sawada and N. Kamada. Relative biological effectiveness and dose rate effect
of tritiated water on chromosomes in human lymphocytes and bone marrow cells. Mutat Res
323(1-2): 53-61 (1994).
T8 Taylor, D.M., J.P. Moroni, J.O. Snihs et al. The metabolism of 3H and 14C with special
reference to radiation protection. Radiat Prot Dosim 30(2): 87-93 (1990).
T9 Taylor, D.M. A biokinetic model for predicting the retention of 3H in the human body after
intakes of tritiated water. Radiat Prot Dosim 105(1-4): 225-228 (2003).
T10 Taylor, D.M. Radiation doses from some [3H]-labelled organic compounds following
ingestion. Radiat Prot Dosim 128(3): 299-308 (2008).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 357

T11 Thierry-Chef, I., M. Marshall, J.J. Fix et al. The 15-Country Collaborative Study of Cancer
Risk among Radiation Workers in the Nuclear Industry: study of errors in dosimetry. Radiat
Res 167(4): 380-395 (2007).
T12 Thierry-Chef, I., D.B. Richardson, R.D. Daniels et al. Dose estimation for a study of nuclear
workers in France, the United Kingdom and the United States of America: methods for the
International Nuclear Workers Study (INWORKS). Radiat Res 183(6): 632-642 (2015).
T13 Trivedi, A. Skin-contact exposure to tritium-gas-contaminated stainless-steel surfaces. Health
Phys 65(5): 514-522 (1993).
T14 Trivedi, A., D. Galeriu and R.B. Richardson. Dose contribution from metabolized organically
bound tritium after acute tritiated water intakes in humans. Health Phys 73(4): 579-586 (1997).
T15 Trivedi, A. and N.E. Gentner. Dosimetry and health effects of tritium in the CANDU business.
Atomic Energy of Canada Limited, Canada, 1999.
T16 Trivedi, A., D. Galeriu and E.S. Lamothe. Dose contribution from metabolized organically
bound tritium after chronic tritiated water intakes in humans. Health Phys 78(1): 2-7 (2000).
T17 Tuschl, H., F. Steger and R. Kovac. Occupational exposure and its effect on some immune
parameters. Health Phys 68(1): 59-66 (1995).
U1 Ueno, A.M., I. Furuno-Fukushi and H. Matsudaira. Induction of cell killing, micronuclei, and
mutation to 6-thioguanine resistance after exposure to low-dose-rate gamma rays and tritiated
water in cultured mammalian cells (L5178Y). Radiat Res 91(3): 447-456 (1982).
U2 Ujeno, Y. Relative biological effectiveness (RBE) of tritium beta rays in relation to dose rate.
Health Phys 45(3): 789-791 (1983).
U3 Ujeno, Y., K. Yamamoto, T. Aoki et al. Tritium content in tissue free water of Japanese
bodies. Radiat Prot Dosim 16 (1-2): 181-183 (1986).
U4 Umata, T., N. Kunugita and T. Norimura. A comparison of the mutagenic and apoptotic
effects of tritiated water and acute or chronic caesium-137 gamma exposure on spleen T
lymphocytes on normal and p53-deficient mice. Int J Radiat Biol 85(12): 1082-1088 (2009).
U5 UNSCEAR. Ionizing Radiation: Sources and Biological Effects. UNSCEAR 1982 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1982 Report to the
General Assembly, with annexes. United Nations sales publication E.82.IX.8. United Nations,
New York, 1982.
U6 UNSCEAR. Sources, Effects and Risks of Ionizing Radiation. UNSCEAR 1988 Report.
United Nations Scientific Committee on the Effects of Atomic Radiation, 1988 Report to the
General Assembly, with annexes. United Nations sales publication E.88.IX.7. United Nations,
New York, 1988.
U7 UNSCEAR. Sources and Effects of Ionizing Radiation. UNSCEAR 1993 Report. United
Nations Scientific Committee on the Effects of Atomic Radiation, 1993 Report to the General
Assembly, with scientific annexes. United Nations sales publication E.94.IX.2. United
Nations, New York, 1993.
U8 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume II: Effects. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.4.
United Nations, New York, 2000.
U9 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.3.
United Nations, New York, 2000.
358 UNSCEAR 2016 REPORT

U10 UNSCEAR. Hereditary Effects of Radiation. UNSCEAR 2001 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation, 2001 Report to the General
Assembly, with scientific annex. United Nations sales publication E.01.IX.2. United Nations,
New York, 2001.
U11 UNSCEAR. Effects of Ionizing Radiation. Volume I: Report to the General Assembly,
Scientific Annexes A and B. UNSCEAR 2006 Report. United Nations Scientific Committee
on the Effects of Atomic Radiation. United Nations sales publication E.08.IX.6. United
Nations, New York, 2008.
U12 UNSCEAR. Effects of Ionizing Radiation. Volume II: Scientific Annexes C, D and E.
UNSCEAR 2006 Report. United Nations Scientific Committee on the Effects of Atomic
Radiation. United Nations sales publication E.09.IX.5. United Nations, New York, 2009.
V1 Virsik, R.P., C. Schafer, D. Harder et al. Chromosome aberrations induced in human
lymphocytes by ultrasoft Al (K) and C (K) X-rays. Int J Radiat Biol Relat Stud Phys Chem
Med 38(5): 545-557 (1980).
V2 Vrijheid, M., E. Cardis, M. Blettner et al. The 15-Country Collaborative Study of Cancer Risk
Among Radiation Workers in the Nuclear Industry: design, epidemiological methods and
descriptive results. Radiat Res 167(4): 361-379 (2007).
V3 Vulpis, N. The induction of chromosome aberrations in human lymphocytes by in vitro
irradiation with beta particles from tritiated water. Radiat Res 97(3): 511-518 (1984).
W1 Wakeford, R., S.C. Darby and M.F. Murphy. Temporal trends in childhood leukaemia
incidence following exposure to radioactive fallout from atmospheric nuclear weapons testing.
Radiat Environ Biophys 49(2): 213-227 (2010).
W2 Wakeford, R. The risk of leukaemia in young children from exposure to tritium and carbon-14
in the discharges of German nuclear power stations and in the fallout from atmospheric
nuclear weapons testing. Radiat Environ Biophys 53(2): 365-379 (2014).
W3 Wang, B. and X.Y. Zhou. Effects of prenatal exposure to low-dose beta radiation from tritiated
water on the neurobehavior of mice. J Radiat Res 36(2): 103-111 (1995).
W4 Wang, B., K. Watanabe, T. Yamada et al. Effects of beta radiation from organically bound
tritium on cultured mouse embryonic mid brain cells. Health Phys 71(6): 915-921 (1996).
W5 Wanigaratne, S., E. Holowaty, H. Jiang et al. Estimating cancer risk in relation to tritium
exposure from routine operation of a nuclear-generating station in Pickering, Ontario. Chronic
Dis Inj Can 33(4): 247-256 (2013).
W6 Wawerna, J.C. Biological implications of the application of tritiated luminous compounds.
pp.356-363 in: Tritium (A.A. Moghissi and M.W. Carter, eds.). Messenger Graphics, Phoenix,
1973.
W7 WHO. Guidelines for drinking-water quality. Fourth edition. World Health Organization,
Geneva, 2011.
W8 Wood, M.J., R.G. McElroy, R.A. Surette et al. Tritium sampling and measurement. Health
Phys 65(6): 610-627 (1993).
W9 Wylie, K.F., W.A. Bigler and G.R. Grove. Biological half-life of tritium. Health Phys 9(9):
911-914 (1963).
Y1 Yamada, T., O. Yukawa, K. Asami et al. Effect of chronic HTO beta or 60Co gamma radiation
on preimplantation mouse development in vitro. Radiat Res 92(2): 359-369 (1982).
Y2 Yamamoto, O., K. Yokoro, T. Seyama et al. HTO oral administration in mice. I: Threshold
dose rate for haematopoietic death. Int J Radiat Biol 57(3): 543-549 (1990).
 ANNEX C: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSTRITIUM 359

Y3 Yamamoto, O., T. Seyama, T. Jo et al. Oral administration of tritiated water (HTO) in mouse.
II. Tumour development. Int J Radiat Biol 68(1): 47-54 (1995).
Y4 Yamamoto, O., T. Seyama, H. Itoh et al. Oral administration of tritiated water (HTO) in
mouse. III: Low dose-rate irradiation and threshold dose-rate for radiation risk. Int J Radiat
Biol 73(5): 535-541 (1998).
Y5 Yin, H., D. Bhattacharjee, G. Roy et al. Tumorigenesis in infant C3H/HeN mice exposed to
tritiated water (HTO). J Radiat Res 43(4): 345-351 (2002).
Y6 Yokoro, K., O. Yamamoto, T. Seyama et al. Acute and chronic effects of tritiated water in
mice with special reference to its carcinogenicity: an interim report. Radiat Prot Dosim
16(1-2): 165-168 (1986).
Y7 Yoon, S., W.H. Ha and S.S. Lee. Tritium analysis of urine samples from the general Korean
public. Appl Radiat Isot 81: 276-278 (2013).
Z1 Zablotska, L.B., J.P. Ashmore and G.R. Howe. Analysis of mortality among Canadian nuclear
power industry workers after chronic low-dose exposure to ionizing radiation. Radiat Res
161(6): 633-641 (2004).
Z2 Zablotska, L.B., R.S. Lane and P.A. Thompson. A reanalysis of cancer mortality in Canadian
nuclear workers (1956-1994) based on revised exposure and cohort data. Br J Cancer 110(1):
214-223 (2014).
Z3 Zamenhof, S. and E. van Marthens. The effects of chronic ingestion of tritiated water on
prenatal brain development. Radiat Res 77(1): 117-127 (1979).
Z4 Zamenhof, S. and E. van Marthens. The effects of pre- and postnatal exposure to tritiated
water for five generations on postnatal brain development. Radiat Res 85(2): 292-301 (1981).
Z5 Zhou, X.Y., J.C. Dong, X.S. Geng et al. Tritium beta-ray and 60CO gamma-ray caused
dominant lethal mutation in mice. Chin Med J 99(5): 420-423 (1986).
Z6 Zhou, X.Y., J.C. Dong, S.Y. Zhou et al. Experimental study on relative biological
effectiveness of tritium and risk estimates of genetic damage. Chin Med J 102(11): 872-878
(1989).
Z7 Zhou, Y. and Y.S. Cheng. Dose assessment for inhaling hafnium particles based on laboratory
rats study. Health Phys 84(4): 469-476 (2003).
Z8 Zhou, Y. and Y.S. Cheng. Dosimetry of metal tritide particles as evaluated by the ICRP 66
model and a biokinetic model from laboratory rats. Health Phys 86(2): 155-160 (2004).
Z9 Zhou, Y., Y.S. Cheng and Y. Wang. Dissolution rate and biokinetic model of zirconium tritide
particles in rat lungs. Health Phys 98(5): 672-682 (2010).
 ANNEX D

BIOLOGICAL EFFECTS OF SELECTED INTERNAL

EMITTERSURANIUM

361
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 363

CONTENTS
I. INTRODUCTION ................................................................................................................................. 365
II. SOURCES AND LEVELS .................................................................................................................... 367
A. Natural sources .......................................................................................................................... 367
B. Artificial sources ........................................................................................................................ 374
III. PHYSICAL, RADIOLOGICAL AND CHEMICAL CHARACTERISTICS ..................................... 378
A. Physical and radiological characteristics .......................................................................... 378
B. Chemical characteristics......................................................................................................... 379
IV. HUMAN EXPOSURE .......................................................................................................................... 380
A. Exposure of members of the public ................................................................................... 380
B. Occupational exposure .......................................................................................................... 382
C. Measurement of uranium ...................................................................................................... 382
V. BIOKINETICS........................................................................................................................................ 383
A. Inhalation..................................................................................................................................... 384
B. Ingestion ...................................................................................................................................... 385
C. Absorption .................................................................................................................................. 386
D. Systemic distribution, retention and excretion ............................................................. 386
E. Materno-fetal transfer ............................................................................................................. 387
VI. DOSIMETRY ......................................................................................................................................... 388
VII. BIOLOGICAL EFFECTS ...................................................................................................................... 390
A. Chemical versus radiological toxicity ................................................................................ 390
B. Kidneys ......................................................................................................................................... 391
C. Bone............................................................................................................................................... 397
D. Lungs ............................................................................................................................................. 399
E. Liver ............................................................................................................................................... 400
F. Brain............................................................................................................................................... 401
G. Reproduction and development ........................................................................................ 405
H. Other organs .............................................................................................................................. 408
I. Relative biological effectiveness ......................................................................................... 414
J. Lethal effects .............................................................................................................................. 414
364 UNSCEAR 2016 REPORT

VIII.EPIDEMIOLOGICAL STUDIES......................................................................................................... 416

A. Studies of occupational exposure ...................................................................................... 416
B. Studies of Gulf War veterans ................................................................................................ 433
C. Studies of environmental exposure................................................................................... 434
IX. RESEARCH NEEDS ............................................................................................................................. 437
X. GENERAL CONCLUSION ................................................................................................................. 438
XI. ACKNOWLEDGEMENTS .................................................................................................................. 439
APPENDIX A. TABLES SUMMARIZING URANIUM LEVELS IN WATER, STUDIES OF
OCCUPATIONAL AND ENVIRONMENTAL EXPOSURE TO URANIUM ...................................... 441
REFERENCES............................................................................................................................................... 467
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 365

I. INTRODUCTION
1. This annex provides a review of the scientific literature on characteristics of uranium, its
biokinetics and dosimetry within the human body for various physical and chemical forms and routes of
intake into the body, radiobiological and toxicological effects of exposure to uranium, and
epidemiological studies of nuclear workers and the public who have been exposed to uranium.

2. Uranium was discovered by Martin Heinrich Klaproth in 1789 and its radioactive properties by
Antoine Henri Becquerel in 1896. Uranium, element 92 in the periodic table, is present naturally in all
rock and soil. Levels of uranium content in soil depend on local geology and range widely from a
few mg/kg up to levels of several per cent in ore bodies. The Committee in its UNSCEAR 2008 Report
[U10] reported a median activity concentration of around 30 Bq/kg (1.2 mg/kg) for uranium in rock and
soil. Uranium is released to the environment through natural events such as forest fires and volcanoes
and released from rock and soil through natural processes. It is distributed through mechanisms such as
leaching to ground and surface water and through wind erosion of soil. In turn, uranium in water, soil
and air is taken up by plants and animals. People may be exposed to uranium by inhalation of airborne
particulates, through skin uptake and through ingestion of uranium in food and water.

3. There are three naturally-occurring, alpha-particle emitting, isotopes of uranium: 238U, 234U and
235
U. Two of these, 238U and 235U, with radioactive half-lives of 4.47 109 and 7.04 108 years
respectively, are the parents of radioactive decay chains that are major contributors to the background
radiation exposure of the human population. Uranium-238 supports 14 decay products. The isotope
234
U, with a half-life of 2.45 105 years, is a member of the 238U decay chain. In natural uranium, 238U
is the most abundant isotope in terms of mass (99.2742%), while 234U and 235U constitute only 0.0054%
and 0.7204%, respectively [N8, S15]. Figure I shows a simplified radioactive decay chains for 238U.
Other isotopes, such as 232U, may be produced in thorium breeder reactors. Further, 236Uranium, with a
half-life 2.35 107 years, is present in spent nuclear fuel and in reprocessed uranium [W24], and occurs
naturally as a very small component of natural uranium (<1011% by mass).

4. Both 238U and 234U, when in secular equilibrium, contribute 48.9% of the total alpha particle
activity of natural uranium, while 235U contributes 2.2%. Some nuclear reactors require fuel that is
enriched to the fissionable isotope 235U. Current technologies for enriching natural uranium are gaseous
diffusion and centrifugation. Enrichment increases the proportion of 235U from its natural levels
(0.72%) to 25%, depending on the design requirements of nuclear power reactors. In addition, higher
enrichment levels (>90% of 235U) are achieved for use in weapons. The term depleted uranium (DU)
refers to isotopic mixtures that contain a lower percentage of 235U than is present in naturally occurring
uranium. It is recovered as a by-product of the enrichment process. The proportion of 235U in DU is
between 0.2 and 0.3%. Reprocessed uranium (especially from earlier military reprocessing) may also be
contaminated with traces of fission products and transuranic elements [W24].

5. Uranium compounds exhibit differences in their chemical and physical properties and, as a result,
also differ in their toxicological properties. For example, uranium compounds vary widely in their
solubility and this can result in differences in bioavailability following intake (via inhalation or
ingestion) into the body [A31, L9, S37, U16]. The biological and health effects of uranium are due to
its chemical and radiological toxicity. In general, this toxicity, as demonstrated in animal studies, is
caused by chemical rather than radiological components, excepting that effects induced by the isotopes
of higher specific activity and by enriched uranium are more probably due to radiation exposure.
366 UNSCEAR 2016 REPORT

6. Since 1949, many animal studies have indicated that the toxicity of uranium is due mainly to
chemical damage to the kidneys [A25]. Other systems or organs may also be affected by exposure to
uranium, such as the skeleton [A26], the lungs [L18], the gonads [A19] and the liver [P6].

Figure I. Radioactive decay chain for 238U [I17]

Half-life is expressed in a = year; d = day; h = hours; m = minutes; s = seconds

7. Uranium concentrations in environmental media may be measured in terms of radioactivity

(measured in Bq/L, e.g. by alpha spectrometry) or mass (measured in g/L, e.g. by high-resolution
inductively coupled plasma mass spectrometry). Consequently, data on uranium levels in soil, air, water
and food are indicated in Bq/L and in g/L.

8. A general concept is the relation between radioactivity and mass. As mentioned above, the
activity of each member of a chain headed by a parent radionuclide would be the same under conditions
of secular equilibrium, but the mass of each member of the chain would be quite different. The
relationship between activity, A, and mass, M, of a radionuclide is given by:

2

/

where
A = activity of a radionuclide, Bq;
A0 = Avogadros constant, 6.023 1023 atoms/mole;
AW = atomic weight of the radionuclide, kg/mole;
= decay constant, dis/(atom s);
T1/2 = half-life of the radionuclide, s;
M = mass of the radionuclide, kg.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 367

For natural uranium, the activity is 25,400 Bq/g (table 1). However, as shown in table 1, the natural
relative abundance of 234U, 235U and 238U can be expressed in terms of either numbers of atoms or
weight, giving slightly different values [K3, M31, M32].

Table 1. Mass activities of the three natural isotopes of uranium [K3]

Relative abundance
Natural uranium
atoms% (wt%) mBq/g U
238
U 99.274 (99.284) 12.40
235
U 0.720 (0.711) 0.60
234
U 0.0054 (0.0053) 12.40

II. SOURCES AND LEVELS

A. Natural sources

1. Levels in soil
9. Naturally occurring radionuclides in the environment affect the levels of background radiation
encountered at different locations around the world [N2, U8]. As mentioned by Cuney [C38], three
types of deposits contain more than three quarters of the worldwide uranium resources: unconformity-
related deposits, iron oxidecoppergold (IOCG) deposits, and sandstone-hosted deposits [L10].

10. The concentration of uranium in soil varies with location and local geology. Its concentration is
relatively low in basic rock, such as basalt, and higher in acid rock, such as the sedimentary rock
saturated with silica. The uranium content of granites is higher still [U10]. For example, the nominal
activity concentration of uranium in soil is about 15 Bq/kg of 238U (1.2 mg/kg) with a typical activity
range of 1050 Bq/kg (0.4 to 2 mg/kg) [N2, U10]. Much higher concentrations are found in uranium
mining areas such as the Northern Saskatchewan in Canada, the Colorado Plateau and central Florida
where phosphate is mined. The uranium content of phosphate rock used for phosphate fertilizers ranges
from about 50 to 2,400 Bq/kg of 238U (4190 mg/kg) [A31, N2, R20, U8]. One of the highest activity
concentrations worldwide is localized in the region of Recife in Brazil, with sedimentary rock that
contains 30500 mg/kg with an average of 150 mg/kg (1,860 (range 3726,200 Bq/kg of 238U)) [S18].
However, since uranium in soil may be more or less tightly bound depending on soil characteristics, the
uranium speciation in soil has an impact on bioaccessibility in the gut.

2. Levels in air
11. Soil particles containing uranium may be transferred into the atmosphere through natural
mechanisms. The natural uranium concentration in air is typically very low, varying from location to
location according to local ground sources [H19]. Airborne uranium can deposit on soil, plants and
open water as dry or wet deposition [A31].
368 UNSCEAR 2016 REPORT

12. Golchert et al. [G16] measured airborne concentrations of 238U of around 0.3 Bq/m3 at a site near
the Argonne National Laboratory (Illinois, United States). Average levels of natural uranium in ambient
air have been reported to be 0.25 Bq/m3 of 238U (0.02 ng/m3) in Tokyo [H19]. Tracy and Prantl [T17]
found the average concentration of 238U in air in a southern Ontario rural environment to be about
1.25 Bq/m3 (0.1 ng/m3), on the basis of measurements of 226Ra in dust and an assumption of
equilibrium between 238U and 226Ra. Taken together, these different values indicate an average uranium
level in air of around 1 Bq/m3.

13. The World Health Organization (WHO) [W14] estimates that an adult of average size inhales
20 m3 of air per day with a nominal natural uranium concentration of 0.6 Bq/m3 of 238U (0.05 ng/m3),
corresponding to 12.4 Bq (1 ng) of 238U. These values lead to a calculated annual intake through
inhalation by adults of approximately 0.0045 Bq of 238U. For comparison, tobacco smoke (from two
packages of cigarettes per day) contributes to 0.11 Bq of 238U (corresponding to 9 g) of inhaled natural
uranium per year [L49].

3. Levels in water
14. As long as uranium is inside undisturbed crystalline rock in secular equilibrium with its progeny,
the ratio of 234U to 238U is expected to be one. Nevertheless, disequilibrium can be observed when rock
is disturbed by chemical or physical processes involving water. As a result, water from any source may
contain 234U/238U ratios greater than unity because of the greater mobility and increased availability of
234
U, generally due to a reducing environment [D1]. Two isotopes in the decay series, 234Th and 234mPa,
separate the two uranium isotopes and their different solubilities in the source rock, permitting 234U to
be released preferentially and leading to variations in the ratio 234U/238U [O9]. In addition, the total
uranium activity present in water may influence the 234U/238U ratio. Indeed, Ortega et al. found that
~80% of samples with a high ratio of disequilibrium (>1.6) were linked to the lowest uranium activities
<50 mBq/L, when the samples with a low activity ratio (<1.5) corresponded to samples with high
concentrations of uranium (>200 mBq/L) [O8].

15. Uranium is present in different water sources (surface water, groundwater and drilled water) at
variable levels [W16]. In oxygenated surface water, uranium levels were found at around 0.02 to 6 g/L
(0.2576 mBq/L of 238U). In sea water, its average content is 3.3 g/L (42 mBq/L of 238U), often bound
by ligands or associated with suspended particles [B42]. Natural uranium levels were found to be
higher in Precambrian rock aquifers (average, 115.6 g/L (1.45 Bq/L of 238U)) than in Palaeozoic
sedimentary rock aquifers (average, 3.5 g/L (0.045 Bq/L of 238U)). Cothern and Lappenbusch [C34]
reviewed the available data on the occurrence of uranium in surface and groundwater supplies in the
United States and reported that surface water samples derived from about 35,000 sources had an
average uranium concentration of 18 mBq/L of 238U (1.45 g/L) (range from 0.18 to about
12,500 mBq/L of 238U (0.014982 g/L) and that about 55,000 samples of groundwater supplies had an
average uranium concentration of 55 mBq/L of 238U (4.4 g/L) (range 0.01812,000 mBq/L of 238U
(0.0014942 g/L)).

16. The WHO indicated values for uranium levels in water generally less than 12.4 mBq/L of 238U
(<1 g/L) [W15, W18]. The Agency for Toxic Substances and Disease Registry (ATSDR) reported an
average concentration of 14.4 mBq/L of 238U (1.16 g/L) [A31]. This was much higher than the
previously reported value by the Committee in its UNSCEAR 1977 Report 0.54 mBq/L of 238U
(0.044 g/L) [U7]. Thus, it might be more appropriate to report median rather than mean values
because of the large variation of uranium concentration in water. The Committee reported a variation of
natural uranium concentrations measured in drinking water samples in 16 countries of about eight
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 369

orders of magnitude [U10]. A global overview of uranium concentration (g/L) and activity (mBq/L) is
given in appendix A, in table A1 for the values measured in groundwater, in table A2 for the values
measured in surface water, in table A3 for the values measured in public water supplies and in table A4
for the values measured in bottled mineral water.

17. In Canada, different surveys aimed to measure uranium levels in drinking water in different
provinces. The mean natural uranium concentration in surface water and groundwater (some treated)
supplies was about 50 mBq/L of 238U (4 g/L) in southern-central British Colombia [P32], 124 mBq/L
of 238U (10 g/L) in south-eastern Manitoba [B21], 65 mBq/L of 238U (5.2 g/L in the Kitigan Zibi First
Nation community in Quebec) [M58], and 5 mBq/L of 238U (0.40 g/L) in Ontario [O3]. In summary,
the analysis of the different values measured in these surveys indicates that mean uranium levels in
drinking water were extremely variable in the different Canadian counties/provinces, from 5 to
750 mBq/L of 238U (0.458.3 g/L), including great internal variations depending on the precise
location. Furthermore, behind these average uranium levels in drinking water, more extreme values
were measured in some Canadian provinces or some counties in the United States as indicated in
appendix A, table A1. In fact, natural uranium concentrations as high as 8,680 mBq/L of 238U
(700 g/L) were found in private groundwater supplies [M64, M65]. A value of 25,048 mBq/L of 238U
(2,020 g/L) was measured in groundwater in south-eastern Manitoba [B21].

18. Concerning water in the United States, official reports indicated an average natural uranium
concentration of 31.6 mBq/L of 238U (2.55 g/L) in drinking water from 978 sites in the 1980s [U14].
These values are higher than those mentioned in a study by Fisenne et al. with mean activity of natural
uranium in drinking water in New York City ranging from 0.62 to 1.25 mBq/L of 238U (0.05 to
0.09 g/L) [F5]. In this study, New York city tap water had 234U, 235U and 238U activities of 1.040.19,
0.0350.010 and 0.870.18 mBq/L, respectively [F5]. Maximum values were measured in Connecticut
86.472 mBq/L of 238U (7,780 g/L). As for Canada, differences may relate to geographical variations
and local geology (appendix A, table A1).

19. In Finland, national and local surveys of uranium content in water distributed by Finnish
waterworks have been conducted. The median value was of 1.9 mBq/L of 238U (0.15 g/L) [T22]
(appendix A, table A1). An extreme value of 114,100 mBq/L of 238U (9,200 g/L) was measured in the
South of Finland [M66]. It is noteworthy that although the uranium concentrations in Finnish wells
drilled in bedrock are among the highest in the world [K26, M66, M67, P28], the uranium
concentration in water distributed by the waterworks is generally low [M67, T22].

20. In France, periodic reports address the levels of the radiological quality of drinking water [I21].
Measurements performed during 2008 and 2009 indicate that 226Ra and uranium isotopes constitute the
main contributors to a total alpha activity above 0.1 Bq/L; in this case, the mean value of uranium
concentration was 2.22 g/L (27.5 mBq/L of 238U) with a range from 0.14 to 114 g/L
(1.8-1,450 mBq/L of 238U) [I21].

21. Activity concentrations of natural radionuclides in soil, food, natural and drinking water were also
measured in China [P2]. However, measurements of uranium were not reported for food and drinking
water. Except for salt water lakes that presented higher uranium levels (22 g/L or 272.5 mBq/L of
238
U), the values for uranium in freshwater lakes, reservoirs, rivers, hot and cold springs, well water and
sea water were similar (2.2 g/L with a range from 0.87 to 3.82 (27.3 mBq/L of 238U with a range from
10.78 to 47.3 mBq/L of 238U).

22. The guidance levels of radionuclide concentration provided in the WHO Guidelines for Drinking-
Water Quality are based on an individual dose criterion (IDC) of 0.1 mSv committed effective dose
from one years consumption of drinking water. They are expressed as activity concentration for a
370 UNSCEAR 2016 REPORT

given isotope (Bq/L) and were calculated by dividing the IDC of 0.1 mSv per year by the product of the
isotope dose conversion factor (Sv/Bq) and an assumed water consumption of 2 L per day (i.e. 730 L
per year). The guidance levels of radioactivity concentration for 238U and 234U have been rounded to
10 Bq/L and 1 Bq/L, respectively [W18].

23. Due to the fact that the uranium chemical toxicity is generally of greater importance than
radiological effects, several national and international guidelines refer to concentrations of uranium in
drinking water, as indicated in table 2. Guideline values (in mg/L or g/L) were derived from the total
tolerable daily intake (TDI) expressed in mg/kg or g/kg of body weight (e.g. 60 kg for an adult used
by WHO), itself based on the no observed adverse effect level (NOAEL)1 or lowest observed adverse
effect level (LOAEL) for kidney toxicity, divided by an uncertainty factor of 100 (for intra- and
interspecies variation), and taking the daily drinking water consumption into account (~2 litres) [W16].

Table 2. National and international guidelines for uranium content in drinking water

Only chemical aspects of uranium toxicity are addressed in these guidelines

Organizations/countries Uranium in drinking water (g/L) Reference

Australia 17 [N5]
Bulgaria 60 [E2]
Canada 20 [H16]
Finland 100 [E2]
Germany 10 [B27]
Slovenia 6.8 [E2]
USA 30 [U17]
WHO 30 [W18]

24. The WHO chemical guideline value for uranium in drinking water significantly increased from
2 g/L in 1998 up to 15 g/L in 2004 and then to 30 g/L in 2011 [W18]. The current WHO chemical
guideline of 30 g/L is still designated as provisional because of scientific uncertainties regarding
uranium toxicity, notably with regard to possible carcinogenic effects of uranium [A17] and specific
sensitivity of some groups, such as children or people with hypertension or osteoporosis [F15].

25. Tables A1A4 in appendix A show that uranium concentrations may exceed guideline values in
several countries, including those of water from public supplies. In a study of 476 Norwegian
groundwater samples, 18% had natural uranium concentrations in excess of 20 g/L (0.25 Bq/L of 238U)
[F12]. Natural uranium concentrations in groundwater in excess of 20 g/L (0.25 Bq/L) have also been
reported in parts of New Mexico, the United States [H5], central Australia [F9] and France [I21]. Some
Finnish studies noted a median uranium concentration of 28 g/L (0.35 Bq/L of 238U) and 285 g/L
(3.5 Bq/L of 238U) in drinking water [K26, P28], respectively. In Canada, one study also reported high
levels of uranium concentration up to 845 g/L (10.5 Bq/L of 238U) in private wells [Z8].

26. In a Canadian study, Zamora et al. [Z6] found that water contributed 3198% of the total daily
intake of uranium from food and water for individuals whose drinking water contained uranium at
concentrations ranging from 2 to 780 g/L (25 to 10,000 mBq/L of 238U). This was similar to values
1
NOAEL: the greatest concentration or amount of a substance that causes no detectable adverse alteration in morphology,
functional capacity, growth, or life span of the target organism under defined conditions of exposure [W13].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 371

obtained in studies by the United States Environmental Protection Agency, which reported that uranium
in drinking water contributed about 31% [U13, U14] of the total daily uranium intake.

27. In summary, uranium average levels in water worldwide are: 2 g/L (15 mBq/L of 238U) for
groundwater (appendix A, table A1), 1 g/L (12.4 mBq/L of 238U) for surface water (appendix A,
table A2), 1 g/L (12.4 mBq/L of 238U) for public water supplies (appendix A, table A3), 0.5 g/L
(6.5 mBq/L of 238U) for bottled mineral water (appendix A, table A4) (for natural uranium of
25,400 Bq/g). These median values hide great variability, notably for groundwater (0.00057.780 g/L
(0.006396,472 mBq/L of 238U)). However, overall only a small proportion of few drinking water
samples (generally <3%) exceed the national or international guidelines. As expected, the values for
uranium content in bottled mineral water are not so scattered.

4. Levels in food
28. The measurement of the bioaccumulation of uranium in animals and plants shows that
concentration factors are dependent on organism characteristics (e.g. species, life stage, physiology),
exposure pathways, and the chemical and physical characteristics of the environment [G4, Q2]. Various
publications have reported that the most available forms of uranium in plants were phosphate,
carbonate, sulphate or citrate forms [L14, L15].

29. Most available data relate to transfer through plants from their roots and the direct contamination of
aquatic organisms. Within plants, uranium concentrates mainly in the roots. Uranium found in meat and
dairy products results from livestock feeding on plants and on food supplement made from natural
phosphates and supplied to dairy cows. Ingestion of soil particles, either directly or through consumption
of grass contaminated with soil, is likely to be a significant component of the total intake by livestock.
Transfer parameters of natural uranium are known for the main meat-producing species (cattle, sheep and
pigs) and also for cows milk [I1]. They vary between 3.9 104 and 7.5 101 day/L in bovine meat and
poultry, respectively.

30. Uranium has been detected in a variety of foodstuffs, with great variability. The 238U activity has
been estimated to be 100-fold higher in root vegetables than in fruit or leafy vegetables as shown, for
example, by measurements for beets and tomatoes (100 vs. 1.13 mBq/kg, respectively) [I2]. A synopsis
of the activity of 238U measured by different authors in several types of foodstuffs is contained in the
Committees 1977 Report and in its 2000 Report [U7, U8]. Meat products have the lowest uranium
activity (between 0.08 and 20 of 238U mBq/kg). A recent report indicated activities of 238U between
1 and 49 mBq/kg for meat products [R11].

31. An estimate of daily uranium ingestion of food was made in Japan for urban residents [K30].
Concentration of 238U varied between 9.9 105 and 5.9 Bq/kg depending on food types: grain vinegar
and boiled and dried hijiki (a brown sea vegetable), respectively. When prepared diets were analysed,
the uranium concentrations observed were, on average, about four times higher than those seen in raw
foodstuffs. Hamilton explained this to the possible addition of uranium in seasonings and to transfer
from the cookware [H8]. It was unclear whether these dietary intakes included those from drinking
water and it was emphasized that the latter had sometimes been found to be equal to that from the diet
[H8]. Wrenn et al. have suggested that in regions where treated surface water was used for cooking and
drinking, food appeared to be the major source of uranium intake [W26].

32. In aquatic animals (crustaceans, molluscs and fish), the bioconcentration factor from water is very
low [I3]. Concentration factors for fish vary from 0.01 to 20. The values depend on the behaviour of the
organisms (pelagic species accumulate approximately 10 times less than benthic species) and the
372 UNSCEAR 2016 REPORT

tissues considered (bone 2008,000 and kidneys > liver and gills > muscles 1.524 > digestive system
> gonads). A study was performed in Japan to determine uranium concentrations in marine organisms
(soft tissues) [M15]. This study showed large differences in uranium levels depending on the marine
species, with a minimum value of 0.077 g/kg (0.97 mBq/kg of 238U) measured in rockfish (kichiji) and
a value of 5,040 g/kg (63.8 Bq/kg of 238U) found in octopus. The values reported by Belles et al. also
indicated that fish and seafood showed the highest uranium concentrations (90 g/kg, 1.1 Bq/kg of
238
U), followed by dairy products (40 g/kg, 0.5 Bq/kg of 238U) [B12].

33. The concentrations of uranium in fish are also dependent on uranium levels in water. The uranium
concentrations in the muscle (dry weight) of fish caught in a Canadian lake receiving effluents from a
uranium mill were 711 times higher than those from fish caught in uncontaminated lakes [S50].
Uranium mines and mills operating between the 1940s and the late 1970s have left behind legacy
contamination due to historic mining and milling practices and incomplete site remediation during
decommissioning. In Beaverlodge Lake, Northern Saskatchewan, Canada, elevated concentrations of
uranium are still present with a mean concentration for the period 20132015 of approximately
135 g/L (range 130142) corresponding to 1.69 Bq/L of 238U (range 1.631.78) [C2, C3, C4].

34. Human health risk assessments for environmental contaminants take soil ingestion rates into
consideration. The value recommended by Richardson and Stantec Consulting Ltd. [R13] of 2040 mg/d
for children is based on mechanistic assessments made by Wilson et al. [W21] and Ozkaynak et al. [O10].
Another report suggested assessing health risk for children with the value of 100 mg of soil per day [U15].
However, uranium bound to soils is not completely bioavailable. Reported bioaccessibility (in vitro
estimate of bioavailability) values are quite variable ranging from 2112% in the gastric phase and
4817% in the gastric and intestinal phase [J4] to less than 5% in the gastro-intestinal phase [T16].

35. The urinary concentrations of several metals, including uranium, were found to be higher than
international reference values in a study of schoolchildren and working children in Lahore, Pakistan
[S44]. The measured urinary concentrations of uranium corresponded well with uranium concentrations
in drinking water.

36. A study of 19 categories of food was performed by Fisenne et al. [F5, F6]. Potatoes, meat, fresh
fish and bakery products were found to contribute more than 70% of the average uranium intake
(1.2 g/d or 14.9 mBq/d of 238U). Dietary levels of uranium in the United Kingdom were reported in a
study of typical diets using both raw and prepared foodstuffs. Analysis of the raw foodstuffs indicated
that 83% of the daily intake of uranium derived from starchy roots, vegetables and fruit, and cereals.

37. The Committee in its 1977 Report [U7] included a summary of 238U concentrations in foodstuffs
in France, Japan and the former Soviet Union along with the results given above for the United States
and the United Kingdom. In areas with typical uranium concentrations in soils, the daily dietary intake
fell within a relatively small range, ~0.9 to 1.5 g natural uranium (11.5 to 19 mBq of 238U). This range
is consistent with values given in several publications calculated from the mass activity of natural
uranium (25,400 Bq/g): 1.32 g/day corresponding to 16.4 mBq/day of 238U for typical diets of adults
in New York City, Chicago and San Francisco in the United States [W11], 1.14 g/day (14.5 mBq/day
of 238U) [K30] and 1.46 g (18.6 mBq/day of 238U) in different cities in Japan [N9]. In the United
States, the average daily per capita intake of natural uranium in foodstuffs was estimated to range from
about 1 to 33 g (i.e. from about 12 mBq 238U/day to 405 mBq 238U/day) determined from excretion
measurements [S26]. The values given by ATSDR indicated that the daily intake of uranium from food
sources ranged from 0.9 to 1.5 g/day [A31, W11]. Similar values were given for European countries
[W15]: intakes ranged from 6 to 22 mBq/day of 238U corresponding to 0.47 and 1.77 g/day,
respectively. This daily intake contributed to a body burden of around 50 g (0.62 Bq of 238U) in
humans [F6].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 373

38. The Codex Alimentarius gives guideline levels applied to radionuclides contained in food,
destined for human consumption and traded internationally, which has been contaminated following a
nuclear or radiological emergency [C29]. These guideline levels apply to food after reconstitution or as
prepared for consumption, i.e. not to dried or concentrated foods, and are based on an intervention
exemption level of 1 mSv in a year. A value of 100 Bq/kg is given for 235U. However, these guideline
levels exclude radionuclides of natural origin such as 238U.

39. In summary, uranium is present in a variety of foodstuffs, with great variability. Potatoes, meat,
fresh fish and bakery products were found to contribute more than 70% of the average uranium daily
intake. The total daily intake in food was found to be around 1.5 g (18.6 mBq of 238U), about twice
that via drinking water, recognizing that levels in diet and drinking water can vary greatly.

5. Levels in milk
40. Some publications report uranium levels in milk, notably from cattle. The uranium concentrations
measured in milk were in a wide range, from 0.012 to 0.41 g/L (0.155.2 mBq/L of 238U), depending
on the species and the technical methodology (table 3). A mean value of 0.26 g/L is given,
corresponding to 3.3 mBq/L of 238U (range 0.0011.20 g/L; 0.01215 mBq/L of 238U), if the highest
values of Santos et al. [S3] are excluded. This mean value is above the reference value of 1 mBq/L of
238
U given in the Committees UNSCEAR 2000 Report [U8]. Furthermore, transfer coefficients of
uranium into milk are also available in the literature [A5, K1, T8, W9, W28] and presented for different
species in table 4.

Table 3. Uranium content (mass and activity) in milk (animals)

Values are expressed in kg of fresh matter, the numbers in italics correspond to calculated data obtained from the mass
activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium of 48.3%

238
[U] in g/L U in mBq/L Reference
0.140.24 1.742.98 [A14]
1.20 14.8 [S43]
0.10 (0.030.24) 1.24 (0.383.0) [F16, M34]
0.720.35 8.94.3 [A5]
0.0010.01 0.010.12 [R11]
0.210.02 2.560.25 [P19]
0.250.06 3.070.74 [P20]
3.09 (0.189.6) 38 (2.2118) [S3]
374 UNSCEAR 2016 REPORT

Table 4. Transfer coefficients of uranium into milk (animals)

n.i.: not indicated

Species Transfer factor (d/L) Range (d/L) Reference

Cattle 2.0 104 6.0 1056.0 104 [T8]
Cow 2.9 103 5.0 1046.1 103 [K1]
3 4 3
Sheep and goat 1.0 10 3.0 10 3.0 10 [T8]
3
Goat 1.4 10 n.i. [K1]
Camel 4.2 103 1.1 1031.5 102 [A5]

B. Artificial sources
41. The main application of uranium is for energy production and military use. Uranium-235 is
naturally a fissile isotope. Uranium is used primarily in most nuclear power plants. Its utilization in
most reactors requires enrichment of natural uranium containing 0.72% by weight of 235U to a 235U
content of 25%. Weapons use high enriched uranium with over 90% 235U. Some research reactors and
naval reactors also use high enriched uranium. Depleted uranium is used as a metal in kinetic energy
penetrators and tank armour.

42. The nuclear fuel cycle leading to the production of electricity from uranium in nuclear power
reactors includes mining, milling, conversion, enrichment, fabrication of nuclear fuel and reprocessing
[G18]. Mining for the extraction of uranium involves both conventional open pit (where deposits are
close to the surface) and underground mining (used for deeper deposits). Currently, most uranium
mining worldwide uses the in situ leach mining process [S4]. The milling process produces a uranium
oxide concentrate, named Yellowcake, which contains more than 80% uranium. This uranium oxide is
then converted to uranium hexafluoride (UF6). It contains only natural uranium, which is enriched via
one of the two major types of enrichment technologies, gaseous diffusion or gas centrifuge. The
enriched uranium hexafluoride is then converted to uranium dioxide (UO2) powder and processed into
ceramic pellets. Finally, these pellets are inserted into tubes of corrosion-resistant metal alloy, called
fuel rods, which are grouped in fuel assemblies for the nuclear fuel core of a power reactor.

43. The uranium remaining after removal of the enriched fraction is DU, containing 0.3% 235U or less
[B22, B26]. This uranium has various civilian applications, such as in counterweights or ballast in
aircraft or counterweights for rudders and flaps [B22], for X-ray radiation shielding in medical
equipment and also for containers for the transport of radioactive material. Moreover, DU has also been
used in glassware, ceramics and dentistry.

44. In addition to exposure to natural uranium in the environment, anthropogenic activities have led to
increasing uranium exposure for humans. For instance, uranium was found to leach into water from
uranium-bearing glass items (maximum uranium in water, 30 g/L (0.38 Bq/L of 238U) and from ceramic-
glazed items in which natural uranium is used as a colouring agent (300 g/L; 3.7 Bq/L of 238U) [L6].

45. Uranium is present in water as a result of leaching from natural deposits and waste from the
mining of uranium and other minerals, releases from the nuclear fuel cycle and the combustion of coal
and other fuel [D22, E6, S26, T1]. Phosphate fertilizers, which may contain uranium at concentrations
as high as 150 mg/kg, may also contribute to the uranium content of groundwater [S26].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 375

46. Contamination of surface water and groundwater by effluents from uranium mining, milling, and
production operations due to in situ leaching methods has been documented [A31, E1, H14, S50].
Table 5 shows some recent data. Except for some specific locations, i.e. at the pit, the values of
uranium concentrations in surface water or groundwater are usually below the WHO guideline of
30 g/L (372 mBq/L of 238U).

47. Since the 1970s, DU is used for kinetic energy penetrators and tank armour, because of its
pyrophoricity [B22]. The military applications of DU led to the significant release of this radionuclide
into the environment during the conflicts in Iraq and Kuwait (321 tons of DU), in Bosnia-Herzegovina
(3 tons of DU), and in the Kosovo (10 tons of DU). More details are given in reports by the United
Nations Environment Programme (UNEP) and the National Defence Research Institute (NDRI) [H13,
U3, U4, U5, U6].

48. The use of reprocessed DU in mixed oxide (MOX) fuels (constituted by 89% of plutonium and
~90% of DU) has been used as a recycling strategy by countries including France and Japan as an
option to reduce the necessity for storage of spent fuel [R6].

49. In a study in Tajikistan, uranium concentrations were shown to vary from more than 1,600 g/L
(>20 Bq/L of 238U) at the pit lake to 90 g/L (1 mBq/L of 238U) in tube supplies and 6.3 g/L
(80 mBq/L of 238U) in drinking water from the neighbouring village [S28]. Another study documented
uranium contamination of groundwater in Arizona, the United States after uranium mining [D13].
Approximately 20% of total uranium concentrations in the water samples exceeded the maximum
concentration level for drinking water of 0.37 Bq/L of 238U [U17]. In another study, the uranium
content was measured in drinking water samples from locations near the uranium mining site at
Jaduguda, India [P8] with uranium concentrations between 0.030.01 and 11.61.3 g/L values
below the WHO guidelines of 30 g/L for uranium level in drinking water [W18].

50. Uranium conversion, uranium enrichment and fuel fabrication facilities are other steps in the
nuclear fuel cycle, which also release small amounts of uranium to the environment [A31]. Tracy and
Meyerhof showed that concentrations of uranium in the air near a uranium refinery were 200 times
higher than background concentrations [T18]. For monitoring stations in Port Hope, Canada, where a
uranium refinery is operating, the annual average concentration varied between 0.001 and 0.0158 g/m3
in 1988 and 1989. Uranium concentrations subsequently decreased and varied between 0.005 and
0.00028 g/m3 in the early 2000s [C27]. In 2009, elevated levels of uranium were registered in storage
reservoirs of liquid radioactive waste at the Mayak facility, Russian Federation: the concentration was
370 mBq/L to 520 Bq/L for 234U, and 260 mBq/L to 520 Bq/L for 238U [T21].

51. Concentrations of uranium in surface waters downstream from currently operating uranium mines
and mills are relatively low and decrease with distance from the point of effluent discharge. For
example, between 2000 and 2012 in the vicinity of Canadian uranium mines and mills within 1 km
from the discharge points the mean concentration values ranged from 17 to 0.92 g/L and decreased to
mean values in the range of 1.44 to 0.096 g/L at distances greater than approximately 10 km from
discharge points [C28].

52. Environmental contamination by uranium caused by DU in ammunition used in military conflicts

was reported in several studies and in UNEP reports [U3, U4, U5, U6]. Uranium in agriculture soil in
Kosovo and BosniaHerzegovina averaged 1.8 and 3 mg/kg, while concentrations in public drinking
water averaged 0.5 g/L (16.3 mBq/L of 238U) and 0.4 g/L (5.1 mBq/L of 238U), respectively [C12].
The average uranium concentrations in soil and water were consistent with natural levels, although
localized areas of greater concentration were measured in the immediate surroundings of the DU
376 UNSCEAR 2016 REPORT

penetrators [D4, E5, S2]. In the UNEP report on Kosovo, a great variability was observed in uranium
concentrations in water samples, with a range between 0.006 and 2.15 g/L [U4].

53. Carvalho and Oliveira [C12] found high, localized contamination of soil with DU at Djakovica
(4,662 Bq/kg of 238U; 376 mg/kg). The water samples collected from public water distribution networks
and river water ranged from 0.2 to 0.76 g/L (2.59.7 mBq/L of 238U) and the air samples ranged from
0.8 to 7.2 Bq/m3. Consistent results from a number of studies indicate that environmental
contamination by DU has been very localized and confined to the areas of ammunition impact.

54. Concentrations of uranium were measured in surface and groundwater at the Semipalatinsk
nuclear test site, where more than 400 nuclear tests were conducted [L28]. The measurements showed
that 238U concentrations in well water within the study area were in the range of 74213 mBq/L. The
results of these studies suggest that diverse human activities involving uranium (from extraction to
application) have led to some localized increase of the concentration in the environment.
Table 5. Overview of uranium content in water close to nuclear fuel facilities worldwide

The numbers in italics correspond to calculated data obtained from the mass activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium (48.3%); * mean
value

238
Sample Total uranium U
Country Location Type of plants Reference
number (g/L) (mBq/L)
Brazil Lagoa Real Mining and ore processing 26 5.45 (0.1259) 67.6 (1.23 212) [C13]
Canada Saskatchewan, Beaverloke Lake Mining 22 100.8 1 250 [Y1]
India Jaduguda Mining and ore processing 33 3.2 (0.0311.6) 39.7 (0.37144) [P8]
India Jharkhand, Narwapahar Mining 103 0.63 (0.103.75) 7.8 (1.2446.5) [R7]
India Jharkhand, Bagjata Mining 10 3.22* (<0.511.2) 40* (<6.2139) [G12]
India Jharkhand, Banduhurang Mining 10 2.15* (<0.527.5) 26.7* (<6.2341) [G12]

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 377

India Jharkhand, Bagjata Mining 40 (<6155.9) (<12.6693) [G13]

Kazakhstan and Kurdai (Kazakhstan) and Mining 10 28.2 (1.935.9) 350 (23.6445) [U12]
Kyrgyzstan Shekaftar, Kavak and Kadji-say (Kyrgyzstan) (1 525 in lake at mining pit) (18 910)
Kyrgyzstan Mailuu Suu Mining and milling 25 0.28 (0.270.34) 3.47 (3.354.2) [C32]
(6 820 for tailings)
Nigeria Jos plateau Abandoned mining 5 0.10 (0.030.27) 1.24 [A22]

Nigeria Abakaliki Mining 20 2 (07) 24.6 (087) [O2]

Portugal Viseu, Quinta do Bispo & Cunha Baixa Mining 12 17.7 (1 km) 220 (1 km) [C11]
0.5 (7 km) 6.2 (7 km)
Tajikistan Taboshar & Digmai Mining and milling 6 6.95 (3.492) 86.2 (42.21 138) [S28]

USA Karnes County, Texas Pana Maria Mining and milling 6 19.7 (14.895) 244 (1831 178) [M18]
USA California Juniper 9 3.35 (0.0252.37) 42.2 (0.25649) [K8]
378 UNSCEAR 2016 REPORT

III. PHYSICAL, RADIOLOGICAL AND CHEMICAL CHARACTERISTICS

A. Physical and radiological characteristics

55. Uranium is an actinide and has one of the highest atomic numbers (92) of any naturally occurring
element. It is a silvery-white metal that is malleable, ductile, slightly paramagnetic, with a very high-
density. In its natural state, crustal uranium occurs as a component of several minerals, including
carnotite, uraninite (pitchblende) and brannerite, but is not found in the metallic state in nature. In
addition, uranium metal is pyrophoric. Due to its pyrophoricity, it is used in military applications,
particularly in armor-piercing projectiles.

56. The three naturally occurring isotopes of uranium (234U, 235U and 238U) behave the same way
chemically but have different radiological properties (table 6). Uranium-238 has the longest half-life
and consequently the lowest specific activity. It is the most abundant naturally occurring uranium
isotope. Among the natural isotopes of uranium, 234U has the highest specific activity and the shortest
half-life [L35, S15]. However, other isotopes of uranium may be produced, such as 233U that has a very
high specific activity (3.57 108 Bq/g) [L43].

57. Natural uranium is made of a mixture of the three isotopes described above with about 0.72% of
235
U in mass (table 7). Depleted uranium refers to isotope mixtures that contain a lower percentage of
235
U (from about 0.20.3%) while enriched uranium contains typically 35% 235U in mass and may
contain up to 90% 235U for military applications (table 7) [L35, S15].

Table 6. Radiological properties of uranium isotopes [N8]

nt: nuclear transformation

Half-life Daughter Emitted energy (MeV/nt)

Isotope
(years) nuclide Alpha Electron Photon Total
234 5 230
U 2.46 10 Th 4.8430 0.0137 0.0020 4.8587
235
U 7.04 108 231
Th 4.4693 0.0530 0.1669 4.6891
238
U 4.47 109 234
Th 4.2584 0.0092 0.0014 4.2691

Table 7. Typical isotopic composition in mass and activity of different types of uranium [L35, S15]

Mass (%) Activity (%) Activity (Bq) for 1 g

Type of uranium
238 235 234 238 235 234 238 235 234
U U U U U U U U U

Natural 99.284 0.711 0.0053 48.2 2.3 49.5 12 400 580 12 474

Depleted 99.807 0.0.199 0.0008 86.1 1.1 12.8 12 400 158 1 843
235
Enriched (3.5% U) 96.481 3.46 0.02831 14.7 3.4 81.8 12 005 2 800 66 703
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 379

58. Uranium in rock and soil is in secular equilibrium with the daughters of the decay chain.
Uranium-238 decays to 234Th and 234mPa reaching secular equilibrium within about one year. The 238U
decay chain ends with the stable isotope 206Pb. However, disequilibrium between the uranium isotopes
can occur through physical and chemical changes involving water. For example, combinations of
physical and chemical processes can lead to a separation of 238U and 234U in groundwater [N2].
Uranium-235 and 238U decays contribute to subsequent formation of 10 or more emitters of , and
(figure I). Due to their short half-life, 234Th and 234mPa (24.1 days and 1.17 min, respectively) are
generally present together with 238U.

B. Chemical characteristics
59. Uranium has four valencies, which represent the number of valence bonds that uranium can form
with other atoms. The most common valencies of uranium in ores are IV and VI. The conditions of
transition from valency IV to VI depend on the redox potential of the medium. Compounds containing
hexavalent uranium are much more soluble than those containing tetravalent uranium. Hexavalent
uranium forms complexes such as uranyl carbonates (UO2CO3) and uranyl sulphates (UO2SO4).

60. Uranium can take many other chemical forms. In nature, it is generally found as uranium dioxide
(UO2) with other compounds, such as in pitchblende. Uranium dioxide (UO2) is the final product in the
manufacture of nuclear fuel pellets used in most reactors, and is also present as DU in MOX. Uranium
metal is generally alloyed with other elements (Si, Cr, Al, Fe, Mo, Sn, Al).

61. Uranium metal is pyrophoric and extremely reactive. It oxidizes readily to form triuranium
octaoxide (U3O8) and uranium dioxide (UO2). Uranium trioxide (UO3.xH2O) and uranium peroxide
(UO4.2H2O) also exist. Triuranium octaoxide (U3O8) is the most stable oxide of uranium and is the form
most commonly found in nature. Both triuranium octaoxide (U3O8) and uranium dioxide (UO2) are solids
that have low solubility in water and are relatively stable over a wide range of environmental conditions.
Through reactions with acids, bases or chelating agents, compounds such as uranyl nitrate, uranyl
carbonate, uranyl chloride, uranyl sulphate and uranyl acetate may also be formed. Ammonium diuranate
(NH4)2U2O7 is a basic product in the uranium fuel cycle, a component of Yellowcake, which is produced
during milling and consists of magnesia or ammonium diuranate (respectively MgU2O7 and (NH4)2U2O7).
These compounds are converted to uranium hexafluoride (UF6) prior to enrichment. During conversion,
diuranate is converted to uranium trioxide (UO3), then UF4, and finally UF6 is then enriched from 235U
~0.7% to 235U at ~4%. This enriched UF6 is then converted into UO2 [D6].

62. As indicated above, different chemical forms of uranium are produced throughout the nuclear fuel
cycle [G18]. Uranium-fluorine compounds are encountered in uranium processing, with uranium
hexafluoride (UF6) and uranium tetrafluoride (UF4) being the two most common. The compound UF6 is
used in the enrichment process of uranium to increase the proportion of 235U, either by gaseous diffusion
or by gas centrifuge. It is prepared industrially by the reaction of UF4 powder with fluorine gas. Uranium
tetrafluoride is obtained by treating UO2 with gaseous fluorhydric acid. It is a non-hygroscopic, non-
volatile compound and very soluble in water. In the presence of water vapour, it undergoes pyrohydrolysis
and becomes UO2. When gaseous UF6 is released into air or as it enters the respiratory tract, it hydrolizes
with moisture in air to produce hydrofluoric gas and particulate UO2F2. The oxidation states and
crystallographic forms of uranium in DU particles have been determined from selected samples collected
at different sites in Kosovo and Kuwait contaminated by DU ammunition during conflicts [L39]. Oxidized
uranium (+6) was found in large, fragile and bright yellow DU particles released during a fire at DU
ammunition storage facilities in Kosovo and Kuwait and crystalline phases such as schoepite
380 UNSCEAR 2016 REPORT

(UO3.2.25H2O), dehydrated schoepite (UO3.0.75H2O) and metaschoepite (UO3.2.0H2O) were identified

[L39]. These DU particles were rapidly dissolved indicating a high degree of potential mobility and
bioavailability. Crystalline phases such as UO2 and metallic uranium or UTi alloy were also determined
in impacted DU particles from Kosovo and Kuwait.

IV. HUMAN EXPOSURE

A. Exposure of members of the public

63. Natural uranium exposure of humans occurs through food, water and inhalation. Reviews of
autopsy data show that the skeleton is the main site of accumulation of uranium (~80% of total) [B53,
K6, S25, U18]. This result, observed in a United States population, was corroborated by measurements
made in the United Kingdom [H7, H8] and Japan [I18]. Publications differ about whether the
distribution of uranium in the skeleton appears to be uniform [H12] or not [S24]. The age dependence
of uranium concentration in the skeleton was also investigated in humans (from six months to 65 years)
from vertebrae bones collected in Canada from population exposed to high uranium levels in drinking
water [L11]. The data indicated higher uranium accumulation at six months.

64. Fisenne et al. [F6] summarized numerous publications from 12 countries concerning uranium
concentration in human tissues, including blood, soft tissue and bone. This analysis demonstrated small
differences in uranium concentration in soft tissue and bone. The authors calculated an average of 30 g
for the skeletal burden. The body burden of uranium in humans was estimated between 50 and 60 g, with
57% in the skeleton, 20% in muscles, 16% in fat, 4% in blood, 2% in the lungs, 1% in the liver and 0.36%
in kidneys as presented in figure II [F6]. One publication mentioned uranium levels in the human brain
[K6] with values between 0.18 and 0.77 g of uranium/kg (n=3), corresponding to 0.4 to 0.99%.

65. The Committee in its UNSCEAR 1977 Report [U7] provides further reference concentrations of
238
U in various human tissues expressed in mBq/kg: 20 mBq/kg in the lungs, 3 mBq/kg in the liver,
30 mBq/kg in the kidneys, 5 mBq/kg in muscles and 100 mBq/kg in the bones. Despite some
differences, these values and those reported by [F6] are similar. High uranium concentrations were also
measured in the kidneys in humans [A2, D20, S25]. Calculations made for four age groups (infant, one-
year-old, ten-year-old, adult) indicated that long-term chronic uranium ingestion would result in a
kidney burden of 6.6% of daily uranium intake for all age groups [C21].

66. Fisenne and Welford [F4] measured an average kidney content of 0.13 g 238U from 12 New York
City dwellers, aged 2060. An average daily New York City diet of 1.3 g yields a blood uptake of
0.026 g. The kidney-to-blood uptake ratio was estimated at 4 and the bone-to-blood uptake ratio at
3.3. Figure III shows the average daily dietary intake of 238U measured in 16 cities in 12 countries [F7].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 381

Figure II. Distribution of 238U in human body [F6]

Figure III. Average daily 238U dietary intake measured in 12 countries [F7]
382 UNSCEAR 2016 REPORT

B. Occupational exposure
67. Exposure to uranium may also be relevant to occupational exposure, notably for workers involved
in electricity production in nuclear power reactors or at any stage of the nuclear fuel cycle. Each stage
of the nuclear fuel cycle is associated with distinct exposure characteristics [D6, D7]. During mining,
workers may be exposed to various uranium compounds exhibiting different solubility such as
Yellowcake. Further, workers in metal mines, such as underground gold mines, may also be exposed to
uranium.

68. Several tissues were collected at the autopsy of workers by the United States Transuranium and
Uranium Registries (USTUR) [F3]. Some cases have been analysed following uranium exposure [A34,
R31]. Accidental exposure to uranium hexafluoride (UF6) led to long-term (65 years) retention of
uranium in the deep lungs and in thoracic lymph nodes [A34]. High concentration of uranium in
tracheobronchial lymph node was also found in other cases without accidental exposure [R31].

69. Often, data on worker exposure are restricted to measurements of external radiation exposure. For
instance, Anderson et al. reported that only 16% of the Oak Ridge Gaseous Diffusion Plant (ORGDP)
workers were monitored between 1948 and 1988 for internal exposure to uranium by urinalysis [A12].
The difficulty is that monitoring programmes for internal exposure need a combination of bioassay
techniques, e.g. urine and faecal analysis, especially in workplaces where compounds of different
solubility are handled and also in cases of accidental intakes [J6].

70. Following inhalation of insoluble forms of uranium, the lungs may retain the highest
concentrations of uranium [A2, I19]. Adams et al. reported measurements of uranium in kidney and
bone following lifetime occupational exposure to uranium aerosols [A2], showing greater retention in
bone than kidneys, as seen in data for natural uranium in tissue samples from members of the public.

C. Measurement of uranium
71. The amount of uranium taken into the body can be assessed from external radiation measurements
or by bioassay sampling (urine or faeces). The choice and efficacy of each procedure is dictated by the
route of intake, the pattern of exposure, the physical and chemical form of the uranium, the time between
intake and measurement, and the detection limit of the analytical procedure used [A31, L9, S30].

72. Exposure to uranium can be assessed through the detection of uranium in the urine [B4, C18,
C31, D21, L18, L21, M62, S37, S39, W12, W23]. After absorption through oral, dermal and inhalation
routes, uranium is excreted in urine mostly as uranyl ions. Uranium urinalysis data have been shown to
correlate with airborne uranium exposure when averaged over time and the contribution from ingested
uranium is insignificant. Thus, urinalysis can be used to verify the adequacy of air sampling and as a
non-invasive method for the estimation of exposure [C18, D9, T7].

73. In vivo external radiation measurements can be used to determine the amount of uranium in the
respiratory tract or whole body. The energy of the main gamma emission from 235U is 186 keV.
Specialized counting systems (e.g. using germanium semiconductor detectors) are required. The
minimum detectable activity (MDA) of uranium in the chest depends on the isotopic mix of 235U and
238
U and also on the total amount of uranium in the chest or in the whole body. However, it is of the
order of 4 Bq of 235U and 100 Bq of 238U [L25]. When 234Th has reached secular equilibrium with 238U,
photon emissions from 234Th may be used in addition to those from 235U to further reduce the detection
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 383

limit, by summing the 234Th and 235U photopeaks. Kramer et al. have reported an MDA of 4 mg (i.e.
49 Bq of 238U) [K13]. The detection limit is a function of the chest wall thickness of the measured
individual. This parameter must be measured in order to interpret in vivo measurements of uranium in
the respiratory tract.

74. Measurement of uranium excreted in urine after exposure is potentially a more sensitive method
than chest monitoring to determine the amount of inhaled uranium. The limit of detection by alpha
spectrometry is approximately 0.1 mBq of 234U, 235U or 238U in a 24-hour urine sample. Counting times of
approximately one week are required to achieve this sensitivity. For natural uranium, a measurement of
0.1 mBq of either 234U or 238U would correspond to about 8 ng of total uranium. Specialized mass
spectrometric techniques (such as multicollector inductively coupled plasma mass spectrometry; high-
resolution inductively coupled plasma mass spectrometry; or thermal-ionization mass spectrometry) can
provide isotopic analysis at levels lower than can be achieved by alpha spectrometry [K12, L20].
Synchrotron-based X-ray techniques (e.g. X-ray fluorescence microscopy, X-ray absorption fine structure,
X-ray diffraction may be also used for uranium measurements [C15, P5].

75. The use of faecal assay is confined to intakes by inhalation of relatively insoluble forms of
uranium, and dose assessments using these data are subject to substantial uncertainties [D8, J6].
Measurement of uranium in hair could be used as an indicator of body burden in contaminated subjects
[B16, I23, J3].

76. Many studies have reported urinary excretion of uranium in humans. The National Report on
Human Exposure to Environmental Chemicals [C16] gives uranium concentrations in both g/L and g/g
creatinine. Expressed in g/L, mean levels of uranium in the general United States population range from
0.0050.009 g/L according to surveys conducted from 1999 to 2012 on 18,266 individuals. Oeh et al.
[O1] measured uranium content in 113 urine samples from 63 occupationally unexposed persons in
Germany. The urinary excretion of uranium per day was in the range of 1.4 to 77.5 ng with a geometric
mean of 13.9 ng and median of 14.4 ng. Hllriegl et al. [H21] measured the uranium content in urine in
the general public of Nigeria with creatinine normalized values from <10.4 to 150 ng/L (median
13.8 ng/L) and from 2.52 to 252.7 ng/g creatinine (median 33.4 ng/g). Maltov et al. [M8] measured
daily excretion of 238U in urine in the general population (mean 0.311 mBq, range 0.0112.88 mBq). The
measured urinary excretion per day among 40 active uranium miners indicate a mean value of 0.56 mBq
with a range of 0.082.77 mBq of 238U normalized to 1.7 g daily creatinine excretion [M7]. A study
performed in Italy indicated that the daily excretion for the Italian volunteers ranges from 8.2 ng to 59 ng
uranium [B1]. The lowest daily excretion was observed for the youngest volunteer (seven years old).

V. BIOKINETICS
77. The main routes of intake of uranium into the human body are ingestion and inhalation. Transfer
through intact skin is a minor route. Wounds require consideration in occupational exposure. In general,
occupational exposure arises primarily through inhalation of dust containing uranium or following
injury. Exposure of the general public arises mainly through ingestion of water or foodstuffs containing
uranium. The extent of transfer of ingested or inhaled uranium to blood depends on its chemical form
[I10, N2].
384 UNSCEAR 2016 REPORT

A. Inhalation
78. The intake of radionuclides is determined by the air concentration and by the respiratory
characteristics of the subjects, particularly the ventilation rate, which changes according to the level of
exercise and determines the volume of air inhaled and the deposition of inhaled radionuclides in the
airways of the lungs.

79. For radionuclides inhaled in particulate form, regional deposition in the respiratory tract is
governed mainly by the size distribution of the aerosol particles [I9]. Deposition fractions of gases and
vapours are determined by their chemical form. After deposition in the respiratory tract, absorption and
transport of radionuclides involve three general processes. Material deposited in the anterior nasal
passage is removed by extrinsic means such as nose blowing. In other regions, clearance is competitive
between upward particle transport out of the lungs and dissolution and absorption to blood from the
lungs. Particles escalated out of the lungs are subsequently swallowed and pass through the alimentary
tract where absorption can occur.

80. The ICRP human respiratory tract model describes the biokinetics and dosimetry of inhaled
material and is used to calculate the inhalation dose coefficients that are in general use for radiological
protection and scientific purposes [I8]. This model represents the deposition of inhaled radionuclides in
the different regions of the respiratory tract, and the clearance of the deposited activity by mechanical
transport and absorption to blood. Mechanical particle transport rates are taken to be the same for all
material, but are altered by factors such as smoking and disease. Absorption into body fluids depends
on the physical and chemical form of the deposited material [I7]. Absorption is modelled as a two-stage
process: dissolution (dissociation of material into body fluids) and uptake of soluble material. Uptake
into blood is usually treated as instantaneous while dissolution is time dependant and modelled by three
parameters: a fraction (fr) of the activity is rapidly dissolved at a rate (sr), the remaining fraction (1 fr)
is dissolved at a slower rate (ss).

81. The absorption rate of a given compound may vary greatly depending on its method of production
and history. The ICRP recommends that the absorption rate of any material should be determined from
the study of the material itself. In the absence of data, ICRP recommends default parameters for three
reference absorption types: Type F (fast), corresponding to rapid and complete absorption of the
radionuclide with a half-time of about 10 min; Type M (moderate), corresponding to the absorption of
20% of the activity with a half-time of 10 min; Type S (slow), corresponding to the absorption of 1%
with a half-time of 10 min [P4].

82. Hodgson et al. [H20] and Ansoborlo et al. [A15] reviewed the absorption kinetics of uranium
compounds handled in the British and French nuclear industry. In vivo experiments in rats and in vitro
dissolution studies led to the classification of UO2 and U3O8 as Type S; mixed oxides, UF4, UO3 and
(NH4)2U2O7 as Type M; and UO4, UO2(NO3)2 and UO2F2 as Type F. In addition, these studies provided
specific absorption parameter values for each of these compounds. Duport et al. [D32] studied the
solubility of radioactive dust present in the workplace atmosphere from three types of Canadian
uranium oresYellowcake, UO2 and UO3using simulated lung fluid and determined their solubility
classification. Solubility studies were conducted in Canada on material from various uranium mining
and production facilities [R17]. Bekov and Maltov [B10] studied the solubility of dust samples
from the underground uranium mine, Ron, using simulated lung fluid and calculated specific
absorption parameter values for 238U, 234U and 230Th. The dissolution parameters calculated for UF6
mixture were higher than the current ICRP dose coefficient for Type F uranium (factor 27) [A34].

83. Models allow dose calculation from the inhalation of uranium particles in different chemical forms
and in several sizes. However, specific data might be needed for remediation and decommissioning
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 385

activities potentially generating uranium nanoparticles [T20]. A study of inhaled nanoparticles of uranium
in rats [P18] showed that 97% of inhaled particles were deposited in the deep lung and partly translocated
to the pulmonary interstitium. Approximately 22% of these deposited particles were rapidly cleared (lung
retention half-life of 2.4 h) and for the 78% remaining, the lung retention half-time was estimated at 1,412
days. The ICRP is currently developing material specific absorption parameter values for compounds of
uranium and is revising the default parameter values for Type F, M and S compounds.

B. Ingestion
84. Material may reach the alimentary tract either directly by ingestion or indirectly by transfer from
the respiratory tract or from the systemic circulation. Absorption takes place largely in the specialized
absorptive region of the small intestine [L2]. The extent of absorption of individual radionuclides is
dependent on the chemical properties of the element, and also the specific form of the intake. It is
quantified by the fraction of element reaching blood following entry in the alimentary tract. The
absorption and retention of radionuclides in the human alimentary tract are described in the human
alimentary tract model produced by ICRP [I14]. This model depicts transfer of ingested material
between alimentary tract regions, faecal elimination and absorption into blood [I14]. In the ICRP
model, the fraction f1 represents the small intestinal absorption, and the symbol fA refers to the total
absorption from the different sectors of the human alimentary tract, including the walls [I14].

85. Soluble uranium is absorbed throughout the small intestine [D25, K9]. Comparative data between
species (rabbit, rat, hamster, dog, baboon, pig and human) have been provided by several authors [F11,
T19, W26]. Tracy et al. reported a gastrointestinal absorption factor of 0.06% in rats and rabbits [T19]
while Frelon et al. reported a value of 0.4% in rats [F11] for uranyl nitrate administrated in drinking water.
Leggett and Harrison [L23], Zamora et al. [Z5, Z7] and Wrenn et al. [W27] reviewed the uptake of
ingested uranium from the alimentary tract in environmentally exposed human subjects and in volunteer
studies. The distribution of fA values was in a range of 0.001 to 0.063, with daily uranium intake varying
from 0.37 to 573 g. The authors estimated that the best estimate for fA was 0.009, with no correlation
with age, sex, duration of exposure, and total uranium intake. These values are in accordance with those of
another study that found fractional absorption (fA) values in a range from less than 0.1% to about 6% for
individual subjects, with the central values from the different studies falling in the range 12% and 4%
from water for both 234U and 238U [S30]. On the basis of available data, the ICRP Publication 69 [I10]
takes the fractional absorption of uranium from diet to be 2% in adults.

86. In newborn infants, fractional absorption may increase by a factor of about two due to the higher
intestinal permeability [I14]. This higher absorption in newborns was measured in different species
(rats, guinea pigs, pigs, dogs) [S45, S47]. On the basis of animal data, ICRP recommended an f1 value
of 0.04 for infants and 0.02 for anyone more than one year of age [I10]. Chen et al. measured uranium
concentrations in 73 bone ash samples of young children residing in a Canadian community known to
have an elevated level of uranium in its drinking water supply [C22] and estimated fractional
absorption with confidence intervals as wide as 0.0930.113 for infants and 0.0500.032 for children of
17 years. In another extended study by Chen et al. [C23], the absorption fractions were estimated to be
0.0300.022 for children and youths of 718, and 0.0210.015 for adults of 1825 years of age.

87. Experimental studies have shown that fractional absorption depends strongly on the ingested
chemical forms. The ICRP is currently revising its Publication 69 [I10] and will adopt an fA value of
0.002 for relatively insoluble compounds (e.g. UO2, U3O8) and an fA value of 0.02 for all other more
soluble chemical forms [P4]. A biokinetic model was recently developed to describe uptake and
retention in hair following ingestion [L34].
386 UNSCEAR 2016 REPORT

C. Absorption
88. Few human data are available on uranium transfer through skin in a report of the National Council
on Radiation Protection and Measurements [N3] and in a review of the Armed Forces Radiobiology
Research Institute [M17]. Accidents involving workers with extensive skin exposure to uranium have
been reported in the other reviews. Lu and Zhao indicated a rapid increase in the uranium level in urine
followed by severe kidney dysfunction [L48] with a return to normal values at post-accident day 30.
These results were similar to the clinical follow-up made on workers with acute uranium compound
intoxication [S16]. Acute renal failure occurred for several days post-accident with recovery one month
later. The result of a 33-year follow-up showed that kidney and liver functions were normal [S49].

89. In the use of DU munition, small particles originating from DU dust can contaminate open
wounds, and embedded DU fragments may be implanted in muscles. Uranium urine concentrations
following accidental intramuscular implantation of metal DU fragments were measured in United States
service members exposed to DU through incidents involving DU munition and vehicles protected by
DU armour [M25]. Uranium values were from 0.001 to 39.955 g/g creatinine.

90. Several in vivo studies of rodents with uranium exposure of intact skin [D10, O4, T23] demonstrate
that very soluble forms of uranium such as uranyl nitrate and ammonium uranyl tricarbonate are able to
diffuse through the skin layer [L47]. The LD50 of uranium depended on the species as follows: rabbits
>rats >guinea pigs >mice. This toxicity increased with the time and the area of exposure. In vitro Franz
diffusion chamber model [T23] and in vivo hairless rat model may be used for evaluation of uranium
passage through intact skin [P16, P17]. An in vivo study showed that a significant uptake of uranium from
a uranyl nitrate solution could occur within the first six hours of exposure after skin contact. Furthermore,
as high uranium concentration remained present at the deposit site for up to 24 hours after contamination,
skin provided a reservoir for uranium that remained bioavailable [P17].

91. Percutaneous diffusion of uranium was also studied on damaged skin from hairless rats, following
an abrasion (stratum corneum removal) [P15, P17]. In vitro study of biopsies showed that the
percutaneous absorption of uranium increased with the impairment of the stratum corneum. Ex vivo
studies with biopsies showed an increase in the diffusion of uranium through skin after abrasion
compared with that through intact skin. These results have been confirmed in vivo in hairless rats [P16].

92. The NCRPs biokinetic model describes the mobilization of radionuclides, including uranium,
entering the body through a wound to blood [N3]. Three wound retention categories were described,
corresponding to contamination with soluble forms, colloids, particles or fragments. The uranyl ion
(UO22+) was classified as a weakly-retained radionuclide, and UO2 oxide particles behaved rather as
strongly-retained radionuclides.

D. Systemic distribution, retention and excretion

93. The translocation of uranium to blood strongly depends on the physical and chemical form of the
initial compound [A15]. After its absorption to blood, uranium is present mainly as uranyl ions
complexed to proteins (e.g. transferrin, albumin) or bicarbonate anions [A16]. The main sites of
deposition of uranium from the circulation are the skeleton, kidneys and general soft tissues. Human
and animal data show that urinary excretion is rapid with about two thirds of uranium reaching blood
excreted in the first 24 hours and a further 10% over the next 5 days. Most of the remaining uranium is
excreted over a period of a few months, but a small percentage of the amount injected may be retained
for a period of years [L22].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 387

94. The work of Leggett [L22] was used by ICRP [I11] to propose a reference model for uranium
biokinetics. This model is constructed within a physiologically-based framework that is also applied to
the alkaline earth elements. Rates of uranium transfer between plasma, red blood cells, skeleton, liver,
kidneys, other soft tissue and excretion pathways are based on measurement of uranium in humans and
animals and consideration of the physiological processes. The model considers age-related changes in
organ and tissue uptake and retention of uranium. The ICRP will use the same model in ongoing
revisions of dose coefficients [P4].

95. The United Kingdom Royal Society [R28] used the ICRPs biokinetic models to estimate the
concentrations of uranium in the kidneys following chronic exposure over one year at a constant daily
uptake of 1 g of uranium to blood. The estimated uranium concentration in the kidneys was
0.0056 g/g kidney after one year and 0.011 g/g kidney after 50 years of contamination. This result is
not in accordance with the result of other studies. For instance, a study measured 238U in diet and
kidney tissue in residents of New York and showed that a daily intake of 1.270.03 g resulted in a
uranium concentration in the kidneys of 0.000430.26 g/g that was constant over ages <20 years to
>60 years old [F4].

96. Several studies have found that uranium can be incorporated into brain tissues [B11, G10, L30,
O7]. It has been demonstrated by an in situ rat brain perfusion method that uranium is able to cross the
blood-brain barrier [L27] but the mechanism by which uranium is transferred to the brain is unknown.
After acute, subchronic or chronic exposure, low uranium concentrations are found in the various
structures of the brain [B6, F8, H24, L38, P11, P12]. The level of uranium in the brain, notably in
olfactory bulbs, was greatest after inhalation [H26, T15]. Such uranium concentration in olfactory bulbs
as compared to other cerebral structures is attributable to the direct transfer of uranium via olfactory
receptor neurons [I6, T15].

97. Mean values of uranium concentrations in human breast milk were reported as 0.03 g/L
(0.76 mBq/L) [W9] and 0.30 g/L (7.6 mBq/L) [C1]. The daily intake of uranium of mothers is
0.030.019 g/kg (meanSD) [W9]. These values are in accordance with values measured in several
animal studies, which suggest similar transfer processes (see also table 3).

E. Materno-fetal transfer
98. Few data are available on materno-fetal transfer in humans and in animals. Some publications
have reported uranium concentrations in fetal samples [B46, B47, L26, S17, W10], in placenta [B47,
S17, W10], in amniotic fluid [C14, S17] and in cord blood [G22].

99. Ham et al. [H7] reported uranium concentrations measured in human fetus and Bradley and
Prosser in placenta samples [B47]. The activities of 238U ranged from 0.1 to 9 mBq/kg in the fetal
samples of human fetus and from <1 to 11 mBq/kg in placenta samples. Uranium concentrations ranged
from 0 to 0.2 g/L (mean=0.024 and median=0.005) in human amniotic fluid [C14] and from 0.003 to
0.834 g/L (mean=0.104 and median=0.057) in cord blood [G22].

100. An experimental study performed by Legrand et al. on pregnant rats exposed to uranium via
drinking water (40 and 120 mg/L) indicated no elevated uranium concentration in exposed fetuses as
compared to control animals [L26]. Another study performed by Sikov and Mahlum on pregnant rats
after injection indicated that only a small fraction of the injected nuclide (148 kBq of 233U) entered the
fetus and the distribution within the fetus was dissimilar from that observed in the dam [S17]. Only
0.06% of injected dose/g body weight was measured in fetal kidneys at 20 days as compared to 5.18 in
388 UNSCEAR 2016 REPORT

the dam. The authors also reported 0.03% injected dose/g body weight in fetus, 0.05% in placenta and
0.001% in the amniotic fluid [S17].

101. The ICRP [I12] has provided biokinetic and dosimetric models and dose coefficients for the
embryo, fetus and newborn as a result of intake of radionuclides by the mother. The term fetus refers to
the developing human offspring after the eighth week of pregnancy. The equivalent dose to the embryo
is assumed to be the same as that to the uterus wall and proportional to the concentration of uranium in
maternal soft tissue. The ICRP uses a simple approach for the calculation of fetal doses for the majority
of elements and their radioisotopes, including uranium, considering data collected from studies of
animals and humans [I12, I13]. Thus, fetal doses from uranium are calculated on the basis of relative
concentrations of uranium averaged over the whole body of the fetus (CF) and that of the mother (CM).
A conservative CF:CM ratio of 1 is used for intakes of uranium during pregnancy. The distribution of
uranium in the fetus is assumed to be 80% to skeleton, 2% to kidneys, and 18% to other tissues [I12].

VI. DOSIMETRY
102. Absorbed doses in tissues are calculated using dosimetric models such as those of ICRP (e.g. [I7,
I10, I11]) and of the Medical Internal Radiation Dose (MIRD) committee [B39]. The absorbed dose is
the fundamental quantity that is estimated and averaged over particular tissues and organs. The
distribution of absorbed dose from internally deposited radionuclides, here the uranium isotopes,
depends on a number of factors, including the distribution of the radionuclides within organs and
tissues, and the penetration and range of radiation emitted from the radionuclides. Dosimetric models
have been developed for this purpose.

103. Absorbed doses from photons and electrons are calculated by applying Monte-Carlo codes of
radiation transport to anthropomorphic computational phantoms representing a reference person [I15].
Alpha particles are considered to be absorbed in the region where they are emitted, except for the
skeleton, lung, urinary bladder and alimentary tract where the respective positions of the alpha-emitting
radionuclides and of the sensitive target cells are taken into account to assess the dose absorbed by the
target cells. In most organs and tissues, local activity and also the radiosensitive target cells are
assumed to be uniformly distributed. However, in these few specific tissues, the identification of the
putative radiosensitive cells allows a more precise definition of the source and target geometry of
irradiation, which is of concern for the short-range alpha particles emitted by uranium isotopes. The
target cells identified in the thoracic airways include basal and secretory cells in the bronchial
epithelium, endothelial cells such as those of capillary walls, and type II epithelial cells in the alveolar-
interstitial region [I8].

104. Following the ICRP, the equivalent dose in a region T, HT, is defined as:

w D
R
R T ,R

where DT,R is the average absorbed dose in region T, due to radiation of type R; wR is the radiation
weighting factor for radiation R and is equal to 1 for photons and beta particles and 20 for alpha
particles [I10, I15].

The effective dose E is defined as a weighted average of equivalent doses to radiosensitive tissues of
the body:
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 389

E wT H T wT wR DT , R
t T R

where wT is the tissue weighting factor for tissue T, with the sum of wT being 1. The committed
effective dose ET() is defined as the effective dose delivered over the time following intake of a
radionuclide. The is usually set to 50 years for adults and for children up to the age of 70 years so as
to cover life-long irradiation [I10].

105. Doses from the inhalation or ingestion of unit mass of uranium can be determined by multiplying
the corresponding dose coefficient (i.e. for 238U, 235U and 234U) by the isotopic activity for each level of
enrichment [I7, I11].

106. The ICRP provides reference dose coefficients per intake of uranium isotopes for workers [I7]
and members of the public [I10] in accordance with the international safety standards of the IAEA [I4].
The values for inhalation by workers and ingestion by members of the public are given in tables 8 and
9, respectively. The f1 values of the ICRP model depend on age, and the model for the systemic
behaviour of uranium is also age-dependent, and so the committed effective dose per unit intake
calculated for ingestion of soluble uranium by members of public exhibits age-dependence (table 9).

Table 8. Committed effective dose per intake (Sv/Bq) for inhalation by workers (for 5 m AMAD
particulates) [I7]

Solubility group of uranium compound: F (fast soluble), M (moderately soluble) and S (slowly soluble)

Absorption type
Isotope
F M S
234 7 6
U 6.4 10 2.1 10 6.8 106
235
U 6.0 107 1.8 106 6.1 106
238
U 5.8 107 1.6 106 5.7 106

Table 9. Committed effective dose per unit intake (Sv/Bq) for ingestion of soluble uranium by
members of public [I10]

Age at intake
Isotope
3 months 1 year 5 years 10 years 15 years Adult
234 7 7 8 8 8
U 3.7 10 1.3 10 8.8 10 7.4 10 7.5 10 5.0 108
235
U 3.5 107 1.3 107 8.5 108 7.1 108 7.0 108 4.7 108
238
U 3.3 107 1.2 107 8.0 108 6.8 108 6.7 108 4.5 108
390 UNSCEAR 2016 REPORT

VII. BIOLOGICAL EFFECTS

A. Chemical versus radiological toxicity

107. Uranium represents a particularly difficult problem for internal emitter studies because of its
chemical and radiological toxicities. It is a radioactive heavy metal, and it is difficult to characterize
differences in responses to the metal component alone, the radioactivity alone, or the possible combined
effects of both. Some studies on DU have attempted to define the metal component, however these are
limited by the presence of radioactivity.

108. Despite the dual toxicity of uranium, few studies investigate the respective parts of its chemical
and radiological toxicities. As with all chemicals, the chemical toxicity of uranium is linked to its
ability to interfere with compounds and biochemical processes in living organisms. The chemical action
of all isotopes and isotopic mixtures of uranium is independent from the specific activity because
chemical action depends only on chemical properties [W13]. However, it is dependent on its physical
and chemical forms. For instance, the NOAEL values measured for uranium effects vary depending on
the absorption type of uranium compound, i.e. F (fast soluble), M (moderately soluble) and S (slowly
soluble). Different reports gave NOAEL values as a function of the administration mode and the
exposure duration [A31, W14].

109. Concerning the radiological hazard of uranium, alpha particles do not penetrate beyond the outer
layer of skin, except in regions of thinner skin (the depth varies typically in the range 20100 m). The
impact on health of alpha particles of uranium is expected mainly after internal contamination and
depends partly on the route of exposure (inhalation or ingestion) [W14].

110. Both natural uranium and DU pose primarily chemical rather than radiological hazards in the
short term. The toxicity of uranium depends on its chemical form and the route of exposure [A31]. The
potential health effects arising from uranium exposure are discussed below. Some effects are related to
the chemical toxicity of uranium, notably renal effects, while other effects are mainly due to the
radiological toxicity of uranium, such as tumorigenic effects.

111. The relative importance of chemical and radiological toxicities of uranium thus depends on the
degree of enrichment of 235U (and 234U), the compound solubility, the chemical speciation and the mode
of incorporation [A31, S46, T9]. Chemical toxicity from uranium exposure appears mainly in the
kidneys and is assumed not to occur below a threshold concentration. The thresholds given in the
literature are most often derived from a NOAEL in animal experiments with the application of an
appropriate safety factor for transposition to humans [A31]. Human autopsy data are used to confirm
these observations [A31, S43].

112. Stradling et al. discussed anomalies between radiological and chemical limits for uranium after
inhalation by workers [S41]. As a consequence of the different procedures used in their calculation,
they are incompatible and adherence to one limit may result in a breach of the other. They concluded
that for chronic intake by members of the public, it can be deduced that a unified exposure level of
0.5 g/kg per day or a daily intake of 35 g would be acceptable in most cases. More recently, Thorne
and Wilson proposed higher limits corresponding to 2 g/kg per day or a daily intake of 140 g [T9].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 391

B. Kidneys
113. The limited human studies suggest that damage to the kidneys can be detected following chronic
exposure that results in uranium concentrations as low as 0.1 g per gram kidney [R28]. Human studies
also suggest that acute intake which leads to a peak uranium level of about 1 g per gram kidney can
lead to detectable kidney dysfunction [R28]. Some authors have aimed to predict renal concentrations
in human populations exposed to uranium for dose assessments [C21, S31]. In the epidemiological
section some human data are presented of populations exposed to high uranium concentrations in
drinking water.

1. Acute exposure
114. Morphological renal modifications induced by uranium (from 0.1 mg/kg after injection) were
reported in rodents in several experimental studies, suggesting that the kidneys are the major target
organ of acute uranium toxicity independent from the route and duration of exposure [B25, D17, G27].
Histopathological changes, including degenerative changes or necrosis of the proximal tubular
epithelium and glomeruli, were reported after acute exposure in rats [D12, M16]. Further, some
histological alterations were noted in renal tubules of rats following chronic exposure with elevated
concentration of uranium [G10, O7]. Thus, acute exposure may lead to alteration of glomerules and
tubules, whereas chronic exposure to uranium seems to affect only tubular functions. Following
intramuscular injection of DU, rats undergo dose-related tubular necrosis and glomerulonephritis. After
30 days, glomerular damage is reversible in rats at all doses (0.11 mg/kg) but there is a dose dependent
delay in the initiation of the regeneration, seen first in the low-dose group (0.1 mg/kg) [Z17].

115. An experimental study performed by Shiquan et al. on rats considered a wide range in uranium
concentrations of 5, 2.5, 1.0, 0.5, 0.25, 0.1, 0.05, 0.025, 0.01, 0.0075, 0.005, 0.0025 and 0.001 mg/kg,
administered by intraperitoneal injection as uranyl nitrate [S14]. Doses from 0.01 to 0.05 mg/kg
induced slight renal damage. The lowest dose that induced renal damage in all rats of one group was
0.1 mg/kg. At the highest dose (5 mg/kg), necrosis of the proximal tubular epithelium was observed
after 612 hours, and half the rats died after 68 days. Primary damage to the kidneys was necrosis of
the proximal tubular epithelium. After repair with regeneration, different degrees of fibrous scarring
were found in the kidneys.

116. The third segment of the proximal convoluted tubule is the most affected site in the kidneys of
rats [G11] and humans [M9]. Experimental studies performed on dogs [M63] demonstrated that the
complexed uranium (VI) is filtered from blood via the glomeruli. In the proximal tubule, water is
reabsorbed and uranium is concentrated. As urinary flow acidifies, uranium (VI) complexes are
dissociated and uranium can bind epithelial membrane proteins. Uranium kidney retention thus
increases with urinary acidity and decreases with uranium complexation.

117. Table 10 summarizes studies of the toxic effects of uranium on the kidneys of rats following acute
exposure. Biological effects indicating toxicity of uranium to the kidney were noticed starting from
0.1 mg/kg [Z17] to 126 mg/kg of body weight [F18]. Most of the studies were done on rats and showed
a decreased creatinine clearance and an increased electrolyte or protein excretion, indicating tubular
alterations of the kidneys [B5, F18] that can also be associated with liver alteration as shown by
increased transaminase level [G29].

118. Shude and Suquin et al. reported dose evaluation and medical follow-up of a case of acute
uranium compound intoxication [S16, S49]. A nuclear worker received both thermal and acid burns. A
392 UNSCEAR 2016 REPORT

solution of uranyl nitrate hexahydrate was spurted on his body, and uranium was absorbed through the
skin to blood. Acute renal failure and toxic hepatitis occurred during several days post-accident and
recovery occurred one month later. The amount of uranium taken into the body as a result of this
incident was calculated to be 116 mg [S16, S49]. The result of 33 years follow-up showed that the
function of kidney and liver were normal. Chromosome aberrations of peripheral lymphocytes were
observed, including ring, fragment and dicentric. The results of other examinations were at normal
levels, including peripheral blood, bone marrow, immune system, cardiovascular and respiratory
function, endocrine and metabolism [S49].

Table 10. Summary of studies of toxicity of acute exposure in kidneys of adult rats

ip: intraperitoneal; im: intramuscular; iv: intravenous; BUN: blood urea nitrogen; GFR: glomerular filtration rate; LDH: lactate
dehydrogenase; AST: aspartate aminotransferase; ALT: alkaline phosphatise; BW: body weight

Post- Sex Study

Uranium Exposure
exposure Biological effects references
compound conditions
follow-up
Uranyl ip 0.0015 mg/kg 8 days Male Necrosis of the proximal tubular [S14]
nitrate epithelium
Damage repair with renal fibrous scar
Half died after 68 days in the highest-
dose group

Uranyl ip 0.11 mg/kg 30 days Male Increased creatinine, BUN and albumin [Z17]
acetate at high dose in serum

Uranyl ip 0.5 mg/kg 5 days Male Increased creatinine, urea, cholesterol, [B5]
nitrate phospholipids in plasma

Uranyl ip 0.66 mg/kg 110 days Male Decreased kidney weights [D12]
fluoride Increased LDH, AST in plasma and
protein, albumin in urine

Uranyl im 0.22 mg/kg 28 days Male Decreased plasma ALT, AST, protein [F18]
nitrate and increased urea, creatinine, ALP

Uranyl ip 10 mg/kg 28 days Male Increased concentrations of sodium [H6]

nitrate and protein in urine
Decreased GFR

Uranyl ip 11.5 mg/kg 3 days Male Increased creatinine, urea, ALT, AST [G25]
nitrate plasma levels

Uranyl im 7.9126 mg/kg 28 days Male Decreased BW and died after 37 days [F18]
nitrate

Uranyl ip 5/10/20 mg/kg 2 days Male Histopathological lesions [A18]

nitrate Increased plasma creatinine and urea
plasma levels

Uranyl ip 50500 g/kg 21 days Female Renal morphological alterations (at the [B17]
nitrate first dose, becoming increasingly
prominent with higher doses)

Uranyl iv 15/25 mg/kg 17 hours Female Changes in endothelial cell [A33]

nitrate morphology for the higher dose
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 393

119. Other high human exposureaccidental or deliberatewere reported. Roszell et al. reported
27 cases of human exposure to uranium and the resulting kidney effects [R24]. The uranium burden
estimated in the kidneys ranged between 10 g/g kidney after accidental inhalation of UF4 powder
[L48] and 100 g/g kidney after deliberate ingestion of 15 g uranium acetate [P9]. Early symptoms of
renal failure were noted in the first weeks after exposure, with renal dysfunction observed in some
cases until 6 [P9] or 18 months [L48] after exposure. Kathren and Moore re-evaluated the intake and
deposition of soluble natural uranium compounds in three men accidentally exposed in an explosion in
1944. One of the three exposed individuals showed an altered clearance pattern for uranium shortly
after the accident, possibly from pulmonary oedema associated with concomitant exposure to acid
fumes. However, medical examinations of two of the men 38 years after the accident revealed no
detectable deposition of uranium [K4].

120. Table 11 provides a summary of the effects of chronic uranium exposure on the kidneys of
experimental animals. Conversely, chronic exposure did not clearly induce a toxic effect on the kidneys
of mammals, with duration of exposure from three to twelve months and exposure level from 0.02 to
200 mg/kg. Tissue alteration of the tubules or glomerules was observed only for uranium doses above
400 mg/kg [G10, Z13]. After a nine-month chronic exposure to uranium via ingestion of uranium-
contaminated drinking water, the kidneys of rats did not show signs of histological lesions for uranium
renal levels >3 g/g (3 g/g for 120 mg/L [D29] and 6 g/L for 600 mg/L [P23]).

121. Gilman et al. [G10] noted that effects on the kidneys could be seen at uranium levels of 0.06 mg of
uranium per kilogram of body weight per day for male rats. Nevertheless, no rise in histopathological
severity with increasing dose was reported: histological lesions starting from the lowest concentration
(0.96 mg/L, 0.09 mg/kg body weight) were not significantly different from the kidney lesions observed for
the highest concentration (600 mg/L) [G10, G11]. No clear dose-dependent effect was observed after
chronic exposure. Indicators of kidney function after acute exposure (0.110 mg/kg) included the
concentrations of blood urea nitrogen (BUN), creatinine in blood or protein in urine [B5, F18, G23, H6,
O7]. Conversely, only limited changes were identified for chronic exposure to uranium in drinking water
(216 mg/kg) [G11, O7] or from implanted DU pellets (200 mg/kg) [P12, Z13, Z14].

122. Various urinary parametersincluding levels of urea, glucose, creatinine, total protein, and albumin
and also activities of LDH and N-acetyl-beta-D-glucosaminidase (NAG), and glucose excretionhad the
most persistent effect after chronic exposure [G10]. Complementary studies showed that chronic low dose
uranium exposure did not modify the nephrotoxic effects of gentamicin renal response as evidenced by the
renal tissue levels of kidney injury molecule (KIM-1), osteopontin, and kallikrein [R27]. The results
observed with osteopontin were different from those found in clinical studies, which indicated a decreased
osteopontin level in urine [P30]. Gentamicin-induced increase in renal levels of KIM-1 was augmented in
rats previously exposed to uranium as compared to uncontaminated animals.

123. In a dose-response study (0.2740 mg/kg) of chronic oral exposure of rats, nephrotoxic and
pro/anti-oxidant effects were analysed after three- and nine-month exposure [P23]. The uranium
content of the kidneys was proportional to uranium intake after three and 9 months of contamination. It
reached 6 g/g of kidneys for the highest uranium exposed group, a nephrotoxic level for acute intake.
Uranium microdistribution analyses showed that it was found mainly in the nucleus of renal proximal
tubular cells and to a lesser extent in other renal structures. Nevertheless, no renal impairment was
observed according to histological analysis or measurements of sensitive kidney biomarkers such as
KIM-1, 2-microglobuline or retinol binding protein. Uranium contamination appeared to reinforce the
antioxidant system in the kidneys as the glutathione pool increased dose-dependently up to tenfold.
394 UNSCEAR 2016 REPORT

Table 11. Summary of studies of biological effect of chronic exposure to uranyl nitrate on kidneys
of male adult animals

PCT: proximal convoluted tubule

Study
Species Exposure conditions Follow-up Biological effects
references
Rat Oral 27 days Histopathological lesions (increased when [G15]
100 mg/kg exposure duration increased)

Rat Oral 91 days Decreased haemoglobin, erythrocytes, [G11]

0.0240 mg/kg glucose not correlated to the concentration
in drinking water
Histopathological lesions in the lowest group

Rat Oral 275 days Decreased plasma vitamin D and vitamin D [T11]
2.7 mg/kg target genes in the kidney tissue

Rat Oral 275 days Following single exposure to [G26]

2.7 mg/kg acetaminophen (paracetamol), increased
PCT necrosis of the kidney

Rat Oral 28 days Increased glycaemia whatever the dose in [O7]

216 mg/kg plasma

Mouse Oral 122 days Decreased urea and increased creatinine in [T4]
1326 mg/kg plasma

Rabbit Oral 91 days No biochemical changes [G10]

0.02400 mg/kg Kidney histopathological lesions in the
highest group

Rat Muscle implantation 91 and 360 days Increased plasma urea, creatinine and [Z14]
200600 mg/kg urinary beta-2 microglobulin and albumin

Rat Oral 275 days No biochemical changes [D29]

0.014-8 mg/kg No renal histopathological lesions

Rat Oral 91 and 275 days Glutathion increase dose-dependently and [P23]
0.2740 mg/kg lipid peroxidation decrease in kidney
No nephrotoxity (biomarkers and histology)

124. A study by Silver et al. provided information regarding non-malignant chronic kidney disease
[S19], which proved to be non-significant. There was a trend in chronic renal failure deaths in the
Colorado uranium miller cohort with duration of uranium milling employment (SMRs of 1.27, 1.33 and
1.53 for 12, 39 and 10 years of employment, respectively) [P21]. However, when treated end-stage
renal disease for those receiving renal dialysis or kidney transplants, the incidence was evaluated using
the ESRD program (End Stage Renal Disease) date, there was no excess (SIR=0.71, 95% CI:
0.26-1.65). In the Fernald cohort, mortality from chronic renal disease was not related to uranium
exposure [S19] (see also table 18). Other studies of uranium worker cohorts indicated no significant
overall excesses of chronic kidney disease death [B38, C20, D33, M26, M27, P24]. Some renal effects
were studied in humans following chronic exposure to uranium via drinking water. These data are
shown in appendix A, table A11, which summarizes the published literature on the health effects of
human exposure to uranium through ingestion of surface or groundwater.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 395

2. Influence of age
125. Few studies have investigated the influence of age at exposure on renal toxicity. The first results
considering this issue were published before 1920 for studies using dogs [M3]. The author
demonstrated that older animals developed more severe uranium poisoning than young animals,
associated with more marked histological alterations, leading to an impaired functional capacity [M3].
This result was corroborated by a more recent experiment, also performed on dogs [P10]. Uranium
effects on renal function (glomerular filtration rate) were more severe in older (3/4 weeks) than in
younger dogs (1/2 weeks). Recent publication of an experiment performed on rats with 0.12 mg/kg of
uranium acetate indicated a higher uranium concentration in the kidneys of neonatal animals than in
prepubertal or adult animals [H22].

126. Magdo et al. studied the effects of the high uranium concentration in drinking water of a private
wells used by a family (two adults and five children). The authors measured concentrations of up to
(1,160 g/L) in the groundwater [M5]. The authors evidenced a nephrotoxicity in the youngest family
member (a three-year-old child), demonstrating the highest sensitivity to uranium exposure. This case
shows potential for significant residential exposure to naturally occurring uranium in well water. It
highlights the special sensitivity of young children to residential environmental exposures, a reflection
of the large amount of time they spend in their homes, the developmental immaturity of their kidneys
and other organ systems, and the large volume of water they consume relative to body mass [M5].
However, this result appears inconsistent with observations made in young rodents.

3. Biological mechanisms
127. According to Leggett [L21], uranium binds to the luminal brush-border membrane of tubules.
This binding decreases reabsorption of sodium and other compounds, resulting in an increased urinary
excretion of proteins, glucose, catalase, phosphate, citrate and sodium without causing cellular damage
[B25]. Uranium can then separate from the luminal membranes by a number of mechanisms including
association with complexing ligands from the urinary tubular flow, detachment of uranium-bound
microvillosities, and elimination of dead cells. The mechanism by which uranium then enters tubular
cells has been investigated. Once in the cytoplasm, uranium accumulates in lysosomes, where it
precipitates with phosphate, forming microcrystals at high concentrations [M47]. This process induced
destruction of the lysosome.

128. The mechanisms of uranium effects on the kidneys have not been fully elucidated. The relations
between uranium penetration into and distribution within cells and its toxicity was analysed in kidney
proximal tubular cells. Some authors posited that uranium did not need to penetrate cells to exert its
toxic effect [L21, M68] while others posited that it did [L1, M47]. Rouas et al. have suggested that the
physical form of uranium (soluble or precipitate) and its intracellular localization play a role in cell
toxicity [R26]. They suggested that uranium may be visualized in the nuclei in kidney cultured cells
exposed to uranyl nitrate. Experimental studies have indicated that uranium exposure of animals (rats or
mice) at 525 mg/kg may induce changes in solute transport [G11, T5], protein biosynthesis-related
genes [T4, T5] or cell signalling [P26, T5].

129. Several studies reported effects of uranium on renal transporters, Na+,K+-ATPase [B48], sugar
transporters [G17, N4, R19], sodium-dependent phosphate cotransporters [M68, M69] and organic
cation transporters [M4, S13] but these effects were studied only in vitro using relatively high uranium
concentrations.
396 UNSCEAR 2016 REPORT

130. Mitochondrial dysfunctions have recently been demonstrated in rats after injection at 0.5, 1 and
2 mg/kg per intraperitoneal injection [S9]. Isolated mitochondria from the uranyl acetate-treated rat kidney
showed a marked elevation in oxidative stress accompanied by mitochondrial membrane potential
collapse as compared to the control group. In addition, direct incubation of isolated kidney mitochondria
with uranyl acetate (50, 100 and 200 M) suggested that uranyl acetate can disrupt the electron transfer
chain at complex II and III.

131. To investigate the influence of uranium speciation on its toxicity, cells representative of rat kidney
proximal epithelium (NRK-52E) have been exposed to uranyl-carbonate and -citrate complexes,
because they are the major complexes transiting through renal tubules after acute in vivo
contamination. When citrate is added to the exposure medium, the predominant species is uranium
(VI)-bicarbonate. Nonetheless, citrate increases uranium (VI) toxicity and accelerates its intracellular
accumulation kinetics without inducing precipitation [C10].

132. Uranium can induce cell death but the exact mechanisms are still unclear. Some proposed
mechanisms include apoptosis or genotoxicity [M54, T6, V5]. The mechanisms of acute toxicity of
uranium (from 50 to 500 M) have been studied in renal cell lines and have shown a specific uranium
signature characterized by the downregulation of tubulin and actin [P27]. The most investigated
mechanism to explain uranium toxicity is the oxidative stress response, investigated both in vitro on
cell cultures and in vivo following acute or chronic exposure in rats and mice [B5, S9, T4, T5, T6]. The
results of these studies suggested uranium-induced oxidative stress imbalance with an increased
reactive oxygen species production associated with a depletion of endogenous cellular antioxidants for
elevated concentrations administrated to cells (>400 M) or to animals (>0.5 mg/kg).

133. Several studies have focused on the interaction between iron metabolism and uranium, in
conjunction with the affinity of these two elements for some proteins such as transferrin, ceruloplasmin
and ferritin [V8]. These studies revealed changes in gene expression and protein carriers of iron, DMT1
(Divalent Metal Transporter Type 1) and Fpn (ferroportin), in the liver and kidneys [B19]. A similar
study [D19] performed on rats chronically exposed to uranium in drinking water (2.7 mg/kg/day)
showed the appearance of iron granules (aggregates) in the kidneys, indicating that chronic
contamination by uranium could cause long-term changes in the regulation of iron metabolism. A
recent study using surface plasmon resonance techniques has shown a strong affinity of fetuin-A
protein for uranyl ions even though this protein is present in a very small amount [B7].

4. Conclusion
134. The kidneys are known as the most sensitive target organ after acute exposure, on the basis of
well-documented studies. Studies on rodents (mice and rats) indicated that injury to the kidneys occurs
from 0.1 mg/kg whole body and renal concentrations of >3 g/g, targeting the third segment of the
proximal convulated tubule. Conversely, data on chronic exposure to low uranium doses are more
recent and have not shown clear nephrotoxic effect at 220 mg/kg and no specific biomarkers have
been identified to date. New blood plasma or urinary biomarkers of the renal function or integrity have
been investigated in experimental or clinical studies, with a view to allowing a precise diagnosis of the
kidney function [G27]. Mitochondrial dysfunction seems to be involved in toxic mechanisms of
uranium, but the mechanisms of its toxicity are not fully elucidated to date, especially after chronic
exposure to low and high doses. Further experimental studies are necessary for a better comprehension
of the renal alteration process and the identification of more specific biomarkers of kidney alteration.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 397

135. Chemical effects on the kidneys are usually assumed not to occur below certain threshold
concentrations of uranium; most often, these findings have been derived from animal experiments with
the appropriate safety factors applied to human exposure [A31]. Maximum concentrations over 3 g of
uranium per gram of kidney have been used as the basis for occupational exposure limits (e.g. [H20,
L21]). Leggett [L21] suggested that the occupational limit based on 3 g of uranium per gram of kidney
is about tenfold too high for non-occupational exposure. Indeed, some human studies suggest that
damage to the kidneys can be detected following chronic exposure that results in uranium
concentrations as low as 0.1 g per gram kidney [R28]. Thorne and Wilson [T9] suggested 30 g U/kg
kidney as a level below which effects will not be observed. Further, following a review of the existing
literature, Leggett et al. have recommended that the concentration of uranium in the kidney should not
exceed 1 g U/g kidney at any time [L25].

C. Bone

1. Acute exposure
136. Several animal experiments demonstrated a high uranium accumulation in bones of rats and dogs
[A23, A24, P31]. Several studies performed on rats have investigated the effects of uranium on bone
physiology. These studies found an impairment of bone growth and bone formation [P33, U1, U2].
These effects were associated with inhibition of endochondral ossification in mice [B45] and in rats
[D15]. Other rat experiments indicated that uranium induced ultra-structural alterations in osteoblasts
[T3]. An acute high dose of uranyl nitrate also delayed both tooth eruption and development in rats
[P33]. However, the retardation observed seven days after acute uranyl nitrate exposure was reversed
completely after 27 days [P34].

137. The in vivo results were confirmed by in vitro studies that showed that human osteoblasts were
sensitive to uranium effects (increased reactive oxygen species production, decreased alkaline
phosphatase activity, modified osteoblast phenotypes, genomic instability) [M38, M39, M42, T3]. An
increase in oxidant stress shown by an increased lipid peroxidation was also observed in vivo in rat
bone at high doses [G8].

2. Chronic exposure
138. On the basis of measured tissue concentrations and organ weights [F4, F8, L22, W26], bone tissue
may contain up to 6675% of the body burden of uranium following chronic exposure to uranium.

139. Several rat studies were designed to determine the uranium content of bones following chronic
contamination [A24, P3, R18]. Rodrigues et al. showed that accumulation of uranium in the rat skeleton
following intake via food increased to reach a plateau after one month [R18]. A long-term study of
Paquet et al. demonstrated that uranium accumulation in bones increased until 18 months [P3].

140. Some experiments investigated the long-term accumulation in bones of young male rats (between
the weaning and the post-puberty periods) [A23, R18]. Experiments performed on females during the
growing period indicate that concentration of uranium in the animals femora increased faster in the
early stages of the animal life, then saturating in adult animals.
398 UNSCEAR 2016 REPORT

141. Contamination of growing rats with uranium in food led to accumulation in bones (0.11.1 g
U/g bone) that exhibited the same pattern as the skeleton growing curve. Despite this accumulation,
there was no change in the bone mineral density (BMD) [R18]. However, uranium exposure of adults
led to a decrease in the BMD.

142. In contrast, a study of chronic exposure of growing rats to natural uranium for nine months via
daily oral ingestion (uranium-contaminated drinking water at 40 mg/L) affected the skeleton by
decreasing messenger RNA (mRNA) expression of genes involved in bone metabolism and decreasing
femoral cortical bone area, while no changes were observed in adult rats [W1]. Chronic contamination
by subcutaneous implantation of powdered uranium dioxide (125 mg/kg) in rats resulted in an
inhibition of bone formation as has also been described for acute poisoning with uranium [D15].

143. Bone metabolism was investigated in human populations receiving high levels of uranium via
drinking water by measurements of biochemical indicators of bone formation (osteocalcin and amino
terminal propeptide of type I procollagen) and a marker for bone resorption (serum type I collagen
carboxy-terminal telopeptide (CTx)) [K27]. The authors showed an elevation of CTx and osteocalcin
that could be associated with increased uranium exposure.

3. In vitro studies: mechanisms

144. The cytotoxic effect of uranium on rat and human osteoblasts is highly dependent on its
speciation. Exposure to non-toxic doses or non-toxic species of uranium induces the activation of two
markers of bone formation and mineralization (osteocalcin and bone sialoprotein), while their inhibition
is observed after toxic exposure [M37, M38]. This study highlights the importance of a controlled
speciation of uranium in toxicological studies.

145. In vitro transcriptomic studies performed on several human cell lines taken from kidneys or lungs
as representative targets have highlighted the involvement of osteopontin in the uranyl toxicity
mechanisms [P29]. This major non-collagenous protein involved in the organo-mineral homeostasis of
the bone presents a specific composition in acidic clusters associated with numerous phosphorylations,
and also a relative plasticity. Qi et al. showed with in vitro models that native phosphorylated
osteopontin binds uranyl with a nanomolar affinity and that this binding induces conformational
changes enabling the formation of a very stable complex of uranium [Q1].

4. Conclusion
146. In conclusion, acute and chronic exposure to uranium-induced biological effects on bone
metabolism such as the impairment of bone growth and of bone formation and the inhibition of
endochondral ossification or the delay of development. Most effects were observed in experimental
models (rodents or isolated humans cells). One study performed on humans exposed to high levels of
uranium in drinking water showed an elevation of type 1 collagen carboxy-terminal telopeptide (CTx),
a plasma marker for bone resorption, and of osteocalcin, indicator of bone formation [K27]. All these
studiesin animals and humanssuggest that uranium affects bone turnover. Osteoblasts appeared to
be the main cell targets of uranium. Some effects were also observed in humans, indicating that bone
may be a target of uranium chemical toxicity in humans.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 399

D. Lungs

1. Acute exposure
147. In general, the more soluble compounds (uranyl fluoride, uranium tetrachloride, uranyl nitrate) are
less toxic to the lungs but more toxic to distal organs due to easier absorption of the uranium from the
lungs into the blood and systemic transport [G1].

148. The behaviour of uranium in the lungs following inhalation has been studied in animal
experiments from various industrial settings [A31, C30, C31, D3, E3, S35, S38, S39, S40, W19, W20].
These studies demonstrated that the behaviour of uranium, its distribution, and its clearance in lungs is
dependent on the solubility of uranium compounds. Results of these experiments were used to confirm
or improve the ICRP models as described in chapter V on biokinetics.

149. Concerning the clinical effects of uranium in the respiratory system, authors reported alveolar
fibrosis in rats [M60], congestion and haemorrhage in rats and guinea pigs [L19] and bronchopneumonia
in rats and rabbits [D36] for uranium levels of between 10 and 100 mg/m3. The comparison of repeated
and acute uranium exposure via inhalation was performed in rats [M55] with aerosols varying from 116 to
375 mg/m3. The results showed that UO2 repeated pre-exposure by inhalation increased the genotoxic
effects of UO4 inhalation, when UO4 exposure alone had no effect. However, it is not clear if these effects
were due to a potentiation of the effect of UO4 by pre-treatment with UO2 or to a cumulative effect of the
two types of exposure.

2. Chronic exposure
150. Experimental studies performed on dogs, monkeys and rats with uranium show that inhalation of
natural uranium dioxide (UO2) at a mean concentration of 5 mg U/m3 for periods as long as five years led
to pulmonary neoplasia and fibrosis [L18]. Pulmonary neoplasia developed in a high percentage of the
dogs examined twosix years after exposure. Pulmonary and tracheobronchial lymph node fibrosis,
consistent with radiation effects, was dose dependent and more marked in monkeys than in dogs.

151. Nose only inhalation in rats showed that chronic inhalation of natural uranium ore dust (without
significant radon content) created a risk of primary malignant and non-malignant lung tumour
formation [M49]. The frequency of primary malignant lung tumours was 0.016, 0.175 and 0.328 and
primary non-malignant lung tumours 0.016, 0.135 and 0.131 in the control, low (19 mg/m3 leading to
an absorbed dose of 0.87 Gy) and high (50 mg/m3 leading to an absorbed dose of 1.64 Gy) aerosol
exposed groups, respectively, without difference in tumour latency between the groups. Despite lymph
node specific burdens ranging from 1 to 60-fold greater than the specific lung burden in the same
animal, no lymph node tumours were observed.

3. In vitro studies: mechanisms

152. Inhalation of soluble uranyl nitrate led to uranium uptake in the lysosomes of alveolar
macrophages and precipitation in the form of insoluble phosphate [B20]. A study by Lizon and Fritsch
on alveolar macrophages with uranium concentrations from 5 105 to 103 M showed that the toxicity
of uranium was concomitant with the presence of insoluble forms in the culture medium [L40].
400 UNSCEAR 2016 REPORT

153. Orona and Tasat analysed rat alveolar macrophages to better understand the pathological effects
associated with DU inhalation, metabolic activity, phagocytosis, genotoxicity and inflammation [O6].
The effects of 12.5200 M DU seemed to be dose-dependent, most observed from 100 M. At low
doses, DU induced phagocytosis and at high doses, provoked the secretion of TNF. Apoptosis was
induced through the whole range of doses tested. The uranium-induced TNF secretion by
macrophages was consistent with results from previous studies [G5, Z11]. Lung fibrosis was correlated
with abnormal expression of TNF and IL6, which could be antagonized by antibodies against TNF
[Z11, Z12].

154. Orona and Tasat suggested that at low doses (12.5 M), DU induced O2, which may act as the
principal mediator of DNA damage, while at higher doses (200 M), the signalling pathway mediated
by O2 may be blocked, and the prevailing DNA damage would be by TNF [O6]. A study by Monleau
et al. indicated that exposure to DU by inhalation resulted in DNA strand breaks in broncho-alveolar
lavage cells and in an increase in inflammatory cytokine expression and production of hydroperoxides
in lung tissue [M56].

155. In addition to effects on pulmonary macrophages, a study showed that uranium (from 0.25 to
1 mM) induced significant oxidative stress in rat lung epithelial cells followed by a concomitant
decrease in the antioxidant potential (glutathione and superoxide dismutase) of the cells [P13]. Further,
some publications indicated that not only soluble (uranium acetate) but also particulate (uranium
trioxide) uranium induced concentration-dependent cytotoxicity in human epithelial cells [W22].

156. Depleted uranium was clastogenic and induced chromosomal aberrations after 48 hours [L4]. Xie
et al. [X1] found a loss of contact inhibition of these cells after particulate DU exposure, with
chromosome instability and a change of cell phenotype, suggesting a neoplastic process.

4. Conclusion
157. Experimental acute and chronic studies demonstrated possible effects of uranium on fibrosis and
tumour formation in lungs. Data obtained in vitro indicated the induction of genotoxic lesions (DNA
strand breaks) and activation of inflammatory pathways on alveolar macrophages with high uranium
concentrations. However, these underlying mechanisms were not investigated in vivo, limiting the
relevance of these molecular effects to understand the link between uranium exposure and observed
pathologies (fibrosis and tumour formation).

E. Liver
158. Accumulation of uranium was observed in the liver of rats after injection, implantation or oral
administration (uranyl nitrate) but at a lower concentration than in the kidneys for the same level of
exposure [C35, P3, P11]. Acute exposure led to decreased liver weight and increased plasma
transaminase levels, both indicators of liver hepatotoxicity in rats [G24, M57, O7] and in mice [O11].

159. More recently, some results were published on hepatic effects of uranium following in vivo
exposure of rats [D29, G29], which demonstrated that some enzymes of xenobiotic metabolism, notably
the cytochrome P450 of type 3A (CYP3A), were modified by chronic contamination with DU. The
time-course study performed at 1, 3, 6, 9 and 18 months after exposure indicated that significant
changes were observed at six and nine months [G29], with a 50% decrease in the mRNA level of
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 401

CYP2C11 at six months and an increase in gene expression of CYP3A at nine months. Concerning the
doseresponse study, the most substantial effects were observed in the liver of rats after nine months of
exposure to 120 mg/L: CYP3A gene and protein expression and enzyme activity all decreased by more
than 40% [D29, G29].

160. Several studies investigated the functional consequences of a co-exposure to uranium (uranyl
nitrate) and the drugs chlorzoxazone [C26, M57], ipriflavone [C25], theophylline [Y4] or paracetamol
[G27, R25] in rats. Altered drug pharmacokinetics was observed with high dose and chronic low dose
exposure to uranium, which could be due to liver dysfunction. Functional toxicity of uranium was also
estimated by measuring xenobiotic detoxification enzymes and their gene expression levels, protein
levels or enzyme activities. Some studies reported altered levels of xenobiotic detoxification enzymes
[G27, P6] while others reported altered pharmacokinetic metabolism of certain drugs after acute [C25,
C26, M57] and chronic exposure to uranium [G27]. Oxidative stress may also occur in hepatocytes
exposed to uranium, as demonstrated by mitochondrial or lysosomal alterations [P26] or
metallothionein involvement [M47].

161. Chronic contamination with DU (uranyl nitrate, 1 mg/rat/day for 9 months) affected cholesterol
metabolism in the rat liver [R2]. Relative mRNA levels of the enzyme cholesterol storage were
modified, and also the proteins involved in the transport and the regulation of cholesterol homeostasis.
One accident involving workers with extensive skin exposure to uranium has been reported in the
literature [L48, S16, S49]. These results were similar to observations from the clinical follow-up of
workers with acute uranium compound intoxication. Toxic hepatitis occurred for several days post-
accident with recovery one month later [S16]. The result of 33-year follow-up showed that liver
functions were normal [S49]. In conclusion, experimental studies performed in rats indicated that
uranium exposure induced some biological effects on the liver, without a high uranium accumulation in
this organ. These biological effects did not lead to the appearance of pathologies in animals. However,
in humans, one study reported a transient toxic hepatitis.

F. Brain

1. Behavioural effects
162. Adult rats exposed to uranium showed subtle but significant behavioural changes. Increases in
locomotor activity, in line crossing and in rearing behaviour were observed after exposure of rats to DU
in drinking water at 2 or 4 mg/kg per day or after inhalation [B52, M54]. Females seemed to be more
resistant: unlike males, they did not show locomotor symptoms [B51]. Exposure to uranium to 0.1, 0.3
or 1 mg uranium/kg also affected working memory with poorer performance (decreased latency) in a
lightdark passive avoidance response system [B6]. The spatial working memory measured by the
percentage of spontaneous alternation was significantly lower after exposure to DU by inhalation and
after ingestion of 1 mg per day of 4% enriched uranium [H24, H25, M54]. Lastly, exposure at this dose
of enriched uranium had a deleterious effect on anxiety and increased the amount of rapid eye
movement in sleep [H24, H25, H26]. However, DU had no significant effect in the same experimental
conditions [H24, H26, L30]. A recent review summarized the different effects of uranium on behaviour.
It can lead to neurobehavioural impairments, including increased locomotor activity, perturbation of the
sleep-wake cycle, decreased memory, and increased anxiety. The mechanisms underlying these
neurobehavioural disturbances are not clearly understood [D16].
402 UNSCEAR 2016 REPORT

163. Mouse experiments were conducted on ApoE/ mice that had been genetically knocked out for
the apolipoprotein E gene, the product of which regulated cholesterol metabolism. These mice had
hypercholesterolemia and expressed biochemical markers of Alzheimers disease. Administration of
DU to these mice resulted in impaired memory compared to unexposed ApoE/ mice [L33]. This
cognitive effect was associated with a trend toward higher total cholesterol content in the cerebral
cortex (+15%). This study demonstrated that some pathological conditions may increase sensitivity to
uranium. Table 12 summarizes the main studies on animal behaviour after uranium exposure.

Table 12. Summary of studies of uranium exposure on rodent behaviour

im: intramuscular; DU: depleted uranium; EU: enriched uranium

Uranium
Species Exposure conditions Main biological effects Reference
compound
Uranyl Adult rat 2 and 4 mg/kg/day (water) DU Increased locomotor activity [B52]
acetate during 2 weeks and 6 months

Uranyl Adult rat 2.5, 5 and 10 mg/kg/day (water) Increased locomotor activity, [B11]
nitrate DU during 3 months decreased spatial memory

Uranyl Adult rat Inhalation DU 30 min at Increased locomotor activity and [M54]
dioxide 197 mg/m3, 4 days a week for decreased spatial memory one
3 weeks day post-exposure

Uranyl Adult rat im. 0.1, 0.3 and 1 mg/kg DU Decreased locomotor activity, [B6]
acetate decreased grip strength,
decreased working memory at 6,
13, 20 and 27 days post-dosing

Uranyl Adult rat 1 mg/kg/day (water) 4% enriched Decreased spatial memory, [H25]
nitrate uranium (EU) during 1.5 months increased of anxiety

Uranyl Adult rat 1 mg/kg/day (water) 4% EU during Increased paradoxical sleep [L29]
nitrate 1.5 months

Uranyl Adult rat i.p. 144 g /kg No change in sleep-wake cycle [L30]
nitrate 1 and 3 days

Uranyl Adult rat 1 mg/kg/day (water) 4% EU during Decreased spatial memory [H24]
nitrate 3 or 9 months

Uranyl Fetal 1, 2 and 4 mg/kg/day DU during Accelerated appearance righting [B50]

acetate mouse gestation and lactation reflex, forelimb placing, grasping,
swimming and weight gain

Uranyl Young 1, 2 and 4 mg/kg/day (water) DU Decreased locomotor activity, [B51]

acetate mouse during 21 days since the birth decreased working memory

Uranyl Fetal rat 10, 20 and 40 mg/kg/day (water) Decreased learning in pups [A10]
acetate DU during 3 months (male)

Uranyl Adult rat 1 mg/kg/day (water) DU during Decreased spatial memory [B15]
nitrate 2 months since the birth

Uranyl Adult 4 mg/kg/day (water) DU during 3.5 Impaired memory [L33]

nitrate mouse months
(ApoE/)
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 403

2. Neurotransmitters
164. Several experimental studies suggest that uranium can induce changes in neurotransmitter levels:
acetylcholine levels were unchanged in the hippocampus after exposure to 1 mg DU per kg and day but
were decreased in the cortex [B14]. Chronic DU contamination induced a fall in the rate of
acetylcholine (ACh) and AChE activity in the entorhinal cortex and cerebellum [B15, B55]. This
disturbance of cholinergic function was associated with a decrease in gene expression of several
proteins [B14]. These studies suggest that the modifications of neurobehavioural tasks following
uranium exposure could be linked with a change of AChE activity in the cerebral cortex [B52].

165. A study performed in Sprague-Dawley rats (following 1.5, 6, or 9 months with 2.7 mg U/kg per
day) reported that AChE activity was not significantly affected in the striatum, hippocampus, or frontal
cortex at any time point, but it was significantly decreased in the cerebellum at six months [B55].
Depleted uranium exposure at 2.7 mg/kg per day also induced a significant decrease in AChE activity
in the striatum and cerebral cortex [B14, B55]. A doseresponse study performed at nine months
following chronic contamination indicated that uranium effects (15% decrease in AChE activity) were
independent of uranium content in drinking water (from 0.2 to 120 mg/L).

166. A 1.5-month ingestion of 1 mg U/kg per day increased the dopamine level in the hypothalamus
[B55]. Chronic exposure produced a significant decrease in the serotonin (5HIAA) level and the
serotoninergic (5HT-ergic) turnover ratio in the frontal cortex and also a decrease in the dopamine
(DOPAC) level and dopaminergic (DA-ergic) turnover ratio in the striatum [B55]. It appeared that
disruption of these systems differed depending on the brain structure considered, the time of exposure
and the degree of uranium enrichment [A1, B14, B55, L29].

3. Oxidative stress
167. One specific mechanism by which uranium leads to neuro-effects could be oxidative stress. The
behavioural changes correlated with lipid peroxidation in the brain induced by uranium [B52, G8]. The
gene expression or enzymatic activity of the main antioxidant enzymes, i.e. superoxide dismutase,
catalase and glutathione peroxidase, increased significantly in the hippocampus and the cerebral cortex
after exposure to DU and decreased after exposure to 1 mg of 4% enriched uranium per kg per day
[L31, L38]. The cell response to DU could be interpreted as a defensive mechanism towards free
radical damage to cerebral tissue (increase of several antioxidant agents in order to counteract the
oxidative stress). The oxidative stress induced by the enriched uranium is possibly too high to be
counteracted by the cell defences.

168. A doseresponse study performed nine months after chronic contamination with DU from 0.2 to
120 mg/L indicated that uranium affected the activity of these enzymes differently (diminution for SOD
and increase for GPx or Catalase). The gene expression of inducible nitric oxide synthase, the enzyme
that synthesizes nitric oxide, increased significantly after chronic exposure to 1 mg DU per kg and day
[L38]. Repeated administration of DU as uranyl acetate during seven days also increased the nitrite
levels in the brains of rats [A1].
404 UNSCEAR 2016 REPORT

4. Exposure of developing animals

169. In rodents, exposures at 1, 2 or 4 mg DU/kg per day during development accelerated the
appearance of several types of behaviour: righting reflex (for example, when the rat is put on its back, it
turns over immediately), placing reactions (for example, the rat is held by the tail over a table until its
whiskers get near, when it puts its paws on the table), grasping (the rat is picked up and the palm is
touched with a wire and the response is to grip the wire), swimming and weight gain [B50].

170. Animals exposed to uranium at 1, 2 or 4 mg U/kg per day during development had a decreased
locomotor activity [B51] and their performance was worse on a test of working memory [B15]. The
spatial learning of the offspring of uranium-exposed male rats at 1, 2 or 4 mg U/kg per day for learning
was also affected [A10]. A doseresponse study (0, 10 or 40 mg/L of uranium in drinking water)
indicated no significant uranium effect on behaviour at 10 mg /L, and an impairment of object
recognition memory (20%) at 40 mg/L [L32].

171. Neurogenesis processes during pre- and postnatal brain development were studied in rats by
investigating the structural morphology of brain, cell death and cell proliferation after chronic exposure
to drinking water contaminated with uranium (40 and 120 mg/L) [L26]. Major changes were observed
at 120 mg/L, both during prenatal and postnatal periods. At the highest dose, DU caused opposite
effects during brain development on cell proliferation and cell death processes, mainly between prenatal
and postnatal development. These modifications did not have a major impact on brain morphology but
they could affect the next steps of neurogenesis and disrupt the organization of the neuronal network.

172. Some studies were published on the comparative effects of depleted and enriched uranium in rats
[H24, L31]. Chronic exposure to 4% enriched uranium for 1.5 months through drinking water increased
the amount of paradoxical sleep, reduced the spatial working memory capacity and increased anxiety
while no effect was observed following exposure to DU [H24]. These cognitive effects were associated
with imbalance of oxidative stress [L31]. Indeed, lipid peroxidation was increased in brain after
enriched uranium exposure but not after DU exposure. Enriched uranium induced a decrease of anti-
oxidative enzymes, and DU induced an increase of these anti-oxidative enzymes.

5. Conclusion
173. Animal studies suggest that uranium can have some negative effects on the behaviour of mature
animals that could be explained at a neurochemical level (neurotransmitters, oxidative stress) for
highest doses. In addition, some results obtained in rats and mice demonstrate differential effects
between depleted and enriched uranium, suggesting the importance of the radiological toxicity of
uranium. The data also suggest that the developing animal may acquire a specific sensitivity to uranium
effects. In humans (workers or public), the correlation between behavioural symptoms and exposure to
uranium was not demonstrated. Thus, data obtained in animals are only suggestive until they can be tied
to meaningful human research.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 405

G. Reproduction and development

1. Female reproductive function

174. Results for measurements of uranium concentrations in gonads were contradictory depending on
the study (species, administration pathway, uranium dose and exposure time). In fact, high uranium
accumulation is found in some fish and birds. For instance, significant uranium accumulation was
measured in female fish gonads (Danio rerio), corresponding to >20% of the relative burden [S20].

175. A dose-dependence of uranium concentration in ovaries was measured in rats and their offspring
after chronic oral exposure via food [H9, H10]. The accumulation of uranium was higher in the second
exposed generation. Other experimental studies performed on mice failed to report uranium
accumulation in the ovary after contamination by nitrate uranyl with doses up to 400 mg/L for uranium
content in drinking water [A21, F2, R9].

176. Some studies found that uranium affected oocyte quality in vivo with a 50% reduction in the
proportion of healthy oocytes from 20 mg/L [F2] and germinal vesicle oocytes cultured in vitro in the
presence of 1 mM uranyl acetate and observed for 72 hours [A21]. In vivo, these morphological effects
were observed from uranium content in drinking water above 5 mg/L [F2]. A study with similar
approaches (uranium ingestion via drinking water in mice) indicated similar results, with increased
dysmorphism of oocytes in contaminated groups (from 2.5, 5 and 10 mg U/kg per day chronically
administered in drinking water for 40 days) in a dose independent manner [K24]. In addition, a study
on mice contaminated in utero by uranium levels in drinking water from 0.5 to 60 g/L, showed an
increase in uterine weight and a decrease in primary follicles [R9].

177. Reproductive effects following chronic oral uranium exposure were observed in rats exposed
during the first and the second generation [H10]. No effect were observed in F0 rats, but pregnancy rate,
normal labour rate, and survival rate were decreased in offspring [H10]. In vitro organ culture system
was used to investigate the effects of uranium on human gonads during the first trimester of gestation
(712 weeks) [A13]. Uranium at 0.05 mM increased the apoptosis rate, decreasing the germ cell density
of human fetal ovaries. The authors also demonstrated that human fetal germ cells were more sensitive
to uranium than mouse germ cells.

2. Male reproductive function

178. Some animal studies indicated accumulation of uranium in testes with 0, 5, 10, and 25 mg/kg/day
of uranyl acetate dehydrate before mating and up to 21 days post birth [P7]. A linear dose-dependence
was found in testes of Japanese quail, with a ratio of accumulation similar in testes and in kidney [K25].
A dose-dependence of uranium concentration in testes was measured in rats and their offspring after
chronic oral exposure via food [H10]. The accumulation of uranium was higher in the second
generation.

179. Concerning the effects of DU on sex hormone levels, experimental studies performed in male rats
showed differing results. One experiment performed in male rats indicated an increase in testosterone
and luteinizing hormone levels and a decrease in follicle stimulating hormone level after a four-month
contaminated food ingestion [H10], while a nine-month contaminated drinking water ingestion did not
lead to changes in testosterone and 17-estradiol levels [G20]. The different uranium doses 4 and
40 mg/kg per day in the studies of Hao et al. [H10], and 2.7 mg/kg per day in the study of Grignard et
al. [G20] were used to explain the different levels. However, changes in testosterone levels were not
406 UNSCEAR 2016 REPORT

observed in depleted-uranium Gulf War veterans [M19, M20]. A nine-month chronic oral exposure to
enriched uranium produced a significant increase in the blood levels of testosterone at 40 mg/L in
drinking water, while no effect was observed with DU [G20].

180. Some reports on rodents highlighted a negative impact of uranium on male reproductive function,
including a decrease in male fertility and in the spermatid number per testis with a few
histopathological effects on the seminiferous tubules and interstitium after chronic exposure [L37,
L41]. Although some abnormal morphological forms and sperm parameters measured were affected by
uranium exposure, these changes were independent of the uranium dose levels from 10 to 40 mg/kg per
day corresponding to ~200 to 800 mg/L in drinking water, respectively [L37]. Further, a dose-
independent decrease in the pregnancy rate was observed in untreated females mated with male mice
exposed to between 10 and 80 mg/kg per day of uranium [L41].

181. Implantation of DU pellets did not change the concentration, motion and velocity of sperm, and
there is no evidence of detrimental effects of uranium on mating success, suggesting that implantation
of up to 20 DU pellets of 12 mm (760 mg) in rats did not have an adverse impact on male
reproductive success, sperm concentration, or sperm velocity [A19]. Testicular histopathological
abnormalities with deformations of seminiferous tubules (marked reduction in the seminiferous tubule
diameter) were observed in mice after high acute exposure [J1]. A seven-day daily intraperitoneal
administration of uranyl nitrate (0.5 mM/kg) induced a marked reduction in the seminiferous tubule
diameter and gametogenic count, with signs of testicular necrosis and exfoliation of germ cells,
including karyolysis and karyorrhexis figures.

182. The contribution of new in vitro models, such as organotypic culture systems, helps the
understanding of the underlying action mechanism of chemicals. This approach was used as a
toxicological test to evaluate the effects of various compounds, including uranium, on gametogenesis
and steroidogenesis in rat, mouse and human testes [H1]. Some effects on germ cell development
(reduction of the number of germ cells) or Leydig cell function were observed for uranium
concentrations above 5 105 M in human testis cells and above 5 104 M in rat testis cells [A13].

183. Uranyl fluoride injected in vivo into mouse testes led to an increase in the frequency of
chromosomal aberrations in spermatogonia and primary spermatocytes [H27]. These results were
confirmed in another study that indicated the highest effects when doses of UO2F2 increased up to
6 mg/kg at 12 days post-exposure [H28]. A study performed with enriched uranium demonstrated
chromosome aberrations in spermatogonia [Z15]. Chromosome fragmentation, translocation,
polyvalence in primary spermatocyte and DNA strand breakage were observed in sperm. Effects of
uranium on human male reproduction studied in Gulf War I veterans did not evidence deleterious
effects on sperm quality, including volume, concentration, total sperm count, and functional parameters
of sperm motility [M20, M21, M22].

3. Effects on embryos and development

184. The effects of uranium on embryotoxicity were studied in rats and mice after acute exposure
[D17]. Subcutaneous injections of uranyl acetate dihydrate (0.5, 1 and 2 mg/kg/day) in mice from day
6 to day 15 of gestation induced various effects [B44]. Although it was not dose-related, embryo-
toxicity also occurred in all uranium-treated groups (significant increases in the number of non-viable
implantations and in the percentage of postimplantation loss). Both the maternal NOAEL and the
NOAEL for embryotoxicity of uranyl acetate dihydrate were below 0.5 mg/kg/day, whereas the
NOAEL for teratogenicity was equal to 0.5 mg/kg/day.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 407

185. Few animal studies investigated the effects of uranium exposure on development and results were
conflicting, depending on the quantity of administrated uranium and the exposure duration.
Subcutaneous injections of uranyl acetate dihydrate from day 6 to day 15 of gestation induced
malformations detected at 1 and 2 mg/kg/day in mice [B44], while no effect was observed at
0.5 mg/kg/day.

186. Paternain et al. demonstrated that embryo lethality could be observed in mice contaminated at
25 mg/kg/day [P7]. Significant increases in the number of dead young per litter were seen at birth and
at day 4 of lactation in the 25 mg/kg/day group. The growth of the offspring was always significantly
lower for the uranium-treated animals. Since no effect was observed at lower doses (5 and
10 mg/kg/day), the results suggest that uranium does not cause adverse effects on fertility, general
reproductive parameters, or offspring survival at the concentrations usually ingested by man.

187. In vitro studies were also conducted on one-cell mouse embryos in culture medium with uranyl
nitrate at concentrations of 26, 52, 104 and 208 g/mL [K23]. The results obtained showed that
concentrations from 26 g U/mL induced the delay of embryo development and the impairment of
blastomere proliferation. A study of acute toxicity was performed in mice receiving 4 mg/kg of DU per
intraperitoneal injections (i.p.) at day 11 of gestation and observations were made 4 days later.
Paradoxically, the authors found an increase in the fetus length and weight [M46].

188. Subcutaneous injections of uranyl acetate (0.415 and 0.830 mg/kg/day) were given to pregnant
rats on days 6 to 15 of gestation [A9]. Maternal toxicity and embryotoxicity were noted at the higher
dose, while fetotoxicity was evidenced at both doses. The fetotoxicity was evidenced by significant
reductions in fetal body weight and increases in the total number of skeletal abnormalities [A9].

189. Some studies indicated developmental toxicity of uranium, including teratogenicity, following an
acute subcutaneous administration of 1 or 2 mg/kg/day [B44, D18]. A study by Zhu et al. [Z15] was
performed with enriched uranium injected in rat testes, causing skeletal abnormalities in fetal rats with
a positive correlation to the injected dose.

190. Hereditary effects of uranium were investigated in rats following the implantation of uranium
pellets in muscle (up to 12 DU pellets corresponding to 360 mg) [A20]. This study indicated no
changes in sperm motility and ribcage malformations, suggesting that uranium was not a significant
reproductive or developmental hazard. This study is in accordance with another study that indicated
that uranium did not cause any adverse effects on fertility, general reproductive parameters, or offspring
survival at the concentrations usually ingested by humans [P7]. One study performed on mice
investigated the transmission of genetic damage to offspring of fathers contaminated with uranium via
depleted-uranium-implanted pellets [M45]. The authors demonstrated a dose-dependent increase in
mutation frequency in the offspring. Congenital malformations [A6, A7, S48] or birth defects [A3, B54,
F1] were reported in some human populations, but these effects were not correlated to a quantification
of uranium exposure.

4. Conclusion
191. Some publications focused on reproduction and development issues after exposure to uranium in
animals (rats and mice). These studies indicated that both male and female reproductive functions
(quality of oocytes and sperm parameters, embryo viability, and the development processes) may be
affected by acute and chronic exposure. However, these effects were observed for exposure levels
greater than either typical occupational or environmental levels of exposure.
408 UNSCEAR 2016 REPORT

H. Other organs

1. Skin
192. A study of acute exposure of rabbits was conducted using different administrations of uranium
oxide: subcutaneous deposit of ~30 mg U3O8; cutaneous deposit of 70 mg of U3O8 or 5 g of uranium
acetate (1.9 kBq of 233U) [W3]. Uranyl nitrate, in ethereal or aqueous solution, produced a superficial
coagulation necrosis within a few hours [O4]. A similar but delayed effect was seen sevennine days
following the application of powdered hygroscopic uranium pentachloride to the skin. Uranium
tetrachloride in suspension in lanolin caused a moderate local erythema of the skin at the site of
application, which disappeared in onetwo days.

193. In vitro studies on primary cultures of rat skin keratinocytes and fibroblasts [P15] showed a
greater decrease in proliferation rate associated with a greater mortality rate in rat skin keratinocytes
than in rat skin fibroblasts after uranium exposure. This can be explained by the three times higher
ability of keratinocytes to incorporate uranium compared to fibroblasts. This greater capacity of
epidermal cells than dermal cells to incorporate uranium was confirmed in vivo in hairless rats
following topical contamination with uranyl nitrate.

194. The consequences of protracted exposure to uranium were investigated using guinea pigs and
rabbits [O4]. Guinea pigs that were repeatedly exposed to uranyl nitrate exhibited a superficial
coagulation necrosis and an inflammation of the epidermis. Their skin was in a constant state of
encrustation and desquamation accompanied by rapid regeneration from beneath. Rabbits that were
repeatedly exposed at multiple dermal sites exhibited an effect similar to that seen after single acute
exposure, but those rabbits that were repeatedly exposed at the same site developed severe dermal
ulcers after five to ten applications of the compound. Another in vivo study in rats revealed that long-
term exposure of the dermis to uranium (U3O8 at 0.012 g/d for 30 daily topical applications) led to an
epidermal atrophy which, in turn, resulted in an increased permeability of the skin [U2].

2. Endocrine system and metabolism

195. Vitamin D is essential for the homeostasis of calcium and phosphorus in the body. A few studies
have determined the effects of acute or chronic uranium contamination on the metabolism of vitamin D
in rats. Contamination by DU via drinking water at 40 mg/L [T11] induced a decrease in the blood
levels of vitamin D in rats following a nine-month chronic exposure. Moreover, uranium targeted key
transcription factors (PPAR, PPAR, HNF-1, HNF-4, LXRr, RXR, and VDR) involved in this
metabolism [T11, T12]. However, these molecular changes did not lead to the emergence of disease
associated with vitamin D metabolism.

196. Chronic contamination by DU (uranyl nitrate, 2.7 mg/kg/day for 9 months) affected cholesterol
metabolism in the liver and brain [R1, R2], mRNA levels of the enzymes involved in the cholesterol
storage were modified in the liver and brain, and mRNA levels of enzyme involved in the cholesterol
synthesis were modified in the brain. Uranium also affected the proteins involved in the transport of
cholesterol and the regulation of cholesterol homeostasis. Thus, a chronic ingestion of uranium
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 409

(40 mg/L in drinking water) caused subtle molecular effects on metabolism in the liver and brain of
rats. However, overall cholesterol levels were unaltered in this study using 40 mg/L uranium in
drinking water.

197. Studies of rats exposed to uranium showed both disruptive effects on the reproductive system and
estrogenic effects. Chronic contamination with DU (560 Bq/L) in drinking water produced no change in
the blood levels of the two principal steroid hormonesoestradiol and 17-testosterone (synthesized by
the testis), whereas contamination with enriched uranium (1,680 Bq/L; 40 mg/L) produced a significant
increase in the blood levels of testosterone [G20]. Consistent with the absence of hormonal changes
with DU, chronic contamination with DU did not induce a change of gene expression. However, the
expression of enzymes involved in the metabolism of hormones was amplified following a nine-month
exposure to enriched uranium. In addition, enriched uranium increased the gene expression of
transcription factors (RXR, LXR, FXR, SHP, SF-1, DAX-1) that positively regulate steroid
metabolism. These results show a differential effect of depleted and enriched uranium contamination on
testicular steroidogenesis [G20].

198. The human body is frequently exposed to potentially toxic compounds and is able to metabolize
them in order to protect itself. Xenobiotic-metabolizing enzymes, including cytochrome P450
(CYP450), play a central role [G25]. The kidneys and liver can metabolize many drugs or other
xenobiotics. Disturbances of this system have been demonstrated in vitro [M43] and in rats exposed to
nephrotoxic [M57] or non-nephrotoxic uranium concentrations [P6, S29]. These studies showed that the
expression and activity of CYP450 can be modified by uranium exposure. The consequences of such
modifications on xenobiotic metabolism were investigated using acetaminophen [G26]. Rats
contaminated with DU presented slower plasma acetaminophen elimination and also more marked renal
histological changes and an increase in blood markers of liver damage (at 40 mg/L during 9 months).
However, only slight effects of uranium were reported on enzymes of xenobiotic metabolism when
acetaminophen was administrated to rats as a single therapeutic treatment [R25].

199. In conclusion, studies performed on animal models (rat and mouse) chronically exposed to
radionuclides showed that the chronic ingestion of uranium resulted in subtle biological effects on
various metabolic systems. These modifications did not lead to the appearance of pathologies, even for
uranium levels in drinking water up to 40 mg/L. The observed biological effects probably resulted from
an adaptive response to the internal contamination.

3. Immune and haematopoietic systems

200. Very few studies have investigated the effects of uranium exposure on the haematological system.
Reduced erythropoiesis might be expected owing to accumulation of uranium in bone in proximity to
bone marrow and renal damage might result in a decrease in the number of red blood cells. A study
using rats found that chronic ingestion of uranium at 40 mg/L in drinking water for nine months led to
kidney deterioration which may have been responsible for an observed decrease in the red blood cell
count; there was an associated modification of spleen erythropoiesis and levels of molecules involved
in erythrocyte degradation [B19].

201. A study by Giglio et al. was performed to assess the effect of uranium (uranyl nitrate) on the rate
of erythropoiesis in rats [G9]. The authors showed that a single injection of uranium at 1 mg/kg induced
a transient depression of the red cell volume between 7 and 14 days. These effects were associated with
a decreased Epo production and direct or indirect damage of erythroid progenitor cells.
410 UNSCEAR 2016 REPORT

202. An experimental study using mice fed uranium-contaminated food [H11] compared the effect of
concentrations of DU of 0, 3, 30 and 300 mg/kg feed for a duration of four months. The most
significant effects were observed in the 300 mg/kg group while the effects were either minor or
indiscernible in the other groups. At high dose, the authors observed decreased immune function,
manifesting as decreased secretion of inflammatory mediators in the peritoneal macrophages, and also
reduced cytotoxicity of the splenic natural killer cells. Moreover, the cellular and humoral immune
functions were abnormal (decreased proliferation of the splenic T cells, proportion of the cluster of
differentiation (CD) 3+ cells, ratio of CD4+/CD8+ cells and delayed-type hypersensitivity, and increased
proliferation of the splenic B cells, total serum immunoglobin (Ig) G and IgE, and proportion of splenic
mIgM+mIgD+ cells). The authors concluded that chronic intake of high doses of DU (300 mg/kg) had a
significant impact on the immune function, most likely due to an imbalance in T helper Th1 and
Th2 cytokines.

203. Studies on rats have addressed the effects of uranium exposure on the mucosal immune system of
the intestine following ingestion of uranium-contaminated drinking water [D26, D27, D28]) and in
lungs following inhalation [M54, M56, P6]. Studies performed on the rat intestine found that Peyers
patches (aggregated structures of gut-associated lymphoid tissue) were a site of retention of uranium
following chronic ingestion (with uranium content in drinking water of 40 mg/L) after 9 months,
without inducing any biological effects on the function of Peyers patches [D26]. However, chronic
ingestion of uranium (nine months at 40 mg/L in drinking water) led, in the long term, to some changes
in the immune cell populations in lamina propria, diffuse gut-associated lymphoid tissue, notably
increase in neutrophil number (+300%) and decrease in macrophage (50%), and mast cell number
(30%) in rats [D28].

204. Dublineau et al. indicated that immune cell populations in the intestine (neutrophils, mast cells,
macrophages) did not vary after 9 months of uranium exposure in rats, at any exposure level (from
0.2 to 120 mg/L in drinking water) [D29]. The authors observed an increase in mRNA levels of
cytokines (IFN-gamma, IL-10, and CCL-2), at uranium levels in drinking water >20 mg/L, and a
decreased protein expression of cytokines (IFN-gamma, IL-10, and TNF). Such effects may be
explained by uranium interaction with proteasome [M6]. Changes in immune cell populations of the
intestinal wall were not reported in the short term following acute contamination with a sublethal dose
of uranium in rats (204 mg/kg of DU per os) [D27].

205. Immunological changes of long-term uranium exposure were investigated in mice fed with
various DU doses (0, 3, 30 and 300 mg/kg food) [H11]. A four-month contamination of animals with
300 mg U/kg food led to a decreased secretion of nitric oxide, interleukin (IL)-1, IL-18, and tumour
necrosis factor (TNF)- by the peritoneal macrophages, a decreased proliferation of the splenic T cells
and increased proliferation of the splenic B cells associated with an increase in total serum
immunoglobin (Ig) G and IgE, and proportion of splenic mIgM(+)mIgD(+) cells. Such effects were not
observed for the lowest groups (3 and 30 mg U/kg food).

206. Some studies reported effects of uranium exposure on pulmonary immune cells following
inhalation in rats [M54, M56, P6]. An uptake of uranium by pulmonary macrophages was found
(phagocytosis), followed by an accumulation of uranium in lysosomes [B20, M61]. This accumulation
was associated with release of pro- and anti-inflammatory cytokines in vivo [M54, M56] or in vitro
[G5, W8, Z12]. Inflammatory cells were observed in proximity to aggregates of uranium particulates
[L9]. This inflammatory cell response may lead to oedema and bronchial inflammation [B8, M59]. An
induction of apoptosis was observed in vitro in macrophages and T lymphocytes (CD4+) [W8].

207. In the follow-up of United States veterans of the first Gulf War published by McDiarmid et al.
[M25], the authors reported for the first time on the analysis of immune function in Gulf War I veterans
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 411

exposed to DU, investigating the lymphocyte response to stimulation by two T-cell mitogens. The
results on released cytokines (IFN, IL-10 and IL-2) in the high- and low-uranium exposure groups did
not provide evidence of an effect of uranium on the immune cells.

208. Haematological parameters were recorded in some human populations exposed to uranium.
Haematocrit, haemoglobin and RBC content from uranium processing site workers remained within the
normal ranges [S10]. Haemoglobin concentrations, erythrocyte counts, haematocrit values, and mean
corpuscular volumes were determined on a group of miners in a uranium mine [V6]. There were small
differences of mean values from those of controls, such as diminution of haemoglobin concentrations
and erythrocyte counts when uranium exposure time increased, but the mean values remained within
normal limits. A clinical study of Gulf War veterans showed that soldiers exposed to uranium had a
reduction in haemoglobin and haematocrit levels [S32].

209. In conclusion, it appears that slight modifications of the immune system, either systemic or
mucosal (including intestines and lungs), were induced following uranium exposure only at high doses,
at concentrations usually not incorporated by humans.

4. Cardiovascular system
210. Few human data are available concerning uranium effects on the cardiovascular system. A study
was performed on people who had used drinking water from drilled wells containing high levels of
uranium (up to 1,500 g/L) for several years. This study suggested that uranium exposure was
associated with greater diastolic and systolic blood pressure [K28]. Two other studies were performed
among residents living near the Fernald Feed Materials Production Center, which functioned as a
uranium processing facility from 1951 to 1989 [P22, W2]. The systolic and diastolic mean blood
pressure levels were higher for population groups living near this uranium plant than those found in the
general population, and for most age- and sex-specific groups. However, neither distance nor direction
from the site influenced the blood pressure measurement results, suggesting that these findings are not
exposure-related [P22]. Another study showed similar results: women with higher uranium exposure
had elevated systolic blood pressure compared to women with lower exposure, but the changes
observed in diastolic blood pressure or hypertension were not related to exposure level [W2].

211. Concerning experimental studies, there is a lack of literature on the effect of uranium exposure on
the development of cardiovascular diseases. One of the consequences of uranium-induced kidney
disease could be adverse effects on the cardiovascular systems by a number of mechanisms, including
changes to the renin-angiotensin pathway. Several publications have focused on the relationship
between kidney and cardiovascular systems in humans [P1, R21, T2]. Consequently, the absence of
relevant publications precludes demonstration of an obvious link between uranium exposure and
induction of cardiovascular pathologies.

5. Effects on DNA
212. Uranyl acetate staining of DNA has been used for staining fixed cells for more than forty years
[H27]. In vitro studies performed with purified DNA showed the presence of a tight binding site for the
uranyl ion (UO22+) in a four-way DNA junction [M53], demonstrating that uranium can be bound to
DNA in vitro. Exposure to uranium may lead to inhibition of DNA-binding proteins [H15, V7]. In vitro
studies demonstrated that uranium can displace transcription factor binding to DNA and that it can bind
to serum proteins accumulation in the nucleus of human cells [H15, V7]. Experimental studies have
412 UNSCEAR 2016 REPORT

shown accumulation of uranium in the nucleus of human cells (renal, hepatic, neuronal cells) treated in
vitro [G30, R26] and in kidney cells of rats treated in vivo [H23].

213. Depleted uranium can induce oxidative DNA damage [B3, K2, M40]. These studies suggested
that DU could induce DNA lesions through interaction with cellular oxygen species. These results were
corroborated by an in vivo study that showed that inhalation of DU resulted in a production of
hydroperoxides in lung tissue of rats [M56]. A recent in vitro experimental study reported the formation
of uraniumDNA adducts and mutations in mammalian cells after direct exposure to a compound of
DU [S34]. The data suggest that uranium could be chemically genotoxic and mutagenic. The unique
mutation spectrum in hypoxanthine-guanine phosphoribosyl transferase (HPRT) locus elicited by
exposure to DU suggests that uranium-generated mutations in ways that are different from spontaneous,
free radical and radiological mechanisms [C33].

214. Another study on rats demonstrated that UO2 repeated pre-exposure by inhalation increased DNA
single-strand breaks induced by a single UO4 inhalation exposure in epithelial nasal cells, broncho-
alveolar lavage cells and kidney cells, whereas a single inhalation exposure to UO4 alone had no such
effect [M55]. In vitro studies indicated that uranium-induced DNA single-strand breaks were catalysed
in the presence of ultraviolet light or ascorbate [M53, N7, Y2]. In vivo [M54, M55] and in vitro [D5,
T6] studies showed that uranium induced DNA double-strand breaks. An in vitro study demonstrated
that DU was a weak clastogen2 and induced aneugenic3 effects while enriched uranium is a more potent
clastogen [D5].

215. In summary, experimental studies performed in vivo in rodents or in vitro in cell cultures
suggested that uranium could be chemically genotoxic and mutagenic through the formation of strand
breaks and covalent uranium-DNA adducts.

6. Cytogenetic damage
216. Uranyl fluoride injected in vivo into mouse testes led to an increase in the frequency of
chromosomal aberrations in spermatogonia (breaks and polyploids) and primary spermatocytes
(fragments, chromatid exchanges, reciprocal translocations) [H27]. An in vitro study performed on
human bronchial fibroblasts indicated no significant increase in chromosomal damage following
chronic exposure to uranyl acetate and a slight increase following exposure to uranium trioxide [W22].
An increase in the frequency of sister chromatid exchange and in the frequency of chromosomal
aberrations was also observed in cell lines [L36] and in human osteosarcoma cells [M41].

217. A study by Hao et al. using rats showed that chronic oral exposure to DU in food led to an
increased frequency of micronuclei in bone marrow cells at 4 and 40 mg U/kg/day in weaning rats
[H9]. Miller et al. investigated the transmission of genetic damage to offspring of males mice
contaminated with uranium via depleted-uranium-implanted pellets. The authors demonstrated a dose-
dependent increase in mutation frequency in the offspring [M45].

218. Chromosome aberrations in Gulf War I veterans following DU exposure were evaluated in several
studies [A8, B2, M24, N6]. Four biomarkers of genotoxicity (micronuclei, chromosome aberrations,
mutant frequencies of HPRT and PigA4) were examined. There were no statistically significant
2
A clastogen is a mutagenic agent that induces disruption of chromosomes.
3
An aneugenic agent promotes aneuploidy in cells during mitosis or meiosis leading to the presence of abnormal number of
chromosomes in cells.
4
PigA gene mutation assay is used to assess genomic instability through the measurement of fluorescently labelled antibodies to
specific membrane proteins by flow-cytometric methodology.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 413

differences in any outcome measure when results were compared between the low- and the high-
uranium groups. However, modelling suggests a possible threshold effect for mutant frequencies
occurring in the highest uranium exposed cohort members [M24]. This study demonstrates a relatively
weak genotoxic effect of the DU exposure. In the second study [B2], the results indicate that ongoing
systemic exposure to DU occurring in Gulf War I veterans with DU embedded fragments does not
induce significant increases in micronuclei in peripheral blood lymphocytes compared to micronuclei
frequencies in veterans with normal uranium body burdens.

219. Some studies investigated the possible effects of uranium exposure on chromosomal damage in
workers and reported an increase in chromosomal aberrations in miners [P25, Z3, Z4] whereas others
suggested no significant effects [B49, K21, L42, M33, M70, W25]. Martin et al. performed a study for
workers from fuel production and fuel enrichment plants to analyse asymmetrical chromosome
aberrations and sister chromatid exchanges [M13]. Both worker groups had higher levels of
chromosome aberrations than the studied controls. This effect appeared not to be linked to external
radiation. Smoking increased the frequency of dicentrics but not the sister chromatid exchanges in the
workers exposed to soluble uranium, suggesting some interaction between the two clastogens.

220. In conclusion, some experimental studies performed on animals or on culture cells at high doses
of uranium demonstrate an increase in the frequency of chromosomal aberrations (fragments, chromatid
exchanges, reciprocal translocations). Human data are inconclusive but consistent with effects at higher
levels of exposure.

7. Tumorigenic potential
221. Information about the carcinogenesis processes induced by uranium in experimental models and
humans was provided by the WHO [I5]. Publications on experimental models are detailed below.

222. Chronic inhalation of natural uranium ore dust alone created a risk of primary malignant and non-
malignant lung tumour formation in rats [M49]. The risk of the induction of malignant tumours was not
directly proportional to dose but was directly proportional to dose rate. A study performed on mice
reported an increase in osteosarcoma and acute myeloid leukaemia with 239Pu and 241Am but no
significant difference between results for 233U exposed animals and controls [E4]. The authors
considered that the observed differences between the radionuclides were due to differences in
irradiation of peripheral and central regions of the bone marrow, with lowest doses from 233U. No renal
and liver carcinomas were noted [E4].

223. A study by Hahn et al. conducted in vivo showed that intramuscular DU fragments induced soft
tissue sarcomas (fibrous histiocytoma, fibrosarcoma and osteosarcoma) at the site of implantation [H3].
However, no direct dose-dependent relationship could be drawn due to the presence of varying
corrosion products. Intravenous injection of murine haematopoietic cells into depleted-uranium-
implanted mice was followed by the development of leukaemia in 76% of mice implanted with DU
pellets in contrast to 12% of control mice [M44].

224. Miller et al. reported the ability of DU-uranyl chloride (10250 M) to transform immortalized
human osteoblastic cells to the tumorigenic phenotype [M39]. The changes in phenotype are
characterized by anchorage-independent growth, tumour formation in nude mice, expression of high
levels of the k-ras oncogene, reduced production of the Rb tumour-suppressor protein, and elevated
levels of sister chromatid exchanges per cell.
414 UNSCEAR 2016 REPORT

225. Some sparse studies have been performed on DU clean-up workers and on populations from
depleted-uranium-contaminated regions [K20, M35, M36]. In clean-up workers, the total number of
DNA alterations was higher immediately after decontamination than before decontamination, but
tumours did not develop in the group of DU clean-up workers during the investigation period of four
years [M36]. This point is not surprising, since four years would be a short latency period for the
development of any tumour. In summary, some publications reported tumorigenic effects of uranium in
animals (mice or rats) or tumorigenic potential in in vitro models.

I. Relative biological effectiveness

226. Relative biological effectiveness is an empirical value that measures the capacity of a specific
type of ionizing radiation to produce a biological effect in a particular biological system (for instance
surviving fraction of irradiated cells). It is the ratio of biological effectiveness of one type of ionizing
radiation relative to a reference radiation (gamma- or X-rays), given the same amount of absorbed
energy. Thus, in the case of uranium isotopes, these values refer exclusively to radiobiological effects
of alpha particles compared to the reference radiation. The chemical toxicity of uranium may influence
observed RBE values. RBE values are used in experimental radiobiology and are also used by ICRP to
derive radiation weighting factors for the calculation of effective dose, which allows the summation of
radiation doses for all radiation types for the control of exposure. The value of the weighting factor
used by ICRP for alpha particles is 20 compared with a value of 1 for all low-LET radiation [I15].
While there are no direct determinations of alpha particle RBE for uranium isotopes, and such studies
would require the use of high specific activity isotopes such as 233U to avoid chemical effects, there is
no reason to consider that alpha particles from uranium will have different relative effectiveness from
alpha particles of similar energy emitted by other radionuclides (e.g. 222Rn and 239Pu [M12]).

J. Lethal effects
227. The lesion that can cause animal death after acute contamination is tubular nephritis, which, as
described above, results from the metallotoxic properties of uranium at high doses. Following
intravenous injection of a soluble form of uranium, the lethal dose (LD50) over a period of three weeks
after injection was measured as 0.1 mg/kg for rabbits and 20 mg/kg for mice [B20]. In rats, death
occurred from 3 to 7 days for doses 8 mg/kg [F17]. Following oral ingestion of uranium acetate, the
LD50 after two weeks was more than 100 mg/kg in rats and mice [D17].

228. One publication reported the deliberate ingestion of uranium (15 g of uranium acetate) [P9]. This
ingestion resulted in an acute renal failure, with persistent renal impairment six months after the initial
exposure. However, no death was reported despite the high uranium level ingested. This suggested that
several grams of uranium for ingestion intakes are necessary for inducing death in humans [K5].

229. As indicated in the recent review of Pernot et al. [P14], the definition and classification of the
different types of biomarkers have varied slightly, depending on the biomedical field considered. A
biomarker has been defined as any measurement reflecting an interaction between a biological system
and an environmental agent, which may be chemical, physical or biological. No real bioindicator is
specific for uranium exposure. Concerning uranium exposure, the analysis of uranium content (total
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 415

concentration or radiological activity in 238U, 234U and 235U) in urine [C18, L21, S37] may provide
information on uranium exposure. More indications are given in the uranium measurements section.

230. Since the kidneys are sensitive to damage by uranium, renal molecules were studied to describe
potential health consequences [G27, G28]. However, standard potential bioindicators have not yet
proven to be sufficiently specific and reproducible. The development and validation of bioindicators
that link uranium exposure to renal damage would be valuable [G27].

231. Currently, urinary levels of albumin, glucose, 2-microglobulin and N-acetyl--D-

glucosaminidase activity are used to evaluate possible renal effects of uranium in humans [K26, K28]
or in animals [G27]. The use of osteopontin as a potential bioindicator of uranium effects was recently
investigated in a combined experiment and human study [P30]. A decreased osteopontin level in urine
was found when the concentration of uranium in urine, after acute exposure, exceeded 30 g/L. Such a
decrease may suggest renal damage induced by uranium exposure. However, no clear relationship
between exposure level, duration of exposure and observed renal effect can be drawn from the available
studies on humans. One experimental study performed on rats at 40 mg/L of uranium in drinking water
failed to demonstrate variations in renal osteopontin mRNA level in the kidneys [R27].

232. Zamora et al. [Z6] studied the effects of chronic ingestion of uranium via drinking water on
human populations. They found statistically significant subtle changes in two of the biological
indicators measured, namely ratios of glucose and LDH to creatinine excreted (table 13) that did not
translate to any observed health effects. In addition, LDH excretion decreased with higher exposure.
The excretion of 2 microglobulin increased but the increase was not statistically significant.

233. Some biochemical parameters were measured in the urine of Gulf War veterans [M20, M22, M23,
M24, M25, S32]. Table 14 summarizes some of the results obtained. None of the levels measured were
statistically significantly different from control values. Glucose concentrations were lower for the high-
exposed veterans whereas the concentrations of -2 microglobulin and retinol binding protein were
higher. The results of the study indicated that there were a few subtle trends in changes of biochemical
parameter levels of exposure.

234. The use of omic techniques (genomics, proteomics and metabolomics) to screen unknown
biological or toxicological effects is expanding to develop new bioindicators. Omic methodology has
recently been used to screen mRNA, proteins or metabolome involved in the response to uranium
exposure in various cell lines or in animals [G21, M6, P27, P29, T4, T5]. A recent study has
demonstrated the relevance of metabolomics in cases of uranium exposure [G21]. The aim of this study
was to assess the biological changes in rats caused by ingestion of natural uranium in drinking water
over 9 months and to identify potential biomarkers related to such contamination. LC-MS
metabolomics identified urine as an appropriate biofluid for discriminating the experimental groups. Of
the 1,376 features detected in urine, the most discriminant molecules were metabolites involved in
tryptophan, nicotinate, and nicotinamide metabolic pathways, in particular N-methylnicotinamide.
These results are in accordance with a previous study that showed the role of N-methylnicotinamide in
rats with experimental renal failure induced by uranium [S12]. The study of Grison et al. thus
establishes the validity of using metabolomics to address chronic low-dose uranium contamination
[G21]. These studies show that while reliable biomarkers of tissue damage are not yet available,
modern techniques have the potential to identify such molecules.
416 UNSCEAR 2016 REPORT

Table 13. Inter-subject variability in bioindicator data (adapted from [Z6])

Low exposurea High exposurea

Urinary parameter
(<1 g U/L (n=20)) (2781 g/L (n=30))
Mass of glucose excreted (mg/d) 55.9 (20.2111) 82.4 (32.7427)
Mass of LDH per unit mass of 25 (7.666) 17 (0410)
creatinine excreted (U/g)
Mass of -2 microglobulin per unit 43 (11270) 56 (15340)
mass of creatinine excreted (g/g)
a
Median value with range provided in brackets. The group Low exposure is considered as the control group.

Table 14. Bioindicator data for Gulf War veterans (adapted from [M23])

Normal Low exposurea High exposurea

Urinary parameter range (<0.1 g uranium per gram (0.1g uranium per gram
of creatinine (n=25)) of creatinine (n=10))
Glucose g/day 0.00.5 6.34.38 0.120.01
-2 microglobulin g/g creatinine 0160 59.087.48 81.7213.28
Urine retinol binding protein g/g <610 31.004.73 48.119.73
creatinine
a
mean standard error.

VIII. EPIDEMIOLOGICAL STUDIES

235. Several population groups have been considered for epidemiological studies on the health effects of
uranium due to occupational (uranium mines, milling and processing plants, facilities involved in the
nuclear fuel cycle) or to environmental exposure (elevated uranium levels in drinking water, vicinity of
uranium processing facilities or areas affected by depleted-uranium-munition use). Clear and comparative
epidemiological information is limited. Nonetheless, in appendix A, tables A1A4 summarize the
principal studies and characterize their value in assessing uranium risks for occupational exposure, and
tables A5 to A7 for environmental exposure.

A. Studies of occupational exposure

236. The preparation of the fuel used in nuclear power plants relies on a complex cycle, including the
steps of mining, crushing of the ore and preparation of Yellowcake (uranium mills and processing
plants), conversion of uranium oxide to UF6, enrichment, and fuel manufacturing. Other activities
include research and reprocessing of the nuclear fuel. These activities involve different types of jobs
and different patterns of uranium exposure. Also, historical methods of individual exposure monitoring
vary (measurement of ambient exposure, bioassay of excreted uranium, personal film dosimeters, in
vivo measurements).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 417

237. Studies of workers involved in the nuclear fuel cycle present a good potential to investigate cancer
and other health effects of internal uranium exposure on the basis of long-term follow-ups, insofar as
they have individual estimates of exposure levels. These studies have been identified as among the most
pertinent ones to quantify a potential exposure-risk relationship [L16].

1. Uranium miners
238. Miners are exposed to internal contamination due to inhalation of long-lived radionuclides (LLR)
from uranium ore dust, but also to external gamma radiation and radon and its progeny. The main
source of radiological exposure of uranium miners is radon and its decay products. Many
epidemiological studies of uranium miners have been performed that demonstrated an association of
accumulated exposure to radon and its decay products with lung cancer risk [I16, L17, N1, U9, W17].

239. A concerted European effort (Alpha risk project) [T10] has furthered the assessment of health
risks associated with uranium exposure based on the Czech, French CEA-COGEMA and German
Wismut cohorts. Individual exposure to uranium was estimated from measurement of ambient
concentration (in Bq/m3). The assessment of uranium contamination in miners was based on the
reconstruction of individual accumulated ore dust exposure over time. The development of dedicated
dosimetric software enabled the estimation of cumulative organ doses due to radon and its progeny,
LLR arising from uranium ore dust, and external gamma ray exposure [M10, M11]. This approach
allowed the estimation of the contribution of uranium exposure to the total organ dose, and the initiative
has resulted in a number of cohort and nested case-control studies of uranium exposure and mortality
by cancers, leukaemia or cardiovascular diseases [D23, D31, K15, K17, M50, M51, M52, R4, R5, T13,
V2]. A difficulty with the uranium miner studies is that the lung doses from short-lived radon decay
products are much larger than and often correlated with the LLR exposure from uranium, which
hampers the assessment of exposure effects for lung cancer.

2. Uranium millers
240. The first steps after uranium extraction are crushing of the ore and preparation of the Yellowcake
by uranium millers or as part of uranium processing. Like the miners, millers are subjected to inhalation
of radon and radon decay products, external gamma radiation and inhalation of LLR from uranium ore
dust. One potential constraint of uranium miller studies is that, despite uranium LLR exposure being a
higher percentage of total radiation dose, uranium miller exposure is typically much lower than
uranium miner total exposure. Individual exposure to uranium among millers has been estimated solely
from measurement of ambient concentration (in Bq/m3).

241. Only a few cohort studies have specifically considered health risks among uranium millers
[B36, K17, P21, Z1]. These studies reported mortality results, but only the German miller study
estimated individual LLR exposure from uranium ore dust and potential health risk [K18]. They
studied 4,054 millers who had never worked in underground or open pit mines and assessed mortality
from various diseases in relation to their exposure to radon, external gamma ray, LLR and silica dust
exposure [K18].
418 UNSCEAR 2016 REPORT

3. Nuclear fuel cycle workers

242. A variety of epidemiological studies enabled the estimation of health risks among workers
involved in the fuel nuclear cycle: in the United States (Fernald; Rocketdyne; Oak Ridge TEC, Y-12
and K-25; Paducah; Linde; Mallinckrodt), in the United Kingdom (UKAEA and AWE composite
cohorts, Sellafield), and in France (CEA; AREVA NC; Eurodif) [C5, Z9, Z10].

243. After milling, the nuclear fuel cycle entails different successive steps, including conversion,
enrichment, fuel manufacturing, reprocessing and research. These steps involve diverse radiological
and chemical exposure to different forms of uranium compounds. However, as these steps are not
distinguished in most of the epidemiological studies, they are considered together in this section.
Individual exposure to uranium has often been estimated solely from measurement of ambient
concentrations. Uranium exposure is preferably estimated from bioassays of individuals excreted
uranium or by other personal measurement (in vivo counting). In most studies exposure estimates for
workers could be based only on ambient measurements at job locations or even extrapolations for time
and location with no available ambient measurements.

4. Results of occupational exposure studies

244. The exposure situations may influence cancer risk in occupational settings, especially for lung
cancers. For miners, the main exposures are from radon (internal exposure), external gamma radiation,
and exposure to uranium dust. Typically, workers in the nuclear fuel production are subjected to less
radon exposure but additional potentially toxic chemical exposure. In addition, for lung cancer,
confounding factors may occur due to variations in smoking habits, about which most studies contain
little or no information. Exposure levels have been assessed differently in various studies depending on
the occupational exposure setting and the historical exposure information available.

245. Characteristics and results of uranium miner, miller and processor studies are summarized in
appendix A, tables A1A4, with the studies in tables A3 and A4 having more quantitative analyses. The
health outcomes are described below by organ systems. Since well-conducted nested case-control
studies provide risk estimates that in principle are comparable to those from the full cohort, the results
of both types of studies are provided together without distinction.

246. Studies that have identified uranium-exposed cohorts have varied in the amount of data they
presented to adduce uranium health effects. Studies that contain a subset of uranium-exposed workers
who are not actually identified as exposed provide essentially no information on uranium effects; most
studies in appendix A, table A1 fall into this category. Others have identified uranium workers but have
no individual exposure levels and therefore provide only standardized mortality ratios (SMRs) or other
overall statistics which yield little valid information on uranium effects because they consider those
with appreciable exposure and little or no exposure as a composite group. Also, SMRs are subject to
healthy worker bias; such studies are shown mainly in appendix A, table A2. Analyses by duration of
employment can also be subject to healthy worker survivor bias and therefore should be viewed
circumspectly. Analyses that show SMRs or relative risks (RRs), including hazard ratios (HRs) and
odds ratios (ORs) by cumulative dose groups are somewhat informative, but difficult to interpret since
the common problem of small numbers within groups leads to inconsistencies in risk estimates across
groups; trend analysis summaries are helpful but fail to provide quantitative overall risk estimates
(appendix A, table A3). Studies that provide quantitative risk estimates per unit of exposure are
presented in appendix A, table A4.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 419

247. Studies that explicitly estimate individual uranium exposure and use those data to calculate
quantitative dose-response associations with specific health end points are valuable for assessing risk.
Typically, dosimetric estimates are made for detailed types of jobs at particular work locations for
specific periods based on ambient monitoring data, and this job exposure matrix (JEM) is applied to
individual work records. Better individual dosimetry is achieved when individual urinalyses of uranium
excretion or other personalized measurements are available, supplemented by ambient monitoring.
However, some quantitative studies had limited ability to distinguish effects from LLR exposure from
uranium from those associated with external irradiation or radon decay products because those doses
were much larger than the LLR doses to organs. That problem applies especially to lung cancer among
miners, since radon decay products usually contribute much more lung dose than LLR, so that in miner
studies the LLR lung dose was often only 12% as great as the radon decay product dose and the two
doses are typically correlated. In that situation, estimates of LLR-associated risk may be inaccurate to
an unknown degree.

248. The following overview of associations of LLR exposure from uranium with diseases thought to
be induced by uranium will concentrate on the most informative studies that have presented quantitative
dose-response data. The most recent reports of study cohorts are shown in preference to older reports to
avoid redundancy of information, since the most recent data with longer follow-up and sometimes
improved dose information provide better risk estimates.

249. The study results will be considered by organ systems. Essentially, all the informative studies of
occupational populations exposed to uranium have been conducted since 1980, some with exposure
dating back to the mid-1940s. These studies considered a large range of potential health effects. Among
the most frequently considered ones are respiratory cancers (lung, laryngeal), urological cancers
(kidney, bladder, prostate), digestive cancers (stomach, intestine, liver, pancreas), cancers of brain and
central nervous system, lympho-haematopoietic malignancies (leukaemia, lymphoma, multiple
myeloma). Non-cancer diseases have also been considered, especially respiratory, circulatory and renal
system diseases. Further details about these studies are presented in appendix A, table A4.

(a) Lung cancer and other respiratory diseases

250. Lung cancer. Studies with quantitative risk estimates for respiratory diseases from LLR exposure
from uranium are presented in table 15. Ten reports were found that had dose-response estimates for
lung cancer risk based on individualized estimates of uranium exposure. The study of CEA-COGEMA
miners in France reported a positive risk coefficient for LLR exposure from uranium, though the radon
and external-radiation co-exposure make the result difficult to interpret. Other miner studies have not
characterized and analysed LLR exposure from uranium and lung cancer. The German study of
uranium millers [K18] reported a dose-response analysis of LRR exposure from uranium and lung
cancer risk. The association was non-significant and in the negative direction. Two other studies of
uranium millers reported no significant elevation in lung cancer mortality [B36, P21].

251. The remaining eight reports of LLR dose-related lung cancer mortality were based on workers in
nuclear processing industries. The nature of the workplaces and the types of measurements and
procedures that generated the individual exposure estimates are given in appendix A, table A4, for the
various studies. The large study of Fernald workers, with 269 lung cancer deaths, showed a positive,
but non-significant, association and most other studies had negative or non-significant coefficients of
risk. Additional uranium worker cohorts showed no significant elevation in overall lung cancer SMRs
[C36, D33, D34, M26, M27]. However, the relatively small French study of the AREVA NC Pierrelatte
plant showed significant positive dose responses [G31]. This is the only study that conducted separate
analyses for exposure to uranium compounds that differ in solubility (Types: F, M and S). They found
420 UNSCEAR 2016 REPORT

the Type M and Type S exposure conferred lung cancer risk (table 15). In a further analysis [G32] that
distinguished exposure to natural uranium from the reprocessed uranium which has a different isotope
composition, they found that risks were especially dose related for reprocessed uranium of the Type M
and Type S. This suggests that insoluble uranium particles with a longer residence time in the lung
confer more risk than the more soluble particles with a shorter residence time. However, these
interesting results were based less than 55 lung cancer deaths and require confirmation in larger studies.

Table 15. Dose-response studies of uranium exposure and risk of respiratory system diseases

No. of Nature of Unit of uranium (LLR) Risk estimate per unit

Study references
deaths uranium work exposure LLR exposurea
LUNG CANCER
France, CEA- 94 Mining kBqh/m3 ERR=0.32 (95% CI: 0.09, 0.73)
COGEMA [R5]
(Also [R4, V2])

Germany, milling 159 Milling 100 kBqh/m3 ERR=0.61 (95% CI: 1.42, 1.9)
[K18]

France, AREVA NC 48 Processing Years of Types F, M, HR=1.01 (95% CI: 0.96, 1.01) [F]
[C6] and S uranium 1.07 (1.01, 1.13) [M]
exposure 1.07 (1.01, 1.14) [S]

France, AREVA NC 53 Processing Years of exposure to HR=1.07 (95% CI: 0.96, 1.19) [F]
[G32] Types F, M and S 1.13 (1.03, 1.25) [M]
reprocessed uranium 1.13 (1.01, 1.25) [S]

USA, TEC/Y-12/ 787 Processing 0.5, 2.5, 5, 25, 50, OR=1.03, 0.57, 0.85, 0.82, 0.64, 2.05
Mallinckrodt/ 250 mGy LLR (95% CI: 0.20, 21), respectivelyc
Fernald [D35]b

USA, Rocketdyne 94 Processing 100 mSv LLR RR=1.01 (95% CI: 0.89, 1.16)
[B38]

USA, Fernald [S19] 269 Processing mGy LLR ERR=0.022 (95% CI: 0.009, 0.06)

USA, Paducah 129 Processing g.y RR=0.91 (95% CI: 0.5, 1.6) (2150 g.y)
gaseous 0.95 (0.6, 1.6) (51125 g.y)
diffusion [C17] 0.51 (0.3, 0.9) (>125 g.y)

USA, Y-12 [R12] 111 Processing 10 mSv LLR ERR=0.77 (95% CI: 2.5, 1.0)

France, Gaseous 100 Processing Low, medium, high Exposure to natural soluble uranium
diffusion [Z10] compounds (with 90% CI):
Medium: RR=0.92 (CI: 0.54, 1.6)
High: RR=0.74 (CI: 0.42, 1.3)
Enriched uranium (n=23):
Medium RR=1.8 (CI: 0.64, 4.6)
High RR=0.69 (CI: 0.21, 1.9)
Depleted uranium (n=23):
Medium RR=1.2 (CI: 0.33, 3.7)
High RR=1.5 (CI: 0.61, 3.9)

EXTRATHORACIC RESPIRATORY CANCERS

Germany, Wismut 554 Miners 10 kBqh/m3 ERR=0.098 (95% CI: 0.11, 0.31)
[M51] (Laryngeal cancer)

Germany, Wismut 234 Miners 100 kBqh/m3 ERR=0.17 (95% CI: 2.50, 2.16)
[K17] (All extra-thoracic airway cancers)
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 421

No. of Nature of Unit of uranium (LLR) Risk estimate per unit

Study references
deaths uranium work exposure LLR exposurea
NON-MALIGNANT RESPIRATORY DISEASE
France, CEA- 37 Miners kBqh/m3 ERR=0.086 (95% CI: n.e.)d
COGEMA [R5]

USA, Y-12 [R12] 50 Processing 100 mSv ERR=0.085 (90% CI: 0.32, 0.15)

USA, Fernald [S19]e 102 Processing 100 Gy LLR ERR=0.0062 (95% CI: 0.007, 0.0006)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio) and RR
(relative risk or rate ratio) are expressed as a multiple of the rate in the baseline (lowest/no exposure) group. (ERR = RR 1).
b
This study [D35] was partially repeated later on [R12, S19] but included two cohorts that were not reported elsewhere.
c
The risk estimate of 2.05 for the 250 mGy group became 0.36 in an analysis with adjustment for smoking.
d
Not estimable from the likelihood profile.
e
Analysis of chronic obstructive pulmonary disease.

252. In summary, there is mixed evidence for the lung carcinogenicity of uranium. The inconsistency
may relate to variations in the amount and types of exposure data available (e.g. extrapolated dose
reconstruction, ambient monitoring, uranium urinalyses or other personalized exposure measures) or to
variations in methods for calculating LLR doses from the raw data. The inconsistency can also relate to
other radiation or chemical exposure that may not have been adequately accounted for in the LLR
analyses, or to dose-dependent variations in smoking behaviours. On the other hand, since the studies
have low statistical power to detect risks, given the relatively small numbers of lung cancers and fairly
low levels of LLR exposure from uranium for most workers, it is notable that two cohorts with dose-
response data showed a statistically significant association [C6, R5].

253. Other respiratory cancers. A study of all extrathoracic airway cancers, based on a well-defined
cohort of the German uranium miners, showed a negative dose-response coefficient for LLR exposure
from uranium [K17]. Another study of German uranium miners had LLR dose-response data for
laryngeal cancer, which yielded a non-significantly positive risk coefficient [M51]. There was no
association of LLR exposure from uranium and laryngeal cancer among French uranium miners [R5].
Other studies have reported overall excess laryngeal or upper airway cancers in worker populations
potentially exposed to uranium [B18, B29, B33, D35, G31], but those are based on small numbers of
cancers, without individual uranium exposure estimates, and with limited information on the risk
factors of smoking and alcohol intake.

254. Non-malignant respiratory disease (NMRD). Four studies with dose-response analyses all
reported negative dose-response coefficients for NMRD [B38, R5, R12, S19]. Pinkerton et al. [P21]
reported an inverse relationship of NMRD with length of time working in uranium milling. SMRs in a
number of other cohorts with uranium exposure were not significantly elevated [C20, C36, D33, D34,
M26, M27, R14, Z10] so there is no evidence that NMRD is associated with uranium exposure.

(b) Lymphatic and haematopoietic cancer

255. Leukaemia. Since leukaemia shows a large excess relative risk from radiation exposure, it is a
strong a priori candidate to investigate regarding uranium exposure effects. Table 16 shows uranium
dose-response related results for leukaemia. A nested case-control study of leukaemia mortality within
the large collection of German Wismut miners [M50] found a positive but non-significant association
of estimated LLR exposure with non-chronic lymphocytic leukaemia (non-CLL) risk. The positive risk
coefficient was driven entirely by the highest dose group (20 kBqh/m3, ERR=1.26, 90% CI: 0.7, 2.2,
422 UNSCEAR 2016 REPORT

n=14). A smaller miner study in Czechia found a statistically significant dose-response association of
leukaemia mortality with total red bone marrow dose, of which the majority was from LLR exposure
[T13]. Three smaller dose-response studies of leukaemia risk in the uranium processing industries did
not support excess risk (table 16). In summary, there is limited dose-response evidence for an
association of uranium exposure with subsequent leukaemia. Atkinson et al. [A29] reported a non-
significant SMR (SMR=1.10; 95% CI: 0.89, 1.33; n=103) in nuclear workers, particularly among
workers monitored for internal radiation exposure. Others likewise have reported non-significant
overall SMRs in uranium worker cohorts [C36, D30, D33, D34, L44, M26, M27, P24, S5, Z2, Z10], but
did not estimate worker uranium doses or conduct LLR dose-related analyses.

256. Other lympho-haematopoietic malignancies. Lymphoma is a biologically plausible outcome of

inhalation exposure to uranium, since uranium deposited in the lung tends to migrate to the thoracic
lymph nodes. However, there are few studies providing dose-dependent analyses of uranium exposure
and lympho-haematopoietic malignancies other than leukaemia (table 16). The cohort of German
uranium millers [K18] showed a non-significant negative risk coefficient for all lympho-haematopoietic
malignancies. In the Rocketdyne cohort there was a non-significant positive dose-response [B38],
though the specifics were not reported. A French study of gaseous diffusion plant workers reported a
non-significant excess risk for all lympho-haematopoietic malignancies other than leukaemia among
those with medium or high exposure to soluble uranium [Z10]. A study in the United States showed for
uranium miners an elevated risk of lympho-haematopoietic malignancies (SMR=1.38, 95% CI: 1.03,
1.82) [S5]. Pinkerton et al [P21] showed a suggestive excess of lymphatic and haematopoietic
malignancies, excluding leukaemia, (SMR=1.44, 95% CI: 0.83, 2.35, n=16) with no dose response.

257. A study of non-Hodgkins lymphoma (NHL) at the Paducah gaseous diffusion plant [C17]
produced high relative risks, but the fact that there was no dose response suggests that the high values
are likely attributable to a deficit of NHL in the baseline comparison group rather than large excesses in
the other dose groups. NHL mortality was non-significantly elevated for exposure to uranium in the
Fernald cohort [S19]. The Czech uranium miner study [T13] indicated a non-significantly elevated
SMR of 1.4 (95% CI: 0.9, 5.1, n=16) for NHL but did not provide a dose-response analysis. Other
uranium worker cohorts reported null overall SMRs for lymphomas or all lympho-haematopoietic
cancers [B38, C9, C36, D30, D33, D34, G31, L44, M27, P21, P24, S5, Z2].

258. In the study at the Oak Ridge K-25 gaseous diffusion plant by Yiin et al. [Y3] with a dose-
response analysis of multiple myeloma and uranium exposure, a significant association was not found.
A significant overall excess of multiple myeloma was found among United States miners (SMR=1.97,
95% CI: 1.05, 3.37, n=13), and an excess based on two cases (SMR=8.38; 90% CI: 1.44, 26.20) was
reported among French CEA-COGEMA uranium processing workers [B9]. A non-significant excess of
multiple myeloma was observed among Mallinckrodt uranium processing workers [D33] (SMR=1.30,
95% CI: 0.42, 3.0, n=5). The SMRs for multiple myeloma were not significantly elevated in other
uranium worker groups [C17, Z2].
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 423

Table 16. Dose-response studies of uranium exposure and risk of lympho-haematopoietic

malignancy mortality

Nature of Unit of uranium Risk estimate per unit LLR

Study references No. of deaths
uranium work (LLR) exposure exposure a
LEUKAEMIA
b
Germany, Wismut 218 (non-CLL) Mining 100 kBqh/m3 ERR=0.76 (90% CI: 1.26, 2.78)
[M50]

Czechia [T13] 30 Mining Svc ERR=2.5 (90% CI: 0.3, 9.3)

USA, Paducah 21 (all types) Processing #1 (020 g.y) Baseline

[C17] #2 (2150 g.y) #2: RR=0.73 (95% CI: 0.2, 3.0)
#3 (51125 g.y) #3: RR=0.49 (CI: 0.2, 2.3)
#4 (>125 g.y) #4: RR=0.77 (CI: 0.2, 2.5)

USA, Rocketdyne 10 (non-CLL) Processing 100 mSv LLR RR=1.06 (95% CI: 0.50, 2.23)
[B38]

USA, Fernald [S19] 28 (non-CLL) Processing 100 Gy LLR HR=0.18 (95% CI: 0.012, 0.80)

OTHER LYMPHO-HAEMATOPOIETIC MALIGNANCIES

Germany [K18] 23 (all lympho- Millers 100 kBqh/m3 ERR= 0.65 (95% CI: 2.78, 1.47)
haematopoietic)

Zhivin, Gaseous 28 (all lympho- Processing Medium and high Natural soluble uranium
diffusion [Z10] haematopoietic) exposure compounds (with 90% CI):
Medium RR=1.4 (CI: 0.52, 3.9)
High RR=1.08 (CI: 0.37, 3.3)

USA, Paducah 26 (non-Hodgkins Processing #1 (020 g.y) Baseline

[C17] lymphoma) #2 (2150 g.y) #2: RR=9.95 (95% CI: 1.2, 81)
#3 (51125 g.y) #3: RR=8.85 (CI: 1.1, 71)
#4 (>125 g.y) #4: RR=5.74 (CI: 0.7, 45)

USA, Fernald [S19] 12 (non-Hodgkins Processing 100 Gy HR=1.2 (95% CI: 0.88, 1.5)
lymphoma)

USA, K-25 [Y3] 98 (multiple Processing 10 Sv OR, 1.04 (95% CI: 1.00, 1.09)
myeloma)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio) and RR
(relative risk or rate ratio) are expressed as a multiple of the rate in the baseline (lowest/no exposure) group. (ERR = RR 1).
b
Non-CLL, all leukaemias except chronic lymphocytic leukaemia.
c
Analysis was for total red bone marrow dose, of which 5264% was estimated to be due to LLR from uranium dust inhalation.

(c) Digestive system cancer

259. Stomach cancer. Only a few studies have conducted dose-dependent analyses of uranium
exposure and digestive system cancers (table 17). Studies of German uranium millers [K18], the
Fernald uranium processors [S19], and Rocketdyne workers with internal exposure monitoring [B38]
all showed weakly positive but non-significant dose-response risk estimates for stomach cancer. The
risk of mortality from stomach cancer was increased with the alpha absorbed stomach dose among
German uranium miners (ERR per Gy=37.3; 95% CI: 3.4, 71.1, n=592), but the contribution of LLR to
424 UNSCEAR 2016 REPORT

the absorbed stomach dose was less than 1% [K15]. Other uranium worker cohorts reported null overall
SMRs for stomach cancer [B38, C9, C17, D34, L44, M26, R5, S5, Z1, Z10].

260. Intestinal cancer. The study of German uranium millers showed non-significant ERR coefficients
for both colon and rectal cancers [K18], and both the French uranium miners study [R5] and the
Rocketdyne study [B38] reported non-significant LRR dose coefficients for colorectal cancers. On the
other hand, a dose-response analysis of combined small intestine and colon (but not rectum) cancer at
the United States Fernald uranium processing plant yielded a statistically significant excess risk [S19].
Since this result was based on relatively small numbers, it is in need of confirmation by larger studies.
Other uranium worker cohorts reported null overall SMRs for colon or colorectal cancers [C9, D33,
D34, G31, L44, M26, P21, Z1, Z10] but did not conduct LLR dose-related analyses.

261. Pancreatic cancer. The United States Paducah [C17] and Fernald [S19] studies provided dose-
response analyses of pancreatic cancer after LLR exposure (table 17). Neither study provided an
indication of an association with uranium exposure, nor did the studies of French uranium miners [R5]
or Rocketdyne workers [B38]. However, the studies had relatively small numbers of pancreatic cancers
and thus limited statistical power. Other reports have indicated non-significant SMRs for pancreatic
cancer in uranium processing workers [C36, D33, D35, L44, M26, Z1, Z10] but did not estimate
worker uranium doses or conduct LLR dose-related analyses.

Table 17. Dose-response studies of uranium exposure and risk of digestive system disease mortality

Nature of Unit of uranium Risk estimate per unit LLR

Study references No. of deaths
uranium work (LLR) exposure exposure a
STOMACH CANCER
Germany, [K18] 49 Milling 100 kBqh/m3 ERR=1.5 (95% CI: 2.9, 5.9)b

USA, Fernald [S19] 29 Processing 100 Gy ERR=0.041 (95% CI: 0.20, 5.6)

INTESTINAL CANCER
Germany, [K18] 22 (colon) Milling 100 kBqh/m3 ERR=0.07 (95% CI: 3.3, 3.2)

Germany, [K18] 26 (rectum) Milling 100 kBqh/m3 ERR=0.56 (95% CI: 3.1, 4.2)

USA, Fernald [S19] 48 (colon & Processing 100 Gy ERR=1.5 (95% CI: 0.12, 4.1)
small intestine)

PANCREATIC CANCER
USA, Paducah [C17] 30 Processing #1 (020 g.y) Baseline
#2 (2150 g.y) #2: RR=1.42 (95% CI: 0.4, 4.7)
#3 (51125 g.y) #3: RR=0.49 (95% CI: 0.1, 1.9)
#4 (>125 g.y) #4: RR=0.97 (95% CI: 0.3, 3.0)

USA, Fernald [S19] 41 Processing 100 Gy HR= 0.61 (95% CI: 0.015, 3.5)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio), RR
(relative risk or rate ratio) and OR (odds ratio for case-control studies) are expressed as a multiple of the rate in the baseline
(lowest/no exposure) group. (ERR = RR 1).
b
Estimate is adjusted for radon exposure levels.

262. Liver cancer. Dufey et al. [D31] observed a non-significant increase in liver cancer mortality risk
associated with high-LET absorbed liver dose among German uranium miners (ERR per Gy=48.3; 95%
CI: 32.0, 128.6 (n=159) after adjustment for low-LET dose). However, the contribution of LLR to the
absorbed liver dose was less than 2%, so the association is not informative regarding uranium risk.
Among French uranium miners there was also a non-significant association of LLR exposure and liver
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 425

cancer risk [R5]. Other uranium worker cohorts reported non-significant SMRs for liver cancer [B38,
C9, M26, M27, P24, R5, Z10] but had no uranium dose-response analyses.

263. In summary, there is no persuasive evidence of an association between uranium exposure and
digestive cancers. The effect, if any, is likely small. However, the relatively small numbers of digestive
cancer cases and consequent limited statistical power to detect effects make any conclusion uncertain.

(d) Kidney and other urological cancers

264. Kidney cancer. The toxicological data suggest that uranium exposure may be related to urological
cancers, especially to the kidney because of the potential for both adverse radiological and metal effects
upon that organ. The available epidemiological studies with dose-response results are shown in
table 18. Dose-response analyses of kidney cancer in relation to LLR exposure have been conducted for
the cohorts of miners in France and Germany [D23]. Neither cohort showed a significant association
with kidney cancer even though the large German miner cohort had a substantial number of renal
cancers. In the French CEA-COGEMA cohort, the overall SMR was elevated for kidney cancer
(SMR=2.0, 95% CI: 1.2, 3.1, n=20), but LLR analyses were not reported [V1].

Table 18. Dose-response studies of uranium exposure and risk of urological system disease mortality

Nature of Unit of uranium Risk estimate per unit LLR

Study references No. of deaths
uranium work (LLR) exposure exposure a
KIDNEY CANCER
France, CEA- 11 Mining kBqh/m3 HR=0.89 (95% CI: 0.55, 1.42)
COGEMA [D23]

Germany, Wismut 174 Mining kBqh/m3 HR=1.009 (95% CI: 0.991, 1.027)
[D23]

Germany [K18] 11 Milling 100 kBqh/m3 ERR=7.36 (95% CI: 11, 26)

USA, Fernald [S19] 15 Processing 100 Gy ERR=0.039 (95% CI: 0.021, 0.55)

OTHER UROLOGICAL DISEASES

Germany [K18] 30 (Prostate Milling 100 kBqh/m3 ERR=0.21 (95% CI: 2.8, 2.4)
cancer)

USA, Fernald [S19] 19 (chronic Processing 100 Gy HR=0.98 (95% CI: 0.78, 1.1)
kidney disease)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio) and RR
(relative risk or rate ratio) are expressed as a multiple of the rate in the baseline (lowest/no exposure) group. (ERR = RR 1).

265. The study of German uranium millers [K18] had a large risk coefficient for kidney cancer but the
association was non-significant because the number of cases was small (n=11) and consequently the
confidence interval of the estimate was very wide. The dose-response coefficients for the Fernald
uranium processing workers [S19], the Rocketdyne workers [B38] and the French uranium miners [R5]
were all non-significant. Overall SMRs for kidney cancer were found to be increased in two uranium-
processing facilities: Y-12 [C20, L44] and Capenhurst [M26]. But at Capenhurst, the increase was
limited to unexposed workers. The UK Nuclear research workers also experienced elevated mortality
from kidney cancer [B18, C8]. However, none of these studies included an investigation of the relation
with internal exposure to uranium. Other uranium worker studies did not report significantly elevated
overall SMRs for kidney cancer [C9, C17, L44, S5, Z10] but had no uranium dose-response analyses.
426 UNSCEAR 2016 REPORT

266. Bladder cancer. French uranium miner studies reported a non-significant LLR dose-response
association with bladder cancer [R5]. An increase in mortality (an elevated overall SMR) from bladder
cancer was reported among Fernald workers [D35]. It was potentially associated with a high exposure
to the cutting fluids used during uranium metal production but not with internal exposure to uranium
compounds [R14]. In a more recent study of the Fernald workers [S19], elevated mortality was
observed only among salaried female workers; for cancer of the bladder and other urinary organs, the
SMR was 5.13 (95% CI: 1.06, 15.0; n=3). However, the small number of cases, no increase in bladder
cancer mortality among males (n=21) or hourly paid female workers (n=0), and no dose-response
information argue against a meaningful excess [S19]. Other reports have indicated non-significant risks
(usually SMRs) for bladder cancer in uranium processing workers [B38, Z1, Z10], but did not estimate
worker uranium doses or conduct LLR dose-related analyses.

267. Prostate cancer. The recent study of German uranium millers analysed uranium dose and prostate
cancer risk and found no evidence of an association [K18], albeit the number of prostate cancer deaths
was small (table 18). The study of United States Rocketdyne workers reported a non-significant dose-
response trend in the negative direction for prostate cancer (n=63), but did not provide a risk coefficient
[B38]. The LLR dose-response trend among French uranium miners was likewise non-significant [R5].

268. Radiation-exposed worker cohorts with excess prostate cancer have been investigated for internal
radionuclide exposure in the UK Atomic Energy Authority cohorts [A28, B18, R22] but with only a
limited examination of uranium exposure. Rooney et al. [R22] conducted a nested case-control
examination of the relation between prostate cancer and occupational exposure in those cohorts,
including 136 prostate cancer cases and 404 controls and examination of 29 radionuclides. Three cases
and twelve controls had an indication of potential uranium exposure, and zero cases and six controls
had documented uranium exposure (RR=0, 95% CI: 0, 2.55), so no effect on prostate cancer mortality
was observed. Other studies have reported non-significant overall SMRs for prostate cancer among
uranium workers [D33, L44, M26, Z1, Z10] but had no uranium dose-response analyses.

269. In summary, except for the German Wismut cohort, the number of kidney cancers was very small,
thus constraining the ability to detect small-to-moderate effects. Since the studies of kidney cancer and
chronic kidney disease are uniformly negative, the risk of uranium exposure for kidney cancer is weak
or absent. The data do not generally support an increase in bladder cancer associated with uranium
exposure; in fact, the only dose-response report regarding bladder cancer risk had a negative coefficient
[B38]. The two studies that have evaluated dose-response data for prostate cancer did not show a
significant association.

(e) Brain and central nervous system cancers

270. Experimental evidence indicates that uranium compounds, particularly more soluble ones, cross
the blood-brain barrier, thereby potentially putting the brain at risk. There have been only a small
number of studies with dose-response analyses of brain and central nervous system (brain/CNS)
tumours (table 19). An analysis of brain/CNS tumours by LLR dose groups was conducted among
Paducah gaseous diffusion uranium enrichment workers [C17], but no trend by dose was seen. A study
of French uranium miners reported a non-significantly positive brain/CNS tumour dose-response risk
for LLR exposure, accompanied by an overall excess (SMR=1.71, 95% CI: 1.00, 2.74) [R5]. Similarly,
three United States studies of uranium processing workers did not find a dose-related excess of
brain/CNS tumours [B38, C17, C20].

271. Carpenter et al. [C7] investigated the possible association between brain/CNS cancers and
exposure to external and internal radiation among Oak Ridge Y-12 workers. The internal dose to the
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 427

lung, calculated as in the study by Checkoway et al. [C20], was used as a surrogate for the internal dose
to the brain for the 47 cases and 120 matched controls. Odds ratios (ORs) were non-significantly
elevated for categories of cumulative lung dose: 150 to 290 mSv (OR=2.8; 95% CI: 0.7, 11.9; n=5);
300 to 450 mSv (OR=2.7; 95% CI: 0.8, 9.3; n=5); and among workers with mean annual lung dose
>150 mSv (OR=1.7; 95% CI: 0.7, 4.2; n=16). No doseresponse trend was observed after adjustment
for possible confounding factors (26 different chemicals, socio-economic status, duration of
employment), using either a 5- or 10-year dose lag. Other studies of uranium worker cohorts indicated
no significant overall excess of brain/CNS tumour deaths [C9, D33, L44, M26, M27, P24, R14, Z10]
but did not estimate worker uranium doses or conduct LLR dose-related analyses.

272. A limiting factor in all of these studies was the very small numbers of tumours, so only quite a
large excess risk would be detectable. Clearly more data are needed to make a better judgment about
brain/CNS tumour risk from uranium exposure.

Table 19. Dose-response studies of uranium exposure and risk of brain and central nervous system
(CNS) tumour mortality

No. of Nature of Unit of uranium Risk estimate per unit LLR

Study references
deaths uranium work (LLR) exposure exposure a
BRAIN/CNS TUMOURS
France, CEA- 17 Mining kBqh/m3 ERR=0.28 (95% CI: n.e. <0, 1.87)b
COGEMA [R5]
[Also [V2]]

USA, Paducah [C17] 14 Processing #1 (020 g.y) Baseline

#2 (2150 g.y) #2: RR=0.66 (95% CI: 0.1, 4.2)
#3 (51125 g.y) #3: RR=1.07 (95% CI: 0.2, 4.8)
#4 (>125 g.y) #4: RR=0.45 (95% CI: 0.1, 2.2)

USA, Y-12 [C20] 14 Processing #1 (09 mSv) Baseline

#2 (1049 mSv) #2: RR=1.10 (95% CI: 0.2, 6.5)
#3 (50 mSv LLR) #3: RR=0.45 (95% CI: 0.1, 3.2)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio) and RR
(relative risk or rate ratio) are expressed as a multiple of the rate in the baseline (lowest/no exposure) group. (ERR = RR 1).
b
n.e. = not estimable from the likelihood profile.

(f) Circulatory diseases

273. Suggestive findings from external radiation studies over the past 1520 years have prompted
examinations of circulatory disease risks as a potential radiation effect of internal radionuclides. The
available studies of uranium exposure have evaluated the risk for all circulatory system diseases (CSD)
and major subcategories of CSD, namely ischaemic heart diseases (IHD) and cerebrovascular diseases
(CeVD) (table 20).

274. Circulatory system diseases. The German Wismut miner cohort is the largest to study uranium
associated CSD risk. They reported a negative CSD risk coefficient for internal LLR exposure on the
basis of 5,417 CSD deaths [K14]. Two reports of CSD in the French CEA-COGEMA miner cohort are
of interest (table 20). A report of the entire cohort indicated a non-significantly positive risk coefficient
[R5], while another report of the subset of the cohort for whom information on radon, external radiation
exposure, and medical risk factors for CSD also was available showed a nearly significant risk of CSD
from LLR exposure after accounting for the other factors [D24].
428 UNSCEAR 2016 REPORT

275. A report of the French AREVA NC uranium processing cohort found that exposure to Type M
and Type S (less soluble) reprocessed uranium conferred statistically significant risk of CSD, as did
Type S natural uranium exposure [G33], but Type F exposure did not. The study found statistically
significant associations of CSD with the number of years exposed to Type S exposure to both
reprocessed and natural uranium, and a near-significant association for Type M reprocessed uranium
(table 20) [G33]. Less quantitative but supportive findings were that CSD mortality was increased
overall among workers exposed to slowly soluble reprocessed (HR=2.13; 95% CI: 0.96, 4.70) and
natural uranium (HR=1.73; 95% CI: 1.11, 2.69). In the subgroup of smokers, the risk estimates were
higher but with larger CIs (HR=1.91; 95% CI: 0.92, 3.98 for natural uranium and HR=4.78; 95% CI:
1.38, 16.50 for reprocessed uranium). The AREVA NC findings suggested that types of uranium with a
long residence time in tissues may confer risk of CSD.

276. McGeoghegan et al. [M29] reported an association between mortality from CSD and radiation
exposure in males performing industrial work in the British Nuclear Fuels Limited cohort (BNFL). This
cohort consisted partially of uranium workers (37% of the cohort were workers employed at the
Springfields uranium processing plant and 6.7% at the Capenhurst uranium enrichment plant). Their
analysis of internal exposure to any radionuclide gave a dose-response ERR per Gy of 0.76 (90% CI:
0.37, 1.23; n=2,275), but a dose-response analysis was not available specifically for uranium exposure.

277. Ischaemic heart diseases. The large German Wismut study of uranium miners found a negative
risk coefficient for LLR exposure and the end point of all heart disease (table 20) [K14]. The other
studies evaluated IHD as a cardiovascular end point. Studies of French uranium miners (CEA-
COGEMA) reported negative risk coefficients for IHD for the entire cohort [R5] and for the subset
where they could account for radon, external radiation, and medical heart disease risk factors [D24].
The German uranium miller study also reported a non-significantly negative risk coefficient for LLR
dose [K18]. McGeoghegan et al. [M29] reported an association between mortality from IHD and
radiation exposure in males performing industrial work at BNFL. Their analysis of internal exposure to
any radionuclide gave a dose-response ERR per Gy of 0.52 (90% CI: 0.09, 1.06; n=1,494), but uranium
exposure was not analysed.

278. Cerebrovascular diseases. The same set of studies also provided risk estimates for CeVD
(table 20). The large German Wismut study of uranium miners found a negative risk coefficient for
LLR exposure and CeVD [K14]. Studies of French uranium miners (CEA-COGEMA) reported non-
significant positive risk coefficients for CeVD for the entire cohort [R5] and for the subset with
adjustment for radon, external radiation, and medical heart disease risk factors [D24]. The German
uranium miller study reported a non-significantly negative risk coefficient for LLR dose [K18]. In the
study of French AREVA NC uranium processing workers, there was a statistically significant risk for
Type S exposure to reprocessed uranium and a near-significant risk for Type M reprocessed uranium
[G33], but little indication of risk from natural uranium exposure. Further, McGeoghegan et al. [M29]
also reported an association between mortality from CeVD and radiation exposure in males performing
industrial work in the BNFL cohort. Their analysis of internal exposure to any radionuclide gave a
dose-response ERR per Gy of 1.47 (90% CI: 0.49, 3.00; n=456).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 429

Table 20. Dose-response studies of uranium exposure and risk of circulatory system disease mortality

Nature of Unit of uranium Risk estimate

Study references No. of deaths
uranium work (LLR) exposure per unit LLR exposure a
ALL CIRCULATORY SYSTEM DISEASES
Germany, Wismut 5 417 Mining 100 kBqh/m3 ERR=0.2 (95% CI: 0.5, 0.06)
[K14]

France, CEA- 185 Mining kBqh/m3 ERR=0.016 (95% CI: 0.06, 0.13)
COGEMA [R5]

France, CEA- 76 Mining kBqh/m3 HR=1.13 (95% CI: 0.97, 1.31)

COGEMA [D24]

France, AREVA NC 111 Processing Reprocessed HR (95% CI)

[G33] uranium, Cumulative exposure duration
absorption Types (per year):
M, S 1.09 (1.02, 1.18) [M]
1.11 (1.03, 1.20) [S]
1.04 (1.00, 1.07) for natural
uranium [S]
High cumulative exposure:
3.40 (1.47, 7.85) [M]
8.79 (1.21, 28) [S]
2.84 (1.38, 5.85) for natural
uranium [S]
Cumulative exposure score:
1.14 (1.05, 1.24) [M]
1.17 (1.07, 1.27) [S]
1.07 (1.02, 1.13) for natural
uranium [S]

France, Gaseous 281 Processing Medium and high Natural soluble uranium
diffusion [Z10] vs. no exposure compounds RR (95% CI):
Medium RR=0.98 (0.71, 1.3)
High RR=1.2 (0.85, 1.6)
Enriched uranium (n=45):
Medium RR=0.96 (0.32, 2.9)
High RR=0.84 (0.28, 2.8)
Depleted uranium (n=45):
Medium RR=0.64 (0.23, 1.7)
High RR=0.84 (0.32, 2.3)

HEART DISEASE
Germany, Wismut 3 719 (all heart Mining 100 kBqh/m3 ERR=0.3 (95% CI: 0.6, 0.02)
[K14] disease)

France, CEA- 72 (Ischaemic Mining kBqh/m3 ERR=0.029 (95% n.e. (<0, 0.14)b
COGEMA [R5] heart disease)

France, CEA- 26 (Ischaemic Mining kBqh/m3 HR=0.94 (95% CI: 0.73, 1.20)
COGEMA [D24] heart disease)

Germany [K18] 341 (Ischaemic Milling 100 kBqh/m3 ERR=0.09 (95% CI: 0.84, 0.65)
heart disease)
430 UNSCEAR 2016 REPORT

Nature of Unit of uranium Risk estimate

Study references No. of deaths
uranium work (LLR) exposure per unit LLR exposure a
France, AREVA NC 48 (Ischaemic Processing Reprocessed HR (95% CI)
[G33] heart disease) uranium, Cumulative exposure duration
absorption Types (per year)
M, S 1.08 (0.97, 1.21) [M]
1.14 (1.03, 1.26) [S]
1.04 (0.99, 1.10) for natural
uranium [S]
High cumulative exposure:
2.05 (0.53, 7.85) [M]
4.38 (0.47, 41) [S]
2.57 (0.82, 8.07) for natural
uranium [S]
Cumulative exposure score:
1.12 (0.99, 1.27) [M]
1.17 (1.03, 1.33) [S]
1.13 (1.05, 1.22) for natural
uranium [S]

France, Gaseous 95 (Ischaemic Processing Medium and high Natural soluble uranium
diffusion [Z10] heart disease) vs. no exposure compounds (with 95% CI):
Medium RR=0.71 (0.39, 1.3)
High RR=0.91 (0.53, 1.5)

CEREBROVASCULAR DISEASE
Germany, Wismut 1 297 Mining 100 kBqh/m3 ERR=0.05 (95% CI: 0.5, 0.6)
[K14]

France, CEA- 41 Mining kBqh/m3 ERR=0.125 (95% CI: 0.06, 0.50)

COGEMA [R5]

France, CEA- 16c Mining kBqh/m3 HR=1.17 (95% CI: 0.90, 1.53)
COGEMA [D24]

Germany [K18] 171 Milling 100 kBqh/m3 ERR=0.17 (95% CI: 1.14, 0.80)

France, AREVA NC 31 Processing Reprocessed HR (95% CI)

[G33] uranium, Cumulative exposure duration
absorption Types (per year)
M, S 1.09 (0.93, 1.27) [M]
1.11 (0.95, 1.29) [S]
1.04 (0.97, 1.11) for natural
uranium [S]
High cumulative exposure:
5.71 (1.48, 22) [M]
3.26 (0.97, 11.0) for natural
uranium [S]
Cumulative exposure score:
1.13 (0.97, 1.31) [M]
1.16 (1.00, 1.35) [S]
1.01 (0.92, 1.12) for natural
uranium [S]
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 431

Nature of Unit of uranium Risk estimate

Study references No. of deaths
uranium work (LLR) exposure per unit LLR exposure a
France, Gaseous 77 Processing Medium and high Natural soluble uranium
diffusion [Z10] compounds (with 95% CI):
Medium RR=1.2 (0.66, 2.3)
High RR=1.07 (CI: 0.6, 1.9)
a
Risk estimation metrics: ERR (excess relative risk) for which zero represents no excess or deficit; HR (hazard ratio) and RR
(relative risk or rate ratio) are expressed as a multiple of the rate in the baseline (lowest/no exposure) group. (ERR = RR 1).
b
n.e. = not estimable from the likelihood profile.
c
A case-control subset of workers for whom information was available on radon and external gamma exposure and medical risk
factors. Risk estimates were adjusted for those factors.

279. Studies of CSD/IHD/CeVD end points typically had greater statistical power than those for most
cancer diseases because the numbers of CSD-related deaths were much larger than for most types of
cancer. Other considerations arise; one might expect lesser statistical power to detect CSD because the
consensus has been that the risk coefficients derived for external radiation of the circulatory system are
several times smaller than for cancer induction. On the other hand, damage to the kidney, which is
thought to be a primary target organ for uranium, affects the risk of heart disease, probably through the
renin-angiotensin pathway. Results regarding CSD end points are also difficult to interpret because of
the numerous medical and lifestyle factors that affect cardiovascular risk. The results of the French
processing workers (AREVA NC) suggesting that uranium compounds with low solubility may induce
CSD more than soluble compounds are somewhat puzzling. While less soluble uranium compounds are
thought to confer more risk to the lung because of longer residence times, it is believed that more
soluble compounds confer larger doses to most other organs than insoluble compounds do because of
differences in biokinetics. If so, the results of the French processing workers (AREVA NC) are contrary
to what one would expect for CSD, IHD and CeVD.

(g) Conclusion
280. From the occupational exposure studies, a weak association of lung cancer risk with uranium
exposure is concluded. However, currently available results are not consistent enough to demonstrate a
causal association with uranium exposure. The results for leukaemia, other lympho-haematopoietic
malignancies, digestive system cancers, kidney and other urological cancers and brain/CNS tumours
did not provide clear evidence of uranium exposure-related risks. The results for non-malignant
diseasesrespiratory, cardiovascular and kidney diseasesalso showed no relationship with uranium
exposure. A number of studies without dose-response analyses for LLR exposure have provided null
overall risk estimates for every health end point considered; while the negative SMRs are not very
specific, at least they suggest the risks are likely to be small.

5. Limitations of occupational exposure studies

281. For uranium miners studies, recent developments enabled the calculation of organ doses and,
therefore, estimation of the contribution of LLR from uranium ore. Studies published up to now have
demonstrated a very small contribution of uranium to miner dose. Organ doses appear to be dominated
by radon and radon decay products for lungs and by external gamma exposure for other organs.
Demonstrating a potential risk associated to uranium, therefore, appears difficult.
432 UNSCEAR 2016 REPORT

282. Studies of uranium millers are limited in size, with the exception of the German Wismut miller
study, which allowed assessment of individual radiation doses from uranium exposure [K18].
Undertaking analysis of combined cohorts after further development of organ dose calculation should
improve results in the future.

283. Most studies of workers in the nuclear fuel cycle are limited by the difficulties in estimating the
doses due to uranium internal contamination. Recent studies are the most meaningful since they have
been based on more accurate exposure assessment and for some of them internal organ-specific
absorbed doses were estimated by implementing the latest updates of the ICRP models and dosimetric
tools. Nevertheless, improvement of dosimetric estimation is still needed to provide pertinent estimates
of potential risks associated with uranium contamination. Segmentation between the different steps of
the nuclear fuel cycle should enable improved assumptions regarding the solubility of the uranium
compounds. The combined use of individual urinalysis dosimetry and of job exposure matrices may
also allow improved characterization of radiation doses from uranium exposure and the quantification
of related risks.

284. Natural uranium is not very radioactive (238U decays very slowly) and its chemical properties are
often such that any inhaled or ingested uranium is excreted rather quickly from the human body. Thus,
studies of exposure to enriched or reprocessed uranium may be more informative. Some publications
indicated thateven when the uranium doses are knownexternal exposure can dominate [L24]. In
addition, most of these studies had major limitations (poor statistical power, no or imprecise estimates
of doses, insufficient accounting for other exposure influences). Moreover, other exposure is mostly not
taken into account, such as exposure to chemicals, heat and noise, which may also contribute to the
increase of certain diseases.

285. Uranium worker data have often been limited to studies of male Caucasians. Quantitative
generalization to women or other population groups is therefore uncertain. No occupational studies
have attempted to examine genetic, epigenetic or metabolic susceptibility factors for uranium related
diseases. Worker studies also provide no information about children, who may be more susceptible to
the effects of uranium exposure than adults.

286. Continued follow-up of the principal uranium worker cohorts that have individualized worker
exposure data will be valuable. Many members of the cohorts were relatively young at their most recent
follow-up. Because mortality rates for many cancers increase as a power of age attained, these cohorts
will become increasingly informative with future follow-up, providing greater ability to detect smaller
effects and generate more precise risk estimates.

287. Since most studies have a limited number of uranium workers and relatively slight exposure for
most of the workers, it is unlikely that epidemiological studies of individual nuclear facilities will have
sufficient statistical power for a reasonable prospect to detect risks. Consequently, international pooled
studies with high-quality, harmonized individual exposure estimates are likely to be necessary to assess
uranium risks with high precision. However, studies of workers have advantages over those on
environmental exposure of the public or of special groups (e.g. military personnel deployed in regions
with potential DU exposure). The advantages particularly centre around having measurements to
estimate individual uranium exposure levels, along with other occupational radiological exposure so
they can potentially distinguish LLR exposure from other exposures. In addition, some of the studies
had data on various chemical exposure at workplaces.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 433

B. Studies of Gulf War veterans

288. Depleted-uranium munition and armour were extensively used by the United States military in the
first Gulf War in Iraq and Kuwait (Desert Storm) and again in the Balkans military action. Military
personnel were exposed to DU via inhalation or wounds, notably due to friendly-fire incidents burning
depleted-uranium-containing tanks and ammunition and clean-up operations [B26].

289. Several authors reported investigations on the Gulf war and the Balkan veterans, especially on
United States, United Kingdom, Canadian, Danish and Dutch veterans [B28, M1, M2, S36, U11],
which were reviewed by others [I20, L5, S6]. For example, the Institute of Medicine [I20] discussed
extensively the results of about 25 studies on health outcomes following exposure to natural uranium
and DU. The review integrated malignant (lung cancer, leukaemia, Hodgkins and non-Hodgkins
lymphoma, bone cancer, renal cancer, bladder cancer, brain and other nervous system cancers, stomach
cancer, prostatic cancer, testicular cancer) and non-malignant (renal disease, respiratory disease,
neurological effects, reproductive and developmental effects, cardiovascular effects, genotoxicity,
haematological effects, immunological effects, and skeletal effects) pathologies. They concluded that
there was insufficient evidence to determine whether an association exists between exposure to uranium
and the health outcomes cited above. However, the two following major limitations of the veteran
studies were identified: (a) short period of follow-up and (b) poor assessment of uranium exposure.

290. The Royal Society comprehensively reviewed the use of DU, especially on the battlefield [R28].
The Royal Society concluded in its report that doses from DU were unlikely to be high, even in the
most unfavourable (battlefield) conditions, so that lung cancer risks were unlikely to be more than
doubled. The report indicated a potential non-radiological risk associated with exposure to DU, in
particular with its nephrotoxicity. A summary of studies of the health status of veterans with potential
or known exposure to DU is given in appendix A, table A5.

291. A recent study by Strand et al. aimed to investigate cancer incidence and also all-cause mortality
in a cohort of Norwegian military present in Kosovo between 1999 and 2011 [S42]. Cancer incidence
and mortality were studied from 1999 to 2011 and compared to national rates. The authors found no
excess incidence of cancer except an elevated SIR for melanoma of the skin in men. All-cause mortality
was half the expected rate (SMR=0.49; 95% CI: 0.35, 0.67).

292. A biennial health surveillance programme established for the United States Gulf War veterans has
shown continuously elevated DU concentrations in urine among those with embedded fragments for
over 20 years [M25]. No differences have been seen between the high- and low-exposure groups with
regard to haematology, clinical chemistries, neuroendocrine parameters, bone metabolism,
neurocognitive function, immune function, pulmonary function or nodules. Regarding renal function
and injury, no high vs. low exposure differences were found for 16 clinical indicators of renal function,
six urine markers for kidney injury, or four urine measures of low molecular weight proteins, except for
two sensitive biomarkers of proximal tubule function that suggested subtle renal injury [M25].

293. In a study by Hines et al., some self-reported respiratory symptoms, mean pulmonary function
values and prevalence of low-dose chest computed tomography abnormalities were compared in two
populations of Gulf War veterans (high body burden group vs. low body burden) [H18]. The authors
found no significant differences between the two groups, suggesting that DU levels inhaled during the
1991 fire incidents probably do not cause long-term adverse pulmonary health effects [H18].

294. In conclusion, several studies on the health pathologies among veterans with potential or no
exposure to DU were published. Up till now, no clinically significant pathology related to DU has been
found in the veterans cohorts. The diversity of these studies in terms of topics has limited their
434 UNSCEAR 2016 REPORT

reproducibility, except for the biennial examinations of a small group of United States veterans with
retained DU shrapnel in whom comprehensive examinations have consistently found no clinically
adverse effects.

C. Studies of environmental exposure

1. Living around uranium processing facilities

295. Numerous, mostly ecological, studies have been carried out to assess whether long-term residence
in the vicinity of nuclear fuel cycle facilities or nuclear power plants affects the health of the residents.
To focus on uranium effects, only studies carried out in population groups living around uranium
processing facilities (after uranium mining and prior to electricity production) were examined. Eleven
published studies were identified over the past ten years, which are presented in appendix A, table A6.

296. Because of potential bias, inability to check the validity of ecological results, and the lack of
sufficient measurements of ambient uranium exposure levels, no firm conclusion could be drawn from
ecological studies. Further, caution is required in interpreting ecological studies in general as causal
inference is not warranted because of numerous limitations in their study design. The major limitation
of ecological studies is the potential of ecological associations to misrepresent, sometimes greatly, the
biological effect at individual level. Thus, an association observed between variables on an aggregate
level does not necessarily mean that the same association will exist at individual level [G19].

297. Lane et al. reported a review of 13 epidemiological studies conducted in Port Hope, Canada in the
past 30 years, including residents and workers [L7]. These studies included environmental
measurements of the radiological and non-radiological contaminants, the estimation of the multi-
pathway of exposure and also the health risks to the population, using environmental monitoring data or
dose reconstruction methods based on a variety of approaches. The authors concluded that, taken
together, the findings of these studies conducted on the Port Hope community indicated that observed
adverse health effects were unlikely to be the result of exposure to environmental contaminants from
radium and uranium processing. Other studies shown in appendix A, table A6 are also consonant with
that conclusion.

2. Living in an environment affected by depleted uranium munition use

298. Some epidemiological studies attempted to determine if the health of populations living in
countries or regions involved in the recent conflicts (i.e. Iraq, Kuwait, Bosnia and Herzegovina,
Kosovo, Serbia and Montenegro) was affected by the use of DU in shrapnel or tanks. Iraq is the most
studied country for investigation on possible effects of DU. Several publications aimed at describing
the incidence and types of congenital malformations [A6, A7, S48] or birth defects [A3, B54, F1].
However, these publications failed to demonstrate a link between the increase in these pathologies and
the environmental exposure to DU, notably due to the absence of evaluation of the exposure levels.

299. Few studies investigated cancer incidence in these populations. Al-Hashimi and Wang used in
their study three sub-periods (19801990, 19912000, and 20012010), corresponding to the three Iraq
wars, the IranIraq war (19801988), the Gulf War I in 1991 and the Gulf War II in 2003 [A4]. The
authors reported increases in the total number of cancer cases. However, the in-depth analysis indicated
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 435

a decrease in incidence rates in most cancer types when they were analysed statistically, considering
population growth in the Ninawa province in the northern part of Iraq.

300. Another study aimed to describe changes in haematological malignancies (leukaemia and
Hodgkins lymphoma) in Croatian counties potentially exposed to DU in comparison to the pre-war
period [L3]. This study did not find a significant difference in the incidence of these haematological
malignancies.

301. In parallel to these health studies, measurements were made of daily urinary uranium excretion in
German peacekeeping personnel (n=1,228) and unexposed subjects coming from the South of Germany
(n=113) to assess potential intakes of DU [O1]. A daily urinary excretion of uranium of
13.92.2 ng/day (3 to 23 ng/day) measured for German peacekeeping personnel was similar to that of
unexposed subjects (12.82.6 ng/day).

3. Drinking water with elevated uranium levels

302. Possible health effects after long-term ingestion of uranium via drinking water was reviewed by
Guseva Canu et al. [G32]. The description and main results for selected studies of the possible impact
of elevated levels of uranium in drinking water are summarized in appendix A, table A7. This table
notes potential uranium effects ascribable to its dual radiological and chemical toxicity. However, some
studies related the effects to chemical toxicity only [K26, K27, K28, M9, S8, Z6], while other studies
related to potential radiation effects [A32, K29, S7].

303. As shown in appendix A, table A7, most of the studies focused on the nephrotoxicity of uranium
using cross-sectional study designs. In total, five studies were carried out: in Canada [M9, Z6, Z8]; in
Finland [K26]; and in Sweden [S8]. The uranium concentrations in water were fairly similar in all the
studies, with median concentrations in the range 2030 g/L among the exposed groups. All these
studies, except the one from Mao et al. [M9], found no glomerular effect of chronic ingestion of
uranium. Among people drinking water from private drilled wells, uranium exposure caused damage to
the proximal tubule, shown by nephron reabsorption alteration [K26, S8, Z6] or tubular cytotoxicity
[S8, Z1] was observed in four of the studies. Several biomarkers were measured in these studies (e.g.
creatinine for glomerular filtration function at the early stage of renal injury) but none was specific for
injury caused by uranium. Kurttio et al. [K29] carried out a casecohort study in Finland of bladder and
kidney cancer after long-term consumption of private well water containing uranium and its decay
products. No association between the prevalence of these cancers and the uranium concentration in well
water was found [K29].

304. Lymphatic and haematopoietic malignancies were considered in three studies [A32, S7, W23].
Seiler [S7] investigated whether 16 children with leukaemia in the City of Fallon, Nevada, United
States, had higher levels of naturally occurring radioactive material in their well water compared to
other inhabitants of the town. Water samples were collected in 2001 for the measurement of uranium,
radon and gross alpha concentrations, and leukaemia cases were identified for 19972000. To resolve
this potential time sequence problem, the authors also retrieved the 1989 citywide water analyses. The
natural origin of the uranium present was confirmed by the calculation of the isotopic ratio. No
difference was indicated in uranium concentration in the water drunk by the children compared to other
inhabitants.
436 UNSCEAR 2016 REPORT

305. Witmans et al. [W23] compared the uranium concentrations in water between non-Hodgkins
lymphoma cases and their matched controls selected from the Saskatchewan (Canada) cancer registry.
The cases had been exposed to significantly higher uranium concentrations in drinking water than the
controls. However, uranium was one of 63 inorganic constituents tested in the study.

306. A casecohort study by Auvinen et al. [A32] of Finnish adults that enrolled 35 cases of leukaemia
also reported a negative result regarding exposure to naturally occurring uranium (and its decay
products) in drinking water and leukaemia [A32]. The statistical power of the study was limited and no
data on potential confounding factors were available. The risk of stomach cancer from exposure to
naturally occurring radionuclides in drinking water was investigated. However, no association was
found in this study.

307. Clinical studies in Nova Scotia, Canada performed on 324 persons exposed to variable amounts of
naturally occurring uranium in drinking water (up to 0.7 mg/L) found no relationship with overt renal
disease. Though there was a trend towards increasing excretion of urinary -2 microglobulin with
increasing concentration of uranium in well water, this was not seen in the group with the highest
uranium well-water concentrations. This group had significantly reduced its consumption of well water
by the time the measurements were made, supporting the hypothesis that the suspected tubular defect
might well be rapidly reversible [M64, M65].

308. A pilot study by Mao et al. of three communities in Saskatchewan with mean uranium levels
ranging from 0.71 (control) to 19.6 g/L found a statistically significant association (p=0.03) between
increasing but normal levels of urine albumin and the uranium exposure [M9]. Another Canadian study
on two groups of subjects with chronic exposure to uranium in drinking water, the first group exposed
to <1 g/L and the other exposed to 2781 g/L found no correlation with alkaline phosphatize and
-2 microglobulin in urine. The authors concluded that the uranium concentrations observed in the
study affected the kidney function at the proximal tubule [Z6].

309. Another study by Zamora et al. [Z8] on chronic ingestion of uranium in drinking water
demonstrated subtle changes in two of the indicators measured that were statistically significant
namely, glucose and LDH excretion concentrations. However, this did not result in any observable
health effects (see also table 13). In addition, the change in LDH excretion was rather beneficial and
was seen only in males. Thus, these changes are not nephrotoxic effects.

310. In conclusion, epidemiological studies of public uranium exposure to drinking water indicate that
chemical toxicity of uranium may occur mainly in the kidneys and, in high concentrations, uranium
may affect the kidney function. However, the functional alterations found in the kidneys were small and
within normal limits, so the clinical significance of the findings may be minimal. The available
literature focused on lymphatic and haematopoietic tissue malignancies is limited to three studies,
which do not support a causal association between uranium exposure and those malignancies.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 437

IX. RESEARCH NEEDS

311. The estimation of organ doses from incorporated uranium isotopes depends on the availability of
reliable biokinetic data and the construction of physiologically realistic biokinetic models. In general,
good human and animal data are available for the construction of models. However, limited information
is available on the age-dependence of organ retention and excretion rates, including information on the
cross-placental transfer of uranium. In addition, more information is required on the distribution of
uranium within tissues and cells, for example in CNS tissue and lungs.

312. Dosimetric models in general apply the same assumptions regarding source and target
distributions within tissues to all internal emitters. For example, uranium isotopes and other
radionuclides deposited in bone are assumed to accumulate on internal bone surfaces and/or in bone
volume and target cells for cancer induction are assumed to reside along bone surfaces (bone cancer)
and throughout red bone marrow (leukaemia). The validity of such assumptions requires further
investigation, with consideration of the inhomogeneity of uranium distribution within tissues and cells.

313. Toxicological studies of uranium exposure are required to distinguish the chemical and
radiological components of damage caused to cells and tissues, including short-term damage to organ
function and longer-term effects including cancer. Comparisons of radionuclide toxicity and RBE
determinations would assist in quantifying the potential health effects of uranium isotopes. Studies of
the age-dependence of chemical and radiological toxicity would be valuable.

314. Future epidemiological studies require careful consideration of the acquisition of dosimetric data
to assess individual organ and tissue doses for cohort subjects. A high priority would be a consortium
effort by investigators to develop pooled data on uranium risks. The result could be considerable gains
in the statistical power and precision of risk estimates that would potentially provide the best overall
answers achievable as to health effects from uranium exposure. Pilot studies to quantify the magnitude
of uncertainties in exposure assessment would ideally be part of this effort, so that sound estimates of
dosimetric uncertainties could be incorporated into the risk modelling [L13, S21].

315. Concerning (molecular) epidemiology, setting up prospective follow-up or case control studies in
selected subgroups, including collection of information on biomarkers, has the potential to provide
more specific doseresponse curves for defined subsets of cohorts and thereby improve knowledge of
health effects in humans, including cancer and non-cancer diseases. High-throughput technologies
(especially the -omics) would be relevant to apply to this field. However, proposed biomarkers will
need to be rigorously evaluated as to their ability to improve exposure and risk assessment.

316. Mixed exposure should be taken into account when studying effects, such as other radionuclides
(e.g. 239Pu, 222 Rn), other chemical carcinogens (e.g. solvents, smoking, dust, silica, asbestos) and also
the physical forms of uranium (solubility), e.g. through further development of exposure matrices in
epidemiological research, and through animal studies.

317. Understanding the molecular mechanism of action of uranium on cells in culture and animal
models, both as a metal and as a radionuclide, would be important in (a) facilitating the identification of
bioindicators; (b) identifying portions of the molecular response that are attributable to the radiation
response, the heavy metal response or both; and (c) defining the possible development of mitigators,
(few mitigators for uranium exposure).
438 UNSCEAR 2016 REPORT

X. GENERAL CONCLUSION
318. This annex provides a detailed review of sources and levels of uranium in the environment,
exposure of the public and workers to uranium, biological effects of uranium, and epidemiological
studies of nuclear workers and the public exposed to uranium.

319. Uranium is a naturally occurring radionuclide and is ubiquitously distributed in the environment.
In daily life, people are exposed to uranium originating mainly from drinking water and foodstuffs.
Average uranium levels in water vary between countries and within countries, with typical values of
around 2 g/L (~25 mBq/L of 238U) in groundwater and 1 g/L (~12.4 mBq/L of 238U) in public water
supplies. Some drinking water samples (<3%) may exceed the national or international guidelines set to
prevent kidney toxicity. Concerning foodstuffs, potatoes, meat, fresh fish and bakery products are the
main sources of uranium ingestion. The total daily intake from water and food consumption is around
1.5 g/d (18.6 mBq/L of 238U).

320. The main routes of entry of uranium into the body are inhalation and ingestion. The absorption to
blood in each case is highly dependent on the chemical form (speciation) of the intake. For example,
human data show that the absorption of ingested uranium is a few per cent of intake for soluble forms
in water compared with substantially less than 1% for insoluble oxides. Human and animal data have
been used to model the behaviour of uranium absorbed to blood, showing that the main site of retention
is the skeleton, with lower amounts in soft tissues and rapid urinary excretion of a large proportion. The
ICRP models make appropriate use of the available data.

321. Uranium is both a radioelement and a metal, and biological effects may result from the combined
effects of the chemical element or species and the radiation. The radiological and chemical
consequences of internal exposure to uranium depend partly on the route of intake (principally
inhalation or ingestion), and the chemical form of the intake. Some effects are likely to be related to the
chemical toxicity of uranium species, namely the renal effects, whereas others are rather related to
radiological toxicity of uranium, including tumorigenic effects such as soft tissue sarcomas in rats and
osteosarcoma in mice. In general, chemical effects are observed with short lag-times after exposure
whereas radiological effects such as carcinogenesis have long lag-times.

322. Considering the chemical effects of uranium species, the kidneys are the most sensitive target
organ. At higher levels, chemical effects of uranium are also observed in bones, indicating that uranium
can induce effects on bone metabolism such as the impairment of bone growth and formation. Chemical
effects of uranium have also been observed, in rodent studies, in liver, gonads, central nervous system,
and the immune system. These experimental studies indicate that uranium induces biological effects in
these organs, but the changes do not lead to the appearance of observable pathologies. While effects in
these tissues may be seen at higher doses, damage to kidneys (and skeleton) is likely to be critical.
Concerning the central nervous system, animal studies suggest that high doses of uranium may have
some negative effects on the behaviour of animals. With the exception of kidney damage, animal
studies showing toxicological effects have used concentrations of uranium substantially above those to
which humans are exposed. No clinically significant pathologies have been found in the veteran cohorts
potentially exposed to DU. Moreover, the biennial examinations of a group of United States veterans
with retained DU shrapnel have found no clinically meaningful adverse effects.

323. Epidemiological studies of uranium miners and millers have included estimates of doses, showing
the small contribution of uranium to overall doses and the dominant contributions of radon and radon
decay products to lung dose and external gamma radiation for other organs. Most studies of nuclear
workers are limited by difficulties in estimating radiation doses due to uranium. A weak association of
lung cancer risk with uranium exposure is suggested but the currently available results are not
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 439

consistent enough to demonstrate a causal association. Results for other malignancies and non-
malignant disease were also negative. The Committee concluded that epidemiological studies of public
exposure to uranium in drinking water have reported small functional alterations in the kidneys, within
normal limits and hence of minimal clinical significance.

XI. ACKNOWLEDGEMENTS
The Committee wishes to acknowledge with gratitude the help of the experts E. Ansoborlo (France),
Y. Gueguen (France), I. Guseva-Canu (France), D. Laurier (France), P. Lestaevel (France), F. Paquet
(France), E. Rage (France), J. Harrison (United Kingdom) and R. Shore (United States), who were
directly involved in preparing the evaluation, and S. Baatout (Belgium) for critically reviewing the
manuscript. In particular, the Committee would like to thank I. Dublineau (consultant), who mainly
elaborated this scientific annex based on a first draft prepared by K. Bundy (Canada), D. Chambers
(Canada), R. Lane (Canada) and B. Thriault (Canada). The views expressed in this scientific annex
remain those of the Committee and do not necessarily represent the views of individual experts, the
United Nations or its Member States.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 441

APPENDIX A. TABLES SUMMARIZING URANIUM LEVELS IN

WATER, STUDIES OF OCCUPATIONAL AND ENVIRONMENTAL
EXPOSURE TO URANIUM
Tables A1 to A4 present uranium levels in water.

Tables A5 to A8 present four groups of occupational studies that provide increasing levels of
information on uranium-specific risks.

Tables A9 to A11 present studies of risks from various sources of potential military and environmental
exposure to uranium.
Table A1. Overview of uranium content in groundwater worldwide
The numbers in italics correspond to calculated data obtained from the mass activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium of 48.3%; * mean
value (not median value); n.i: information not included in the study

238
Continent Country Location Sample number Total uranium (g/L) U (mBq/L) Reference
Europe Finland n.i. 396 0.15 (<0.0110) 1.84 (0.12123) [T22]
Finland South 13 285 (5.63 410) 3 504 (14341 924) [P28]
Finland South 288 1.5 (0.3800) 19 (49 990) [V4]
Finland South 194 28 (0.0011 920) 344 (0.0123 605) [K26]
Finland South 167 2 (<0.011 770) 24.6 (<0.1221 761) [M67]
Sweden Vrmland 153 6.7 (<0.20470) 82.4 (<2465 778) [S8]
Sweden South 328 14.2 (<2425) 177 (<275,293) [I22]
Norway South 476 2.5 (750) 31 (9 300) [F12]

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 443

Poland Swieradow 51 1.26 (0.22.4) 15.55 (2.429.4) [K10]
Spain Catalonia 37 4.88 (<0.4156.1) 60 (<5690) [O8]
Switzerland n.i. 5 548 0.77 (0.0592.02) 9.5 (0.611 131) [S33]

France Vals les Bains n.i. 1.85 (0.553.6) 22.7 (6.844.3) [M34]
Greece North 10 2.02 (0.157.66) 25.06 (1.8295.3) [S1]

North America Canada n.i. n.i. 0.31 (0.166.23) 3.8 (2.076) [L12]
Canada Kitigan Zibi, Quebec 32 39.25 (0.4845.33) 470 (4.910 392) [Z8]
Canada Southeastern Manitoba 287 10 (<0.022 020) 124 (<0.2525 048) [B21]

Canada Nova Scotia 20 0.39 (0.0641.08) 4.8 (0.8505) [K19]

USA Connecticut 11 16.3 (0.211 166) 200 (2.614 335) [M5]
USA Connecticut 35 157 (1.87 780) 1 930 (22.195 649) [O5]

USA Cities with the largest 55 433 1.04 (0.031 945) 12.95 (0.3724 124) [C34]
average concentration

South America Brazil Sao Paulo et Santa Catarina 78 0.28 (0.00815.0) 3.4 (0.1184.3) [B43]
Brazil n.i. 358 1.2* (<0.017.5) 14.8 (0.1292.2) [G14]
 444
238
Continent Country Location Sample number Total uranium (g/L) U (mBq/L) Reference

UNSCEAR 2016 REPORT

Africa Morocco n.i. 15 4.5 (0.3725.1) 55 (4.5309) [H4]
Ethiopia Rift Valley 138 0.55 (0.00548) 6.8 (0.06590) [R10]
Egypt Eastern desert 12 5.21 (1.19519.4) 64.1 (14.66 386) [D2]
Ghana North and coast 195 0.114 (<0.0011.99) 1.4 (<0.0124.4) [R23, Z16]

Asia Fujian Province China 110 and 552 0.54 (0.0313.4) 6.6 (0.4164) [Z16]
Iran Caspian Sea 27 2.2 (0.245.4) 27 (366) [J5]
Jordan n.i. n.i. 1.3 (0.56.7) 16.0 (6.682.4) [V3]
Bangladesh West 67 2.5 (0.210) 30.7 (2.525.8) [F14]
Japan Niigata 23 0.001 (0.00050.03) 0.018 (0.0050.383) [T14]
India Hisar 38 33.9* (5.3113.5) 417 (65.21 395) [G3]
India Kula area 15 0.83 (0.262.56) 10.2 (3.231.5) [S27]
India Punjab 25 22 (2.6574.98) 271 (32.6922) [K22]

Table A2. Overview of uranium content in surface water worldwide

The numbers in italics correspond to calculated data obtained from the mass activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium of 48.3%;
n.i: information not included in the study

238
Continent Country Location Sample number Total uranium (g/L) U (mBq/L) Reference
Europe Finland Southern Finland 184 0.18 (0.0834) 2 (1.0420) [V4]
Finland n.i. 152 0.099 (<0.010.92) 1.2 (<0.1211.3) [T22]

North America USA Cities with the largest average 34.561 0.45 (0.031 737) 5.55 (0.3721 534) [C34]
concentration

South America Argentina n.i. 92 1.9 (<0.0150) 23.4 (<0.12615) [B41]

Asia India Upper Siwaliks and Punjab 34 3.84 (1.0819.68) 47.2 (13.3242) [S23]
Iran Ardabil 22 4.2 (2.113.6) 51.6 (25.8167) [H2]
Table A3. Overview of uranium content in public water supplies worldwide

The numbers in italics correspond to calculated data obtained from the mass activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium of 48.3%; * mean
value (not median); n.i: information not included in the study

238
Continent Country Location Sample number Total uranium (g/L) U (mBq/L) Reference
Europe Finland Southern Finland 951 1.25 (<0.011 770) 15.4 (<0.1221 761) [M67]
Germany Bavaria 461 0.9 (<0.0139) 11.1 (<0.12479) [R3]
Germany National scale 564 3.2 (<0.7320) 39.3 (8.63 934) [B23]
Germany National scale 164 0.073 (0.001159.0) 0.90 (0.001111) [B24]
Germany National scale 36 2.0 (<0.1660.2) 24 (<2740) [G2]
Austria n.i. 41 0.91 (0.0679.2) 11.2 (0.72975) [G6]
Austria Waldviertel 48 1.1 (0.157.5) 14.05 (0.7707) [W5]

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 445

Greece Aksios, Kalikratia 23 3.46 (0.06110.02) 42.5 (0.75123) [K7]

Poland Centre 26 0.39 (0.031.94) 4.8 (0.423.9) [P20]

Spain Biscay 4 0.07 (0.0030.24) 0.8* (0.042.9) [H17]
Italy Rome 9 1.46 (0.028.37) 18 (0.3103) [J2]

Norway, Sweden, Finland, Iceland Several places 22 0.107 (0.004956.2) 1.32 (0.06691) [F13]

North America USA Illinois, Minnesota, Texas, Wisconsin 24 0.25 (0.113.2) 3.1 (1.23162) [L50]

South America Argentina 12 provinces 145 0.4 (<0,0121) 4.9 (<0.12258) [B41]

Africa Morocco National scale 6 0.75 (0.201.28) 9.25 (2.515.7) [H4]

Asia Islamic Rep. of Iran Ardabil 3 7.6 (4.711.7) 93.4 (57.8144) [H2]

India Punjab 3.84 (1.1119.68) 47.2 (13.6242) [S22]

India Punjab 45 10.4 (1.2445.42) 260 (301 150) [R8]
India Punjab 29 (2.5313) 357 (30.73 848) [S11]

India Himachal Pradesh 46 1.34 (0.5610.11) 30 (10260) [R8]

India Western Haryana 23 17.03 (6.3743.31) 209.4 (78.3532) [K1]

Oceania Australia West 23 0.19 (01.16) 2.3* (014.3) [W7]

Australia West 173 0.06* (<0.0011.40) 0.74 (<0.0117.2) [C1]
 446
UNSCEAR 2016 REPORT
Table A4. Overview of uranium content in bottled mineral water worldwide

The numbers in italics correspond to calculated data obtained from the mass activity of natural uranium of 25,400 Bq/g and from the relative proportion of 238U (12.4 mBq/g) in natural uranium of 48.3%; * mean
value (not median); n.i: information not included in the study

238
Continent Country Sample number Total uranium (g/L) U (mBq/L) Reference
Europe Germany 908 0.17 (<0.000516.0) 2.09 (<0.006197) [B24]
Germany 21 0.41 (<0.0811.4) 5 (<1140) [G2]
Spain 32 0.48 (0.045.8) 5.9 (0.570.9) [D14]
Slovenia 11 0.42 (0.094.3) 5.2 (1.153) [B13]
Italy 21 1.38 (0.209.92) 17 (2.5122) [R30]
Italy 51 0.73 (<0.017.2) 8.97* (<0.1789) [D11]
Poland 22 0.59 (0.060.86) 7.26 (0.7510.54) [C19]
Croatia 12 0.53 (0.171.19) 6.55 (2.114.6) [R29]

France, Portugal, Spain 14 3.58 (1.7940.7) 44 (22500) [M14]

France 106 0.2 (<0.1019) 2.25 (<1.3230) [A27]
28 countries 132 0.23 (0.000227.5) 2.8 (0.002338) [K11]

Austria 10 0.16 (0.0125.4) 2.0 (0.1566.4) [W4]

Norway, Sweden, Finland, Iceland n.i. 0.102 (0.005532.4) 1.25 (0.07698) [F13]

South America Argentina 62 1.9 (0.0411) 23.4 (0.5135) [B40]

Africa Tunisia 10 1.01 (0.132.14) 12.36 (1.5626.36) [G7]

Morocco 10 0.54 (0.340.70) 6.6 (4.28.6) [M48]

Asia Kuwait 23 0.22 (0.052.04) 2.74 (0.6325.07) [A11]

Table A5. Studies of workforces that include workers (potentially) exposed to uranium though not explicitly identified

Study references Summary of study Summary of findings relating to uranium Relevance for this report
Atkinson et al. UKAEA employees, 19461996. 10,249 deaths. Dosimetry: All cancer: external radiation exposure trend tests in Not possible to derive uranium-specific risks,
[A29] external radiation doses. Neutron and tritium doses included those monitored for any internal exposure. Trends because uranium workers not analysed
when available. Internal doses noted but not quantitative not significant for all cancer, stomach, colon, liver, separately, and uranium-specific doses not used
pancreas, lung, bladder, kidney, brain
Prostate cancer: dose-response trend before 1980, but
not 19801997 among those with internal
monitoring, but uranium not examined

Atkinson et al. Further analysis of an extended UKAEA dataset [A29], conducted by Prostate cancer: exposure levels at heavy-water Not possible to derive uranium-specific risks,
[A30] time period to examine internal exposure for pre/post 1980 and reactors fairly constant over time, but no indication because uranium workers not analysed
prostate cancer. Excess associated with work with heavy-water of elevated risk after 1980. So earlier excess with separately, and uranium-specific doses not used
reactors. Radionuclides of concern: 3H, 59Fe, 51CR, 60Co, 65Zn. Case- internal exposure probably not meaningful
control prostate cancer substudy conducted

Carpenter et al. Cancer mortality 19461988 among 75,006 UKAEA, AWE and All cancer: analyses of external radiation plus tritium Not possible to derive uranium-specific risks,
[C8] BNFL employees. Uranium exposure not assessed exposure because uranium workers not analysed
separately, and uranium-specific doses not used

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 447

Carpenter et al. Cancer mortality 19461988 among 40,761 UKAEA, AWE or BNFL All cancer: separate analyses conducted for tritium, Not possible to derive uranium-specific risks,
[C9] employees who had radionuclide monitoring plutonium and other radionuclides. Insufficient detail because uranium workers not analysed
about other radionuclides, so no uranium analyses separately, and uranium-specific doses not used

Cragle et al. Savannah River Plant (USA) conducted uranium processing, Analyses by time of first employment and years of Not possible to derive uranium-specific risks,
[C36] nuclear fuel fabrication and processing. Follow-up to 1980, employment. Suggestion of elevated leukaemia risk because uranium workers not analysed
9,860 white male employees. 85% of exposure to external in small subgroup of early workers, but uranium separately, and uranium-specific doses not used
radiation; exposure to numerous internal radionuclides exposure not reported

Fraser et al. [F10] Cancer mortality and morbidity in UKAEA cohort of 39,718 during Cancer analyses conducted for external exposure, Not possible to derive uranium-specific risks,
19461986. Internal exposure noted for tritium, plutonium or tritium, plutonium and monitored for any because uranium workers not analysed
other unspecified radionuclides radionuclide separately, and uranium-specific doses not used

Loomis and Wolf Cancer mortality (19471990) among 6,591 white males at USA Oak No measurements or estimates of uranium exposure, Not possible to derive uranium-specific risks,
[L44] Ridge Y-12 nuclear material production plant. Plant converted UF6 so no relevant analyses because uranium workers not analysed
to UF4 to uranium metal which was fabricated and milled. Other separately, and uranium-specific doses not used
exposure: beryllium, solvents, machine oils, mercury, lead

McGeoghegan Mortality and cancer morbidity in 2,209 radiation workers at UK No measurements reported of internal exposure: only Not possible to derive uranium-specific risks,
and Binks Chapelcross plant, 19551995. Main activity: operation of external radiation analysed because uranium workers not analysed
[M28] 450 MW Magnox gas cooled reactors separately, and uranium-specific doses not used

McGeoghegan 38,779 radiation and 15,040 non-radiation male workers at UK Investigated circulatory and other non-cancer Not possible to derive uranium-specific risks,
et al. [M29] BNFL facilities (Sellafield, Springfields, Capenhurst, diseases. No measurements of internal exposure, because uranium workers not analysed
Chapelcross). Some workers had exposure to uranium, analysed external radiation exposure only separately, and uranium-specific doses not used
plutonium, tritium and other radionuclides
 448
Table A6. Studies of groups of workers identified and investigated as uranium workers but not monitored specifically for potential exposure to uranium, so uranium-

UNSCEAR 2016 REPORT

specific doses were not available
Abbreviations: LLR, long lived radionuclide exposure, primarily from uranium

Study references Summary of study Summary of findings relating to uranium Relevance for this report
Baysson et al. Metallurgy department workers (N=356) of French CEA, 255 handled radionuclides, principally natural Study small, low statistical power. Analysis only for
[B9] 19501968 were studied for excess cancers (19501990), uranium, mean exposure duration 11 years exposure to any radionuclide. Potential confounding
since workers believed there was a cancer cluster. All cancer: suggestion of risk (p=0.13) per year by chemical exposure. Multiple myeloma results:
Department research primarily on uranium metallurgy. handling radionuclides, but stronger trend for small numbers, maybe a chance finding from
Reconstructed external radiation doses; internal handling chemicals multiple comparisons. Not possible to derive
radionuclide and chemical exposure noted. Radionuclides: uranium-specific risks, because uranium doses not
thorium, natural uranium/enriched uranium, some Multiple myeloma: suggestive excess, but only 2 cases used
activation and fission products (0.2 expected, SMR=8.4, 90% CI: 1.4, 26). No evidence
of cancer cluster

Boice et al. [B36] Mortality in uranium miners and millers, Grants, New Mexico, Increased mortality in underground miners: lung One of few studies of uranium millers. Study suggests
19792005: 1,735 underground uranium miners and 904 cancer, non-malignant respiratory disease, liver uranium exposure effects are small or absent but not
non-mining uranium millers. No measurements available cirrhosis. No significant excess among non-mining possible to derive uranium-specific risks, because
on either radon or uranium exposure levels millers.Among uranium millers: uranium doses not used
Total cancer: SMR=0.89, n=65
Lung cancer: SMR=0.85, n=21
Cerebrovascular disease: SMR=1.06, n=14
Heart disease: SMR=0.84, n=73
Non-malignant respiratory disease: SMR=1.07, n=25
No suggestive excesses of kidney cancer, liver cancer
or lymphoma, but small numbers

Dupree-Ellis et Mortality (19421993) investigated among 2,514 white male All cancer: SMR=1.05 (95% CI: 0.93, 1.07). Some Only total dose analysed, mainly external radiation.
al. [D33] workers at Mallinckrodt (USA) uranium processing plant. evidence of excess kidney cancer (ERR per Sv=10.5, Not possible to derive uranium-specific risks,
Mean cumulative total dose, 47.8 mSv. For ~11 years plant 90% CI: 0.6, 57; n=10) in relation to external radiation because uranium doses not used
also processed pitchblende, which increased external exposure
radiation exposure

Kreuzer et al. Circulatory system disease (CSD) mortality (19462008, Circulatory disease: ERR per Sv for external gamma Analysis for external radiation exposure only. Not
[K16] n=9,039 CSD deaths) in 58,982 male German Wismut radiation: 0.13 for CSD, 0.03 for ischaemic heart possible to derive uranium-specific risks, because
uranium miners. External radiation estimated using a job disease, and 0.44 (95% CI: 0.16, 0.44) for uranium doses not used
exposure matrix cerebrovascular disease
Study references Summary of study Summary of findings relating to uranium Relevance for this report
Lane et al. [L8] 17,660 Canadian Eldorado uranium workers (Beaverlodge and Significant associations of radon decay product Analysis for radon decay products only. Not possible
Port Radium miners, Port Hope uranium refinery/processing). exposure and lung cancer mortality (n=618) in each to derive uranium-specific risks, because uranium
Radon decay product exposure, mortality (19501999) and subcohort. No associations for any other cancers. No doses not used
cancer incidence (19691999) estimates of LLR risk

McGeoghegan Mortality and cancer incidence at UK Springfields uranium No measurements of internal exposure; all analyses for No analyses of internal exposure. Not possible to
and Binks production plant, 19461995. Main activities, uranium fuel external radiation exposure derive uranium-specific risks, because uranium doses
[M27] fabrication and UF6 production. 13,960 radiation workers not used

McGeoghegan Mortality and cancer incidence studied at UK Capenhurst No measurements of internal exposure; all analyses for No analyses of internal exposure. Not possible to
and Binks plant, 19461995. Main activities, uranium enrichment for external radiation exposure derive uranium-specific risks, because uranium doses
[M26] military or power plant purposes. 12,540 employees not used

McGeoghegan 407 workers involved in 1957 Windscale uranium pile fire. No measurements of internal exposure; only external No analyses of internal exposure. Not possible to
et al. [M30] Mortality and cancer incidence 19572007. Estimated radiation analyses derive uranium-specific risks, because uranium doses
plutonium, but not uranium, doses not used

Pinkerton et al. Mortality of 1,484 men employed in 7 uranium mills in No individual estimates of radiation exposure were No data available on internal exposure. Not possible to
[P21] Colorado Plateau, USA (19401998). Mortality (SMRs) made derive uranium-specific risks, because uranium doses
examined by duration of employment and time since first not used

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 449

employment

Rooney et al. Nested case-control study in five UKAEA facilities of incident Prostate cancer: risk increased with duration and Negative association with uranium exposure (yes/no).
[R22] and fatal prostate cancer and exposure to radionuclides. concentrations of exposure to the targeted Not possible to derive uranium-specific risks,
136 prostate cancer cases diagnosed 19461986 and 404 radionuclides (possible exposure RR: 2.36, 95% CI: because uranium doses not used
matched controls. 28 (21%) prostate cancer cases and 46 1.26, 4.43; documented exposure RR: 5.32, 95% CI:
(11%) controls had potential exposure to 3H, 51Cr, 59Fe, 60Co 1.87, 17.2)
and/or 65Zn Indicated prostate cancer not associated with uranium
exposure; but uranium exposure was rare. Found
0 cases and 6 controls with documented uranium
exposure (RR=0, 95% CI: 0, 2.55)

Vacquier et al. French CEA-COGEMA uranium miners (19461990) followed Elevated SMRs for total cancer (SMR=1.19), lung cancer Not possible to derive uranium-specific risks, because
[V1] up through 1999, mean 30.1 years. Radon exposure (SMR=1.43) and kidney cancer (SMR=2.00), but not uranium doses not used
estimates from ambient monitoring and worker job, leukaemia. Significant radon dose response for lung
location and year. LLR exposure not estimated cancer, but not kidney cancer or leukaemia. No
analyses with regard to LLR exposure

Walsh et al. [W6] Radiation dose and prostate cancer mortality examined in External gamma dose response: ERR per Gy=1.18 Study analysed external dose and prostate cancer
the 19701990 subset (55,435 miners) of German Wismut (95% CI: 2.4, 0.02) mortality. Not possible to derive uranium-specific
uranium mining cohort. Follow-up, 19702003, n=263 risks, because uranium doses not used
prostate cancer deaths. Only gamma dose analysed
 450
Table A7. Studies of workers monitored for potential exposure to uranium, with occupational dose records, but uranium-specific doses not explicitly analysed

UNSCEAR 2016 REPORT

Study references Summary of study Summary of findings relating to uranium Relevance for this report
Boice et al. and 46,970 employees, Rocketdyne, USA (19481999); 5 801 For those monitored for internal exposure, no SMR excesses Sufficient data to estimate internal radionuclide
Ritz et al.[B38, had radiation exposure, 2,232 monitored for internal seen for any cause: all cancers (except leukaemia); exposure on basis of urinalyses. Uranium was largest
R16] radionuclides. Mortality follow-up 19482008 (mean leukaemia; cancers of lung, kidney, stomach, liver, contributor to internal dose but other radionuclides
33.9 years). Activities: operating research nuclear prostate, brain; heart disease and cerebrovascular disease also present. Study limitations: relatively low career
reactors, fabricating nuclear fuel, disassembling and Uranium doses, RR at 100 mSv: doses, incomplete information on smoking. Study does
decontaminating reactor facilities, decladding spent not suggest any strong uranium risk but has
nuclear fuel and storing nuclear material. Intakes of 14 All cancer except leukaemia: 0.98 (95% CI: 0.82, 1.17, n=266) uncertainties regarding uranium doses and small
radionuclides calculated for 16 organs using ICRP Lung cancer: 1.01 (CI: 0.89, 1.16, n=94) numbers of deaths
biokinetic models; >30,000 urinalyses. Most significant
Non-CLL leukaemia: 1.06 (CI: 0.50, 2.23, n=10)
internal exposue was from enriched uranium, especially
for lung and kidney. A few workers received high lung Other trends for internal (mainly uranium) exposure: non-
doses (~0.3 Sv) but 87% of workers had committed significant increasing trends, cancers of stomach, kidney,
equivalent dose to all tissues well below 10 mSv brain/CNS, lymphomas; Non-significant decreasing trends,
cancers of colorectum, pancreas, prostate, bladder, and
non-malignant respiratory disease

Dufey et al. Leukaemia nested case-control study in cohort of 58,987 Leukaemia: analyses for total dose to red bone marrow. Strengths included large cohort, long and high quality
[D30] German Wismut male uranium miners, 19462003; 128 For a 2-years lag, linear ERR per Gy was 1.39 (90% CI: follow-up. However, no assessment of LLR uranium
leukaemia cases (40 CLL and 88 non-CLL) and 0.77, 3.56) for all leukaemia and 2.08 (0.84, 4.99) for exposure, so study uninformative regarding uranium
unspecified number of controls. Mining performed non-CLL. Suggestion of increased non-CLL only for the effects
19461989. Cohort mean dose 48.8 mGy, of which highest dose group: for 0.4, 5.0, 25.6 and 103.7 mGy
external gamma contributed 40.9 mGy RRs were 0.53, 0.89, 0.67 and 1.25 (90% CI: 0.69, 2.20),
respectively

Dufey et al. Liver cancer mortality (n=159) in cohort of 58,987 male Liver cancer: the analysis by high-LET dose categories Analysis adjusted for low-LET radiation exposure, but did
[D31] German Wismut uranium miners, 19462003. Mining did not reveal any statistically significant elevations in not account for radon decay product exposure.
19461989. Average liver dose, 47.9 mGy low-LET and risk, and dose-response analysis, adjusting for low-LET Contribution of LLR to total absorbed liver dose was
2.4 mGy high-LET irradiation; mean high-LET liver dose: dose, age and calendar years, yielded ERR per Gy=48.3 <2%, so study provides little information regarding
2.1 mGy from radon/progeny and 0.8 from LLR. Arsenic (95% CI: 32, 129). Examined confounding factors uranium risk
measurements available including arsenic exposure and alcoholism

Dupree et al. 995 white male employees (19431949) of Linde, USA Elevated SMRs found for laryngeal cancer (SMR=4.47, No quantitative analyses by uranium exposure levels, so
[D34] uranium processing company followed up 19431979. 95% CI: 1.4, 10.4, n=5), arteriosclerotic heart disease uninformative regarding uranium risk. Given the
Doses reconstructed from ambient monitoring data, (SMR=1.19, CI: 1.01, 1.39, n=159) and non-malignant exposure information developed, cohort has some
surface contamination, urinalysis and film badges. respiratory diseases (SMR=1.52, CI: 1.04, 2.14, n=32). No potential to contribute to future uranium risk
Exposure mainly to uranium with low solubility. Job excess risk seen for lung, colorectal or lympho- assessment
exposure was categorized as <10, 10100 and haematopoietic malignancies. No analyses by uranium
>100 mSv/y of internal exposure (which was greater exposure levels
than external exposure levels)
Study references Summary of study Summary of findings relating to uranium Relevance for this report
Guseva Canu et Cancer mortality (19682005) examined in 2,709 male All cancer: the SMR=0.70 (95% CI: 0.60, 0.81, n=193) Mortality in relation to internal radiation dose not
al. [G31] workers at French AREVA NC Pierrelatte uranium Lymphoma and rectal cancer: non-significant increases reported, so study uninformative regarding uranium
enrichment and conversion plant. 15 former uranium in rectal cancer (SMR=1.48, n=10) and non-Hodgkins risk
miners excluded. Uranium assessments: individual lymphoma (SMR=1.32, n=8). Trend analyses by time
dosimetry badges; faecal/urine bioassays and in vivo first employment and length of employment not
measurements performed but not available significant for all cancer, lung cancer, upper aero-
digestive tract cancer, and lympho-haematopoitic
malignancy

Kreuzer et al. Stomach cancer mortality (19462003; n=592) in 58,677 Stomach cancer: for alpha irradiation, ERR per Gy=22.5 Not analysed for LLR exposure. Only <1% of the alpha
[K15] male German Wismut uranium miners and exposure to (95% CI: 27, 72) with statistical adjustment for other dose to the stomach due to LLR, so uninformative
external radiation, alpha radiation, fine dust with silica exposure variables. RR in highest alpha dose category regarding a uranium risk
and arsenic. Mean estimated LLR exposure, (1026 mGy) not significant: 1.59 (CI: 0.69, 2.49)
4.1 kBqh/m3 (<0.05 mGy), was substantially correlated
with arsenic exposure

Mohner et al. Nested case-control study of leukaemia mortality (1953 Added analysis of medical X-ray exposure to Report is uninformative for assessing uranium risk since
[M52] 1998) among ~360,000 male German Wismut uranium occupational radiation sources. Analyses were analyses were of all radiation exposure combined
miners. 377 leukaemia deaths and 980 controls conducted only for total occupational RBM radiation

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 451

matched on age. Cumulative red bone marrow (RBM) exposure, both internal and external and for medical
doses from external radiation, radon decay products, X-rays. Case-control analyses for LLR already reported
LLR and occupational medical diagnostic radiation in [M50]
(including 17,578 X-ray examinations), using a detailed
job exposure matrix. Mean cumulative LLR RBM dose
estimated to be <0.05 mGy

Polednak and 18,869 white males worked at the Oak Ridge TEC Elevation in lung cancer (SMR=1.22, 95% CI: 1.10, 1.36) Individual measurements of uranium exposure levels
Frome [P24] uranium conversion and enrichment plant (operated but not higher among those working in areas with available for only an undefined subsample of workers,
19431947) but not at the Y-12 plant which succeeded more uranium dust or those with longer employment. so dose-response analyses not conducted. Smoking
it. Workers in some departments (e.g. chemical dept.) Mortality not elevated for stomach cancer (SMR=0.73), information not available
exposed to high ambient uranium dust. In 1945 kidney cancer (SMR=0.75), bone cancer (SMR=0.90) or
average levels of uranium in air in various departments leukaemia (SMR=0.92)
ranged from 25 to 300 g/m3. Among 226 men with
urine samples, 72% had >0.01 g/ml and 33% >0.05.
Mortality 19431977
 452
Study references Summary of study Summary of findings relating to uranium Relevance for this report

UNSCEAR 2016 REPORT

Rage et al. [R4] Lung cancer mortality (19561999) studied among 3,377 Lung cancer: significant risk lung cancer mortality Statistical analysis of LLR provided only weak
French uranium miners hired 1955 when LLR and associated with total absorbed lung dose (ERR per information; since LLR were correlated with and a small
gamma ray measurements became available. Among Gy=2.94, 95% CI: 0.80, 7.53, n=66) and the alpha- percentage of total exposure the LLR risk estimate may
2,745 with exposure to uranium, mean was particle absorbed dose (ERR per Gy=4.48, CI: 1.27, 10.9). be inaccurate. No information on smoking habits.
1.63 kBqh/m3 (maximum 10.36). LLR contributed only Assuming RBE=20 alpha-particles, ERR per Gy for total Therefore study provides little information regarding
1.3% of total alpha-particle lung dose. Annual lung weighted lung dose was 0.22 (CI: 0.06, 0.53). LLR ERR uranium risk
dose due to LLR significantly correlated with doses was 5.0 (CI: 1.2, 12.3) per 10 mGy
from low-LET radiation (r=0.49), radon gas (r=0.53), and
radon decay products (r=0.50)

Zablotska et al. Mortality (19501999) and cancer incidence (19691999) No significant elevations in various cancer SMRs. No Study of uranium workers was negative, other than a
[Z1] of Port Hope, Canada radium and uranium process excess cancer incidence seen for a number of cancer weak association of RDP exposure and lung cancer
workers. 2,472 (87% males) worked only with uranium. types or all cancer. Dose-response analyses reported incidence (but not mortality). Had no LLR exposure
Gamma was predominant radiation exposure, so for RDP and external gamma. In uranium workers, lung estimates, so analyses of uranium effects could not be
analyses were of gamma and radon decay products cancer RDP risk estimate non-significantly elevated. presented. Substantial uncertainties: limited or no
(RDP), not of LLR. Urinalysis for uranium begun in early Other malignancies and circulatory diseases: no exposure information for early workers, lack of
1960s; alpha counting of urine samples for workers significant dose related elevations in risk for either RDP smoking information. Study uninformative regarding
exposed to enriched uranium conducted on a limited or external exposure uranium exposure risk
basis, so not used in dose assessment
Table A8. Studies of workers monitored for potential exposure to uranium for whom uranium-specific doses have been used in analyses so that uranium risks can be
explicitly examined

Abbreviations: LLR, long lived radionuclides; n.e., not estimable

Study references Summary of study Summary of findings relating to uranium Relevance for this report
Carpenter et al. Case-control study of brain/central nervous system (CNS) CNS cancer: 63% of brain/CNS tumours were malignant Only semi-quantitative imputation of amount of
[C7] cancer deaths in workers (19431977) at 2 nuclear glial tumours. Ever exposed to uranium, odds ratio exposure. No elevated risk was apparent, but the
facilities at Oak Ridge (ORNL and TEC/Y-12): (OR)=1.06 (95% CI: 0.5, 2.3, p=0.88) with no exposure reliability of dose categories unclear. Limited
enrichment of 235U and conversion to UF4 (TEC, 1943 lag, or 0.94 with a 10 years lag. Lagged levels of graded information regarding uranium risk for brain/CNS
1947); fabrication and testing of components for uranium exposure (grades 1, 2 and 3, with 0=no tumours
nuclear weapons (Y-12); nuclear energy technology exposure as referent) had non-significant ORs of 0.88,
R&D (ORNL). 72 male and 17 female brain/CNS cancer 1.01 and 0.70, respectively. Analysis by duration of
deaths (19431979). 4 matched controls per case. Work heavier uranium exposure (grades 23, 10 years lag)
locations/years rated by industrial hygienist for levels showed: OR=0.86 for 13 years; 0.79, 310 years; 0.99,
of 26 agents, including uranium compounds 1020 years; 1.63, >20 years (n=3), not significant

Chan et al. (and Mortality among 6,759 workers at Paducah Gaseous RRs for different uranium exposure quartiles compared Provides grouped quantitative information about the
supplement) Diffusion Plant, USA studied for 19522003. Workers to exposure quartile (<21 g.years) are provided in health effects of uranium exposure, though LLR linear
[C17] had potential exposure to external and internal table 15 (lung cancer); table 17 (leukaemia and non- dose-response risk coefficients were not given.

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 453

radiation, uranium, several other metals, Hodgkins lymphoma); table 18 (pancreatic cancer); Possible confounding by chemicals and smoking.
trichloroethylene and other chemicals. Urinalyses of table 20 (brain/CNS cancer) Provides semi-quantitative information regarding
uranium used to characterize the cumulative dose of uranium risk for several cancer sites
internally deposited radionuclides as g.years

Checkoway et al. Mortality (19471979) investigated among 6,781 white Lung cancer: (n=45) Analysis with 10-years dose lag for Provides some information on lung cancer among those
[C20] male workers at Y-12 uranium fabrication plant (Oak alpha irradiation, compared to 0<10 mSv group, with measured uranium exposure; showed little
Ridge, USA). 3,490 monitored for internal exposure. 049 mSv, RR=0.93 (95% CI: 0.41, 2.12) apparent risk. Small numbers of lung and other cancers
Internal dosimetry: urine analyses begun in 1950, fully limit the quantitative estimates. Smoking information
implemented by 1953, and in vivo measurements 5099, RR=0.66 (CI: 0.23, 1.90) not available. Provides semi-quantitative information
added in 1961. Internal lung doses calculated using 100, RR=1.12 (CI: 0.47, 2.65, n=11) regarding uranium risk for lung and brain/CNS cancers
metabolic models. For monitored workers, mean lung
Brain/CNS cancer: (n=14) with no lag,
dose 82.1 mSv. Mean external dose, 9.6 mSv. Other
exposure: beryllium, solvents, machine oils, mercury, 1049 mSv, RR=1.10 (0.19, 6.5)
lead. 45 lung cancer deaths in those monitored for 50 mSv, RR=0.45 (0.06, 3.2)
uranium exposure
Other cancers: no bone cancers observed. Trend
analyses for kidney cancer (n=6) or other a priori
cancers not reported for uranium monitored cohort
 454
Study references Summary of study Summary of findings relating to uranium Relevance for this report

UNSCEAR 2016 REPORT

Drubay et al. A case-control study of circulatory system disease (CSD, Circulatory disease: statistically significant association of Had a substantial set of individual measurements of LLR
[D24] n=442) mortality, particularly ischaemic heart disease radon exposure with both CSD (hazard ratio exposure. After adjusting for radon and external
(IHD, n=167) and cerebrovascular disease (CeVD, (HR)=1.11/100 WLM) and CeVD (HR=1.25/100 WLM) risk. radiation, detected no significant LLR risk for CSD, IHD
n=105), and 237 matched controls, nested among Records contained information on a number of medical or CeVD. Confounding by medical CSD risk factors
5,086 French CEA-COGEMA uranium miners who were CSD risk factors for a subset of cases and controls. After proved to be small. Uncertainties: exposure prior to
first employed after 1955, followed up through 2007, adjusting for radon and external gamma exposure and 1983 estimated from ambient measurements or with
mean 35.4 years. Individual exposure estimated from for empirically the main medical risk factors, found LLR no measurements may have had appreciable
ambient monitoring, 19591982 and dose HRs per kBqh/m3 of: measurement error. Relevant for uranium risk
reconstruction for 19561958. Since 1983, individual CSD: 1.13 (95% CI: 0.97, 1.31, n=76) assessment of CSD
LLR exposure estimated with film dosimeters. Mean
cumulative LLR was 1.2 kBqh/m3 (max=7.6) IHD: 0.94 (CI: 0.73, 1.20, n=26)
CeVD: 1.17 (CI: 0.90, 1.53, n=16)
LLR risks in entire nested case-control sample, but without
being able to adjust for medical risk factors, nearly identical

Drubay et al. Kidney cancer mortality in 3,377 French uranium miners Kidney cancer: SMRs were 1.49 (95% CI: 0.73, 2.67, n=11) Lack of association with estimated LLR exposure
[D23] 19562007, and 58,986 German uranium miners 1946 for French cohort and 0.91 (CI: 0.77, 1.06, n=174) for suggests kidney cancer effect is likely small.
2007. For respective French and German cohorts, German cohort. A 10-years lagged LLR dose-response Uncertainties: no information about smoking; French
median durations of follow-up were 30.0 and 34.8 analysis showed hazard ratio (HR) per kBqh/m3 with cohort was small; workers also had radon and external
years; respective median kidney doses were 26.7 mSv 10-years lagged cumulative dose of 0.89 (CI: 0.55, 1.42) gamma exposure
(range 0, 498) and 34.4 mSv (range 0, 2 905) for the French and 1.009 (CI: 0.991, 1.027) for the German
cohort

Dupree et al. Nested case-control study of lung cancer mortality (787 Lung cancer: analysis of internal radiation (primarily from Analysis had large number of lung cancer deaths, a
[D35] cases with 787 matched controls) in cohorts at 4 USA ambient uranium), using 10-years lagged dose with fraction had urinalyses and/or whole body counting in
uranium processing facilites: TEC (operated 19431947 <0.5 mGy as baseline, showed no increased risk for addition to ambient monitoring data. Suggests that
only) and Y-12 (19471982), Mallinckrodt (MCW, 1942 workers exposed below 250 mGy. Lung cancer odds effect of uranium exposure must be small. Estimate of
1966) and Fernald (FMPC, 19471982). Maximum ratios (ORs) for 0.5, 2.5, 5, 25, 50, >250 mGy were 1.03, ERR per Gy not reported, but the grouped dose data do
follow-up period, 19431983. Primary radiation hazard 0.57, 0.85, 0.82, 0.64 and 2.05 (95% CI: 0.20, 21), not suggest a statistically significant elevated lung
was from airborne dust of mainly insoluble natural respectively for LLR. However, for the 166 case-control cancer risk. Limitations: uncertain dose estimates for
uranium compounds. For FMPC, MCW and TEC, pairs with smoking data, no elevation in risk (odds ratio early workers, concomitant exposure to radon and
ambient uranium monitoring to estimate internal of 0.36 in the highest internal dose group) external radiation, and limited smoking information.
radiation doses. Y-12 also had whole body counting Provides semi-quantitative information regarding
and urinalysis. Conversion to doses assumed Type S uranium risk for lung cancer
uranium exposure. Smoking status available for 48% of
cases and 39% of controls
Study references Summary of study Summary of findings relating to uranium Relevance for this report
Guseva Canu et Lung cancer mortality in 2,709 workers at the French Lung cancer: (n=48) for durations of exposure >1 year to Suggestions of elevated lung cancer risk after exposure
al. [G31] AREVA NC Pierrelatte uranium reprocessing plant Type F, M and S uranium compounds found hazard to slowly soluble (Types M and S) uranium compounds.
during 19602005. The plant enriched uranium via ratios (HRs) of 1.05 (95% CI: 0.43, 2.52), 2.61 (0.87, 7.8), However, substantial uncertainties: small number of
gaseous diffusion caused uranium chemical 2.58 (0.76, 8.8), respectively. For duration of exposure lung cancer cases, only an ordinal scale of exposure
conversion. Uranium was only radioactive material as a continuous variable, HRs of 1.01 (0.96, 1.01), 1.07 intensity, limited smoking information, no
used at the plant. Semi-quantitative JEM to (1.01, 1.13) and 1.07 (1.01, 1.14) per year exposure, individualized information on chemical exposure.
characterize uranium exposure (duration and intensity respectively Provides semi-quantitative information on uranium risk
specific for each job and calendar year, on a 4 point for lung cancer, with analyses by uranium solubility,
ordinal scale). Smoking data on 6% of cohort but uncertainty due to small numbers

Guseva Canu et Quantified uranium exposure using a job exposure Lung cancer: (n=53) For natural uranium exposure, no This exploration of chemical types and radioactivity level
al. [G32] matrix for 2,897 workers at French AREVA NC significant associations for Type F, M or S. For of various forms of uranium suggests that natural
Pierrelatte uranium reprocessing plant. Classified reprocessed uranium, significant hazard ratios (HRs) in uranium has little or no lung carcinogenic effect.
exposure by natural vs. reprocessed uranium, and by the highest cumulative exposure group of 4.35 (95% CI: However, the less soluble forms of reprocessed
solubility Types F, M and S. To model cumulative 1.25, 15) for Type M and 10.5 (CI: 2.3, 48) for Type S. uranium dust, with their greater radioactivity and
exposure for individuals, estimated duration and Dose-response analysis for reprocessed uranium relatively long residence time, may induce lung cancer
intensity of exposure to Types F, M and S exposure duration gave HRs of 1.13 (1.03, 1.25) and Limitations: study size was small. Only limited smoking
1.13 (1.01, 1.25) for Types M and S, respectively. data available. Results require confirmation in larger
Analysis in subgroup of 345 workers with smoking

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 455

independent study
information suggested no confounding by smoking
Lympho-haematopoietic malignancies: (n=21) Found
association with insoluble reprocessed uranium, but on
the basis of only 3 exposed cases

Guseva Canu et Mortality from ischaemic heart disease (IHD, n=48), Circulatory disease: CSD mortality was increased among Job exposure matrix was carefully done, but provides only
al. [G33] cerebrovascular disease (CeVD, n=31) and total workers exposed to Type S RPU (HR=2.13, 95% CI: 0.96, an approximate quantitation of uranium exposure.
circulatory system diseases (CSD, n=111) after chronic 4.70) and Type S natural uranium (HR=1.73, CI: 1.11, Concerns with study are: small study size, limited
exposure to uranium among 2,897 workers at the 2.69). Additional information on risk by duration and smoking data. Possible confounding: heat and
French AREVA NC Pierrelatte uranium processing plant intensity of exposure for CSD, IHD and CeVD is given in trichloroethylene exposure were correlated with
(19602006). Cumulative exposure to various uranium the text of table 20 uranium exposure. The analyses of Types M and S RPU
compounds was classified by isotopic composition and For the subset of workers with available smoking data exposure did not adjust for the common exposure to
solubility type and quantified for individual job they found nominally higher CSD HRs for RPU Type M Type F and natural uranium. Provides semi-quantitative
histories via a job-exposure matrix (natural vs. and S exposure among smokers than non-smokers, but information regarding uranium risk for CSD, including
reprocessed uranium (RPU), and absorption Types F, M numbers of cases were small analyses by uranium solubility and isotopic composition,
and S) but with uncertainty due to small numbers
 456
Study references Summary of study Summary of findings relating to uranium Relevance for this report

UNSCEAR 2016 REPORT

Kreuzer et al. Circulatory system disease (CSD) mortality (19461998), Circulatory disease: analyses for cumulative LLR The largest systematically defined cohort of uranium
[K14] German Wismut male uranium miner cohort in relation exposure lagged by 5 years. The LLR risk estimates (ERR workers available. Uncertainties: no confirmation of
to external radiation, radon and LLR exposure. per 100 kBqh/m3) were JEM by urinalyses, correlation of LLR exposure with
Exposure estimated via a job exposure matrix JEM for CSD: 0.2 (95% CI: 0.5, 0.06) external radiation and radon exposure not accounted
each radiation type,19461989. 5,417 deaths from CSD for, lack of dosimetric models to estimate LLR exposure
(19461998), including 3,719 from heart disease and Heart disease: 0.3 (CI: 0.6, 0.02) of the heart and major arteries
1,297 from cerebrovascular disease. Mean LLR Cerebrovascular disease: 0.05 (CI: 0.5, 0.6)
exposure, 3.5 kBqh/m3 (maximum 132)
In no case did the highest dose category show significantly
elevated risk. Ischaemic heart disease (n=2,690) also did
not show a statistically significant elevation

Kreuzer et al. Mortality from cancer of the extra-thoracic airways Cancer of extra-thoracic airways (n=234): non-significant Estimate of risk for uranium exposure had potential
[K17] among 58,690 male German Wismut uranium miners increase with radon exposure: ERR/100WLM=0.036, confounding by external radiation levels, arsenic and
(19462008) in relation to radon and cumulative LLR 95% CI: 0.009, 0.08 silica dust exposure and smoking habit. Quantitative
exposure. LLR exposure estimates derived by a job No increase with LLR exposure: ERR per 100 kBqh/m3= risk estimate is relevant for uranium risk assessment:
exposure matrix based on ambient measurements as 0.17, 95% CI: 2.50, 2.16 (adjusted for radon suggests little risk for extra-thoracic airways
described in Kreuzer et al. [K14] exposure)

Kreuzer et al. Mortality in 4,054 male German uranium millers (1946 All cancer: LRR ERR per 100 kBqh/m3 = 0.43 (95% CI: Well conducted study, suggesting little/no association of
[K18] 2008) who had never worked as uranium miners, so 1.31, 0.44, n=457), adjusted for radon exposure uranium exposure with various health outcomes. Study
radon exposure was low, mean 8 WLM. Estimated Lung cancer: LRR ERR=0.61 (CI: 1.42, 1.9, n=159), (not size was small and no smoking information available.
exposure to radon, external gamma radiation, LLR and adjusted for radon) Quantitative risk estimates are relevant for uranium risk
silica. Exposure estimates derived via a job exposure assessment
matrix of intensity (from ambient monitoring) by Additional LLR risk coefficients given in text of table 16
location, job, calendar year. Mean LLR: 3.9 kBqh/m3. (lympho-haematopoietic), table 17 (colon and rectal),
Preliminary organ dose calculations for alpha-emitting table 18 (kidney and prostate), table 20 (circulatory)
LLR averaged 3 mGy for lung, and 1 mGy for liver and
red bone marrow

Mohner et al. Nested case-control study of leukaemia mortality (1953 Leukaemia: non-chronic lymphocytic leukaemia (non- Provides evidence that LLR exposure has little
[M50] 1989) among ~360,000 German Wismut male uranium CLL) risk not associated with RDP, but showed association with leukaemia risk. Limitations: prior to
miners: 377 leukaemia deaths and 980 controls suggestive association with LLR exposure. ERR per 1955 little data on exposure levels so dose
matched on age. Job exposure matrix JEM by location, 100 kBqh/m3 for LLR was 1.04 (90% CI: 0.64, 2.73, uncertainties. Mortality may have been
job and year used to estimate red bone marrow (RBM) n=377) for all leukaemia, 0.76 (CI: 1.26, 2.78, n=218) underascertained because inadequate identifiers in
exposure. JEM for radon and decay products (RDP), for non-CLL and 1.35 (CI: 1.54, 4.24, n=159) for CLL. early years limited mortality linkage. Underlying cohort
external radiation and LLR exposure estimates for >500 Suggestion that the highest/longest LRR doses may and numbers in it rather loosely defined, though it is
different workplaces, 750 job titles, 44 calendar years. increase risk: for 20 kBqh/m3, (OR=1.26, 90% CI: 0.71, the largest uranium worker cohort. Quantitative risk
Mean cumulative RBM dose was 23.6 mGy; only 2% 2.22) for non-CLL. For acute myelogenous leukaemia, estimates are relevant for uranium risk assessment
from inhalation of LLR the LRR ERR per 100 kBqh/m3 was 0.83 (CI: 1.9, 3.6)
Study references Summary of study Summary of findings relating to uranium Relevance for this report
Mohner et al. Nested case-control study of laryngeal cancer among Laryngeal cancer: elevated risk in highest cumulative LLR The same limitations for this study as for Mohner et al.
[M51] ~360,000 German Wismut male uranium miners ever exposure category (10 kBqh/m3), OR=1.63 (95% CI: [M50]. Follow-up successful for only 72.8% of potential
employed, 19501989. Tumour registry follow-up, 1.03, 2.59, n=56), adjusted for smoking and alcohol controls. Quantitative risk estimate for laryngeal cancer
19611989. Two matched controls per case. Crude intake. For continuous LLR cumulative exposure: is relevant for uranium risk assessment
information on smoking habits available for many ERR=0.098 (CI: 0.11, 0.31) per 10 kBqh/m3, unadjusted,
workers and anecdotal information on alcohol or ERR=0.156 (CI: 0.11, 0.41) adjusted for smoking and
consumption from medical records. Included 554 alcohol intake
laryngeal cancer cases and 929 controls

Rage et al. [R5] 5,086 uranium miners employed by CEA-COGEMA in Internal LLR dose-response analyses were conducted, LLR exposure related to various mortality end points.
France; followed up 19462007 (mean 32.8 years). doses lagged 5 years. LRR results expressed as Limitations: LLR correlated with and a small percentage
Cohort included 3,377 miners first employed after ERR/kBqh/m3: of total radiation exposure; smoking information
1955, for whom radon, LLR and external -ray exposure All cancer, 0.022 (95% CI: 0.049, 0.12, n=315); unavailable; lung cancer analyses did not adjust for
was recorded. Assessment of LLR exposure based on Lung, 0.32 (0.09, 0.73, n=94); silica exposure. Quantitative risk estimates are relevant
ambient measurements 19591982 and individual for uranium risk assessment
measurements thereafter. Doses retrospectively All cancers except lung, 0.065 (n.e. 0.019, n=221);
reconstructed for the period 19561958 [R4]. Post-1955 Other LLR risk coefficients given in text of table 15
workers had mean of 1.64 kBqh/m3 of LLR (range 0.01 (respiratory), 19 (kidney), 20 (brain/CNS) and 21
10.4) (circulatory). In summary, only lung cancer showed a

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 457

significant positive association with LLR exposure

Richardson and Nested case-control study of lung cancer among 3,864 Nested case-control analyses were conducted, with Provides evidence that uranium risk for lung cancer is
Wing [R12] Y-12 (Oak Ridge, USA) workers hired 19471974. Y-12 matched controls. Exposure lagged by 5 years likely small. Strength: measured exposure (both urine
was a nuclear material fabrication plant. Internal Lung cancer: LLR dose-response negative (ERR per assays and in vivo monitoring) and the fact that radon
exposure primarily LLR from ambient uranium dust. 100 mSv=0.077, 90% CI: 0.23, 0.07) exposure did not overshadow the LLR exposure of the
Individual monitoring for external radiation exposure lung. LLR exposure had substantial uncertainties, in
began in 1948 and became plant-wide in 1961. Smoking-related diseases other than lung cancer: LRR part because 58% of exposure person-years had
Urinalysis monitoring increased in coverage through (ERR per 100 mSv=(0.089) was negative, as was non- imputed rather than measured doses. Limitations: dose
the 1950s and in vivo monitoring begun in 1961. Mean malignant respiratory disease (0.085), but with wide uncertainties, inadequate information on smoking and
external lung dose (10.1 mSv) was fourfold lower than confidence intervals workplace chemicals, statistical power limited.
mean cumulative internal dose (44.7 mSv). Other Quantitative risk estimate for lung cancer is relevant for
exposure: beryllium, solvents, machine oils, mercury, uranium risk assessment
lead
 458
Study references Summary of study Summary of findings relating to uranium Relevance for this report

UNSCEAR 2016 REPORT

Silver et al. [S19] Cohort of 6,409 uranium workers at Fernald (USA) Analyses took into account pay code, birth year, This study has longer follow-up and better exposure
(Prior reports: employed (19511985), and followed through 2004 trichloroethylene exposure, radon and external assessment than previous ones. Uranium internal
[D33, R14, (mean follow-up 37 years). Used urine uranium radiation. Overall: hourly males showed excess lung doses estimated for several different organs and linear
R15]) concentration data (>250,000 urine samples) from cancer (SMR=1.25, 95% CI: 1.09, 1.42, n=297). LRR ERRs ERR estimates adjusted for other radiation exposure.
1952 forward to estimate exposure to internally calculated for Caucasian males per organ-specific Sole positive finding related to intestinal cancer, which
deposited uranium compounds. Mean cumulative 100 Gy: is not a very high a priori suspect, so requires
doses to the lung for hourly and salaried workers were Intestinal cancer (small intestine and colon, not confirmation. Limitations: no smoking data, limited
1,552 Gy and 388 Gy for LLR, respectively. Mean LLR rectum): had a significant elevation in the highest dose data on exposure to chemicals and other hazardous
cumulative organ dose ranged from 1.1 mGy (lung) to group (>36 Gy, ERR=1.7, CI: 0.17, 5.7) and a significant substances, limited statistical power. Quantitative risk
6.7 mGy (pancreas) dose response (ERR 100 per Gy=1.5, CI: 0.12, 4.1, estimates are relevant for uranium risk assessment
n=48). Other dose-response estimates at 100 Gy for
internal doses were null. Additional LLR risk coefficients
given in text of table 15 (lung, respiratory), table 16
(leukaemia, lymphoma), table 17 (stomach, pancreas),
table 18 (kidney)

Tomasek and 9,973 Czech uranium miners studied for leukaemia and Leukaemia: using 2 years dose lag, for 119 years since Dose uncertainties were probably large, especially for
Malatova [T13] non-Hodgkins lymphoma risk. Two cohorts: S, 4,348 1st exposure leukaemia SMR=1.0 (95% CI: 0.4, 2.1, n=7). earlier years (that contributed the highest exposure)
exposed 19481963; N, 5,625 exposed 19681986. For >19 years since 1st exposure SMR=1.8 (CI: 1.2, 2.7, and the number of malignancies was small. Analyses of
Though had limited exposure measurements, derived n=23). For total follow-up period, SMR=1.5 (CI: 1.1, 2.2, LLR exposure were by average SMRs and not dose
location-job-year estimates of dose rates for hewers n=30, mean RBM LLR dose 145 mSv). Due to small responses. Study suggestive of leukaemia risk from
and then proportionately scaled for other jobs to numbers, did not separate out non-CLL leukaemias. uranium exposure, but study limitations weaken the
develop a job exposure matrix JEM. Estimated 5264% Leukaemia risk slope for total RBM dose (external, conclusions
of the red bone marrow (RBM) dose was from LLR. radon progeny and LLR), ERR per Sv=2.5 (90% CI: 0.3,
Mean LLR RBM dose of 160 mSv for cohort S and 9.3)
37 mSv for cohort N Non-Hodgkins lymphoma (NHL): dose response not
significant (p=0.16), though a nominal overall excess
(SMR=1.5, CI: 0.9, 2.2)

Vacquier et al. French cohort of 3,377 uranium miners first employed LLR exposure was correlated r=0.52 with cumulative LLR dose-responses could not be calculated with
[V2] 19551990 when exposure to external radiation, radon radon exposure, 0.47 with external exposure. adjustment for radon or external exposure. Substantial
and uranium dust (LLR) could be estimated. 3,240 had Combined cumulative radiation exposure showed correlation of LLR with radon and gamma exposure, so
internal exposure. Follow-up through 1999, a mean of significant dose response only for lung cancer. Linear LLR risk estimates not well defined but suggestive
26.5 years; mean LLR exposure, 1,632 kBqh/m3. LLR ERR risk coefficients per (kBqh.m-3) for LLR exposure, small association with lung cancer. Limitations:
exposure estimates: reconstructed before 1959; lagged 5 years: uncertainties in exposure assessments, correlated
ambient measurements 19591982; since 1983 All cancer: 0.001 (95% CI: 0.08, 0.11); exposure, lack of smoking data. Quantitative risk
personal film dosimeters. Since external radiation, LLR estimates are relevant for uranium risk assessment
and radon exposure instances were correlated, Lung cancer: 0.25 (CI: 0.02, 0.70);
determined which were associated with cancer risks Brain/CNS cancer: 0.17 (CI: n.e., 2.0)
Study references Summary of study Summary of findings relating to uranium Relevance for this report
Yiin et al. [Y3] Nested case-control study of multiple myeloma among Multiple myeloma: 98 multiple myeloma death cases High quality study. Included detailed uranium exposure
47,941 workers at the K-25 Oak Ridge, USA gaseous and 490 controls. Analyses adjusted for birth cohort, assessment and analyses adjusted for external and
diffusion uranium enrichment plant (operated 1945 external and medical irradiation, mercury, nickel and medical radiation doses and prevalent chemicals. Weak
1985). Five matched controls per case. Exposure to trichloroethylene exposure. For those with the best association found between uranium bone marrow
soluble and insoluble uranium compounds. Individual estimated uranium doses to the bone marrow dose and multiple myeloma risk; requires confirmation
uranium dose estimates based on ambient (Group I), the odds ratio (OR) at 10 Gy was 1.04 (95% by other studies. Limitations: less measurement data
measurements and urinalysis. Formed groups CI: 1.00, 1.09). For total case-control group (Group III), were available for workers in the earlier days.
according to the strength of the dosimetry: grouping I, OR was identical: 1.04 (CI: 1.00, 1.09). Indicated a weak Quantitative multiple myeloma risk estimate is relevant
multiple urinanalyses and extensive ambient association of bone marrow dose from uranium with for uranium risk assessment
measurements; groupings II also included those with multiple myeloma risk
fewer measurements; grouping III, all cases and
controls. External radiation doses and medical
radiation exposure also compiled

Zhivin et al. Studied 4,688 French gaseous diffusion uranium From job exposure matrices, grouped workers into no, Study developed quantitative estimates of uranium
[Z10] enrichment workers (AREVA NC, CEA and Eurodif) with low, medium, high exposure for analysis. Analysed exposure, but analysed only low, medium and high
exposure to mainly soluble uranium compounds external radiation, and natural, enriched and DU. grouped uranium exposure. Valuable because it
(Type F). Used plant-specific job exposure matrices Analysed all cancer, lung cancer, lympho- considered mainly highly soluble uranium and
JEMs to estimate cumulative exposure. The AREVA NC haematopoietic malignancies, and circulatory diseases. compared natural, enriched and depleted isotopic

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 459

job exposure matrices showed 64% sensitivity and 80% Results by exposure group presented in text of table 15 forms. Provides semi-quantitative information
specificity in validation against bioassay data. Median (lung), table 16 (lympho-haematopoietic), table 20 regarding uranium risk for CSD, including analyses of
follow-up, 30.2 years, 19682008. Had estimates of (circulatory) isotopic composition
potential confounding exposure situations:
trichloroethylene, heat, noise
 460
Table A9. Studies of groups with potential military uranium exposure

UNSCEAR 2016 REPORT

Study references Summary of study Summary of findings related to uranium Relevance for this report
Bogers et al. Following lay-press reports of alleged excess leukaemia All cancer: total cancer incidence 17% lower among No information about exposure to DU available, so study
[B28] among Dutch Balkan veterans, study examined cancer Balkan than non-Balkan personnel (hazard ratio (HR) non-informative about uranium effects
incidence, comparing 18,175 Balkan-deployed military 0.83, 95% CI: 0.69, 1.00)
male personnel with 135,355 non-Balkan deployed Miscellaneous cancers: rates of digestive, respiratory,
military males and with general population rates. urogenital and haematological cancers non-
Maximum follow-up was for nearly 15 years. Some significantly lower in the Balkan vs. non-Balkan group
differences between the Balkan and non-Balkan
cohorts military status, e.g. conscripted soldiers (15% Leukaemia: HR could not be calculated for leukaemia
vs. 81%, respectively) because of the small number of cases (Balkan n=5)

Hines et al. [H18] 37 US Gulf War I veterans who had inhalation exposure Non-malignant respiratory disease: no differences Study strengths: had clinical examinations, spirometric
to (and sometimes retained fragments of) DU from between low and high exposure groups for any of a list testing, symptom reporting, and smoking information.
friendly fire incidents were examined. Compared those of pulmonary symptoms or for history of steroid DU findings do not indicate any non-malignant
with high vs. low body burdens of DU, as measured by prescriptions. No significant differences regarding DU pulmonary effects. Provided limited information
urine assay. DU remaining in the body is 40% less exposure levels in pulmonary function parameters or regarding uranium effects due to small sample size and
radioactive but chemically similar to natural uranium. chest CT findings likelihood that exposure levels were low
Low and high exposure groups similar in age, race, BMI
and smoking

Labar et al. [L3] Ecological study to examine childhood haematological Childhood haematological malignancies: in the 10 No evidence for an effect of DU on childhood
cancers in Croatian counties with DU (10 counties), counties with DU exposure, no significant increases haematological malignancies. Exposure very low, and
chemical plant damage (2 counties) or "population during or after the War were found for lymphatic ecological data are susceptible to various unidentified
mixing" (4 counties). Compared disease rates for leukaemia, myeloid leukaemia, Hodgkins lymphoma, biases, so study provides no meaningful information
children ages 014 before (19861990), during (1991 or non-Hodgkins lymphoma on uranium risk
1995) or after (19961999) the Croatian war in those
counties

Macfarlane et al. A 13 year follow-up was conducted of 51,753 UK 7% of those with questionnaires responded they had DU exposure was based on unverified self-reports and
[M2] veterans deployed in the Gulf War and 50,808 other received DU exposure, among whom there were 9 the risk estimate was for overall disease deaths, not a
veterans matched for age-group, sex, rank, service and disease-related deaths. The DU exposed vs. unexposed priori causes. The small number of deaths (n=9) among
level of fitness, who were not deployed to the Gulf. yielded a RR=1.00 (95% CI: 0.99, 4.04) after adjustment those reportedly exposed to DU is a very weak finding
57% responded to a questionnaire about deployment for age, sex, smoking and alcohol intake
experiences and morbidity

McDiarmid et al. US Gulf War veterans with DU exposure were followed The extensive clinical examination did not show any Among the several dozen parameters measured, only
[M22] up (19912005) for clinical and laboratory end points. differences between the high and low exposed group. chromosome aberrations showed a (suggestive)
On basis of repeated urine uranium measurements, Other high/low exposure comparisons: no difference in difference. The multiple comparisons and small sample
10 were designated as high DU exposure and 24 as low urine retinol binding protein, a biomarker of renal size limited the statistical power and meaningfulness
exposure. Exposure resulted from inhalation, wound proximal tubule function. No differences found on: of the comparisons
contamination and/or embedded fragments (for other renal measures, a neurocognitive test battery,
~30%) neuroendocrine parameters, semen parameters, or
HPRT mutations. A borderline increase in chromosome
aberrations in the high exposure group
Study references Summary of study Summary of findings related to uranium Relevance for this report
McDiarmid et al. 35 US Gulf War veterans with DU exposure in 1991 were Only two parameters showed marginal differences Among the several dozen parameters measured, only
[M23] again evaluated in 2007 with numerous clinical and between the high- and low exposure groups: 2 two showed a suggestive difference. The multiple
laboratory measures microglobulin (81.7 vs. 69 g/g creatinine, respectively; comparisons and small sample size limited the
p=0.11) and retinol binding protein (48.1 vs. 31 g/g statistical power and meaningfulness of the
creatinine; p=0.07). No differences were seen in rates of comparisons
chromosome aberrations or HPRT mutations

McDiarmid et al. 37 US Gulf War veterans with DU exposure in 1991 were No differences between high- and low-exposure groups Two sensitive markers of kidney tubular injury suggested
[M25] again evaluated 20 years after exposure (2011) for for: haematology, clinical chemistries, neuroendocrine subtle renal injury, but this found only after the main
numerous clinical and laboratory end points. Report parameters, bone metabolism, neurocognitive categories of high vs. low exposure showed no
focused on acute renal toxicity and included three new function, immune function, pulmonary function or differences. Multiple comparisons and small sample
sensitive markers of kidney tubular injury nodules. Regarding renal function and injury, no high size limit the inferences that can be drawn from the
vs. low exposure differences were found for 16 clinical study
indicators of renal function, 6 urine markers for kidney
injury, or 4 urine measures of low molecular weight
proteins, although a re-analysis using a different
definition of high exposure showed elevations in two
kidney injury markers

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 461

Strand et al. Cancer risk and all cause mortality studied among 6,076 69 cancer cases observed (SIR=1.04, 95% CI: 0.81, 1.33). With no data on DU exposure, study is uninformative
[S42] Norwegian military UN peacekeepers in Kosovo, 1999 Suggestion of elevation in melanoma (SIR=1.90, CI: regarding uranium risk
2011. No information available on DU exposure. Mean 0.95, 3.4, n=11). No elevation in stomach, liver, lung,
follow-up, 10.2 years; 4.4% women prostate, kidney, bladder, brain cancers or lympho-
haematopoietic malignancies
 462
Table A10. Studies of general population groups with potential environmental exposure to uranium by inhalation

UNSCEAR 2016 REPORT

Study references Summary of study Summary of findings related to uranium Relevance for this report
Boice et al. [B31] Study evaluated cancer incidence (19931997) among All cancer: found 581 incident cancers when 574 were Earlier investigation had obtained soil measurements of
residents near Apollo (began operations 1957) and expected (SIR=1.01, 95% CI: 0.93, 1.10) uranium, plutonium and other isotopes and air
Parks uranium-plutonium processing plants in A priori tumour sites: for tumour sites with potentially measurements of gamma radiation; all levels well
Pennsylvania, USA. Study population included about greater exposure they found: lung (SIR=0.88), kidney below Nuclear Regulatory Commission release
17,000 individuals in 8 nearby municipalities (SIR=1.05), non-Hodgkins lymphoma (SIR=1.10), liver guidances. Those measurements too sparse to be used
(SIR=0.61) and bone (2 observed, 1.19 expected), none to directly assess uranium effects associated with
of which were statistically significant elevations. Also exposure of individuals. Limited environmental
thyroid and female breast cancer rates were not measurements and negligibly low exposure levels
elevated, nor was leukaemia mean study is not informative as to uranium effects

Boice et al. [B30] People in two counties proximal to the Apollo and Parks, All cancer: during 19501995, 39,287 cancer deaths Strengths: the mortality ascertainment was high, sample
Pennsylvania, USA former uranium/plutonium material occurred in the proximal counties and 77,382 in the size was large. Most likely uranium-related cancer types
processing plants were concerned regarding possible control counties. Compared to control counties, RRs in examined
elevated rates of cancer, especially childhood proximal counties for all cancer deaths before (1950 Weaknesses: no individual or even county-level
leukaemia. Study compared cancer mortality rates in 1965), during (19651980) and after (19801995) were estimates of uranium exposure levels. Proximal county
those two counties (population ~443,000) with six virtually identical0.95, 0.95 and 0.98, respectively areas were rather broad, further diluting possible
other counties (population ~864,000) matched on age, indicating no effect of potential uranium/plutonium exposure, though nearly all inhabitants lived within
race, urbanization and socioeconomic factors. exposure 20 miles of a processing plant
Comparisons made before, during and after operations A priori tumour sites: for childhood leukaemia (total n=119
of the uranium-plutonium plants Conclusion: because of low exposure levels and
proximal county cases and n=272 control cases) before, ecological nature of the study, does not adequately
during, after RRs=1.02, 0.81, 0.57, respectively. Lung address the health risks of uranium
cancer (RR=0.85, 0.99, 0.95), bone (RR=0.96, 1.00, 1.01),
liver (RR=0.98, 1.07, 1.01) and kidney (RR=1.00, 1.08, 1.02)
not significantly elevated in the proximal counties

Boice et al. [B32] Cancer mortality rates investigated in Karnes County, All cancer: Karnes County RRs of 1.0 in 19501964 Strengths: mortality ascertainment was high. A priori
Texas, USA, a county with uranium mining and milling (before/beginning of mining-milling), 0.9 in 19651979 cancer types specifically examined
activities from 1959 to early 1990s, with 3 mills and >40 (early operations), 1.1 in 1980-1989 (later operations Weaknesses: limited uranium exposure measurements
mines. No uranium enrichment activities. Karnes cancer and latency period) and 1.0 in 19902001 (few/no available to use in analysis. Cancer mortality
mortality rates before, during, after that period (1950 operations) misdiagnoses, especially for liver cancer
2001) compared with four match control counties. A priori tumour sites: for prime exposure periods (1965
1,223 cancer deaths observed in Karnes County (1,392 Relevance: because of low, unknown exposure levels
1979 and 19801989), Karnes county RRs were 1.0, 1.2, and ecological nature of the study, does not
expected) and 3,857 in control counties. Texas respectively, for lung cancer, 0.8, 0.9 for kidney cancer,
Department of Health monitored Karnes radiation adequately address the health risks of uranium
1.0, 0.8 for liver cancer, and 1.3 (n=20) and 1.7 (n=17)
levels, found no elevations in radioactive material for leukaemia. no RRs significantly elevated. Childhood
in/near homes cancer mortality 1965-2001, non-significant RR of 1.3
(n=8 cases)
Study references Summary of study Summary of findings related to uranium Relevance for this report
Boice et al. [B34] Mortality evaluated during 19782004 for 19361984 All cancer: no significant elevation in overall cancer Study uninformative regarding uranium effects because
residents of Uravan, Colorado USA, a uranium mill mortality or cancers of lung, kidney, breast; leukaemia; no uranium exposure data and has low statistical
town. Mining and milling activities during mid-1930s to non-malignant respiratory, renal or liver disease power
1984. The mean follow-up time since first Uravan among females or the 622 uranium mill workers, but
residence, 38.1 years excess lung cancer found among underground
uranium miners. Had no quantitative information on
exposure levels of mill workers

Boice et al. [B35] Comparison of 19502000 mortality in Montrose County, All cancer: no difference in total cancer Since no information on who was exposed to uranium
Colorado, USA (Uravan and other mining/milling A priori tumour sites: montrose elevation of lung cancer in and exposure levels, study uninformative regarding
operations) with five comparison counties males (RR=1.19, 95% CI: 1.06, 1.33), thought to be due to uranium risk
underground miner radon exposure and heavy smoking.
No excess of breast, kidney, bone, liver or childhood
cancer, leukaemia, non-Hodgkins lymphoma, renal
disease or non-malignant respiratory disease

Boice et al. [B37] Cancer incidence (19822004) and mortality (1950 Lung cancer: found increased mortality from lung cancer Lung cancer excess was likely due to miner radon
2004) in Grants, New Mexico, USA residents: Grants among men (SMR=1.57, 95% CI: 1.21, 1.99) exposure and smoking. Uranium exposed individuals
mining during early 1950s to 1990; milling operations not identified, so it is uninformative

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 463

Stomach cancer: stomach cancer mortality among
19581990 women was high (SMR=1.30, 95% CI: 1.03, 1.63), but
elevated mainly in the early years before milling
operations began

Chen et al. [C24] Ecological study of cancer incidence in Port Hope, Standardized incidence ratio (SIR) for leukaemia, 0.86 Provides some information on average air uranium
Ontario, Canada residents, 19922007. In 19811982, (95% CI: 0.60, 1.21), with no elevation of childhood levels, but analyses were ecological and not specific for
air uranium concentrations averaged 0.02 g m-3 leukaemia rate SIRs were <1.0 for a number of other uranium exposure. Study useful insofar as it rules out
leading to a committed effective dose of 0.16 mSv, but cancer sites. A significant elevation of lung cancer large uranium effects
by 19881989 were reduced to 0.00105 g m-3. Larger incidence, perhaps related to smoking habits
doses received from gamma and radon exposure.
Population ~16,500

Report of the Ecological study of cancer mortality in municipalities With reconstructed external doses, reported increasing Report is uninformative regarding uranium exposure
Consejo de near seven nuclear power plants and five fuel cycle dose-response trends for kidney cancer around nuclear effects, as it is an ecological study and only estimated
Seguridad facilities (chemical conversion of uranium concentrate) power plants and for lung and bone cancers around external radiation exposure
Nuclear [C37] in Spain. Cancer mortality (19752003) of fuel cycle facilities, but had not estimates of uranium
municipalities within 30 km of facilities compared to exposure
similar municipalities 50100 km distant

Lopez-Abente et Examined solid cancer mortality (19751993) in 283 They concluded that lung cancer and kidney cancer Various inconsistencies in the results depending on how
al. [L46] towns in Spain within 30 km of one of four nuclear mortality rates were higher in the 30 km area, but other the analyses were performed. They chose analyses that
power plants or four nuclear fuel facilities, compared to types of cancer were not showed positive effects. No information presented on
275 towns 50100 km away, matched on various uranium exposure levels, so study is uninformative
sociodemographic variables regarding uranium risk
 464
Study references Summary of study Summary of findings related to uranium Relevance for this report

UNSCEAR 2016 REPORT

Lopez-Abente et Examined lympho-haematopoieticmalignancies (LHMs) No excess LHM found in towns near nuclear power Selecting a few significant results from a large number
al. [L45] in 489 towns within 30 km of Spains seven nuclear plants. Reported excess leukaemia mortality near two of statistical tests is questionable. The results for all
power plants and five nuclear fuel facilities (exposed), nuclear fuel facilities and excess myeloma mortality nuclear fuel facilities combined, or all nuclear power
compared to 477 towns 50100 km away near one nuclear fuel facility. No exposed town showed plants, do not indicate elevated risks for lympho-
(unexposed). Exposed towns reported 610 excess leukaemia for the under age 25 group. Analyses haematological malignancies
leukaemias, 198 lymphomas and 122 multiple of all nuclear fuel facilities combined did not yield
myeloma deaths during 19751993 statistically significant excess for any end point
(leukaemia, leukaemia <25 years, myeloma, Hodgkins
lymphoma, non-Hodgkins lymphoma)

Pinney et al. Examined prevalence ratios of diseases among 8,496 Reported a number of elevated prevalences of kidney Distance from the plant and/or possible exposure to
[P22] residing within 2 miles (3.2 km) from Fernald, USA and bladder diseases/conditions compared to NHIS plant runoff used as surrogates for uranium doses.
uranium plant, or within 5 miles (8 km) and in data, but screening questions or coding sometimes Actual measured exposure levels very low. Since
groundwater runoff direction, or with well/cistern. differed between the two datasets. Found no perceived residential risks from Fernald were current in
Medical conditions obtained by questionnaire and differences for diabetes, thyroid diseases or respiratory the population, results based on self reports may have
screening examination. Prevalences were compared to diseases. Several clinical laboratory variables showed been biased. Inconsistencies among comparisons of
NHIS/NHANES data (national standardized surveys) small but significant differences between those within laboratory findings create uncertainty in the
2 miles or more distant, and a different set of variables interpretation
were significant for those using wells/cisterns

Table A11. Summary of literature review on health effects of human exposure to uranium through ingestion of surface or groundwater

Study Study design Country Effect Effect measurement Relevance for this report
Mao et al. [M9] Cross-sectional Canada Chemical toxicity of Comparison of biomarker levels in urine (microalbuminuria) Positive association between uranium cumulative
urinary system and serum (creatinine) exposure index and albumin level

Zamora et al. Cross-sectional Canada Chemical toxicity of Comparison of biomarker (glucose, creatinine, protein, beta2- Alkaline phosphatase, beta2-microglobulin levels
[Z6] urinary system microglobulin, alkaline phosphatase, -glutamyl transferase, correlated with uranium level in water
lactate dehydrogenase, N-acetyl--D-glucosaminidase) levels

Kurttio et al. Cross-sectional Finland Chemical toxicity of Comparison of biomarker (calcium, phosphate, glucose, Significantly increased calcium, fractional excretion
[K26] urinary system albumin, creatinine, beta2-microglobulin) levels No association between uranium exposure and
other parameters

Kurttio et al. Cross-sectional Finland Chemical toxicity of Comparison of renal damage indicators (glucose, creatinine, No statistically significant association between
[K28] urinary system alkaline phosphatase, -glutamyl transferase, lactate uranium concentrations in urine and any of the
dehydrogenase, N-acetyl--D-glucosaminidase, calcium, renal damage indicators, except glucose excretion
phosphate, cystatin C, glutathione-S-transferase) in urine in urine and diastolic blood pressure
 Study Study design Country Effect Effect measurement Relevance for this report
Selden et al. Cross-sectional Sweden Chemical toxicity of Comparison of biomarker (albumin, beta2-microglobulin, Significant increase in urinary excretion of -2
[S8] urinary system protein HC, kappa and lambda chains, N-acetyl--D- microglobuline, kappa and lambda chains, and HC
glucosaminidase) levels in urine protein with medium to high uranium
concentrations in urine. Doseresponse
relationships observed after exclusion of subjects
with diabetes

Kurttio et al. Case-cohort Finland Urinary system as target Comparison of risk of bladder cancer by uranium level and No excess of bladder cancer with increased level of
[K29] of radio-toxicity radiation dose uranium or radiation dose

Kurttio et al. Cross-sectional Finland Bone as target of Correlation between uranium exposure and biomarkers Marginal positive association of uranium
[K27] chemical toxicity associated with bone (osteocalcin, aminoterminal concentrations in drinking water with serum type I
propeptide of type I procollagen, serum type I collagen collagen and carboxy-terminal telopeptide only in
carboxy-terminal telopeptide) men (p=0.05)
No significant association between uranium
exposure and bone turnover indicators in women

Seiler [S7] Ecological USA Lympho-haematopoietic Comparison of uranium concentration in wells used by case No significant difference between uranium
system families and other wells concentrations in wells used by families of

ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 465

leukaemia cases (median=3.4 g/L) and the
uranium concentrations in other wells (1.6 g/L)
No differences in concentrations of gross activity
or of Rn (617 vs. 563 pCi/L)

Auvinen et al. Case-cohort Finland Lympho-haematopoietic Comparison of risk of leukaemia according to uranium level No excess of leukaemia according to uranium level
[A32] system of drinking water

Witmans et al. Case-control Canada Lympho-haematopoietic Comparison of Uw and Thw exposure between cases and Cases had higher uranium concentrations in
[W23] system controls drinking water than controls (p=0.001)
No significance difference in Th w (p=0.22)

Auvinen et al. Case-cohort Finland Digestive system Stomach cancer risk according to uranium level No excess of stomach cancer by uranium level
[A32]
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 467

REFERENCES
A1 Abou-Donia, M.B., A.M. Dechkovskaia, L.B. Goldstein et al. Uranyl acetate-induced
sensorimotor deficit and increased nitric oxide generation in the central nervous system in rats.
Pharmacol Biochem Behav 72(4): 881-890 (2002).
A2 Adams, N. and N.L. Spoor. Kidney and bone retention functions in the human metabolism of
uranium. Phys Med Biol 19(4): 460-471 (1974).
A3 Al-Hadithi, T.S., J.K. Al-Diwan, A.M. Saleh et al. Birth defects in Iraq and the plausibility of
environmental exposure: A review. Confl Health 6(1): 3 (2012).
A4 Al-Hashimi, M.M. and X. Wang. Comparing the cancer in Ninawa during three periods (1980-
1990, 1991-2000, 2001-2010) using Poisson regression. J Res Med Sci 18(12): 1026-39
(2013).
A5 Al-Masri, M.S., A. Al-Hamwi, Y. Amin et al. Radionuclide transfer from feed to camel milk.
J Environ Radioact 132: 8-14 (2014).
A6 Alaani, S., M. Tafash, C. Busby et al. Uranium and other contaminants in hair from the parents
of children with congenital anomalies in Fallujah, Iraq. Confl Health 5: 15 (2011).
A7 Alaani, S., M.A. Al-Fallouji, C. Busby et al. Pilot study of congenital anomaly rates at birth in
Fallujah, Iraq, 2010. J IMA 44(1): 1-7 (2012).
A8 Albertini, R.J., P.M. Vacek, E.W. Carter et al. Mutagenicity monitoring following battlefield
exposures: Longitudinal study of HPRT mutations in Gulf War I veterans exposed to depleted
uranium. Environ Mol Mutagen 56(7): 581-593 (2015).
A9 Albina, M.L., M. Belles, M. Gomez et al. Influence of maternal stress on uranium-induced
developmental toxicity in rats. Exp Biol Med (Maywood) 228(9): 1072-1077 (2003).
A10 Albina, M.L., M. Belles, V. Linares et al. Restraint stress does not enhance the uranium-
induced developmental and behavioral effects in the offspring of uranium-exposed male rats.
Toxicology 215(1-2): 69-79 (2005).
A11 Alrefae, T. Investigation of 238U content in bottled water consumed in Kuwait and estimates
of annual effective doses. Health Phys 102(1): 85-89 (2012).
A12 Anderson, J.L., H.B. Spitz and J.H. Yiin. Characterization of internal exposure to enriched
uranium at a former gaseous diffusion plant. Health Phys 93(6): 636-644 (2007).
A13 Angenard, G., V. Muczynski, H. Coffigny et al. In vitro effects of uranium on human fetal
germ cells. Reprod Toxicol 31(4): 470-476 (2011).
A14 Anke, M., O. Seeber, R. Mller et al. Uranium transfer in the food chain from soil to plants,
animals and man. Chemie der Erde - Geochemistry 69(Suppl 2): 75-90 (2009).
A15 Ansoborlo, E., V. Chazel, M.H. Henge-Napoli et al. Determination of the physical and
chemical properties, biokinetics, and dose coefficients of uranium compounds handled during
nuclear fuel fabrication in France. Health Phys 82(3): 279-289 (2002).
A16 Ansoborlo, E., O. Prat, P. Moisy et al. Actinide speciation in relation to biological processes.
Biochimie 88(11): 1605-1618 (2006).
468 UNSCEAR 2016 REPORT

A17 Ansoborlo, E., L. Lebaron-Jacobs and O. Prat. Uranium in drinking-water: A unique case of
guideline value increases and discrepancies between chemical and radiochemical guidelines.
Environ Int 77: 1-4 (2015).
A18 Anthony, M.L., K.P. Gartland, C.R. Beddell et al. Studies of the biochemical toxicology of
uranyl nitrate in the rat. Arch Toxicol 68(1): 43-53 (1994).
A19 Arfsten, D.P., K.R. Still and G.D. Ritchie. A review of the effects of uranium and depleted
uranium exposure on reproduction and fetal development. Toxicol Ind Health 17(5-10):
180-191 (2001).
A20 Arfsten, D.P., K.R. Still, E.R. Wilfong et al. Two-generation reproductive toxicity study of
implanted depleted uranium (DU) in CD rats. J Toxicol Environ Health A 72(6): 410-427 (2009).
A21 Arnault, E., M. Doussau, A. Pesty et al. Natural uranium disturbs mouse folliculogenesis in
vivo and oocyte meiosis in vitro. Toxicology 247(2-3): 80-87 (2008).
A22 Arogunjo, A.M., V. Hollriegl, A. Giussani et al. Uranium and thorium in soils, mineral sands,
water and food samples in a tin mining area in Nigeria with elevated activity. J Environ Radioact
100(3): 232-240 (2009).
A23 Arruda-Neto, J.D., M.V. Guevara, G.P. Nogueira et al. Long-term accumulation of uranium in
bones of Wistar rats as a function of intake dosages. Radiat Prot Dosim 112(3): 385-393 (2004).
A24 Arruda-Neto, J.D., M.V. Manso Guevara, G.P. Nogueira et al. Long-term accumulation and
microdistribution of uranium in the bone and marrow of beagle dog. Int J Radiat Biol 80(8):
567-575 (2004).
A25 Arzuaga, X., S.H. Rieth, A. Bathija et al. Renal effects of exposure to natural and depleted
uranium: A review of the epidemiologic and experimental data. J Toxicol Environ Health -
Part B: Critical Reviews 13(7-8): 527-545 (2010).
A26 Arzuaga, X., M. Gehlhaus and J. Strong. Modes of action associated with uranium induced
adverse effects in bone function and development. Toxicol Lett 236(2): 123-130 (2015).
A27 ASN, DGS and IRSN. La qualit radiologique des eaux conditionnes produites en France.
Autorit de sret nuclaire, Direction gnrale de la sant, Institut de Radioprotection et de
Sret Nuclaire, France, 2013. (French).
A28 Atkinson, W., R. Bull, M. Marshall et al. The association between prostate cancer and
exposure to 65-Zn in UKAEA employees. J Radiol Prot 14(2): 109-114 (1994).
A29 Atkinson, W.D., D.V. Law, K.J. Bromley et al. Mortality of employees of the United Kingdom
Atomic Energy Authority, 1946-97. Occup Environ Med 61(7): 577-585 (2004).
A30 Atkinson, W.D., D.V. Law and K.J. Bromley. A decline in mortality from prostate cancer in
the UK Atomic Energy Authority workforce. J Radiol Prot 27(4): 437-445 (2007).
A31 ATSDR. Toxicological profile for uranium. Agency for Toxic Substances and Disease
Registry, Atlanta, Georgia, 2013.
A32 Auvinen, A., P. Kurttio, J. Pekkanen et al. Uranium and other natural radionuclides in drinking
water and risk of leukemia: a case-cohort study in Finland. Cancer Causes Control 13(9):
825-829 (2002).
A33 Avasthi, P.S., A.P. Evan and D. Hay. Glomerular endothelial cells in uranyl nitrate-induced
acute renal failure in rats. J Clin Invest 65(1): 121-127 (1980).
A34 Avtandilashvili, M., M. Puncher, S.L. McComish et al. US Transuranium and Uranium
Registries case study on accidental exposure to uranium hexafluoride. J Radiol Prot 35(1):
129-151 (2015).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 469

B1 Bagatti, D., M.C. Cantone, A. Giussani et al. Regional dependence of urinary uranium baseline
levels in non-exposed subjects with particular reference to volunteers from Northern Italy.
J Environ Radioact 65(3): 357-364 (2003).
B2 Bakhmutsky, M.V., M.S. Oliver, M.A. McDiarmid et al. Long term depleted uranium
exposure in Gulf War I veterans does not cause elevated numbers of micronuclei in peripheral
blood lymphocytes. Mutat Res 720(1-2): 53-57 (2011).
B3 Bal, W., A.M. Protas and K.S. Kasprzak. Genotoxicity of metal ions: chemical insights. Met
Ions Life Sci 8: 319-373 (2011).
B4 Ballou, J.E., R.A. Gies, A.C. Case et al. Deposition and early disposition of inhaled
233UO2(NO3)2 and 232UO2(NO3)2 in the rat. Health Phys 51(6): 755-771 (1986).
B5 Banday, A.A., S. Priyamvada, N. Farooq et al. Effect of uranyl nitrate on enzymes of
carbohydrate metabolism and brush border membrane in different kidney tissues. Food Chem
Toxicol 46(6): 2080-2088 (2008).
B6 Barber, D.S., M.F. Ehrich and B.S. Jortner. The effect of stress on the temporal and regional
distribution of uranium in rat brain after acute uranyl acetate exposure. J Toxicol Environ Health
A 68(2): 99-111 (2005).
B7 Basset, C., O. Averseng, P.J. Ferron et al. Revision of the biodistribution of uranyl in serum: Is
fetuin-A the major protein target? Chem Res Toxicol 26(5): 645-653 (2013).
B8 Batchelor, A.L., T.J. Jenner and D.G. Papworth. Influence of macrophages on
microdistribution of inhaled UO2 aerosol in rat lung. Phys Med Biol 27(7): 949-957 (1982).
B9 Baysson, H., D. Laurier, M. Tirmarche et al. Epidemiological response to a suspected excess
of cancer among a group of workers exposed to multiple radiological and chemical hazards.
Occup Environ Med 57(3): 188-194 (2000).
B10 Beckova, V. and I. Malatova. Dissolution behaviour of 238U, 234U and 230Th deposited on filters
from personal dosemeters. Radiat Prot Dosim 129(4): 469-472 (2008).
B11 Belles, M., M.L. Albina, V. Linares et al. Combined action of uranium and stress in the rat. I.
Behavioral effects. Toxicol Lett 158(3): 176-185 (2005).
B12 Belles, M., V. Linares, G. Perello et al. Human dietary exposure to uranium in Catalonia,
Spain. Biol Trace Elem Res 152(1): 1-8 (2013).
B13 Benedik, L. and Z. Jeran. Radiological of natural and mineral drinking waters in Slovenia.
Radiat Prot Dosim 151(2): 306-313 (2012).
B14 Bensoussan, H., L. Grancolas, B. Dhieux-Lestaevel et al. Heavy metal uranium affects the brain
cholinergic system in rat following sub-chronic and chronic exposure. Toxicology 261(1-2):
59-67 (2009).
B15 Bensoussan, H., G. Resve, C.M. Vacher et al. Memory and cholinergic system impaired by
sub-chronic exposure to uranium in young rats. Toxicol Lett 189: 226-227 (2009).
B16 Bentley, K.W., J.H. Wyatt, D.J. Wilson et al. Uranium and plutonium in hair as an indicator of
body burden in mice of different age and sex. Bull Environ Contam Toxicol 28(6): 691-696
(1982).
B17 Bentley, K.W., D.R. Stockwell, K.A. Britt et al. Transient proteinuria and aminoaciduria in
rodents following uranium intoxication. Bull Environ Contam Toxicol 34(3): 407-416 (1985).
B18 Beral, V., P. Fraser, L. Carpenter et al. Mortality of employees of the Atomic Weapons
Establishment, 1951-82. Br Med J 297(6651): 757-770 (1988).
B19 Berradi, H., J.M. Bertho, N. Dudoignon et al. Renal anemia induced by chronic ingestion of
depleted uranium in rats. Toxicol Sci 103(2): 397-408 (2008).
470 UNSCEAR 2016 REPORT

B20 Berry, J.P., L. Zhang, P. Galle et al. Role of alveolar macrophage lysosomes in metal
detoxification. Microsc Res Tech 36(4): 313-323 (1997).
B21 Betcher, R.N., M. Gascoyne and D. Brown. Uranium in groundwaters of southeastern
Manitoba, Canada. Can J Earth Sci 25(12): 2089-2103 (1988).
B22 Betti, M. Civil use of depleted uranium. J Environ Radioact 64(2-3): 113-119 (2003).
B23 Beyermann, M., T. Bnger, K. Schmidt et al. Occurrence of natural radioactivity in public
water supplies in Germany: 238U, 234U, 235U, 228Ra, 226Ra, 222Rn, 210Pb, 210Po and gross
activity concentrations. Radiat Prot Dosim 141(1): 72-81 (2010).
B24 Birke, M., U. Rauch, H. Lorenz et al. Distribution of uranium in German bottled and tap water.
J Geochem Explor 107(3): 272-282 (2010).
B25 Blantz, R.C. The mechanism of acute renal failure after uranyl nitrate. J Clin Invest 55(3): 621-35
(1975).
B26 Bleise, A., P.R. Danesi and W. Burkart. Properties, use and health effects of depleted uranium
(DU): a general overview. J Environ Radioact 64(2-3): 93-112 (2003).
B27 BMJ. Verordnung ber die Qualitt von Wasser fr den menschlichen Gebrauch
(Trinkwasserverordnung - TrinkwV 2001). Bundesministeriums der Justiz und fuer
Verbraucherschutz. [Internet] Available from (http://www.gesetze-im-internet.de/bundesrecht/
trinkwv_2001/gesamt.pdf) on 11.11.2013. (German).
B28 Bogers, R.P., F.E. van Leeuwen, L. Grievink et al. Cancer incidence in Dutch Balkan veterans.
Cancer Epidemiol 37(5): 550-555 (2013).
B29 Boice, J.D., S.S. Cohen, M.T. Mumma et al. Mortality among radiation workers at Rocketdyne
(Atomics International), 1948-1999. Radiat Res 166(1 Pt 1): 98-115 (2006).
B30 Boice, J.D., Jr., W.L. Bigbee, M.T. Mumma et al. Cancer mortality in counties near two
former nuclear materials processing facilities in Pennsylvania, 1950-1995. Health Phys 85(6):
691-700 (2003).
B31 Boice, J.D., Jr., W.L. Bigbee, M.T. Mumma et al. Cancer incidence in municipalities near two
former nuclear materials processing facilities in Pennsylvania. Health Phys 85(6): 678-90 (2003).
B32 Boice, J.D., Jr., M. Mumma, S. Schweitzer et al. Cancer mortality in a Texas county with prior
uranium mining and milling activities, 1950-2001. J Radiol Prot 23(3): 247-262 (2003).
B33 Boice, J.D., Jr., R.W. Leggett, E.D. Ellis et al. A comprehensive dose reconstruction methodology
for former rocketdyne/atomics international radiation workers. Health Phys 90(5): 409-30 (2006).
B34 Boice, J.D., Jr., S.S. Cohen, M.T. Mumma et al. Mortality among residents of Uravan,
Colorado who lived near a uranium mill, 1936-84. J Radiol Prot 27(3): 299-319 (2007).
B35 Boice, J.D., Jr., M.T. Mumma and W.J. Blot. Cancer and noncancer mortality in populations
living near uranium and vanadium mining and milling operations in Montrose County,
Colorado, 1950-2000. Radiat Res 167(6): 711-726 (2007).
B36 Boice, J.D., Jr., S.S. Cohen, M.T. Mumma et al. A cohort study of uranium millers and miners
of Grants, New Mexico, 1979-2005. J Radiol Prot 28(3): 303-325 (2008).
B37 Boice, J.D., Jr., M.T. Mumma and W.J. Blot. Cancer incidence and mortality in populations
living near uranium milling and mining operations in Grants, New Mexico, 1950-2004. Radiat
Res 174(5): 624-636 (2010).
B38 Boice, J.D., Jr., S.S. Cohen, M.T. Mumma et al. Updated mortality analysis of radiation
workers at Rocketdyne (Atomics International), 1948-2008. Radiat Res 176(2): 244-258
(2011).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 471

B39 Bolch, W.E., K.F. Eckerman, G. Sgouros et al. MIRD pamphlet No. 21: a generalized schema for
radiopharmaceutical dosimetry--standardization of nomenclature. J Nucl Med 50(3): 477-84 (2009).
B40 Bomben, A.M., H.E. Equillor and A.A. Oliveira. 226Ra and natural uranium in Argentinian
bottled mineral waters. Radiat Prot Dosim 67(3): 221-224 (1996).
B41 Bomben, A.M. and M.A. Palacios. Natural radionuclides in drinking water in Argentina.
pp.P-1a-4 in: IRPA-10: International Congress of the International Radiation Protection
Association, Hiroshima, Japan, 14-19 May 2000. IRPA, Hiroshima, 2000.
B42 Bonin, B. and P.-L. Blanc. Luranium dans le milieu naturel, des origines jusqu la mine.
pp.7-41 in: Luranium, de lenvironnement lhomme (H. Mtivier, ed.) EDP Sciences, IRSN,
Les Ulis, France, 2001.
B43 Bonotto, D.M. and T.O. Bueno. The natural radioactivity in Guarani aquifer groundwater,
Brazil. Appl Radiat Isot 66(10): 1507-1522 (2008).
B44 Bosque, M.A., J.L. Domingo, J.M. Llobet et al. Embryotoxicity and teratogenicity of uranium
in mice following subcutaneous administration of uranyl acetate. Biol Trace Elem Res 36(2):
109-118 (1993).
B45 Bozal, C.B., A.B. Martinez, R.L. Cabrini et al. Effect of ethane-1-hydroxy-1,1-bisphosphonate
(EHBP) on endochondral ossification lesions induced by a lethal oral dose of uranyl nitrate.
Arch Toxicol 79(8): 475-481 (2005).
B46 Bradley, E.J. and L.W. Ewers. The transfer and resulting radiation dose from polonium,
thorium and other naturally-occurring radionuclides to the human fetus. pp.19-22 in: Health
Effects of Internally Deposited Radionuclides: Emphasis on Radium and Thorium, an
International Seminar held in Heidelberg, Germany, 18-21 April 1994. World Scientific
Publishing Co., Singapore, 1995.
B47 Bradley, J. and L. Prosser. Radionuclides in human fetal tissues. Radiol Prot Bull 148: 28-31
(1993).
B48 Brady, H.R., B.C. Kone, R.M. Brenner et al. Early effects of uranyl nitrate on respiration and
K+ transport in rabbit proximal tubule. Kidney Int 36(1): 27-34 (1989).
B49 Brandom, W.F., G. Saccomanno, V.E. Archer et al. Chromosome aberrations in uranium
miners occupationally exposed to 222 radon. Radiat Res 52(1): 204-215 (1972).
B50 Briner, W. and K. Byrd. Effects of depleted uranium on development of the mouse.
pp.459-461 in: Proceedings of the 6th International Symposium on Metal Ions in Biology and
Medicine (J. Centeno et al., eds.). John Libby Eurotext, Paris, 2000.
B51 Briner, W. and B. Abboud. Behavior of juvenile mice chronically exposed to depleted uranium.
pp.353-356 in: Metal Ions in Biology and Medicine (L. Khassanova et al., eds.). John Libbey
Eurotext, Paris, 2002.
B52 Briner, W. and J. Murray. Effects of short-term and long-term depleted uranium exposure on
open-field behavior and brain lipid oxidation in rats. Neurotoxicol Teratol 27(1): 135-144 (2005).
B53 Broadway, J.A. and A.B. Strong. Radionuclides in human bone samples. Health Phys 45(3):
765-768 (1983).
B54 Busby, C., M. Hamdan and E. Ariabi. Cancer, infant mortality and birth sex-ratio in Fallujah,
Iraq 2005-2009. Int J Environ Res Public Health 7(7): 2828-2837 (2010).
B55 Bussy, C., P. Lestaevel, B. Dhieux et al. Chronic ingestion of uranyl nitrate perturbs
acetylcholinesterase activity and monoamine metabolism in male rat brain. Neurotoxicology
27(2): 245-252 (2006).
C1 Callan, A.C., A.L. Hinwood, M. Ramalingam et al. Maternal exposure to metals-
concentrations and predictors of exposure. Environ Res 126: 111-117 (2013).
472 UNSCEAR 2016 REPORT

C2 Cameco Corporation. Beaverlodge Project Annual Report: Year 28 - Transition phase

monitoring. Cameco Corporation, Saskatoon, Saskatchewan, 2013.
C3 Cameco Corporation. Beaverlodge Project Annual Report: Year 29 - Transition phase
monitoring. Cameco Corporation, Saskatoon, Saskatchewan, 2014.
C4 Cameco Corporation. Beaverlodge Project Annual Report: Year 30 - Transition phase
monitoring. Cameco Corporation, Saskatoon, Saskatchewan, 2015.
C5 Canu, I.G., E.D. Ellis and M. Tirmarche. Cancer risk in nuclear workers occupationally
exposed to uranium-emphasis on internal exposure. Health Phys 94(1): 1-17 (2008).
C6 Canu, I.G., S. Jacob, E. Cardis et al. Reprocessed uranium exposure and lung cancer risk.
Health Phys 99(3): 308-313 (2010).
C7 Carpenter, A.V., W.D. Flanders, E.L. Frome et al. Chemical exposures and central nervous
system cancers: a case-control study among workers at two nuclear facilities. Am J Ind Med
13(3): 351-362 (1988).
C8 Carpenter, L., C. Higgins, A. Douglas et al. Combined analysis of mortality in three United
Kingdom nuclear industry workforces, 1946-1988. Radiat Res 138(2): 224-238 (1994).
C9 Carpenter, L.M., C.D. Higgins, A.J. Douglas et al. Cancer mortality in relation to monitoring for
radionuclide exposure in three UK nuclear industry workforces. Br J Cancer 78(9): 1224-32 (1998).
C10 Carriere, M., C. Thiebault, S. Milgram et al. Citrate does not change uranium chemical
speciation in cell culture medium but increases its toxicity and accumulation in NRK-52E
cells. Chem Res Toxicol 19(12): 1637-1642 (2006).
C11 Carvalho, F.P., J.M. Oliveira and I. Faria. Alpha emitting radionuclides in drainage from
Quinta do Bispo and Cunha Baixa uranium mines (Portugal) and associated radiotoxicological
risk. Bull Environ Contam Toxicol 83(5): 668-673 (2009).
C12 Carvalho, F.P. and J.M. Oliveira. Uranium isotopes in the Balkan's environment and foods
following the use of depleted uranium in the war. Environ Int 36(4): 352-360 (2010).
C13 Carvalho, I.G., R. Cidu, L. Fanfani et al. Environmental impact of uranium mining and ore
processing in the Lagoa Real District, Bahia, Brazil. Environ Sci Technol 39(22): 8646-8652
(2005).
C14 Caserta, D., A. Mantovani, F. Ciardo et al. Heavy metals in human amniotic fluid: a pilot
study. Prenat Diagn 31(8): 792-796 (2011).
C15 Catalano, J.G., S.M. Heald, J.M. Zachara et al. Spectroscopic and diffraction study of uranium
speciation in contaminated vadose zone sediments from the Hanford site, Washington State.
Environ Sci Technol 38(10): 2822-2828 (2004).
C16 CDC. Fourth national report on human exposure to environmental chemicals, Updated tables,
February 2015. Department of Health and Human Services, Centers for Disease Control and
Prevention, Atlanta, GA, 2015.
C17 Chan, C., T.S. Hughes, S. Muldoon et al. Mortality patterns among Paducah gaseous diffusion
plant workers. J Occup Environ Med 52(7): 725-732 (2010).
C18 Chase, G.R. Correlation between airborne U exposure and U urinalysis results at Bear Creek
uranium. Health Phys 56(2): 195-199 (1989).
C19 Chau, N.D. and B. Michalec. Natural radioactivity in bottled natural spring, mineral and
therapeutic waters in Poland. J Radioanal Nucl Chem 279(1): 121-129 (2009).
C20 Checkoway, H., N. Pearce, D.J. Crawford-Brown et al. Radiation doses and cause-specific
mortality among workers at a nuclear materials fabrication plant. Am J Epidemiol 127(2):
255-266 (1988).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 473

C21 Chen, J., D.P. Meyerhof and B.L. Tracy. Model results of kidney burdens from uranium
intakes. Health Phys 86(1): 3-11 (2004).
C22 Chen, J., D. Lariviere, R. Timmins et al. Uranium GI absorption coefficients for young
children. IRPA12: 12th Congress of the International Radiation Protection Association:
Strengthening Radiation Protection Worldwide - Highlights, Global Perspective and Future
Trends, Buenos Aires, Argentina, 19-24 October 2008. IRPA, 2008.
C23 Chen, J., D. Lariviere, R. Timmins et al. Estimation of uranium GI absorption fractions for
children and adults. Radiat Prot Dosim 144(1-4): 379-383 (2011).
C24 Chen, J., D. Moir, R. Lane et al. An ecological study of cancer incidence in Port Hope, Ontario
from 1992 to 2007. J Radiol Prot 33(1): 227-242 (2013).
C25 Chung, H.J. and M.G. Lee. Pharmacokinetic changes of ipriflavone in rats with acute renal
failure induced by uranyl nitrate. Biopharm Drug Dispos 27(7): 345-351 (2006).
C26 Chung, W., E.J. Kim, I. Lee et al. Effects of recombinant human growth hormone on the
pharmacokinetics of intravenous chlorzoxazone in rats with acute renal failure induced by
uranyl nitrate. Life Sci 73(3): 253-263 (2003).
C27 CNSC. Understanding health studies and risk assessments conducted in the Port Hope
Community from the 1950s to the present. INFO-0781. Canadian Nuclear Safety Commission,
Ottawa, Ontario, 2009.
C28 CNSC. Environmental performance of a uranium mine or mill regulated under the Nuclear
Safety and Control Act. Based on environmental data associated with operating uranium mines
and mills (2000-2012). NAT24_VA. Canadian Nuclear Safety Commission, Ottawa, Ontario,
2014.
C29 Codex. Codex general standard for contaminants and toxins in food and feed. Codex Standard
193-1995. 2013.
C30 Cooke, N. and F.B. Holt. The solubility of some uranium compounds in simulated lung fluid.
Health Phys 27(1): 69-77 (1974).
C31 Cooper, J.R., G.N. Stradling, H. Smith et al. The behaviour of uranium-233 oxide and uranyl-
233 nitrate in rats. Int J Radiat Biol Relat Stud Phys Chem Med 41(4): 421-433 (1982).
C32 Corcho Alvarado, J.A., B. Balsiger, S. Rllin et al. Radioactive and chemical contamination of
the water resources in the former uranium mining and milling sites of Mailuu Suu
(Kyrgyzstan). J Environ Radioact 138: 1-10 (2014).
C33 Coryell, V.H. and D.M. Stearns. Molecular analysis of hprt mutations generated in Chinese
hamster ovary EM9 cells by uranyl acetate, by hydrogen peroxide, and spontaneously. Mol
Carcinog 45(1): 60-72 (2006).
C34 Cothern, C.R. and W.L. Lappenbusch. Occurrence of uranium in drinking water in the U.S.
Health Phys 45(1): 89-99 (1983).
C35 Craft, E.S., A.W. Abu-Qare, M.M. Flaherty et al. Depleted and natural uranium: Chemistry and
toxicological effects. J Toxicol Environ Health - Part B: Critical Reviews 7(4): 297-317 (2004).
C36 Cragle, D.L., R.W. McLain, J.R. Qualters et al. Mortality among workers at a nuclear fuels
production facility. Am J Ind Med 14(4): 379-401 (1988).
C37 CSN. Epidemiological study of the possible effect of ionising radiations deriving from the
operation of Spanish nuclear fuel cycle facilities on the health of the population living in their
vicinity. Summary of the final report. Nuclear Safety Council, Spain, 2009.
C38 Cuney, M. The extreme diversity of uranium deposits. Miner Deposita 44(1): 3-9 (2009).
474 UNSCEAR 2016 REPORT

D1 Da Silva, M.L. and D.M. Bonotto. Uranium isotopes in groundwater occurring at Amazonas
State, Brazil. Appl Radiat Isot 97: 24-33 (2015).
D2 Dabous, A.A., J.K. Osmond and Y.H. Dawood. Uranium/thorium isotope evidence for ground-
water history in the Eastern Desert of Egypt. J Arid Environ 50(2): 343-357 (2002).
D3 Damon, E.G., A.F. Eidson, F.F. Hahn et al. Comparison of early lung clearance of yellowcake
aerosols in rats with in vitro dissolution and IR analysis. Health Phys 46(4): 859-866 (1984).
D4 Danesi, P.R., A. Markowicz, E. Chinea-Cano et al. Depleted uranium particles in selected
Kosovo samples. J Environ Radioact 64(2-3): 143-154 (2003).
D5 Darolles, C., D. Broggio, A. Feugier et al. Different genotoxic profiles between depleted and
enriched uranium. Toxicol Lett 192(3): 337-348 (2010).
D6 Davesne, E. Optimisation des programmes de surveillance de la contamination interne par
ltude des incertitudes lies lvaluation dosimtrique. Thse. Rapport DRPH/SDI/2010-11,
ISRN/IRSN-2010/135. Laboratoire d'Evaluation de la Dose Interne, IRSN/DRPH/SDI/LEDI,
Fontenay-aux-Roses Cedex, 2010.
D7 Davesne, E. and E. Blanchardon. Physico-chemical characteristics of uranium compounds: A
review. Int J Radiat Biol 90(11): 975-988 (2014).
D8 Davesne, E., N. Blanchin, E. Chojnacki et al. Collective dosimetry to distinguish occupational
exposure to natural uranium from alimentary uranium background in bioassay measurements.
Int J Radiat Biol 90(11): 1048-1054 (2014).
D9 Davis, M.W. Correlation analysis of first phase monitoring results for uranium mill workers.
Prepared by Monserco Limited under contract to the Atomic Energy Control Board.
INFO-0098. Atomic Energy Control Board, Ottawa, 1983.
D10 De Rey, B.M., H.E. Lanfranchi and R.L. Cabrini. Percutaneous absorption of uranium
compounds. Environ Res 30(2): 480-491 (1983).
D11 Desideri, D., M.A. Meli, L. Feduzi et al. 238U, 234U, 226Ra, 210Po concentrations of bottled
mineral waters in Italy and their dose contribution. J Environ Radioact 94(2): 86-97 (2007).
D12 Diamond, G.L., P.E. Morrow, B.J. Panner et al. Reversible uranyl fluoride nephrotoxicity in
the Long Evans rat. Fundam Appl Toxicol. 13(1): 65-78 (1989).
D13 Dias da Cunha, K.M., H. Henderson, B.M. Thomson et al. Ground water contamination with
238
U, 234U, 235U, 226Ra and 210Pb from past uranium mining: Cove wash, Arizona. Environ
Geochem Health 36(3): 477-487 (2014).
D14 Diz-Francs, I., J. Mantero, G. Manjn et al. 210Po and 238U Isotope concentrations in
commercial bottled mineralwater samples in Spain and their dose contribution. Radiat Prot
Dosim 156(3): 336-342 (2013).
D15 Diaz Sylvester, P.L., R. Lopez, A.M. Ubios et al. Exposure to subcutaneously implanted
uranium dioxide impairs bone formation. Arch Environ Health 57(4): 320-325 (2002).
D16 Dinocourt, C., M. Legrand, I. Dublineau et al. The neurotoxicology of uranium. Toxicology
337: 58-71 (2015).
D17 Domingo, J.L., J.M. Llobet, J.M. Tomas et al. Acute toxicity of uranium in rats and mice. Bull
Environ Contam Toxicol 39(1): 168-174 (1987).
D18 Domingo, J.L., A. Ortega, J.L. Paternain et al. Evaluation of the perinatal and postnatal effects
of uranium in mice upon oral administration. Arch Environ Health 44(6): 395-398 (1989).
D19 Donnadieu-Claraz, M., M. Bonnehorgne, B. Dhieux et al. Chronic exposure to uranium leads
to iron accumulation in rat kidney cells. Radiat Res 167(4): 454-464 (2007).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 475

D20 Donoghue, J.K., E.D. Dyson, J.S. Hislop et al. Human exposure to natural uranium. A case
history and analytical results from some postmortem tissues. Br J Ind Med 29(1): 81-89 (1972).
D21 Downs, W.L., H.B. Wilson, G.E. Sylvester et al. Excretion of uranium by rats following
inhalation of uranium dioxide. Health Phys 13(5): 445-453 (1967).
D22 Dreesen, D.R., J.M. Williams, M.L. Marple et al. Mobility and bioavailability of uranium mill
tailings contaminants. Environ Sci Technol 16(10): 702-709 (1982).
D23 Drubay, D., S. Ancelet, A. Acker et al. Kidney cancer mortality and ionizing radiation among
French and German uranium miners. Radiat Environ Biophys 53(3): 505-513 (2014).
D24 Drubay, D., S. Caer-Lorho, P. Laroche et al. Mortality from circulatory system diseases among
French uranium miners: A nested case-control study. Radiat Res 183(5): 550-562 (2015).
D25 Dublineau, I., S. Grison, C. Baudelin et al. Absorption of uranium through the entire
gastrointestinal tract of the rat. Int J Radiat Biol 81(6): 473-482 (2005).
D26 Dublineau, I., S. Grison, L. Grandcolas et al. Absorption, accumulation and biological effects
of depleted uranium in Peyer's patches of rats. Toxicology 227(3): 227-239 (2006).
D27 Dublineau, I., S. Grison, C. Linard et al. Short-term effects of depleted uranium on immune
status in rat intestine. J Toxicol Environ Health A 69(17): 1613-1628 (2006).
D28 Dublineau, I., L. Grandcolas, S. Grison et al. Modifications of inflammatory pathways in rat
intestine following chronic ingestion of depleted uranium. Toxicol Sci 98(2): 458-468 (2007).
D29 Dublineau, I., M. Souidi, Y. Gueguen et al. Unexpected lack of deleterious effects of uranium
on physiological systems following a chronic oral intake in adult rat. Biomed Res Int 2014:
181989 (2014).
D30 Dufey, F., L. Walsh, A. Tschense et al. Occupational doses of ionizing radiation and leukemia
mortality. Health Phys 100(5): 548-550; author reply 550-541 (2011).
D31 Dufey, F., L. Walsh, M. Sogl et al. Radiation dose dependent risk of liver cancer mortality in
the German uranium miners cohort 1946-2003. J Radiol Prot 33(1): 175-185 (2013).
D32 Duport, P., R. Robertson, K. Ho et al. Flow-through dissolution of uranium-thorium ore dust,
uranium concentrate, uranium dioxide, and thorium alloy in simulated lung fluid. Radiat Prot
Dosim 38(1-3): 121-133 (1991).
D33 Dupree-Ellis, E., J. Watkins, J.N. Ingle et al. External radiation exposure and mortality in a
cohort of uranium processing workers. Am J Epidemiol 152(1): 91-95 (2000).
D34 Dupree, E.A., D.L. Cragle, R.W. McLain et al. Mortality among workers at a uranium
processing facility, the Linde Air Products Company Ceramics Plant, 1943-1949. Scand J
Work Environ Health 13(2): 100-107 (1987).
D35 Dupree, E.A., J.P. Watkins, J.N. Ingle et al. Uranium dust exposure and lung cancer risk in
four uranium processing operations. Epidemiology 6(4): 370-375 (1995).
D36 Dygert, H. Concepts in inhalation toxicology. pp.603-700 in: Pharmacology and Toxicology
of Uranium Compounds (C. Voegtlin and H. Hodge, eds.). McGraw-Hill Book Co., Inc., New
York, 1949.
E1 Eadie, G.G. and R.F. Kaufmann. Radiological evaluation of the effects of uranium mining and
milling operations on selected ground water supplies in the Grants Mineral Belt, New Mexico.
Health Phys 32(4): 231-241 (1977).
E2 EFSA. Uranium in foodstuffs, in particular mineral water. Scientific opinion of the panel on
contaminants in the food chain. EFSA Journal 7(4): 1-59 (2009).
E3 Ellender, M. The clearance of uranium after deposition of the nitrate and bicarbonate in
different regions of the rat lung. Hum Toxicol 6(6): 479-482 (1987).
476 UNSCEAR 2016 REPORT

E4 Ellender, M., J.D. Harrison, H. Pottinger et al. Induction of osteosarcoma and acute myeloid
leukaemia in CBA/H mice by the alpha-emitting nuclides, uranium-233, plutonium-239 and
amercium-241. Int J Radiat Biol 77(1): 41-52 (2001).
E5 Esposito, M., P. Poli, P. Bartolomei et al. Survey of natural and anthropogenic radioactivity
in environmental samples from Yugoslavia. J Environ Radioact 61(3): 271-282 (2002).
E6 Essien, I.O., D.N. Sandoval and P.K. Kuroda. Deposition of excess amount of natural U from
the atmosphere. Health Phys 48(3): 325-331 (1985).
F1 Fathi, R.A., L.Y. Matti, H.S. Al-Salih et al. Environmental pollution by depleted uranium in
Iraq with special reference to Mosul and possible effects on cancer and birth defect rates. Med
Confl Surviv 29(1): 7-25 (2013).
F2 Feugier, A., S. Frelon, P. Gourmelon et al. Alteration of mouse oocyte quality after a
subchronic exposure to depleted uranium. Reprod Toxicol 26(3-4): 273-277 (2008).
F3 Filipy, R.E. and J.J. Russell. The United States Transuranium and Uranium Registries as
sources for actinide dosimetry and bioeffects. Radiat Prot Dosim 105(1-4): 185-187 (2003).
F4 Fisenne, I.M. and G.A. Welford. Natural U concentrations in soft tissues and bone of New
York City residents. Health Phys 50(6): 739-746 (1986).
F5 Fisenne, I.M., P.M. Perry, K.M. Decker et al. The daily intake of 234,235,238U, 228,230,232
Th and
226,228
Ra by New York City residents. Health Phys 53(4): 357-363 (1987).
F6 Fisenne, I.M., P.M. Perry and N.H. Harley. Uranium in humans. Radiat Prot Dosim 24(1-4):
127-131 (1988).
F7 Fisenne, I.M. The natural radiation environment. Proceedings of the Fifth International
Symposium on the Natural Radiation Environment, 22-28 September 1991, Salzburg.
Salzburg, Austria, 1991.
F8 Fitsanakis, V.A., K.M. Erikson, S.J. Garcia et al. Brain accumulation of depleted uranium in
rats following 3- or 6-month treatment with implanted depleted uranium pellets. Biol Trace
Elem Res 111(1-3): 185-197 (2006).
F9 Fitzgerald, J. Groundwater quality and environmental health implications, Anangu
Pitjantjatjara lands, South Australia. Bureau of Rural Sciences, Canberra, 2000.
F10 Fraser, P., L. Carpenter, N. Maconochie et al. Cancer mortality and morbidity in employees of
the United Kingdom Atomic Energy Authority, 1946-86. Br J Cancer 67(3): 615-624 (1993).
F11 Frelon, S., P. Houpert, D. Lepetit et al. The chemical speciation of uranium in water does not
influence its absorption from the gastrointestinal tract of rats. Chem Res Toxicol 18(7):
1150-1154 (2005).
F12 Frengstad, B., A.K. Skrede, D. Banks et al. The chemistry of Norwegian groundwaters: III.
The distribution of trace elements in 476 crystalline bedrock groundwaters, as analysed by
ICP-MS techniques. Sci Total Environ 246(1): 21-40 (2000).
F13 Frengstad, B.S., K. Lax, T. Tarvainen et al. The chemistry of bottled mineral and spring waters
from Norway, Sweden, Finland and Iceland. J Geochem Explor 107(3): 350-361 (2010).
F14 Frisbie, S.H., E.J. Mitchell, L.J. Mastera et al. Public health strategies for western Bangladesh
that address arsenic, manganese, uranium, and other toxic elements in drinking water. Environ
Health Perspect 117(3): 410-416 (2009).
F15 Frisbie, S.H., E.J. Mitchell and B. Sarkar. World Health Organization increases its drinking-
water guideline for uranium. Environ Sci Process Impacts 15(10): 1817-1823 (2013).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 477

F16 Froidevaux, P., J.J. Geering, L. Pillonel et al. 90Sr, 238U, 234U, 137Cs, 40K and 239/240Pu in
Emmental type cheese produced in different regions of Western Europe. J Environ Radioact
72(3): 287-298 (2004).
F17 Fukuda, S., H. Iida, M. Ikeda et al. Toxicity of uranium and the removal effects of CBMIDA
and EHBP in simulated wounds of rats. Health Phys 89(1): 81-88 (2005).
F18 Fukuda, S., M. Ikeda, M. Chiba et al. Clinical diagnostic indicators of renal and bone damage in
rats intramuscularly injected with depleted uranium. Radiat Prot Dosim 118(3): 307-314 (2006).
G1 Galle, P. Toxiques Nuclaires. Masson, 1997. (French).
G2 Gans, I. Natural radionuclides in mineral waters. Sci Total Environ 45: 93-99 (1985).
G3 Garg, V.K., A. Yadav, K. Singh et al. Uranium concentration in groundwater in Hisar city,
India. Int J Occup Environ Med 5(2): 112-114 (2014).
G4 Garnier-Laplace, J., R. Gilbin, C. Della-Vedova et al. Bioavailability in ecological risk.
Assessment for radionuclides. pp.41-57 in: Scientific Trends in Radiological Protection of the
Environment: ECORAD 2004 - International Conference on the Scientific Basis for
Environment Protection Against Ionising Radiation, Aix-en-Provence, France. TEC&DOC,
Lavoisier, 2005.
G5 Gazin, V., S. Kerdine, G. Grillon et al. Uranium induces TNF alpha secretion and MAPK
activation in a rat alveolar macrophage cell line. Toxicol Appl Pharmacol 194(1): 49-59 (2004).
G6 Gegner, M. and K. Irlweck. Uranium concentrations and isotopic ratios in Austrian drinking
waters. Radioact Environ 7: 135-139 (2005).
G7 Gharbi, F., S. Baccouche, W. Abdelli et al. Uranium isotopes in Tunisian bottled mineral
waters. J Environ Radioact 101(8): 589-590 (2010).
G8 Ghosh, S., A. Kumar, B.N. Pandey et al. Acute exposure of uranyl nitrate causes lipid
peroxidation and histopathological damage in brain and bone of Wistar rat. J Environ Pathol
Toxicol Oncol 26(4): 255-261 (2007).
G9 Giglio, M.J., N. Brandan, T.L. Leal et al. The mechanism of the transient depression of the
erythropoietic rate induced in the rat by a single injection of uranyl nitrate. Toxicol Appl
Pharmacol 99(2): 260-265 (1989).
G10 Gilman, A.P., M.A. Moss, D.C. Villeneuve et al. Uranyl nitrate: 91-day exposure and recovery
studies in the male New Zealand white rabbit. Toxicol Sci 41(1): 138-151 (1998).
G11 Gilman, A.P., D.C. Villeneuve, V.E. Secours et al. Uranyl nitrate: 28-day and 91-day toxicity
studies in the Sprague-Dawley rat. Toxicol Sci 41(1): 117-128 (1998).
G12 Giri, S., G. Singh and V.N. Jha. Evaluation of radionuclides in groundwater around proposed
uranium mining sites in Bagjata and Banduhurang, Jharkhand (India). Radioprotection 46(1):
39-57 (2011).
G13 Giri, S., V.N. Jha, G. Singh et al. Estimation of annual effective dose due to ingestion of
natural radionuclides in foodstuffs and water at a proposed uranium mining site in India. Int J
Radiat Biol 89(12): 1071-1078 (2013).
G14 Godoy, M.L., J.M. Godoy, R. Kowsmann et al. 234U and 230Th determination by FIA-ICP-
MS and application to uranium-series disequilibrium in marine samples. J Environ Radioact
88(2): 109-117 (2006).
G15 Goel, K.A., V.K. Garg and V. Garg. Histopathology of kidney of albino rat poisoned with
uranyl nitrate. Bull Environ Contam Toxicol 24(1): 9-12 (1980).
G16 Golchert, N.W., T.L. Duffy and J. Sedlet. Environmental monitoring at Argonne National
Laboratory. ANL-85-17. Argonne National Laboratory, Illinois, USA, 1985.
478 UNSCEAR 2016 REPORT

G17 Goldman, M., A. Yaari, Z. Doshnitzki et al. Nephrotoxicity of uranyl acetate: effect on rat
kidney brush border membrane vesicles. Arch Toxicol 80(7): 387-393 (2006).
G18 Gonalves, M.M., M. Gerenutti and M.M.D. Carvalho Vila. Uranium: Natural occurrence,
physical and chemical characteristics, environmental contamination and human exposure.
pp.185-209 in: Uranium: Characteristics, Occurrence and Human Exposure. Nova Science
Publishers, 2012.
G19 Greenland, S. and H. Morgenstern. Ecological bias, confounding, and effect modification. Int J
Epidemiol 18(1): 269-274 (1989).
G20 Grignard, E., Y. Gueguen, S. Grison et al. Contamination with depleted or enriched uranium
differently affects steroidogenesis metabolism in rat. Int J Toxicol 27(4): 323-328 (2008).
G21 Grison, S., G. Fave, M. Maillot et al. Metabolomics identifies a biological response to chronic
low-dose natural uranium contamination in urine samples. Metabolomics 9: 1168-1180 (2013).
G22 Guan, H. and F. Piao. Maternal and fetal exposure to uranium in Dalian, China. Acta Obstet
Gynecol Scand 90(8): 924-925 (2011).
G23 Gueguen, Y., F. Paquet, P. Voisin et al. Effects of chronic contamination with depleted
uranium on xenobiotic biotransformation enzymes in the rat. in: Proceedings of the 14th
International Conference on Cytochromes P450: Biochemistry, Biophysics and
Bioinformatics, Dallas, Texas, 2005. Medimond International Proceedings, Bologna, Italy,
2005.
G24 Gueguen, Y., K. Mouzat, L. Ferrari et al. Cytochromes P450: xenobiotic metabolism,
regulation and clinical importance. Ann Biol Clin (Paris) 64(6): 535-548 (2006). (French).
G25 Gueguen, Y., M. Souidi, C. Baudelin et al. Short-term hepatic effects of depleted uranium on
xenobiotic and bile acid metabolizing cytochrome P450 enzymes in the rat. Arch Toxicol
80(4): 187-195 (2006).
G26 Gueguen, Y., L. Grandcolas, C. Baudelin et al. Effect of acetaminophen administration to rats
chronically exposed to depleted uranium. Toxicology 229(1-2): 62-72 (2007).
G27 Gueguen, Y. and C. Rouas. New data on uranium nephrotoxicity. Radioprotection 47(3):
345-359 (2012). (French).
G28 Gueguen, Y., C. Rouas and F.A. Leblond. Kidney injury biomarkers. Nephrologie et
Therapeutique 8(3): 146-155 (2012).
G29 Gueguen, Y., C. Rouas, A. Monin et al. Molecular, cellular, and tissue impact of depleted
uranium on xenobiotic-metabolizing enzymes. Arch Toxicol 88(2): 227-239 (2014).
G30 Gueguen, Y., D. Suhard, C. Poisson et al. Low-concentration uranium enters the HepG2 cell
nucleus rapidly and induces cell stress response. Toxicol In Vitro 30(1 Pt B): 552-560 (2015).
G31 Guseva Canu, I., E. Cardis, C. Metz-Flamant et al. French cohort of the uranium processing
workers: mortality pattern after 30-year follow-up. Int Arch Occup Environ Health 83(3):
301-308 (2010).
G32 Guseva Canu, I., S. Jacob, E. Cardis et al. Uranium carcinogenicity in humans might depend
on the physical and chemical nature of uranium and its isotopic composition: results from pilot
epidemiological study of French nuclear workers. Cancer Causes Control 22(11): 1563-1573
(2011).
G33 Guseva Canu, I., J.P. Garsi, S. Car-Lohro et al. Does uranium induce circulatory diseases?
First results from a French cohort of uranium workers. Occup Environ Med 69(6): 404-409
(2012).
H1 Habert, R., V. Muczynski, T. Grisin et al. Concerns about the widespread use of rodent models
for human risk assessments of endocrine disruptors. Reproduction 147(4): R119-R129 (2014).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 479

H2 Hadad, K. and R. Doulatdar. U-series concentration in surface and ground water resources of
Ardabil province. Radiat Prot Dosim 130(3): 309-318 (2008).
H3 Hahn, F.F., R.A. Guilmette and M.D. Hoover. Implanted depleted uranium fragments cause
soft tissue sarcomas in the muscles of rats. Environ Health Perspect 110(1): 51-59 (2002).
H4 Hakam, O.K., A. Choukri, J.L. Reyss et al. Determination and comparison of uranium and
radium isotopes activities and activity ratios in samples from some natural water sources in
Morocco. J Environ Radioact 57(3): 175-189 (2001).
H5 Hakonson-Hayes, A.C., P.R. Fresquez and F.W. Whicker. Assessing potential risks from
exposure to natural uranium in well water. J Environ Radioact 59(1): 29-40 (2002).
H6 Haley, D.P., R.E. Bulger and D.C. Dobyan. The long-term effects of uranyl nitrate on the
structure and function of the rat kidney. Virchows Arch B Cell Pathol Incl Mol Pathol 41(1-2):
181-192 (1982).
H7 Ham, G.J., S.A. Hodgson, M.J. Youngman et al. Review of autopsy, in vivo and bioassay
measurements on members of the public in the UK. NRPB-W42. National Radiological
Protection Board, Chilton, Didcot, 2003.
H8 Hamilton, E.I. The concentration of uranium in man and his diet. Health Phys 22(2): 149-153
(1972).
H9 Hao, Y., R. Li, Y. Leng et al. A study assessing the genotoxicity in rats after chronic oral
exposure to a low dose of depleted uranium. J Radiat Res 50(6): 521-528 (2009).
H10 Hao, Y., R. Li, Y. Leng et al. The reproductive effects in rats after chronic oral exposure to
low-dose depleted uranium. J Radiat Res 53(3): 377-384 (2012).
H11 Hao, Y., J. Ren, J. Liu et al. Immunological changes of chronic oral exposure to depleted
uranium in mice. Toxicology 309: 81-90 (2013).
H12 Harley, N.H. and I.M. Fisenne. Distribution and alpha radiation dose from naturally occurring
U, Th, and Ra in the human skeleton. Health Phys 58(4): 515-518 (1990).
H13 Harley, N.H., E.C. Foulkes, L.H. Hilborne et al. Depleted uranium, Volume 7. in: A Review of
the Scientific Literature as it Pertains to Gulf War Illness. National Defense Research Institute,
Washington, 1999.
H14 Hart, D.R., P.M. McKee and A.J. Burt. Benthic community and sediment quality assessment
of Port Hope Harbour, Lake Ontario. J Great Lakes Res 12: 206-220 (1986).
H15 Hartsock, W.J., J.D. Cohen and D.J. Segal. Uranyl acetate as a direct inhibitor of DNA-
binding proteins. Chem Res Toxicol 20(5): 784-789 (2007).
H16 HC. Guidelines for Canadian drinking water quality: Guideline technical document -
Radiological parameters. Radiation Protection Bureau, Healthy Environments and Consumer
Safety Branch, Health Canada, Ottawa, Ontario 2009.
H17 Herranz, M., A. Abelairas and F. Legarda. Uranium contents and associated effective doses in
drinking water from Biscay (Spain). Appl Radiat lsot 48(6): 857-886 (1997).
H18 Hines, S.E., P. Gucer, S. Kligerman et al. Pulmonary health effects in Gulf War I service
members exposed to depleted Uranium. J Occup Environ Med 55(8): 937-944 (2013).
H19 Hirose, K. and Y. Sugimura. Concentration of uranium and the activity ratio of 234U/238U in
surface air: effect of atmospheric burn-up of Cosmos-954. Meteorol Geophys 32(4): 317-322
(1981).
H20 Hodgson, A., J.C. Moody, G.N. Stradling et al. Application of the ICRP respiratory tract
model to uranium compounds produced during the manufacture of nuclear fuel.
NRPB-M1156. Chilton, Didcot, 2000.
480 UNSCEAR 2016 REPORT

H21 Hollriegl, V., A.M. Arogunjo, A. Giussani et al. Daily urinary excretion of uranium in
members of the public of Southwest Nigeria. Sci Total Environ 412-413: 344-350 (2011).
H22 Homma-Takeda, S., T. Kokubo, Y. Terada et al. Uranium dynamics and developmental
sensitivity in rat kidney. J Appl Toxicol 33(7): 685-694 (2013).
H23 Homma-Takeda, S., K. Kitahara, K. Suzuki et al. Cellular localization of uranium in the renal
proximal tubules during acute renal uranium toxicity. J Appl Toxicol 35(12): 1594-1600 (2015).
H24 Houpert, P., P. Lestaevel, C. Bussy et al. Enriched but not depleted uranium affects central
nervous system in long-term exposed rat. Neurotoxicology 26(6): 1015-1020 (2005).
H25 Houpert, P., S. Frelon, P. Lestaevel et al. Parental exposure to enriched uranium induced
delayed hyperactivity in rat offspring. Neurotoxicology 28(1): 108-113 (2007).
H26 Houpert, P., S. Frelon, M. Monleau et al. Heterogeneous accumulation of uranium in the brain
of rats. Radiat Prot Dosim 127(1-4): 86-89 (2007).
H27 Hu, Q.Y. and S.P. Zhu. Induction of chromosomal aberrations in male mouse germ cells by
uranyl fluoride containing enriched uranium. Mutat Res 244(3): 209-214 (1990).
H28 Hu, Q.Y. and S.P. Zhu. Detection of DNA damage in spermiogenic stages of mice treated with
enriched uranyl fluoride by alkaline elution. Radiat Environ Biophys 29(3): 161-167 (1990).
I1 IAEA. Effects of ionising radiation on plants and animals at levels implied by current radiation
protection standards. Technical Reports Series No. 332. International Atomic Energy Agency,
Vienna, 1992.
I2 IAEA. Radiological conditions in areas of Kuwait with residues of depleted uranium.
Radiological assessment reports series. International Atomic Energy Agency, Vienna, 2003.
I3 IAEA. Handbook of parameter values for the prediction of radionuclide transfer to wildlife.
Technical Reports Series No. 479. International Atomic Energy Agency, Vienna, 2014.
I4 IAEA. Radiation protection and safety of radiation sources: International basic safety
standards. IAEA Safety Standards Series No. GSR Part 3. International Atomic Energy
Agency, Vienna, 2014.
I5 IARC. A review of human carcinogens. Part D: Radiation. IARC Monographs on the Evaluation
of Carcinogenic Risks to Humans, Volume 100D. International Agency for Research on Cancer,
Lyon, 2012.
I6 Ibanez, C., D. Suhard, C. Tessier et al. Intranasal exposure to uranium results in direct transfer
to the brain along olfactory nerve bundles. Neuropathol Appl Neurobiol 40(4): 477-488
(2014).
I7 ICRP. Dose coefficients for intakes of radionuclides by workers. ICRP Publication 68. Annals
of the ICRP 24(4). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1994.
I8 ICRP. Human respiratory tract model for radiological protection. ICRP Publication 66. Annals
of the ICRP 24(1-3). International Commission on Radiological Protection, Pergamon Press,
Oxford, 1994.
I9 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 4.
Inhalation dose coefficients. ICRP Publication 71. Annals of the ICRP 25(3-4). International
Commission on Radiological Protection, Pergamon Press, Oxford, 1995.
I10 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 3.
Ingestion dose coefficients. ICRP Publication 69. Annals of the ICRP 25(1). International
Commission on Radiological Protection, Pergamon Press, Oxford, 1995.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 481

I11 ICRP. Age-dependent doses to members of the public from intake of radionuclides: Part 5.
Compilation of ingestion and inhalation dose coefficients. ICRP Publication 72. Annals of the
ICRP 26(1). International Commission on Radiological Protection, Pergamon Press, Oxford,
1995.
I12 ICRP. Doses to the embryo and fetus from intakes of radionuclides by the mother. ICRP
Publication 88. Annals of the ICRP 31(1-3). International Commission on Radiological
Protection, Pergamon Press, Oxford, 2001.
I13 ICRP. Doses to infants from ingestion of radionuclides in mothers' milk. ICRP Publication 95.
Annals of the ICRP 34(3-4). International Commission on Radiological Protection, Elsevier
Ltd., 2004.
I14 ICRP. Human alimentary tract model for radiological protection. ICRP Publication 100. Annals
of the ICRP 36(1-2). International Commission on Radiological Protection, Elsevier Ltd., 2006.
I15 ICRP. The 2007 Recommendations of the International Commission on Radiological
Protection. ICRP Publication 103. Annals of the ICRP 37(2-4). International Commission on
Radiological Protection, Elsevier Ltd., 2007.
I16 ICRP. Lung cancer risk from radon and progeny and statement on radon. ICRP Publication
115. Annals of the ICRP 40(1). International Commission on Radiological Protection, Elsevier
Ltd., 2010.
I17 ICRP. Radiological protection against radon exposure. ICRP Publication 126. Annals of the
ICRP 43(3). International Commission on Radiological Protection, Elsevier Ltd., 2014.
I18 Igarashi, Y., A. Yamakawa and N. Ikeda. Plutonium and uranium in Japanese human tissues.
Radioisotopes 36(9): 433-439 (1987).
I19 Igarashi, Y., A. Yamakawa, C. Kim et al. Distribution of uranium in human lungs.
Radioisotopes 36(10): 501-504 (1987).
I20 IOM. Gulf War and Health: Updated Literature Review of Depleted Uranium. Institute of
Medicine, The National Academies Press, Washington, D.C., 2008.
I21 IRSN. Deuxime bilan 2008-2009 de la qualit radiologique des eaux du robinet en France. in:
La qualit radiologique de leau du robinet en France, 2008-2009. Institut de Radioprotection
et de Sret Nuclaire, 2011. (French).
I22 Isam Salih, M.M., H.B.L. Pettersson and E. Lund. Uranium and thorium series radionuclides
in drinking water from drilled bedrock wells: Correlation to geology and bedrock radioactivity
and dose estimation. Radiat Prot Dosim 102(3): 249-258 (2002).
I23 Israelsson, A. and H. Pettersson. Measurements of 234U and 238U in hair, urine, and drinking
water among drilled bedrock well water users for the evaluation of hair as a biomonitor of
Uranium intake. Health Phys 107(2): 143-149 (2014).
J1 Jadon, A. and R. Mathur. Gametogenic count and histopathological effect of thorium nitrate
and uranyl nitrate on mice testes. Andrologia 15(1): 40-43 (1983).
J2 Jia, G., M. Belli, U. Sansone et al. Concentration and characteristics of depleted uranium in
biological and water samples collected in Bosnia and Herzegovina. J Environ Radioact 89(2):
172-187 (2006).
J3 Joksi, A.S. and S.A. Katz. Efficacy of hair analysis for monitoring exposure to uranium: A
mini-review. J Environ Sci Health A Tox Hazard Subst Environ Eng 49(13): 1578-1587
(2014).
J4 Jovanovic, S.V., P. Pan and L. Wong. Bioaccessibility of uranium in soil samples from Port
Hope, Ontario, Canada. Environ Sci Technol 46(16): 9012-9018 (2012).
482 UNSCEAR 2016 REPORT

J5 Jowzaee, S. Determination of selected natural radionuclide concentrations in southwestern

caspian groundwater using liquid scintillation counting. Radiat Prot Dosim 157(2): 234-241
(2013).
J6 Julio, L.M.Q.C., D.R. Melo, W.D.O. Sousa et al. Uncertainty on faecal analysis on dose
assessment. Radiat Prot Dosim 127(1-4): 421-424 (2007).
K1 Kansal, S., R. Mehra and N.P. Singh. Uranium concentration in ground water samples
belonging to some areas of Western Haryana, India using fission track registration technique.
J Public Health Epidemiol 3(8): 352-357 (2011).
K2 Karpas, Z., O. Paz-Tal, A. Lorber et al. Urine, hair, and nails as indicators for ingestion of
uranium in drinking water. Health Phys 88(3): 229-242 (2005).
K3 Karpas, Z. Analytical Chemistry of Uranium: Environmental, Forensic, Nuclear, and
Toxicological Applications. CRC Press, Taylor & Francis Group, Boca Raton, 2015.
K4 Kathren, R.L. and R.H. Moore. Acute accidental inhalation of U: a 38-year follow-up. Health
Phys 51(5): 609-619 (1986).
K5 Kathren, R.L. and R.K. Burklin. Acute chemical toxicity of uranium. Health Phys 94(2):
170-179 (2008).
K6 Kathren, R.L. and S.Y. Tolmachev. Natural uranium tissue content of three Caucasian males.
Health Phys 109(3): 187-197 (2015).
K7 Katsoyiannis, I.A., S.J. Hug, A. Ammann et al. Arsenic speciation and uranium concentrations
in drinking water supply wells in Northern Greece: Correlations with redox indicative
parameters and implications for groundwater treatment. Sci Total Environ 383(1-3): 128-140
(2007).
K8 Kayzar, T.M., A.C. Villa, M.L. Lobaugh et al. Investigating uranium distribution in surface
sediments and waters: a case study of contamination from the Juniper Uranium Mine,
Stanislaus National Forest, CA. J Environ Radioact 136: 85-97 (2014).
K9 Konietzka, R. Gastrointestinal absorption of uranium compounds--a review. Regul Toxicol
Pharmacol 71(1): 125-133 (2015).
K10 Kozowska, B., A. Walencik and J. Dorda. Natural radioactivity and dose estimation in
underground water from the Sudety Mountains in Poland. Radiat Prot Dosim 128(3): 331-335
(2008).
K11 Krachler, M. and W. Shotyk. Trace and ultratrace metals in bottled waters: Survey of sources
worldwide and comparison with refillable metal bottles. Sci Total Environ 407(3): 1089-1096
(2009).
K12 Kraiem, M., S. Richter, N. Erdmann et al. Characterizing uranium oxide reference particles for
isotopic abundances and uranium mass by single particle isotope dilution mass spectrometry.
Anal Chim Acta 748: 37-44 (2012).
K13 Kramer, G.H., B.M. Hauck, S.A. Allen et al. Lung counting: summing techniques to reduce
the MDA. Health Phys 85(2): 220-227 (2003).
K14 Kreuzer, M., M. Kreisheimer, M. Kandel et al. Mortality from cardiovascular diseases in the
German uranium miners cohort study, 1946-1998. Radiat Environ Biophys 45(3): 159-166
(2006).
K15 Kreuzer, M., K. Straif, J.W. Marsh et al. Occupational dust and radiation exposure and
mortality from stomach cancer among German uranium miners, 1946-2003. Occup Environ
Med 69(3): 217-223 (2012).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 483

K16 Kreuzer, M., F. Dufey, M. Sogl et al. External gamma radiation and mortality from
cardiovascular diseases in the German WISMUT uranium miners cohort study, 1946-2008.
Radiat Environ Biophys 52(1): 37-46 (2013).
K17 Kreuzer, M., F. Dufey, J.W. Marsh et al. Mortality from cancers of the extra-thoracic airways
in relation to radon progeny in the Wismut cohort, 1946-2008. Int J Radiat Biol 90(11):
1030-1035 (2014).
K18 Kreuzer, M., F. Dufey, D. Laurier et al. Mortality from internal and external radiation
exposure in a cohort of male German uranium millers, 1946-2008. Int Arch Occup Environ
Health 88(4): 431-441 (2015).
K19 Kronfeld, J., D.I. Godfrey-Smith, D. Johannessen et al. Uranium series isotopes in the Avon
Valley, Nova Scotia. J Environ Radioact 73(3): 335-352 (2004).
K20 Krunic, A., S. Haveric and S. Ibrulj. Micronuclei frequencies in peripheral blood lymphocytes
of individuals exposed to depleted uranium. Arh Hig Rada Toksikol 56(3): 227-232 (2005).
K21 Kryscio, A., W.U. Ulrich Muller, A. Wojcik et al. A cytogenetic analysis of the long-term
effect of uranium mining on peripheral lymphocytes using the micronucleus-centromere assay.
Int J Radiat Biol 77(11): 1087-1093 (2001).
K22 Kumar, M., A. Kumar, S. Singh et al. Uranium content measurement in drinking water
samples using track etch technique. Radiat Meas 36: 479-481 (2003).
K23 Kundt, M., A.M. Ubios and R.L. Cabrini. Effects of uranium poisoning on cultured
preimplantation embryos. Biol Trace Elem Res 75(1-3): 235-244 (2000).
K24 Kundt, M.S., C. Martinez-Taibo, M.C. Muhlmann et al. Uranium in drinking water: Effects on
mouse oocyte quality. Health Phys 96(5): 568-574 (2009).
K25 Kupsh, C.C. and G.A. Robinson. Linearity of the accumulation of various dosages of uranium
in the major organs of mature male Japanese quail. Effect of various doses of estradiol-17 beta.
Comp Biochem Physiol C 92(1): 55-59 (1989).
K26 Kurttio, P., A. Auvinen, L. Salonen et al. Renal effects of uranium in drinking water. Environ
Health Perspect 110(4): 337-342 (2002).
K27 Kurttio, P., H. Komulainen, A. Leino et al. Bone as a possible target of chemical toxicity of
natural uranium in drinking water. Environ Health Perspect 113(1): 68-72 (2005).
K28 Kurttio, P., A. Harmoinen, H. Saha et al. Kidney toxicity of ingested uranium from drinking
water. Am J Kidney Dis 47(6): 972-982 (2006).
K29 Kurttio, P., L. Salonen, T. Ilus et al. Well water radioactivity and risk of cancers of the urinary
organs. Environ Res 102(3): 333-338 (2006).
K30 Kuwahara, C., K. Koyama and H. Sugiyama. Estimation of daily uranium ingestion by urban
residents in Japan. J Radioanal Nucl Chem 220(2): 161-165 (1997).
L1 L'Azou, B., M.H. Henge-Napoli, L. Minaro et al. Effects of cadmium and uranium on some in
vitro renal targets. Cell Biol Toxicol 18(5): 329-340 (2002).
L2 La Touche, Y.D., D.L. Willis and O.I. Dawydiak. Absorption and biokinetics of U in rats
following an oral administration of uranyl nitrate solution. Health Phys 53(2): 147-162 (1987).
L3 Labar, B., I. Rudan, D. Ivankovic et al. Haematological malignancies in childhood in Croatia:
Investigating the theories of depleted uranium, chemical plant damage and 'population mixing'.
Eur J Epidemiol 19(1): 55-60 (2004).
L4 LaCerte, C., H. Xie, A.M. Aboueissa et al. Particulate depleted uranium is cytotoxic and
clastogenic to human lung epithelial cells. Mutat Res 697(1-2): 33-37 (2010).
484 UNSCEAR 2016 REPORT

L5 Lagorio, S., E. Grande and L. Martina. Review of epidemiological studies of cancer risk
among Gulf War and Balkans veterans. Epidemiol Prev 32(3): 145-155 (2008). (Italian).
L6 Landa, E.R. and T.B. Councell. Leaching of uranium from glass and ceramic foodware and
decorative items. Health Phys 63(3): 343-348 (1992).
L7 Lane, R., P. Thompson, M. Ilin et al. Use of a weight of evidence approach to determine the
likelihood of adverse effects on human health from the presence of uranium facilities in Port
Hope, Ontario. J Environ Prot 2(9): 1149-1161 (2011).
L8 Lane, R.S., S.E. Frost, G.R. Howe et al. Mortality (1950-1999) and cancer incidence
(1969-1999) in the cohort of Eldorado uranium workers. Radiat Res 174(6): 773-785 (2010).
L9 Lang, S., V.M. Kosma, T. Kumlin et al. Distribution and short-term effects of intratracheally
instilled neutron-irradiated UO2 particles in the rat. Environ Res 65(1): 119-131 (1994).
L10 Langmuir, D. Uranium solution-mineral equilibria at low temperatures with applications to
sedimentary ore deposits. Geochim Cosmochim Acta 42(6): 547-569 (1978).
L11 Lariviere, D., A.P. Packer, L. Marro et al. Age dependence of natural uranium and thorium
concentrations in bone. Health Phys 92(2): 119-126 (2007).
L12 Larivire, D., S.Y. Tolmachev, V. Kochermin et al. Uranium bone content as an indicator of
chronic environmental exposure from drinking water. J Environ Radioact 121: 98-103 (2013).
L13 Laurent, O., M. Gomolka, R. Haylock et al. Concerted Uranium Research in Europe (CURE):
toward a collaborative project integrating dosimetry, epidemiology and radiobiology to study
the effects of occupational uranium exposure. J Radiol Prot 36(2): 319-345 (2016).
L14 Laurette, J., C. Larue, I. Llorens et al. Speciation of uranium in plants upon root accumulation
and root-to-shoot translocation. Environ Exp Bot 77: 87-95 (2012).
L15 Laurette, J., C. Larue, C. Mariet et al. Influence of uranium speciation on its accumulation and
translocation in three plant species: oilseed rape, sunflower and wheat. Environ Exp Bot 77: 96-
107 (2012).
L16 Laurier, D., I. Guseva Canu, S. Baatout et al. DoReMi workshop on multidisciplinary approaches
to evaluate cancer risks associated to low dose internal contamination. Radioprotection 47(1):
119-148 (2012).
L17 Laurier, D. Progress in understanding radon risk. pp.65-89 in: Radiation Protection No 168. EU
Scientific Seminar 2010 'Issues with internal emitters'. European Commission, Luxembourg, 2013.
L18 Leach, L.J., C.L. Yuile, H.C. Hodge et al. A five year inhalation study with natural uranium
dioxide (UO 2) dust. II. Postexposure retention and biologic effects in the monkey, dog and
rat. Health Phys 25(3): 239-258 (1973).
L19 Leach, L.J., R.M. Gelein, B.J. Panner et al. The acute toxicity of the hydrolysis products of
uranium hexafluoride (UF6) when inhaled by the rat and guinea pig. Final report. K/SUB-81-
9039/3; NTIS DE84011539. Rochester University Medical Center, New York, 1984.
L20 Lee, S.H., P.P. Povinec, E. Wyse et al. Ultra-low-level determination of 236U in IAEA marine
reference materials by ICPMS and AMS. Appl Radiat Isot 66(6-7): 823-828 (2008).
L21 Leggett, R.W. The behavior and chemical toxicity of U in the kidney: a reassessment. Health
Phys 57(3): 365-383 (1989).
L22 Leggett, R.W. Basis for the ICRP's age-specific biokinetic model for uranium. Health Phys
67(6): 589-610 (1994).
L23 Leggett, R.W. and J.D. Harrison. Fractional absorption of ingested uranium in humans. Health
Phys 68(4): 484-498 (1995).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 485

L24 Leggett, R.W., K.F. Eckerman and J.D. Boice, Jr. A respiratory model for uranium aluminide
based on occupational data. J Radiol Prot 25(4): 405-416 (2005).
L25 Leggett, R.W., K.F. Eckerman, C.W. McGinn et al. Controlling intake of uranium in the
workplace: applications of biokinetic modeling and occupational monitoring data. ORNL/TM-
2012/14. Oak Ridge National Laboratory, TN, USA, 2012.
L26 Legrand, M., C. Elie, J. Stefani et al. Cell proliferation and cell death are disturbed during
prenatal and postnatal brain development after uranium exposure. Neurotoxicology 52: 34-45
(2016).
L27 Lemercier, V., X. Millot, E. Ansoborlo et al. Study of uranium transfer across the blood-brain
barrier. Radiat Prot Dosim 105(1-4): 243-245 (2003).
L28 Leon Vintro, L., P.I. Mitchell, A. Omarova et al. Americium, plutonium and uranium
contamination and speciation in well waters, streams and atomic lakes in the Sarzhal region of
the Semipalatinsk Nuclear Test Site, Kazakhstan. J Environ Radioact 100(4): 308-314 (2009).
L29 Lestaevel, P., C. Bussy, F. Paquet et al. Changes in sleep-wake cycle after chronic exposure to
uranium in rats. Neurotoxicol Teratol 27(6): 835-840 (2005).
L30 Lestaevel, P., P. Houpert, C. Bussy et al. The brain is a target organ after acute exposure to
depleted uranium. Toxicology 212(2-3): 219-226 (2005).
L31 Lestaevel, P., E. Romero, B. Dhieux et al. Different pattern of brain pro-/anti-oxidant activity
between depleted and enriched uranium in chronically exposed rats. Toxicology 258(1): 1-9 (2009).
L32 Lestaevel, P., H. Bensoussan, B. Dhieux et al. Cerebral cortex and hippocampus respond
differently after post-natal exposure to uranium. J Toxicol Sci 38(5): 803-811 (2013).
L33 Lestaevel, P., F. Airault, R. Racine et al. Influence of environmental enrichment and depleted
uranium on behaviour, cholesterol and acetylcholine in apolipoprotein E-deficient mice. J Mol
Neurosci 53(3): 469-479 (2014).
L34 Li, W.B., Z. Karpas, L. Salonen et al. A compartmental model of uranium in human hair for
protracted ingestion of natural uranium in drinking water. Health Phys 96(6): 636-645 (2009).
L35 Lide, D.R. Uranium. pp.3132 in: Handbook of Chemistry and Physics. CRC Press, Boca
Raton, 1994.
L36 Lin, R.H., L.J. Wu, C.H. Lee et al. Cytogenetic toxicity of uranyl nitrate in Chinese hamster
ovary cells. Mutat Res 319(3): 197-203 (1993).
L37 Linares, V., M.L. Albina, M. Belles et al. Combined action of uranium and stress in the rat. II.
Effects on male reproduction. Toxicol Lett 158(3): 186-195 (2005).
L38 Linares, V., D.J. Sanchez, M. Belles et al. Pro-oxidant effects in the brain of rats concurrently
exposed to uranium and stress. Toxicology 236(1-2): 82-91 (2007).
L39 Lind, O.C., B. Salbu, L. Skipperud et al. Solid state speciation and potential bioavailability of
depleted uranium particles from Kosovo and Kuwait. J Environ Radioact 100(4): 301-307 (2009).
L40 Lizon, C. and P. Fritsch. Chemical toxicity of some actinides and lanthanides towards alveolar
macrophages: An in vitro study. Int J Radiat Biol 75(11): 1459-1471 (1999).
L41 Llobet, J.M., J.J. Sirvent, A. Ortega et al. Influence of chronic exposure to uranium on male
reproduction in mice. Fundam Appl Toxicol 16(4): 821-829 (1991).
L42 Lloyd, D.C., J.N. Lucas, A.A. Edwards et al. A study to verify a reported excess of
chromosomal aberrations in blood lymphocytes of Namibian uranium miners. Radiat Res
155(6): 809-817 (2001).
L43 LNHB. Mini table de Radionuclides. Laboratoire National Henri Becquerel, EDP Sciences,
2007. (French).
486 UNSCEAR 2016 REPORT

L44 Loomis, D.P. and S.H. Wolf. Mortality of workers at a nuclear materials production plant at
Oak Ridge, Tennessee, 1947-1990. Am J Ind Med 29(2): 131-141 (1996).
L45 Lopez-Abente, G., N. Aragones, M. Pollan et al. Leukemia, lymphomas, and myeloma
mortality in the vicinity of nuclear power plants and nuclear fuel facilities in Spain. Cancer
Epidemiol Biomarkers Prev 8(10): 925-934 (1999).
L46 Lopez-Abente, G., N. Aragones and M. Pollan. Solid-tumor mortality in the vicinity of
uranium cycle facilities and nuclear power plants in Spain. Environ Health Perspect 109(7):
721-729 (2001).
L47 Lopez, R., P.L. Diaz Sylvester, A.M. Ubios et al. Percutaneous toxicity of uranyl nitrate: its
effect in terms of exposure area and time. Health Phys 78(4): 434-437 (2000).
L48 Lu, S. and F.Y. Zhao. Nephrotoxic limit and annual limit on intake for natural U. Health Phys
58(5): 619-623 (1990).
L49 Lucas, H.F. and F. Markun. Thorium and uranium in blood, urine and cigarettes. pp.47-52 in:
ANL-7760. Argonne National Laboratory Radiation Physics Division Annual Report, Part 2.
Argonne National Laboratory, Argonne, 1970.
L50 Lytle, D.A., T. Sorg, L. Wang et al. The accumulation of radioactive contaminants in drinking
water distribution systems. Water Res 50: 396-407 (2014).
M1 Macfarlane, G.J., A.M. Biggs, N. Maconochie et al. Incidence of cancer among UK Gulf war
veterans: cohort study. Br Med J 327(7428): 1373 (2003).
M2 Macfarlane, G.J., M. Hotopf, N. Maconochie et al. Long-term mortality amongst Gulf War
Veterans: is there a relationship with experiences during deployment and subsequent
morbidity? Int J Epidemiol 34(6): 1403-1408 (2005).
M3 MacNider, W.D. A consideration of the relative toxicity of uranium nitrate for animals of
different ages. I. J Exp Med 26(1): 1-17 (1917).
M4 Maeng, H.J., W.S. Shim, S.J. Ahn et al. Differential changes in functional activity of organic
cation transporters in rats with uranyl nitrate-induced acute renal failure. Arch Pharm Res
35(8): 1441-1448 (2012).
M5 Magdo, H.S., J. Forman, N. Graber et al. Grand rounds: nephrotoxicity in a young child
exposed to uranium from contaminated well water. Environ Health Perspect 115(8):
1237-1241 (2007).
M6 Malard, V., O. Prat, E. Darrouzet et al. Proteomic analysis of the response of human lung cells
to uranium. Proteomics 5(17): 4568-4580 (2005).
M7 Malatova, I., V. Beckova, L. Tomasek et al. Reassessment of individual dosimetry of long-
lived alpha radionuclides of uranium miners through experimental determination of urinary
excretion of uranium. Radiat Prot Dosim 154(2): 198-206 (2013).
M8 Malatova, I., V. Beckova and L. Kotik. Urinary excretion of uranium in adult inhabitants of
the Czech Republic. J Environ Radioact 152: 92-96 (2016).
M9 Mao, Y., M. Desmeules, D. Schaubel et al. Inorganic components of drinking water and
microalbuminuria. Environ Res 71(2): 135-140 (1995).
M10 Marsh, J.W., Y. Bessa, A. Birchall et al. Dosimetric models used in the Alpha-Risk project to
quantify exposure of uranium miners to radon gas and its progeny. Radiat Prot Dosim 130(1):
101-106 (2008).
M11 Marsh, J.W., E. Blanchardon, D. Gregoratto et al. Dosimetric calculations for uranium miners
for epidemiological studies. Radiat Prot Dosim 149(4): 371-383 (2012).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 487

M12 Marsh, J.W., J.D. Harrison, D. Laurier et al. Doses and lung cancer risks from exposure to
radon and plutonium. Int J Radiat Biol 90(11): 1080-1087 (2014).
M13 Martin, F., R. Earl and E.J. Tawn. A cytogenetic study of men occupationally exposed to
uranium. Br J Ind Med 48(2): 98-102 (1991).
M14 Martn Snchez, A., M.P. Rubio Montero, V. Gmez Escobar et al. Radioactivity in bottled
mineral waters. Appl Radiat Isot 50(6): 1049-1055 (1999).
M15 Matsuba, M., T. Ishii, M. Nakahara et al. The concentrations of uranium in marine organisms.
Radioisotopes 49(7): 346-353 (2000).
M16 Maynard, E.A. and H.C. Hodge. Studies of the toxicity of various uranium compounds when
fed to experimental animals. pp.309-376 in: The Pharmacology and Toxicology of Uranium
Compounds, First edition (C. Voegtlin and H.C. Hodge, eds.). McGraw-Hill, New York, 1949.
M17 McClain, D.E., K.A. Benson, T.K. Dalton et al. Biological effects of embedded depleted
uranium (DU): summary of armed forces radiobiology research institute research. Sci Total
Environ 274(1-3): 115-118 (2001).
M18 McConnell, M.A., V.M. Sadagopa Ramanujam, N.W. Alcock et al. Distribution of uranium-
238 in environmental samples from a residential area impacted by mining and milling
activities. Environ Toxicol Chem 17(5): 841-850 (1998).
M19 McDiarmid, M.A., J.P. Keogh, F.J. Hooper et al. Health effects of depleted uranium on
exposed Gulf War veterans. Environ Res 82(2): 168-180 (2000).
M20 McDiarmid, M.A., S. Engelhardt, M. Oliver et al. Health effects of depleted uranium on
exposed Gulf War veterans: a 10-year follow-up. J Toxicol Environ Health A 67(4): 277-296
(2004).
M21 McDiarmid, M.A., S.M. Engelhardt, M. Oliver et al. Biological monitoring and surveillance
results of Gulf War I veterans exposed to depleted uranium. Int Arch Occup Environ Health
79(1): 11-21 (2006).
M22 McDiarmid, M.A., S.M. Engelhardt, M. Oliver et al. Health surveillance of Gulf War I
veterans exposed to depleted uranium: updating the cohort. Health Phys 93(1): 60-73 (2007).
M23 McDiarmid, M.A., S.M. Engelhardt, C.D. Dorsey et al. Surveillance results of depleted
uranium-exposed Gulf War I veterans: sixteen years of follow-up. J Toxicol Environ Health A
72(1): 14-29 (2009).
M24 McDiarmid, M.A., R.J. Albertini, J.D. Tucker et al. Measures of genotoxicity in Gulf War I
veterans exposed to depleted uranium. Environ Mol Mutagen 52(7): 569-581 (2011).
M25 McDiarmid, M.A., J.M. Gaitens, S. Hines et al. The Gulf War depleted uranium cohort at 20
years: bioassay results and novel approaches to fragment surveillance. Health Phys 104(4):
347-361 (2013).
M26 McGeoghegan, D. and K. Binks. The mortality and cancer morbidity experience of workers at
the Capenhurst uranium enrichment facility 1946-95. J Radiol Prot 20(4): 381-401 (2000).
M27 McGeoghegan, D. and K. Binks. The mortality and cancer morbidity experience of workers at
the Springfields uranium production facility, 1946-95. J Radiol Prot 20(2): 111-137 (2000).
M28 McGeoghegan, D. and K. Binks. The mortality and cancer morbidity experience of employees at
the Chapelcross plant of British Nuclear Fuels plc, 1955-95. J Radiol Prot 21(3): 221-250 (2001).
M29 McGeoghegan, D., K. Binks, M. Gillies et al. The non-cancer mortality experience of male
workers at British Nuclear Fuels plc, 1946-2005. Int J Epidemiol 37(3): 506-518 (2008).
488 UNSCEAR 2016 REPORT

M30 McGeoghegan, D., S. Whaley, K. Binks et al. Mortality and cancer registration experience of
the Sellafield workers known to have been involved in the 1957 Windscale accident: 50 year
follow-up. J Radiol Prot 30(3): 407-431 (2010).
M31 Meija, J., T.B. Coplen, M. Berglund et al. Atomic weights of the elements 2013 (IUPAC
technical report). Pure Appl Chem 88(3): 265-291 (2016).
M32 Meija, J., T.B. Coplen, M. Berglund et al. Isotopic compositions of the elements 2013 (IUPAC
technical report). Pure Appl Chem 88(3): 293306 (2016).
M33 Meszaros, G., G. Bognar and G.J. Koteles. Long-term persistence of chromosome aberrations
in uranium miners. J Occup Health 46(4): 310-315 (2004).
M34 Michard, A., C. Beaucaire and G. Michard. Uranium and rare earth elements in CO2-rich
waters from Vals-les-Bains (France)*. Geochim Cosmochim Acta 51(4): 901-909 (1987).
M35 Milacic, S., D. Petrovic, D. Jovicic et al. Examination of the health status of populations from
depleted-uranium-contaminated regions. Environ Res 95(1): 2-10 (2004).
M36 Milacic, S. Health investigations of depleted-uranium clean-up workers. Med Lav 99(5):
366-370 (2008).
M37 Milgram, S., M. Carriere, L. Malaval et al. Cellular accumulation and distribution of uranium
and lead in osteoblastic cells as a function of their speciation. Toxicology 252(1-3): 26-32
(2008).
M38 Milgram, S., M. Carriere, C. Thiebault et al. Cytotoxic and phenotypic effects of uranium and
lead on osteoblastic cells are highly dependent on metal speciation. Toxicology 250(1): 62-69
(2008).
M39 Miller, A.C., W.F. Blakely, D. Livengood et al. Transformation of human osteoblast cells to the
tumorigenic phenotype by depleted uranium-uranyl chloride. Environ Health Perspect 106(8):
465-471 (1998).
M40 Miller, A.C., M. Stewart, K. Brooks et al. Depleted uranium-catalyzed oxidative DNA
damage: absence of significant alpha particle decay. J Inorg Biochem 91(1): 246-252 (2002).
M41 Miller, A.C., J. Xu, M. Stewart et al. Potential late health effects of depleted uranium and
tungsten used in armor-piercing munitions: comparison of neoplastic transformation and
genotoxicity with the known carcinogen nickel. Mil Med 167(2 Suppl): 120-122 (2002).
M42 Miller, A.C., K. Brooks, M. Stewart et al. Genomic instability in human osteoblast cells after
exposure to depleted uranium: delayed lethality and micronuclei formation. J Environ
Radioact 64(2-3): 247-259 (2003).
M43 Miller, A.C., K. Brooks, J. Smith et al. Effect of the militarily-relevant heavy metals, depleted
uranium and heavy metal tungsten-alloy on gene expression in human liver carcinoma cells
(HepG2). Mol Cell Biochem 255(1-2): 247-256 (2004).
M44 Miller, A.C., C. Bonait-Pellie, R.F. Merlot et al. Leukemic transformation of hematopoietic
cells in mice internally exposed to depleted uranium. Mol Cell Biochem 279(1-2): 97-104
(2005).
M45 Miller, A.C., M. Stewart and R. Rivas. Preconceptional paternal exposure to depleted uranium:
Transmission of genetic damage to offspring. Health Phys 99(3): 371-379 (2010).
M46 Mirderikvand, N., B. Mohammadzadeh Asl, P. Naserzadeh et al. Embryo toxic effects of
depleted uranium on the morphology of the mouse fetus. Iran J Pharm Res 13(1): 199-206
(2014).
M47 Mirto, H., M.H. Henge-Napoli, R. Gibert et al. Intracellular behaviour of uranium(VI) on renal
epithelial cell in culture (LLC-PK1): influence of uranium speciation. Toxicol Lett 104(3):
249-256 (1999).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 489

M48 Misdaq, M.A. and A. Chaouqi. 238U, 232Th, 222Rn, and 220Rn concentrations measured in
various bottled mineral waters and resulting radiation doses to the members of the European
population living in the City of Marrakech (Morocco). Health Phys 94(3): 279-291 (2008).
M49 Mitchel, R.E.J., J.S. Jackson and B. Heinmiller. Inhaled uranium ore dust and lung cancer risk
in rats. Health Phys 76(2): 145-155 (1999).
M50 Mohner, M., M. Lindtner, H. Otten et al. Leukemia and exposure to ionizing radiation among
German uranium miners. Am J Ind Med 49(4): 238-248 (2006).
M51 Mohner, M., M. Lindtner and H. Otten. Ionizing radiation and risk of laryngeal cancer among
German uranium miners. Health Phys 95(6): 725-733 (2008).
M52 Mohner, M., J. Gellissen, J.W. Marsh et al. Occupational and diagnostic exposure to ionizing
radiation and leukemia risk among German uranium miners. Health Phys 99(3): 314-321 (2010).
M53 Mollegaard, N.E., A.I. Murchie, D.M. Lilley et al. Uranyl photoprobing of a four-way DNA
junction: evidence for specific metal ion binding. EMBO J 13(7): 1508-1513 (1994).
M54 Monleau, M., C. Bussy, P. Lestaevel et al. Bioaccumulation and behavioural effects of
depleted uranium in rats exposed to repeated inhalations. Neurosci Lett 390(1): 31-36 (2005).
M55 Monleau, M., M. De Meo, S. Frelon et al. Distribution and genotoxic effects after successive
exposure to different uranium oxide particles inhaled by rats. Inhal Toxicol 18(11): 885-894 (2006).
M56 Monleau, M., M. De Meo, F. Paquet et al. Genotoxic and inflammatory effects of depleted
uranium particles inhaled by rats. Toxicol Sci 89(1): 287-295 (2006).
M57 Moon, Y.J., A.K. Lee, H.C. Chung et al. Effects of acute renal failure on the pharmacokinetics
of chlorzoxazone in rats. Drug Metab Dispos 31(6): 776-784 (2003).
M58 Moore, J., M. Limson-Zamora, L. Whiteduck et al. Elevated uranium levels in groundwater
and the potential bioeffects. Presented at the 7th National Conference on Drinking Water,
Charlottetown, 10-13 August 1996.
M59 Morris, K.J., K.M. Townsend and A.L. Batchelor. Studies of alveolar cell morphometry and
mass clearance in the rat lung following inhalation of an enriched uranium dioxide aerosol.
Radiat Environ Biophys 28(2): 141-154 (1989).
M60 Morris, K.J., P. Khanna and A.L. Batchelor. Long-term clearance of inhaled UO2 particles
from the pulmonary region of the rat. Health Phys 58(4): 477-485 (1990).
M61 Morris, K.J., C.L. Barker, A.L. Batchelor et al. Dosimetric implications of pulmonary
macrophage clusters observed within lungs of rats that have inhaled enriched UO2 particles.
Environ Health Perspect 97: 201-208 (1992).
M62 Morrow, P., R. Gelein, H. Beiter et al. Inhalation and intravenous studies of UF6/UO2F2 in
dogs. Health Phys 43(6): 859-873 (1982).
M63 Morrow, P.E., L.J. Leach, F.A. Smith et al. Metabolic fate and evaluation of injury in rats and
dogs following exposure to the hydrolysis products of uranium hexafluoride: implications for a
bioassay program related to potential releases of uranium hexafluoride, July 1979-October
1981. NUREG/CR-2268. U.S. Nuclear Regulatory Commission, Washington, D.C., 1982.
M64 Moss, M.A., R.F. McCurdy, K.C. Dooley et al. Uranium in drinking water - report on clinical
studies in Nova Scotia. pp.149-152 in: Chemical Toxicology and Clinical Chemistry of Metals
(S.S. Brown and J. Savory, eds.). Academic Press, London, 1983.
M65 Moss, M.A. Chronic low level uranium exposure via drinking water - clinical investigations in
Nova Scotia. M.Sc. Thesis, Dalhousie University, Halifax (1985).
490 UNSCEAR 2016 REPORT

M66 Muikku, M., T. Heikkinen, M. Puhakainen et al. Assessment of occupational exposure to

uranium by indirect methods needs information on natural background variations. Radiat Prot
Dosim 125(1-4): 492-495 (2007).
M67 Muikku, M., M. Puhakainen, T. Heikkinen et al. The mean concentration of uranium in
drinking water, urine, and hair of the occupationally unexposed Finnish working population.
Health Phys 96(6): 646-654 (2009).
M68 Muller, D., P. Houpert, J. Cambar et al. Role of the sodium-dependent phosphate co-
transporters and of the phosphate complexes of uranyl in the cytotoxicity of uranium in
LLC-PK1 cells. Toxicol Appl Pharmacol 214(2): 166-177 (2006).
M69 Muller, D.S., P. Houpert, J. Cambar et al. Role of the sodium-dependent phosphate
cotransporters and absorptive endocytosis in the uptake of low concentrations of uranium and
its toxicity at higher concentrations in LLC-PK1 cells. Toxicol Sci 101(2): 254-262 (2008).
M70 Muller, W.U., A. Kryscio and C. Streffer. Micronuclei in lymphocytes of uranium miners of
the former Wismut SDAG. Cytogenet Genome Res 104(1-4): 295-298 (2004).
N1 National Research Council Committee on Health Risks of Exposure to Radon. Health Effects
of Exposure to Radon: BEIR VI. National Academies Press, Washington, D.C., 1999.
N2 NCRP. Exposure of the population in the United States and Canada from natural background
radiation. NCRP Report No. 94. National Council on Radiation Protection and Measurements,
Bethesda, 1988.
N3 NCRP. Development of a biokinetic model for radionuclide contaminated wounds and
procedures for their assessment, dosimetry and treatment. NCRP Report No. 156. National
Council on Radiation Protection and Measurements, Bethesda, 2006.
N4 Newey, H., P.A. Sanford and D.H. Smyth. The effect of uranyl nitrate on intestinal transfer of
hexoses. J Physiol 186(3): 493-502 (1966).
N5 NHMRC and NRMMC. Australian drinking water guidelines paper 6. National water quality
management strategy. National Health and Medical Research Council, National Resource
Management Ministerial Council, Commonwealth of Canberra, Australia, 2011.
N6 Nicklas, J.A., R.J. Albertini, P.M. Vacek et al. Mutagenicity monitoring following battlefield
exposures: Molecular analysis of HPRT mutations in Gulf War I veterans exposed to depleted
uranium. Environ Mol Mutagen 56(7): 594-608 (2015).
N7 Nielsen, P.E., N.E. Mollegaard and C. Jeppesen. DNA conformational analysis in solution by
uranyl mediated photocleavage. Nucleic Acids Res 18(13): 3847-3851 (1990).
N8 NNDC. Chart of nuclides. National Nuclear Data Center. [Internet] Available from
(http://www.nndc.bnl.gov/chart/) on 12.03.2014.
N9 Nozaki, T., M. Ichikawa, T. Sasuga et al. Neutron activation analysis of uranium in human
bone, drinking water and daily diet. J Radioanal Chem 6(1): 33-40 (1970).
O1 Oeh, U., N.D. Priest, P. Roth et al. Measurements of daily urinary uranium excretion in
German peacekeeping personnel and residents of the Kosovo region to assess potential intakes
of depleted uranium (DU). Sci Total Environ 381(1-3): 77-87 (2007).
O2 Okogbue, C.O. and S.N. Ukpai. Evaluation of trace element contents in groundwater in
Abakaliki metropolis and around the abandoned mine sites in the southern part, Southeastern
Nigeria. Environ Earth Sci 70(7): 3351-3362 (2013).
O3 OME. Rationale for development of Ontario air standards for uranium and uranium
compounds. Standards Development Branch, Ontario Ministry of Environment, 2009.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 491

O4 Orcutt, J.A. The toxicology of compounds of uranium following application to the skin. Chapter
VIII in: Pharmacology and Toxicology of Uranium Compounds (J.A. Orcutt, ed.). McGraw-Hill,
New York, 1949.
O5 Orloff, K.G., K. Mistry, P. Charp et al. Human exposure to uranium in groundwater. Environ
Res 94(3): 319-326 (2004).
O6 Orona, N.S. and D.R. Tasat. Uranyl nitrate-exposed rat alveolar macrophages cell death:
Influence of superoxide anion and TNF mediators. Toxicol Appl Pharmacol 261(3): 309-316
(2012).
O7 Ortega, A., J.L. Domingo, J.M. Llobet et al. Evaluation of the oral toxicity of uranium in a
4-week drinking-water study in rats. Bull Environ Contam Toxicol 42(6): 935-941 (1989).
O8 Ortega, X., I. Valls and I. Serrano. Natural radioactivity in drinking water in Catalonia
(Spain). Environ Int 22(Suppl 1): 347-354 (1996).
O9 Osmond, J.K., J.B. Cowart and M. Ivanovich. Uranium isotopic disequilibrium in ground
water as an indicator of anomalies. Int J Appl Radiat Isot 34(1): 283-308 (1983).
O10 Ozkaynak, H., J. Xue, V.G. Zartarian et al. Modeled estimates of soil and dust ingestion rates
for children. Risk Anal 31(4): 592-608 (2011).
O11 Ozmen, M. and M. Yurekli. Subacute toxicity of uranyl acetate in Swiss-Albino mice. Environ
Toxicol Pharmacol 6(2): 111-115 (1998).
P1 Palazzuoli, A., S. Masson, C. Ronco et al. Clinical relevance of biomarkers in heart failure and
cardiorenal syndrome: the role of natriuretic peptides and troponin. Heart Fail Rev 19(2):
267-284 (2014).
P2 Pan, Z. and S. Liu. Radiation Level in China. China Atomic Energy Publishing and Media Co.,
Ltd., 2011.
P3 Paquet, F., P. Houpert, E. Blanchardon et al. Accumulation and distribution of uranium in rats
after chronic exposure by ingestion. Health Phys 90(2): 139-147 (2006).
P4 Paquet, F., G. Etherington, M.R. Bailey et al. ICRP Publication 130: Occupational intakes of
radionuclides: Part 1. Ann ICRP 44(2): 5-188 (2015).
P5 Parrish, R.R., M. Horstwood, J.G. Arnason et al. Depleted uranium contamination by
inhalation exposure and its detection after 20 years: Implications for human health
assessment. Sci Total Environ 390(1): 58-68 (2008).
P6 Pasanen, M., S. Lang, A. Kojo et al. Effects of simulated nuclear fuel particles on the
histopathology and CYP enzymes in the rat lung and liver. Environ Res 70(2): 126-133 (1995).
P7 Paternain, J.L., J.L. Domingo, A. Ortega et al. The effects of uranium on reproduction,
gestation, and postnatal survival in mice. Ecotoxicol Environ Saf 17(3): 291-296 (1989).
P8 Patra, A.C., S. Mohapatra, S.K. Sahoo et al. Age-dependent dose and health risk due to intake
of uranium in drinking water from Jaduguda, India. Radiat Prot Dosim 155(2): 210-216
(2013).
P9 Pavlakis, N., C.A. Pollock, G. McLean et al. Deliberate overdose of uranium: toxicity and
treatment. Nephron 72(2): 313-317 (1996).
P10 Pelayo, J.C., P.M. Andrews, A.K. Coffey et al. The influence of age on acute renal toxicity of
uranyl nitrate in the dog. Pediatr Res 17(12): 985-992 (1983).
P11 Pellmar, T.C., A.F. Fuciarelli, J.W. Ejnik et al. Distribution of uranium in rats implanted with
depleted uranium pellets. Toxicol Sci 49(1): 29-39 (1999).
492 UNSCEAR 2016 REPORT

P12 Pellmar, T.C., D.O. Keyser, C. Emery et al. Electrophysiological changes in hippocampal
slices isolated from rats embedded with depleted uranium fragments. Neurotoxicology 20(5):
785-792 (1999).
P13 Periyakaruppan, A., F. Kumar, S. Sarkar et al. Uranium induces oxidative stress in lung
epithelial cells. Arch Toxicol 81(6): 389-395 (2007).
P14 Pernot, E., J. Hall, S. Baatout et al. Ionizing radiation biomarkers for potential use in
epidemiological studies. Mutat Res 751(2): 258-286 (2012).
P15 Petitot, F., A.M. Moreels and F. Paquet. In vitro evaluation of percutaneous diffusion of uranyl
nitrate through intact or excoriated skin of rat and pig. Can J Physiol Pharmacol 82(2):
133-139 (2004).
P16 Petitot, F., S. Frelon, A.M. Moreels et al. Incorporation and distribution of uranium in rats
after a contamination on intact or wounded skin. Health Phys 92(5): 464-474 (2007).
P17 Petitot, F., C. Gautier, A.M. Moreels et al. Percutaneous penetration of uranium in rats after a
contamination on intact or wounded skin. Radiat Prot Dosim 127(1-4): 125-130 (2007).
P18 Petitot, F., P. Lestaevel, E. Tourlonias et al. Inhalation of uranium nanoparticles: respiratory tract
deposition and translocation to secondary target organs in rats. Toxicol Lett 217(3): 217-225 (2013).
P19 Pietrzak-Flis, Z., M.M. Suplinska and L. Rosiak. The dietary intake of 238U, 234U, 230Th, 232Th,
228
Th and 226Ra from food and drinking water by inhabitants of the Wallbrzych region.
J Radioanal Nucl Chem 222(1-2): 189-193 (1997).
P20 Pietrzak-Flis, Z., L. Rosiak, M.M. Suplinska et al. Daily intakes of 238U, 234U, 232Th, 230Th, 228Th
and 226Ra in the adult population of central Poland. Sci Total Environ 273(1-3): 163-169 (2001).
P21 Pinkerton, L.E., T.F. Bloom, M.J. Hein et al. Mortality among a cohort of uranium mill
workers: an update. Occup Environ Med 61(1): 57-64 (2004).
P22 Pinney, S.M., R.W. Freyberg, G.E. Levine et al. Health effects in community residents near a
uranium plant at Fernald, Ohio, USA. Int J Occup Med Environ Health 16(2): 139-153 (2003).
P23 Poisson, C., J. Stefani, L. Manens et al. Chronic uranium exposure dose-dependently induces
glutathione in rats without any nephrotoxicity. Free Radic Res 48(10): 1218-1231 (2014).
P24 Polednak, A.P. and E.L. Frome. Mortality among men employed between 1943 and 1947 at a
uranium-processing plant. J Occup Med 23(3): 169-178 (1981).
P25 Popp, W., U. Plappert, W.U. Muller et al. Biomarkers of genetic damage and inflammation in
blood and bronchoalveolar lavage fluid among former German uranium miners: a pilot study.
Radiat Environ Biophys 39(4): 275-282 (2000).
P26 Pourahmad, J., M. Ghashang, H.A. Ettehadi et al. A search for cellular and molecular
mechanisms involved in depleted uranium (DU) toxicity. Environ Toxicol 21(4): 349-354
(2006).
P27 Prat, O., F. Berenguer, V. Malard et al. Transcriptomic and proteomic responses of human
renal HEK293 cells to uranium toxicity. Proteomics 5(1): 297-306 (2005).
P28 Prat, O., T. Vercouter, E. Ansoborlo et al. Uranium speciation in drinking water from drilled
wells in southern Finland and its potential links to health effects. Environ Sci Technol 43(10):
3941-3946 (2009).
P29 Prat, O., F. Berenguer, G. Steinmetz et al. Alterations in gene expression in cultured human
cells after acute exposure to uranium salt: Involvement of a mineralization regulator. Toxicol
In Vitro 24(1): 160-168 (2010).
P30 Prat, O., E. Ansoborlo, N. Sage et al. From cell to man: evaluation of osteopontin as a possible
biomarker of uranium exposure. Environ Int 37(4): 657-662 (2011).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 493

P31 Priest, N.D., G.R. Howells, D. Green et al. Uranium in bone: metabolic and autoradiographic
studies in the rat. Hum Toxicol 1(2): 97-114 (1982).
P32 Province of British Columbia. Variations in uranium and radioactivity levels in surface and
ground water at selected sites in British Columbia, April 1980-March 1981. Ministry of Energy,
Mines and Petroleum Resources and Ministry of Health, British Columbia, Victoria, 1981.
P33 Pujadas Bigi, M.M., L. Lemlich, P.M. Mandalunis et al. Exposure to oral uranyl nitrate delays
tooth eruption and development. Health Phys 84(2): 163-169 (2003).
P34 Pujadas Bigi, M.M. and A.M. Ubios. Catch-up of delayed tooth eruption associated with
uranium intoxication. Health Phys 92(4): 345-348 (2007).
Q1 Qi, L., C. Basset, O. Averseng et al. Characterization of UO2 2+ binding to osteopontin, a
highly phosphorylated protein: Insights into potential mechanisms of uranyl accumulation in
bones. Metallomics 6(1): 166-176 (2014).
Q2 Qin, S., S. Li and et al. Concentration factors of 90Sr, 137Cs, 210Po, natural uranium and 226Ra,
radionuclides in aquatic organisms in freshwater environment. Fushe Fanghu 15(5): 330-337
(1995). (Chinese).
R1 Racine, R., Y. Gueguen, P. Gourmelon et al. Modifications of the expression of genes
involved in cerebral cholesterol metabolism in the rat following chronic ingestion of depleted
uranium. J Mol Neurosci 38(2): 159-165 (2009).
R2 Racine, R., L. Grandcolas, S. Grison et al. Cholesterol 7alpha-hydroxylase (CYP7A1) activity
is modified after chronic ingestion of depleted uranium in the rat. J Steroid Biochem Mol Biol
120(1): 60-66 (2010).
R3 Radespiel-Trger, M. and M. Meyer. Association between drinking water uranium content and
cancer risk in Bavaria, Germany. Int Arch Occup Environ Health 86(7): 767-776 (2013).
R4 Rage, E., B. Vacquier, E. Blanchardon et al. Risk of lung cancer mortality in relation to lung
doses among French uranium miners: follow-up 1956-1999. Radiat Res 177(3): 288-297 (2012).
R5 Rage, E., S. Caer-Lorho, D. Drubay et al. Mortality analyses in the updated French cohort of
uranium miners (1946-2007). Int Arch Occup Environ Health 20: 20 (2014).
R6 Ramon Ramirez, J., P. Enriquez, R. Castillo et al. MOX fuel use in a BWR with extended
power up-rate. Ann Nucl Energy 50: 63-70 (2012).
R7 Rana, B.K., R.M. Tripathi, S.K. Sahoo et al. Assessment of natural uranium and 226Ra
concentration in ground water around the uranium mine at Narwapahar, Jharkhand, India and
its radiological significance. J Radioanal Nucl Chem 285(3): 711-717 (2010).
R8 Rani, A., S. Singh, V. Duggal et al. Uranium estimation in drinking water samples from some
areas of punjab and Himachal Pradesh, India using ICP-MS. Radiat Prot Dosim 157(1):
146-151 (2013).
R9 Raymond-Whish, S., L.P. Mayer, T. O'Neal et al. Drinking water with uranium below the U.S.
EPA water standard causes estrogen receptor-dependent responses in female mice. Environ
Health Perspect 115(12): 1711-1716 (2007).
R10 Reimann, C., K. Bjorvatn, B. Frengstad et al. Drinking water quality in the Ethiopian section of
the East African Rift Valley I--data and health aspects. Sci Total Environ 311(1-3): 65-80 (2003).
R11 Renaud, P. Doses lies l'incorporation de radionuclides naturels par l'alimentation en France
mtropolitaine. PRP-ENV-SESURE/2014-18. Rapport Ple Radioprotection, Environnement,
Dchets et Crise, Service d'tude et de surveillance de la radioactivit dans l'environnement,
2014. (French).
R12 Richardson, D.B. and S. Wing. Lung cancer mortality among workers at a nuclear materials
fabrication plant. Am J Ind Med 49(2): 102-111 (2006).
494 UNSCEAR 2016 REPORT

R13 Richardson, G.M. and Stantec Consulting Ltd. 2013 Canadian exposure factors handbook.
Toxicology Centre, University of Saskatchewan, Saskatoon, SK Canada, 2013.
R14 Ritz, B. Radiation exposure and cancer mortality in uranium processing workers.
Epidemiology 10(5): 531-538 (1999).
R15 Ritz, B. Cancer mortality among workers exposed to chemicals during uranium processing. J
Occup Environ Med 41(7): 556-566 (1999).
R16 Ritz, B., H. Morgenstern, D. Crawford-Brown et al. The effects of internal radiation exposure
on cancer mortality in nuclear workers at Rocketdyne/Atomics International. Environ Health
Perspect 108(8): 743-751 (2000).
R17 Robertson, R. and D.C. Stuart. Physical characteristics and solubility of long-lived airborne
particulates in uranium producing and manufacturing facilities; Phase IV-Part III. INFO-0363-9.
Atomic Energy Control Board, Ottawa, Canada, 1995.
R18 Rodrigues, G., J.D. Arruda-Neto, R.M. Pereira et al. Uranium deposition in bones of Wistar
rats associated with skeleton development. Appl Radiat Isot 82: 1056110 (2013).
R19 Rodriguez-Yoldi, M.J. and F. Ponz. Uranyl action on sugar transport across rat jejunum. Rev
Esp Fisiol 42(3): 359-364 (1986).
R20 Roessler, C.E., Z.A. Smith, W.E. Bolch et al. Uranium and radium-226 in Florida phosphate
materials. Health Phys 37(3): 269-277 (1979).
R21 Ronco, C., C.Y. Chionh, M. Haapio et al. The cardiorenal syndrome. Blood Purif 27(1): 114-126
(2009).
R22 Rooney, C., V. Beral, N. Maconochie et al. Case-control study of prostatic cancer in
employees of the United Kingdom Atomic Energy Authority. Br Med J 307(6916): 1391-1397
(1993).
R23 Rossiter, H.M., P.A. Owusu, E. Awuah et al. Chemical drinking water quality in Ghana: water
costs and scope for advanced treatment. Sci Total Environ 408(11): 2378-2386 (2010).
R24 Roszell, L.E., F.F. Hahn, R.B. Lee et al. Assessing the renal toxicity of Capstone depleted
uranium oxides and other uranium compounds. Health Phys 96(3): 343-351 (2009).
R25 Rouas, C., M. Souidi, L. Grandcolas et al. Acetaminophen induces xenobiotic-metabolizing
enzymes in rat: Impact of a uranium chronic exposure. Environ Toxicol Pharmacol 28(3):
363-369 (2009).
R26 Rouas, C., H. Bensoussan, D. Suhard et al. Distribution of soluble uranium in the nuclear cell
compartment at subtoxic concentrations. Chem Res Toxicol 23(12): 1883-1889 (2010).
R27 Rouas, C., J. Stefani, S. Grison et al. Effect of nephrotoxic treatment with gentamicin on rats
chronically exposed to uranium. Toxicology 279(1-3): 27-35 (2011).
R28 Royal Society. The health hazards of depleted uranium munitions. Part II. The Royal Society,
London, 2002.
R29 Romari, M., M. Rogi, L. Benedik et al. Natural radionuclides in bottled drinking waters
produced in Croatia and their contribution to radiation dose. Sci Total Environ 437: 53-60
(2012).
R30 Rusconi, R., M. Forte, G. Abbate et al. Natural radioactivity in bottled mineral waters: A
survey in Northern Italy. J Radioanal Nucl Chem 260(2): 421-427 (2004).
R31 Russell, J.J. and R.L. Kathren. Uranium deposition and retention in a USTUR whole body
case. Health Phys 86(3): 273-284 (2004).
S1 Samaropoulos, I., M. Efstathiou, I. Pashalidis et al. Determination of uranium concentration in
ground water samples of Northern Greece. EPJ Web of Conferences 24(03005): 2012.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 495

S2 Sansone, U., L. Stellato, G. Jia et al. Levels of depleted uranium in Kosovo soils. Radiat Prot
Dosim 97(4): 317-320 (2001).
S3 Santos, E.E., D.C. Lauria, E.C.S. Amaral et al. Daily ingestion of 232Th, 238U, 226Ra, 228Ra and
210
Pb in vegetables by inhabitants of Rio de Janeiro City. J Environ Radioact 62(1): 75-86
(2002).
S4 Schnell, H. Uranium processing practices, innovations, and trends. pp.457-465 in: Mineral
Processing and Extractive Metallurgy: 100 Years of Innovation. Society for Mining,
Metallurgy and Exploration (SME), Colorado, 2014.
S5 Schubauer-Berigan, M.K., R.D. Daniels and L.E. Pinkerton. Radon exposure and mortality
among white and American Indian uranium miners: an update of the Colorado Plateau cohort.
Am J Epidemiol 169(6): 718-730 (2009).
S6 Scientific Advisory Committee on Veterans Health. Depleted uranium and Canadian veterans
- a review of potential exposure and health effects. Ministry of Veterans Affairs, Ottawa, 2013.
S7 Seiler, R.L. Temporal changes in water quality at a childhood leukemia cluster. Ground Water
42(3): 446-455 (2004).
S8 Selden, A.I., C. Lundholm, B. Edlund et al. Nephrotoxicity of uranium in drinking water from
private drilled wells. Environ Res 109(4): 486-494 (2009).
S9 Shaki, F., M.J. Hosseini, M. Ghazi-Khansari et al. Toxicity of depleted uranium on isolated rat
kidney mitochondria. Biochim Biophys Acta 1820(12): 1940-1950 (2012).
S10 Shawky, S., H.A. Amer, M.I. Hussein et al. Uranium bioassay and radioactive dust
measurements at some uranium processing sites in Egypt--health effects. J Environ Monit 4(4):
588-591 (2002).
S11 Shenoy, N.S., A. Verma, S.A. Kumar et al. A comparative analysis of uranium in potable waters
using laser fluorimetry and ICPMS techniques. J Radioanal Nucl Chem 294(3): 413-417 (2012).
S12 Shim, C.K., Y. Sawada, T. Iga et al. Estimation of renal secretory function for organic cations by
endogenous N1-methylnicotinamide in rats with experimental renal failure. J Pharmacokinet
Biopharm 12(1): 23-42 (1984).
S13 Shim, W.S., J.H. Park, S.J. Ahn et al. Testosterone-independent down-regulation of Oct2 in the
kidney medulla from a uranyl nitrate-induced rat model of acute renal failure: effects on
distribution of a model organic cation, tetraethylammonium. J Pharm Sci 98(2): 739-747 (2009).
S14 Shiquan, S., Z. Sizhong and Y. Zhanyun. Pathological studies of uranium compound
intoxication. Nucl Prot 3: 1-54 (1976). (Chinese).
S15 Shleien, B., L.A. Slaback Jr. and B.K. Birky. Handbook of Health Physics and Radiological
Health. Third edition. Lippincott Williams & Wilkins, Baltimore, Maryland, 1998.
S16 Shude, L. Evaluation of uranium burden of a case with soluble uranium compound overdose
intake. Nucl Prot 3: 71-81 (1976). (Chinese).
S17 Sikov, M.R. and D.D. Mahlum. Cross-placental transfer of selected actinides in the rat. Health
Phys 14(3): 205-208 (1968).
S18 Silva Filho, C.A., E.J. Frana, E.M. Souza et al. Radioactive risk evaluation of mineral water
in the Metropolitan Region of Recife, Northeastern Brazil. J Radioanal Nucl Chem 295(2):
1215-1220 (2013).
S19 Silver, S.R., S.J. Bertke, M.J. Hein et al. Mortality and ionising radiation exposures among
workers employed at the Fernald Feed Materials Production Center (1951-1985). Occup
Environ Med 70(7): 453-463 (2013).
496 UNSCEAR 2016 REPORT

S20 Simon, O., E. Mottin, B. Geffroy et al. Effects of dietary uranium on reproductive endpoints--
fecundity, survival, reproductive success--of the fish Danio rerio. Environ Toxicol Chem
30(1): 220-225 (2011).
S21 Simon, S.L., F.O. Hoffman and E. Hofer. The two-dimensional Monte Carlo: a new
methodologic paradigm for dose reconstruction for epidemiological studies. Radiat Res 183(1):
27-41 (2015).
S22 Singh, J., L. Singh and S. Kher. A comparison of fission track and laser fluorometry
techniques for uranium analysis in water samples. Radiat Meas 36(1-6): 517-519 (2003).
S23 Singh, J., H. Singh, S. Singh et al. Estimation of uranium and radon concentration in some
drinking water samples of Upper Siwaliks, India. Environ Monit Assess 154(1-4): 15-22 (2009).
S24 Singh, N.P., D.B. Bennett and M.E. Wrenn. Macro-distribution of naturally occurring alpha-
emitting isotopes of U in the human skeleton. Health Phys 52(6): 769-773 (1987).
S25 Singh, N.P. Thorium, uranium and plutonium in human tissues of world-wide general
population. J Radioanal Nucl Chem 138(2): 347-364 (1990).
S26 Singh, N.P., D.P. Burleigh, H.M. Ruth et al. Daily U intake in Utah residents from food and
drinking water. Health Phys 59(3): 333-337 (1990).
S27 Singh, S., R. Malhotra, J. Kumar et al. Uranium analysis of geological samples, water and
plants from Kulu area, Himachal Pradesh, India. Radiat Meas 34(1-6): 427-431 (2001).
S28 Skipperud, L., G. Strmman, M. Yunusov et al. Environmental impact assessment of
radionuclide and metal contamination at the former U sites Taboshar and Digmai, Tajikistan.
J Environ Radioact 123: 50-62 (2013).
S29 Souidi, M., Y. Gueguen, C. Linard et al. In vivo effects of chronic contamination with depleted
uranium on CYP3A and associated nuclear receptors PXR and CAR in the rat. Toxicology
214(1-2): 113-122 (2005).
S30 Spencer, H., D. Osis, I.M. Fisenne et al. Measured intake and excretion patterns of naturally
occurring 234U, 238U, and calcium in humans. Radiat Res 124(1): 90-95 (1990).
S31 Squibb, K.S., R.W. Leggett and M.A. McDiarmid. Prediction of renal concentrations of depleted
uranium and radiation dose in Gulf War veterans with embedded shrapnel. Health Phys 89(3):
267-273 (2005).
S32 Squibb, K.S. and M.A. McDiarmid. Depleted uranium exposure and health effects in Gulf War
veterans. Philos Trans R Soc Lond B Biol Sci 361(1468): 639-648 (2006).
S33 Stalder, E., A. Blanc, M. Haldimann et al. Occurrence of uranium in Swiss drinking water.
Chemosphere 86(6): 672-679 (2012).
S34 Stearns, D.M., M. Yazzie, A.S. Bradley et al. Uranyl acetate induces hprt mutations and
uranium-DNA adducts in Chinese hamster ovary EM9 cells. Mutagenesis 20(6): 417-423
(2005).
S35 Stokinger, H.E., L.T. Steadman, H.B. Wilson et al. Lobar deposition and retention of inhaled
insoluble particulates. AMA Arch Ind Hyg Occup Med 4(4): 346-353 (1951).
S36 Storm, H.H., H.O. Jorgensen, A.M. Kejs et al. Depleted uranium and cancer in Danish Balkan
veterans deployed 1992-2001. Eur J Cancer 42(14): 2355-2358 (2006).
S37 Stradling, G.N., H. Smith, J.R. Cooper et al. Factors affecting the abundance of uranium
isotopes in body tissues and excreta following the deposition of enriched uranium dioxide in
the lungs. The radiological implications. Health Phys 46(2): 434-438 (1984).
S38 Stradling, G.N., J.W. Stather, J.C. Strong et al. Metabolism of some industrial uranium
tetrafluorides after deposition in the rat lung. Hum Toxicol 4(2): 159-168 (1985).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 497

S39 Stradling, G.N., J.W. Stather, S.A. Gray et al. Metabolism of uranium in the rat after inhalation
of two industrial forms of ore concentrate: the implications for occupational exposure. Hum
Toxicol 6(5): 385-393 (1987).
S40 Stradling, G.N., J.W. Stather, S.A. Gray et al. The metabolism of ceramic and non-ceramic
forms of uranium dioxide after deposition in the rat lung. Hum Toxicol 7(2): 133-139 (1988).
S41 Stradling, N., A. Hodgson, E. Ansoborlo et al. Anomalies between radiological and chemical
limits for uranium after inhalation by workers and the public. Radiat Prot Dosim 105(1-4):
175-178 (2003).
S42 Strand, L.A., J.I. Martinsen and E.K. Borud. Cancer risk and all-cause mortality among
Norwegian military United Nations peacekeepers deployed to Kosovo between 1999 and
2011. Cancer Epidemiol 38(4): 364-368 (2014).
S43 Strok, M. and B. Smodis. Transfer of natural radionuclides from hay and silage to cow's milk
in the vicinity of a former uranium mine. J Environ Radioact 110: 64-68 (2012).
S44 Sughis, M., T.S. Nawrot, A. Riaz et al. Metal exposure in schoolchildren and working
children. A urinary biomonitoring study from Lahore, Pakistan. Int J Hyg Environ Health
217(6): 669-677 (2014).
S45 Sullivan, M.F. Absorption of actinide elements from the gastrointestinal tract of neonatal
animals. Health Phys 38(2): 173-185 (1980).
S46 Sullivan, M.F. Absorption of actinide elements from the gastrointestinal tract of rats, guinea
pigs and dogs. Health Phys 38(2): 159-171 (1980).
S47 Sullivan, M.F. and L.S. Gorham. Further studies on the absorption of actinide elements from
the gastrointestinal tract of neonatal animals. Health Phys 43(4): 509-519 (1982).
S48 Sumanovic-Glamuzina, D., V. Saraga-Karacic, Z. Roncevic et al. Incidence of major
congenital malformations in a region of Bosnia and Herzegovina allegedly polluted with
depleted uranium. Croat Med J 44(5): 579-584 (2003).
S49 Suquin, L., L. Aiyun and R. Xue. A 33-year medical follow-up to patient with acute uranyl
nitrate hexahydrate poisoning. Radiat Prot Bull 29(1): 18-25 (2009). (Chinese).
S50 Swanson, S.M. Food-chain transfer of U-series radionuclides in a northern Saskatchewan
aquatic system. Health Phys 49(5): 747-770 (1985).
T1 Tadmor, J. Atmospheric release of volatilized species of radioelements from coal-fired plants.
Health Phys 50(2): 270-273 (1986).
T2 Tan, K. and S.K. Sethi. Biomarkers in cardiorenal syndromes. Transl Res 164(2): 122-134
(2014).
T3 Tasat, D.R., N.S. Orona, P.M. Mandalunis et al. Ultrastructural and metabolic changes in
osteoblasts exposed to uranyl nitrate. Arch Toxicol 81(5): 319-326 (2007).
T4 Taulan, M., F. Paquet, C. Maubert et al. Renal toxicogenomic response to chronic uranyl
nitrate insult in mice. Environ Health Perspect 112(16): 1628-1635 (2004).
T5 Taulan, M., F. Paquet, A. Argiles et al. Comprehensive analysis of the renal transcriptional
response to acute uranyl nitrate exposure. BMC Genomics 7: 2 (2006).
T6 Thiebault, C., M. Carriere, S. Milgram et al. Uranium induces apoptosis and is genotoxic to
normal rat kidney (NRK-52E) proximal cells. Toxicol Sci 98(2): 479-487 (2007).
T7 Thind, K.S. A comparison of ICRP Publication 30 lung model-based predictions with
measured bioassay data for airborne natural UO2 exposure. Health Phys 53(1): 59-66 (1987).
T8 Thorne, M.C. Estimation of animal transfer factors for radioactive isotopes of iodine,
technetium, selenium and uranium. J Environ Radioact 70(1-2): 3-20 (2003).
498 UNSCEAR 2016 REPORT

T9 Thorne, M.C. and J. Wilson. Generally applicable limits on intakes of uranium based on its
chemical toxicity and the radiological significance of intakes at those limits. J Radiol Prot
35(4): 743-762 (2015).
T10 Tirmarche, M., D. Laurier, F. Bochicchio et al. Final scientific report of the Alpha Risk Project
funded by the European Commission EC FP6. FI6R-CT-2005-516483. European Commission
Brussels, 2010.
T11 Tissandie, E., Y. Gueguen, J.M. Lobaccaro et al. In vivo effects of chronic contamination with
depleted uranium on vitamin D3 metabolism in rat. Biochim Biophys Acta 1770(2): 266-72 (2007).
T12 Tissandie, E., Y. Gueguen, J.M. Lobaccaro et al. Enriched uranium affects the expression of
vitamin D receptor and retinoid X receptor in rat kidney. J Steroid Biochem Mol Biol
110(3-5): 263-268 (2008).
T13 Tomasek, L. and I. Malatova. Leukaemia and lymphoma among Czech uranium miners.
Medical Radiology and Radiation Safety 51(5): 74-79 (2006).
T14 Tomita, J., H. Satake, T. Fukuyama et al. Radium geochemistry in Na-Cl type groundwater in
Niigata Prefecture, Japan. J Environ Radioact 101(3): 201-210 (2010).
T15 Tournier, B.B., S. Frelon, E. Tourlonias et al. Role of the olfactory receptor neurons in the
direct transport of inhaled uranium to the rat brain. Toxicol Lett 190(1): 66-73 (2009).
T16 Traber, S.C., V. Hollriegl, W.B. Li et al. Estimating the absorption of soil-derived uranium in
humans. Environ Sci Technol 48(24): 14721-14727 (2014).
T17 Tracy, B.L. and F.A. Prantl. Radiological impact of coal-fired power generation. J Environ
Radioact 2: 145-160 (1985).
T18 Tracy, B.L. and D.P. Meyerhof. Uranium concentrations in air near a Canadian uranium
refinery. Atmos Environ, Part A: General Topics 21(1): 165-172 (1987).
T19 Tracy, B.L., J.M. Quinn, J. Lahey et al. Absorption and retention of uranium from drinking
water by rats and rabbits. Health Phys 62(1): 65-73 (1992).
T20 Trelenberg, T.W., S.C. Glade, J.G. Tobin et al. The production and oxidation of uranium
nanoparticles produced via pulsed laser ablation. Surf Sci 600(11): 2338-2348 (2006).
T21 Triapitsyna, G.A., S.S. Andreev, D.I. Osipov et al. Assessment of radiation exposure level for
hydrobionts in some special industrial ponds at the "Mayak" PA. Radiats Biol Radioecol
52(2): 207-214 (2012). (Russian).
T22 Turtiainen, T., M. Muikku, P. Vesterbacka et al. Uranium and 226Ra in drinking water supplied
by Finnish waterworks. Radioprotection 46(06): S255-S263 (2011).
T23 Tymen, H., P. Gerasimo and D. Hoffschir. Contamination and decontamination of rat and
human skin with plutonium and uranium, studied with a Franz's chamber. Int J Radiat Biol
76(10): 1417-1424 (2000).
U1 Ubios, A.M., M.B. Guglielmotti, T. Steimetz et al. Uranium inhibits bone formation in
physiologic alveolar bone modeling and remodeling. Environ Res 54(1): 17-23 (1991).
U2 Ubios, A.M., M. Marzorati and R.L. Cabrini. Skin alterations induced by long-term exposure
to uranium and their effect on permeability. Health Phys 72(5): 713-715 (1997).
U3 UNEP. Post-conflict environmental assessment - FYR of Macedonia. United Nations
Environment Programme, Geneva, 2000.
U4 UNEP. Depleted uranium in Kosovo. Post-conflict environmental assessment. United Nations
Environment Programme, Geneva, 2001.
U5 UNEP. Depleted uranium in Bosnia and Herzegovina. Post-conflict environmental assessment.
Revised edition: May 2003. United Nations Environment Programme, Geneva, 2003.
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 499

U6 UNEP. Assessment of environmental hot spots in Iraq. United Nations Environment

Programme, Geneva, 2005.
U7 UNSCEAR. Sources and Effects of Ionizing Radiation. UNSCEAR 1977 Report. United
Nations Scientific Committee on the Effects of Atomic Radiation, 1977 Report to the General
Assembly, with annexes. United Nations sales publication E.77.IX.1. United Nations, New
York, 1977.
U8 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume II: Effects. UNSCEAR 2000
Report. United Nations Scientific Committee on the Effects of Atomic Radiation, 2000 Report
to the General Assembly, with scientific annexes. United Nations sales publication E.00.IX.4.
United Nations, New York, 2000.
U9 UNSCEAR. Effects of Ionizing Radiation. Volume II: Scientific Annexes C, D and E.
UNSCEAR 2006 Report. United Nations Scientific Committee on the Effects of Atomic
Radiation. United Nations sales publication E.09.IX.5. United Nations, New York, 2009.
U10 UNSCEAR. Sources and Effects of Ionizing Radiation. Volume I: Sources: Report to the
General Assembly, Scientific Annexes A and B. UNSCEAR 2008 Report. United Nations
Scientific Committee on the Effects of Atomic Radiation. United Nations sales publication
E.10.XI.3. United Nations, New York, 2010.
U11 Unwin, C., N. Blatchley, W. Coker et al. Health of UK servicemen who served in Persian Gulf
War. Lancet 353(9148): 169-178 (1999).
U12 Uralbekov, B.M., B. Smodis and M. Burkitbayev. Uranium in natural waters sampled within
former uranium mining sites in Kazakhstan and Kyrgyzstan. J Radioanal Nucl Chem 289(3):
805-810 (2011).
U13 USEPA. Occurence and exposure assessment for uranium in public drinking water supplies.
EPA contract no. 68-03-3514. Report prepared by Wade Miller Associates, Inc. for the Office of
Drinking Water, April 26. United States Environmental Protection Agency, Washington, 1990.
U14 USEPA. Review of RSC (relative source contribution) analysis. Report prepared by Wade
Miller Associates, Inc., 9 May 1991 (follow-up to USEPA 1990). United States Environmental
Protection Agency, Washington, D.C., 1991.
U15 USEPA. Child-specific exposure factors handbook. Office of Research and Development, U.S.
Environmental Protection Agency, Washington, D.C., 2002.
U16 USEPA. Uranium (natural and soluble salts). Integrated Risk Information System (IRIS),
United States Environmental Protection Agency, Washington, D.C., 2007.
U17 USEPA. Drinking water regulations and contaminants - National primary drinking water
regulations (NPDWRs). EPA 816-F-09-0004. United States Environmental Protection
Agency, Washington, D.C., 2009.
U18 USTUR. United States Transuranium and Uranium Registries: Annual report April 1, 2012 -
March 31, 2014 (S.L. McComish and S.Y. Tolmachev, eds.). USTUR-0382-15. College of
Pharmacy, Washington State University, Richland, WA, 2015.
V1 Vacquier, B., S. Caer, A. Rogel et al. Mortality risk in the French cohort of uranium miners:
extended follow-up 1946-1999. Occup Environ Med 65(9): 597-604 (2008).
V2 Vacquier, B., E. Rage, K. Leuraud et al. The influence of multiple types of occupational
exposure to radon, gamma rays and long-lived radionuclides on mortality risk in the French
"post-55" sub-cohort of uranium miners: 1956-1999. Radiat Res 176(6): 796-806 (2011).
V3 Vengosh, A., D. Hirschfeld, D. Vinson et al. High naturally occurring radioactivity in fossil
groundwater from the middle east. Environ Sci Technol 43(6): 41769-41775 (2009).
500 UNSCEAR 2016 REPORT

V4 Vesterbacka, P., I. Mkelinen and H. Arvela. Natural radioactivity in drinking water in

private wells in Finland. Radiat Prot Dosim 113(2): 223-232 (2005).
V5 Vicente-Vicente, L., Y. Quiros, F. Perez-Barriocanal et al. Nephrotoxicity of uranium: patho-
physiological, diagnostic and therapeutic perspectives. Toxicol Sci 118(2): 324-347 (2010).
V6 Vich, Z. and J. Kriklava. Erythrocytes of uranium miners: the red blood picture. Br J Ind Med
27(1): 83-85 (1970).
V7 Vidaud, C., A. Dedieu, C. Basset et al. Screening of human serum proteins for uranium
binding. Chem Res Toxicol 18(6): 946-953 (2005).
V8 Vidaud, C., S. Gourion-Arsiquaud, F. Rollin-Genetet et al. Structural consequences of binding
of UO2(2+) to apotransferrin: can this protein account for entry of uranium into human cells?
Biochemistry 46(8): 2215-2226 (2007).
W1 Wade-Gueye, N.M., O. Delissen, P. Gourmelon et al. Chronic exposure to natural uranium via
drinking water affects bone in growing rats. Biochim Biophys Acta 1820(7): 1121-1127 (2012).
W2 Wagner, S.E., J.B. Burch, M. Bottai et al. Hypertension and hematologic parameters in a
community near a uranium processing facility. Environ Res 110(8): 786-797 (2010).
W3 Walinder, G. Metabolism and sites of effects of uranium after incorporation along different
routes in mice, rabbits and piglets. Radiat Prot Dosim 26(1-4): 89-95 (1989).
W4 Wallner, G., R. Wagner and C. Katzlberger. Natural radionuclides in Austrian mineral water
and their sequential measurement by fast methods. J Environ Radioact 99(7): 1090-1094
(2008).
W5 Wallner, G. and T. Jabbar. Natural radionuclides in Austrian bottled mineral waters.
J Radioanal Nucl Chem 286(2): 329-334 (2010).
W6 Walsh, L., F. Dufey, A. Tschense et al. Prostate cancer mortality risk in relation to working
underground in the Wismut cohort study of German uranium miners, 1970-2003. BMJ Open
2(3) (2012).
W7 Walsh, M., G. Wallner and P. Jennings. Radioactivity in drinking water supplies in western
australia. J Environ Radioact 130: 56-62 (2014).
W8 Wan, B., J.T. Fleming, T.W. Schultz et al. In vitro immune toxicity of depleted uranium: effects
on murine macrophages, CD4+ T cells, and gene expression profiles. Environ Health Perspect
114(1): 85-91 (2006).
W9 Wappelhorst, O., I. Kuhn, H. Heidenreich et al. Transfer of selected elements from food into
human milk. Nutrition 18(4): 316-322 (2002).
W10 Weiner, R.E., J.F. McInroy and A.V. Wegst. Determination of environmental levels of Pu,
Am, U and Th in human fetal tissue. Thirtieth annual meeting of the Health Physics Society,
26-31 May 1985, Chicago, IL. Health Phys 49(1): 141 (1985).
W11 Welford, G.A. and R. Baird. Uranium levels in human diet and biological materials. Health
Phys 13(12): 1321-1324 (1967).
W12 West, C.M. and L.M. Scott. Uranium cases showing long chest burden retention--an updating.
Health Phys 17(6): 781-791 (1969).
W13 WHO. Depleted uranium: sources, exposure and health effects. WHO/SDE/PHE/01.1. World
Health Organization, Geneva, 2001.
W14 WHO. Depleted uranium: sources, exposure and health effects. J Pharm Belg 56(3): 75-78
(2001). (French).
 ANNEX D: BIOLOGICAL EFFECTS OF SELECTED INTERNAL EMITTERSURANIUM 501

W15 WHO. Uranium in drinking-water. Background document for development of WHO

guidelines for drinking-water quality. WHO/SDE/WSH/03.04/118/Rev/1 (updated 2012).
World Health Organization, Geneva, 2003.
W16 WHO. Guidelines for drinking-water quality. (Vol. 1) Recommendations. Third edition. World
Health Organization, Geneva, 2006.
W17 WHO. WHO Handbook on indoor radon: A public health perspective (H. Zeeb and F. Shannoun,
eds.). World Health Organization, Geneva, 2009.
W18 WHO. Guidelines for drinking-water quality. 4th edition. World Health Organization, Geneva,
2011.
W19 Wilson, H.B., G.E. Sylvester, S. Laskin et al. Relation of particle size of uranium dioxide dust
to toxicity following inhalation by animals; II. Arch Ind Hyg Occup Med 6(2): 93-104 (1952).
W20 Wilson, H.B., G.E. Sylvester, S. Laskin et al. Relation of particle size of U3O8 dust to toxicity
following inhalation by animals. AMA Arch Ind Health 11(1): 11-16 (1955).
W21 Wilson, R., H. Jones-Otazo, S. Petrovic et al. Revisiting dust and soil ingestion rates based on
hand-to-mouth transfer. Human and Ecological Risk Assessment 19(1): 158-188 (2013).
W22 Wise, S.S., W.D. Thompson, A.M. Aboueissa et al. Particulate depleted uranium is cytotoxic
and clastogenic to human lung cells. Chem Res Toxicol 20(5): 815-820 (2007).
W23 Witmans, M.R., H.H. McDuffie, C. Karunanayake et al. An exploratory study of chemical
elements in drinking water and non-Hodgkin's lymphoma. Toxicol Environ Chem 90(6): 1227-
1247 (2008).
W24 WNA. Processing of used nuclear fuel. World Nuclear Association. [Internet] Available from
(http://www.world-nuclear.org/info/Nuclear-Fuel-Cycle/Fuel-Recycling/Processing-of-Used-
Nuclear-Fuel/) on 6.10.2015.
W25 Wolf, G., D. Arndt, N. Kotschy-Lang et al. Chromosomal aberrations in uranium and coal
miners. Int J Radiat Biol 80(2): 147-153 (2004).
W26 Wrenn, M.E., P.W. Durbin, B. Howard et al. Metabolism of ingested U and Ra. Health Phys
48(5): 601-633 (1985).
W27 Wrenn, M.E., N.P. Singh, H. Ruth et al. Gastrointestinal absorption of soluble uranium from
drinking water by man. Radiat Prot Dosim 26(1-4): 119-122 (1989).
W28 Wnschmann, S., S. Frnzle, I. Khn et al. Transfer of selected elements from food into
human milk. Umweltwissenschaften und Schadstoff-Forschung 16(3): 168-174 (2004).
(German).
X1 Xie, H., C. LaCerte, W.D. Thompson et al. Depleted uranium induces neoplastic
transformation in human lung epithelial cells. Chem Res Toxicol 23(2): 373-378 (2010).
Y1 Yao, H., R. Kidd and M. Hashino. Long-term trend of uranium concentrations in Beaverlodge
Lake, Saskatchewan, Canada, under mine decommissioning. IAHS Publication 314: 137-144
(2007).
Y2 Yazzie, M., S.L. Gamble, E.R. Civitello et al. Uranyl acetate causes DNA single strand breaks
in vitro in the presence of ascorbate (vitamin C). Chem Res Toxicol 16(4): 524-530 (2003).
Y3 Yiin, J.H., J.L. Anderson, R.D. Daniels et al. A nested case-control study of multiple myeloma
risk and uranium exposure among workers at the Oak Ridge Gaseous Diffusion Plant. Radiat
Res 171(6): 637-645 (2009).
Y4 Yu, S.Y., H.C. Chung, E.J. Kim et al. Effects of acute renal failure induced by uranyl nitrate
on the pharmacokinetics of intravenous theophylline in rats: the role of CYP2E1 induction in
1,3-dimethyluric acid formation. J Pharm Pharmacol 54(12): 1687-1692 (2002).
502 UNSCEAR 2016 REPORT

Z1 Zablotska, L.B., R.S. Lane and S.E. Frost. Mortality (1950-1999) and cancer incidence
(1969-1999) of workers in the Port Hope cohort study exposed to a unique combination of
radium, uranium and gamma-ray doses. BMJ Open 3(2) (2013).
Z2 Zablotska, L.B., R.S. Lane, S.E. Frost et al. Leukemia, lymphoma and multiple myeloma
mortality (1950-1999) and incidence (1969-1999) in the Eldorado uranium workers cohort.
Environ Res 130: 43-50 (2014).
Z3 Zaire, R., C.S. Griffin, P.J. Simpson et al. Analysis of lymphocytes from uranium
mineworkers in Namibia for chromosomal damage using fluorescence in situ hybridization
(FISH). Mutat Res 371(1-2): 109-113 (1996).
Z4 Zaire, R., M. Notter, W. Riedel et al. Unexpected rates of chromosomal instabilities and
alterations of hormone levels in Namibian uranium miners. Radiat Res 147(5): 579-584 (1997).
Z5 Zamora, L.M., J.M. Zielinski, D. Meyerhof et al. Uranium gastrointestinal absorption: the f1
factor in humans. Radiat Prot Dosim 105(1-4): 55-60 (2003).
Z6 Zamora, M.L., B.L. Tracy, J.M. Zielinski et al. Chronic ingestion of uranium in drinking
water: a study of kidney bioeffects in humans. Toxicol Sci 43(1): 68-77 (1998).
Z7 Zamora, M.L., J.M. Zielinski, D.P. Meyerhof et al. Gastrointestinal absorption of uranium in
humans. Health Phys 83(1): 35-45 (2002).
Z8 Zamora, M.L., J.M. Zielinski, G.B. Moodie et al. Uranium in drinking water: renal effects of
long-term ingestion by an aboriginal community. Arch Environ Occup Health 64(4): 228-241
(2009).
Z9 Zhivin, S., D. Laurier and I. Guseva Canu. Health effects of occupational exposure to uranium:
do physicochemical properties matter? Int J Radiat Biol 90(11): 1104-1113 (2014).
Z10 Zhivin, S., I. Guseva Canu, E. Samson et al. Mortality (1968-2008) in a French cohort of
uranium enrichment workers potentially exposed to rapidly soluble uranium compounds.
Occup Environ Med (2015).
Z11 Zhou, J., L. Zhou and J. Yang. Studies on effects of cytokine released from alveolar
macrophage induced by mineral dust on lung fibroblast. Zhonghua Yu Fang Yi Xue Za Zhi
[Chin J Prev Med] 32(6): 336-339 (1998). (Chinese).
Z12 Zhou, L.R., J.H. Zhou and J.C. Yang. Effects of cytokines induced by mineral dust on lung
fibroblasts in vitro. J Occup Health 41: 144-148 (1999).
Z13 Zhu, G., M. Tan, Y. Li et al. Accumulation and distribution of uranium in rats after
implantation with depleted uranium fragments. J Radiat Res 50(3): 183-192 (2009).
Z14 Zhu, G., X. Xiang, X. Chen et al. Renal dysfunction induced by long-term exposure to
depleted uranium in rats. Arch Toxicol 83(1): 37-46 (2009).
Z15 Zhu, S.P., Q.Y. Hu and M.Y. Lun. Studies on reproductive toxicity induced by enriched
uranium. Zhonghua Yu Fang Yi Xue Za Zhi 28(4): 219-222 (1994). (Chinese).
Z16 Zhuo, W., T. Iida and X. Yang. Occurrence of 222Rn, 226Ra, 228Ra and U in groundwater in
Fujian Province, China. J Environ Radioact 53(1): 111-120 (2001).
Z17 Zimmerman, K.L., D.S. Barber, M.F. Ehrich et al. Temporal clinical chemistry and
microscopic renal effects following acute uranyl acetate exposure. Toxicol Pathol 35(7):
1000-1009 (2007).
 unscear.org

This publication contains:

REPORT OF THE UNITED NATIONS SCIENTIFIC COMMITTEE ON THE EFFECTS

OF ATOMIC RADIATION TO THE GENERAL ASSEMBLY

Scientific Annexes
Annex A. Methodology for estimating public exposures due to radioactive discharges
Annex B. Radiation exposures from electricity generation
Annex C. Biological effects of selected internal emittersTritium
Annex D. Biological effects of selected internal emittersUranium

Supplementary materials (only available online at www.unscear.org)

E VA L U AT I N G R A D I AT I O N S C I E N C E F O R I N F O R M E D D E C I S I O N - M A K I N G

In 1955 the United Nations General Assembly established the Scientific Committee on the Effects
of Atomic Radiation (UNSCEAR) in response to concerns about the effects of ionizing radiation
on human health and the environment. At that time fallout from atmospheric nuclear weapons
tests was reaching people through air, water and food. UNSCEAR was to collect and evaluate
information on the levels and effects of ionizing radiation. Its first reports laid the scientific grounds
on which the Partial Test Ban Treaty prohibiting atmospheric nuclear weapons testing was
negotiated in 1963.

Over the decades, UNSCEAR has evolved to become the world authority on the global level
and effects of atomic radiation. UNSCEARs independent and objective evaluation of the science
are to provide forbut not addressinformed policymaking and decision-making related to
radiation risks and protection.

ISBN 978-92-1-142316-7
V.17-00439