Go back to opening screen
525
Chapter 8
Radioactivity
Contents
External exposure from anthropogenic sources 555
8.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
Internal doses from anthropogenic sources . . 555
8.4.3. Intakes of 137Cs through various dietary components . . . . 558
8.2. Fundamentals and definitions . . . . . . . . . . . . . . . . . . . . . . . . 526
Variations in sources of 137Cs intake. . . . . . . . . . . . . . 559
8.2.1. Radioactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
Temporal variations in 137Cs intake . . . . . . . . . . . . . . 560
8.2.1.1. Natural radioactivity . . . . . . . . . . . . . . . . . . . . . . 526
Changes in the relative importance of dietary
8.2.1.2. Artificial radioactivity . . . . . . . . . . . . . . . . . . . . . 527
components with time . . . . . . . . . . . . . . . . . . . . . . . . 561
8.2.2. Effects of radionuclides . . . . . . . . . . . . . . . . . . . . . . . . . . 527
8.4.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
8.2.2.1. The concept of risk . . . . . . . . . . . . . . . . . . . . . . . 527
8.2.2.2. Health effects and units of dose . . . . . . . . . . . . . 527
8.5. Source-related assessments of past and present releases . . 562
Natural radiation and exposures . . . . . . . . . . 528
8.5.1. Nuclear explosions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
8.2.3. The system of radiological protection . . . . . . . . . . . . . . . . 529
8.5.1.1. Atmospheric nuclear weapons tests . . . . . . . . . . . 562
8.2.3.1. Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
8.5.1.2. Underground nuclear explosions . . . . . . . . . . . . . 564
Individual doses, dose limits and dose con-
8.5.1.2.1. Underground explosions carried out
straints. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
in the Arctic by the former Soviet Union 564
Collective doses . . . . . . . . . . . . . . . . . . . . . . 530
8.5.1.2.2. Underground explosions carried out
Exclusion and exemption . . . . . . . . . . . . . . . 530
in the Arctic by the United States . . . . 565
8.2.3.2. Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 530
8.5.2. Operational releases from the nuclear fuel cycle . . . . . . . . 565
8.2.3.3. Radiological assessments . . . . . . . . . . . . . . . . . . 530
8.5.2.1. Nuclear power plants . . . . . . . . . . . . . . . . . . . . . 565
8.2.3.4. The basis for intervention . . . . . . . . . . . . . . . . . . 531
8.5.2.1.1. Nuclear power plants in the Arctic . . . 565
8.2.3.5. Other issues relevant to radiological assessment . 531
Atmospheric releases . . . . . . . . . . 566
8.2.3.5.1. Relationship between radiation expo-
Liquid releases . . . . . . . . . . . . . . . 566
sure and risk of adverse health effects . 531
Other wastes . . . . . . . . . . . . . . . . . 566
8.2.3.5.2. Transport processes and exposure
8.5.2.1.2. Nuclear power plants in the vicinity
pathways . . . . . . . . . . . . . . . . . . . . . . 532
of the Arctic . . . . . . . . . . . . . . . . . . . . 566
Atmospheric transport . . . . . . . . . 532
Russian NPPs . . . . . . . . . . . . . . . . 566
Marine transport. . . . . . . . . . . . . . 532
Finnish NPPs . . . . . . . . . . . . . . . . 567
Terrestrial transport . . . . . . . . . . . 532
Swedish NPPs . . . . . . . . . . . . . . . . 568
Interception. . . . . . . . . . . . . . . . . . 533
8.5.2.2. Russian civilian nuclear fleet . . . . . . . . . . . . . . . . 568
Soil-to-plant transfer . . . . . . . . . . . 533
8.5.2.3. The Russian Northern Fleet . . . . . . . . . . . . . . . . 570
Plant-to-animal transfer. . . . . . . . . 533
8.5.2.3.1. Nuclear-powered vessel operations . . . 570
Freshwater pathways . . . . . . . . . . 533
8.5.2.3.2. Decommissioning . . . . . . . . . . . . . . . . 570
Marine pathways . . . . . . . . . . . . . 534
8.5.2.3.3. Storage of the spent nuclear fuel
8.2.4. Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
and radioactive waste . . . . . . . . . . . . . 570
Integrated transfer factors . . . . . . . . . . . . . . . . . . . . . 535
8.5.2.3.4. Shipyards . . . . . . . . . . . . . . . . . . . . . . 571
Aggregated transfer coefficients (T
8.5.2.4. European nuclear fuel reprocessing plants . . . . . . 571
ags) . . . . . . . . . . . . 535
8.2.5. The AMAP assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . 536
8.5.2.4.1. British nuclear fuels plant at Sella-
field, UK . . . . . . . . . . . . . . . . . . . . . . . 572
8.3. Past and present radioactive contamination of the Arctic 536
8.5.2.4.2. La Hague, France . . . . . . . . . . . . . . . . 573
8.3.1. Geographical distribution of radioactive contamination . 537
8.5.2.4.3. Dounreay, UK . . . . . . . . . . . . . . . . . . . 573
8.3.1.1. Widespread contamination of land and sea . . . . . 537
8.5.2.4.4. Dose reconstruction for releases from
Terrestrial contamination . . . . . . . . . . . . . . . 537
Western European reprocessing plants 574
Marine contamination . . . . . . . . . . . . . . . . . 539
8.5.2.5. Russian nuclear fuel reprocessing plants . . . . . . . 575
8.3.1.2. Localized contamination . . . . . . . . . . . . . . . . . . . 541
8.5.2.5.1. Mayak . . . . . . . . . . . . . . . . . . . . . . . . 575
8.3.1.2.1. Short-range fallout from Novaya
8.5.2.5.2. Tomsk-7 . . . . . . . . . . . . . . . . . . . . . . . 576
Zemlya tests . . . . . . . . . . . . . . . . . . . . 541
8.5.2.5.3. Krasnoyarsk-26 . . . . . . . . . . . . . . . . . 576
8.3.1.2.2. Chernaya Bay . . . . . . . . . . . . . . . . . . . 542
8.5.2.5.4. Assessment of river transport and
8.3.1.2.3. The Thule accident . . . . . . . . . . . . . . . 542
associated doses . . . . . . . . . . . . . . . . . 576
Plutonium in Bylot Sound seawater 542
8.5.2.6. Mining activities . . . . . . . . . . . . . . . . . . . . . . . . . 577
Plutonium in Bylot Sound sediments 543
8.5.3. Accidental releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577
8.3.1.2.4. Contamination at sea dumping sites . . 543
8.5.3.1. The accidents at the Mayak weapons production
8.3.1.2.5. Sunken Komsomolets submarine . . . . 544
plant in 1957 and at Lake Karachay in 1967 . . . 577
8.3.2. Time dependence of radioactive contamination . . . . . . . . 544
The Kyshtym accident, 1957 . . . . . . . . . . . . 577
8.3.2.1. Air and deposition . . . . . . . . . . . . . . . . . . . . . . . . 545
Lake Karachay, 1967. . . . . . . . . . . . . . . . . . . 577
8.3.2.2. Terrestrial and freshwater ecosystems . . . . . . . . . 547
8.5.3.2. The Thule nuclear weapons accident in 1968 . . . 577
8.3.2.2.1. Lichen . . . . . . . . . . . . . . . . . . . . . . . . . 547
8.5.3.3. The Cosmos-954 satellite re-entry in 1978. . . . . . 578
8.3.2.2.2. Reindeer meat . . . . . . . . . . . . . . . . . . . 548
8.5.3.4. The Chernobyl accident in 1986 . . . . . . . . . . . . . 579
8.3.2.2.3. Freshwater ecosystems . . . . . . . . . . . . 549
8.5.3.4.1. The accident and associated source term 579
8.3.2.3. Marine ecosystems . . . . . . . . . . . . . . . . . . . . . . . 550
8.5.3.4.2. Radiological consequences at temperate
8.3.2.3.1. Seawater . . . . . . . . . . . . . . . . . . . . . . . 550
latitudes . . . . . . . . . . . . . . . . . . . . . . . 579
8.3.2.3.2. Fish and marine mammals . . . . . . . . . 550
8.5.3.4.3. Transport and deposition in the Arctic 579
8.3.3. Human wholebody measurements . . . . . . . . . . . . . . . . . . 550
Marine transport to the Arctic seas 580
8.3.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
8.5.3.4.4. Food chain and human contamination 580
8.4. Individual doses to man
Lichen . . . . . . . . . . . . . . . . . . . . . . 580
estimated from environmental measurements . . . . . . . . . . 552
Reindeer . . . . . . . . . . . . . . . . . . . . 580
8.4.1. Natural radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
Human body . . . . . . . . . . . . . . . . 580
External exposures from natural sources . . . . . . . . . . 552
8.5.3.4.5. Countermeasures . . . . . . . . . . . . . . . . 581
Internal exposures from natural sources . . . . . . . . . . . 552
8.5.3.4.6. Human dose estimations . . . . . . . . . . . 581
8.4.2. Radionuclide contamination . . . . . . . . . . . . . . . . . . . . . . 553
8.5.3.5. Accidents involving nuclear-powered vessels . . . . 581
8.4.2.1. Information base for individual dose estimates . . 553
8.5.3.5.1. Sunken Komsomolets submarine . . . . 582
Finnish Lapland . . . . . . . . . . . . . . . . . . . . . . 553
8.5.3.5.1.1. Accident and source term 582
Greenland (Kalaallit Nunaat) . . . . . . . . . . . . 553
8.5.3.5.1.2. Radiological assessments of
Northern Canada . . . . . . . . . . . . . . . . . . . . . 553
the Komsomolets accident 583
Northern Russia . . . . . . . . . . . . . . . . . . . . . . 554
8.5.4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585
Northern Norway . . . . . . . . . . . . . . . . . . . . . 554
8.6. Source-related assessments of potential releases . . . . . . . . 585
Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554
8.6.1. Nuclear power plant reactor accidents . . . . . . . . . . . . . . . 585
Iceland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555
Safety criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585
Arctic Sweden . . . . . . . . . . . . . . . . . . . . . . . . 555
8.6.1.2. Probabilistic safety assessment (PSA) . . . . . . . . . 586
Diet intakes by Arctic populations . . . . . . . . 555
8.6.1.3. Studies to assess the consequences of major
8.4.2.2. External and internal doses to humans . . . . . . . . 555
reactor accidents . . . . . . . . . . . . . . . . . . . . . . . . . 587
526
AMAP Assessment Report
8.6.2. Potential accidental releases from nuclear vessels and
evaluation of radiological vulnerability in the Arctic. The
nuclear storage sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 588
chapter ends with conclusions and recommendations.
8.6.3. Potential releases from reprocessing plants . . . . . . . . . . . . 590
8.6.3.1. Mobilisation of radionuclides released to the
The chapter was prepared under the guidance of an as-
terrestrial environment . . . . . . . . . . . . . . . . . . . . 590
sessment group comprising scientists from the contracting
8.6.3.2. Mayak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
parties to the international Arctic Environmental Protection
8.6.3.3. Tomsk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
8.6.3.4. Krasnoyarsk . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
Strategy, or Rovaniemi Agreement. Several other individuals
8.6.4. Radioactive wastes dumped at sea . . . . . . . . . . . . . . . . . . 591
have made substantial contributions to the report and the
8.6.4.1. Surveys of dumped objects . . . . . . . . . . . . . . . . . 592
data upon which the report is based. In the preparation of
8.6.4.2. International Arctic Seas Assessment Project
(IASAP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592
this chapter, the explanatory text, data assembly and prepa-
8.6.4.2.1. Source term reconstruction . . . . . . . . . 593
ration of individual-related radiological assessments were
8.6.4.2.2. Consideration of possible criticality . . 594
provided by the assessment group and national staff. Most
8.6.4.2.3. Pathway modeling and radiological
assessment . . . . . . . . . . . . . . . . . . . . . 594
of the source-related assessments in the document, on the
8.6.4.2.4. Effects on marine organisms . . . . . . . . 595
other hand, are based on studies carried out under the aegis
8.6.4.2.5. Remediation . . . . . . . . . . . . . . . . . . . 595
of other agencies, either national or international. The inter-
8.6.4.2.6. Conclusions of IASAP . . . . . . . . . . . . . 595
8.6.5. Nuclear weapons
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595
pretation and representation of these latter studies have been
8.6.6. Radionuclide thermoelectric generators . . . . . . . . . . . . . . 596
carried out by the assessment group in connection with the
8.6.7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597
preparation of this document.
8.7. Spatial analysis of vulnerability of Arctic ecosystems . . . . 597
To the extent that appropriate data and information has
8.7.1. Sources of radionuclide intake by humans . . . . . . . . . . . . 597
8.7.2. Spatial distribution of Arctic communities . . . . . . . . . . . . 598
been made available to the assessment group, the assessment
8.7.3. Spatial differences in transfer through pathways . . . . . . . 599
goal has been achieved. Inevitably, however, because of the
8.7.4. Changes with time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600
heterogeneity and varying comprehensiveness of the infor-
8.7.5. Transfer coefficients and relationships . . . . . . . . . . . . . . . 600
8.7.5.1. UNSCEAR transfer coefficients . . . . . . . . . . . . . . 600
mation available, some sections of the document are more
8.7.5.2. Spatial and temporal variations in transfer to
complete and detailed than others.
Arctic food products using aggregated transfer
The assessment serves to document what is currently
coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602
8.7.5.2.1. Spatial variation in total production . . 603
known about radioactivity from sources in the Arctic and
Reindeer production . . . . . . . . . . . 603
associated risks and effects. It also identifies where addi-
Milk production . . . . . . . . . . . . . . 604
tional efforts are required to obtain more information or
8.7.5.2.2. Spatial variation in fluxes . . . . . . . . . . 604
Reindeer . . . . . . . . . . . . . . . . . . . . 604
conduct additional assessments to improve the characterisa-
Milk . . . . . . . . . . . . . . . . . . . . . . . 604
tion of the risks associated with specific human and indus-
Radioiodine contamination of milk 606
trial activities.
8.7.7. Sensitivity to uncertainties: radiocaesium in fungi and berries 606
8.7.8. Flux vulnerability of Arctic Norway . . . . . . . . . . . . . . . . 606
8.7.8.1. Production data . . . . . . . . . . . . . . . . . . . . . . . . . 606
8.7.8.2. Aggregated transfer coefficients . . . . . . . . . . . . . . 606
8.7.8.3. Total 137Cs output . . . . . . . . . . . . . . . . . . . . . . . . 607
8.2. Fundamentals and definitions
8.7.8.4. Spatial distribution of the Norwegian Arctic
8.2.1. Radioactivity
population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 608
8.7.8.5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 608
Radioactivity is the property of spontaneous disintegration,
8.7.9. Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609
or decay, of atomic nuclei accompanied by the emission of
8.8. Conclusions and recommendations . . . . . . . . . . . . . . . . . . 609
8.8.1. Conclusions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609
ionizing radiation. Activity corresponds to the number of
8.8.2. General recommendations . . . . . . . . . . . . . . . . . . . . . . . . 610
disintegrations per second of an isotope (with dimensions
8.8.3. Specific recommendations . . . . . . . . . . . . . . . . . . . . . . . . 611
T1). The SI (Standards Internationaux) unit of activity is the
8.8.3.1. Recommendations regarding storage of spent
nuclear fuel and radioactive waste . . . . . . . . . . . 611
reciprocal second (s1) with the name Becquerel (Bq). The
8.8.3.2. Recommendations regarding monitoring . . . . . . 611
older, non-SI, unit Curie (Ci) that was derived from the (pre-
8.8.3.3. Recommendations for further study to correct
sumed) activity of one gram of radium and is still used in
information deficiencies . . . . . . . . . . . . . . . . . . . 611
some fora, corresponds to 3.7
1010 Bq. The major forms
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 611
of ionizing radiation emitted during radioactive decay are
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 612
alpha particles, which are essentially charged helium nuclei,
Annex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 615
beta particles, which are electrons, and gamma rays, which
are photons or electromagnetic waves. The nature, energy,
charge and penetrating power of radiation is of relevance to
8.1. Introduction
the consequences of biological exposures. This is dealt with
This chapter deals with the assessment of radioactive conta-
in more detail later in this introductory section.
mination of the environment, radiation sources and associ-
The term `radionuclide' applies to all radioactive isotopes
ated radiological consequences within the Arctic. The pur-
of all elements. The term `radioisotope' strictly refers to the
pose of this chapter is to provide a balanced appreciation of
radioactive isotope of an element having other isotopes of
the nature and risks posed by radionuclides in the Arctic de-
similar chemical properties but differing nuclear properties.
rived from all relevant and known sources. Initially, a sim-
These may include both stable and radioactive isotopes. The
plified explanation of the basis of radiological protection
physical half-life of a radionuclide defines the time required
and the procedures for estimating radiological doses and
for the activity of that radionuclide to decay, by purely phys-
risks is provided. The chapter subsequently deals with doses
ical processes, by a factor of two.
associated with existing radioactive contamination of the
Arctic environment, routine releases from nuclear operations
8.2.1.1. Natural radioactivity
within, and close to, the Arctic, previous accidents in civil
and military nuclear activities that result in exposures to
Natural radioactivity is derived from the decay of nuclei in
Arctic residents, and potential releases from both such in-
the Earth's crust and by the bombardment of the Earth by
stallations and the various packages of high-level waste
cosmic radiation producing radionuclides in the Earth's at-
reposing in the environment, such as those dumped in the
mosphere. These natural radionuclides fall into three cate-
Kara Sea by the former Soviet Union. This is followed by an
gories: the very long-lived primordial radionuclides (40K,
Chapter 8 · Radioactivity
527
238U, 232Th, 235U) formed at the time the Earth was created;
ical benefits in diagnosis and treatment of disease as well as
decay chain radionuclides (radionuclides in the uranium,
in several industrial applications. Both facets of the existence
thorium and actinium decay series) that are the products of
and use of radionuclides and radiation have led to the cre-
decay of primordial nuclides; and cosmogenic nuclides pro-
ation of a major discipline called radiological protection.
duced by the interaction of high energy cosmic radiation
Other terms, such as `radiology' and `health physics', origi-
with the Earth's atmosphere (e.g., 3H, 7Be, 14C, 22Na).
nally used in a wider context, are now almost exclusively
used in connection with nuclear medicine. The entire focus
of radiological protection is the effects of radiation on living
8.2.1.2. Artificial radioactivity
tissues and organisms, and mechanisms for the adequate
In the early days of the 20th Century, human abilities to cre-
protection of both deliberately and accidentally exposed hu-
ate artificial radioactive sources were limited to chemical iso-
mans and populations of other organisms. The remaining
lation and the concentration of natural radionuclides. Later
text of this section is intended to provide a synopsis of the
in this century, linear accelerators were developed for pro-
basis and nature of health protection from the effects of ra-
ducing beams of particles that could be used to artificially
diation and the generation and use of radionuclides, includ-
transmute nuclei. With the application of nuclear fission, for
ing the regulation of the nuclear power industry. Intention-
both peaceful and military purposes in the 1940s, the ability
ally, this synopsis does not go into great detail it merely
of humans to produce large quantities of artificial radionu-
serves as background to much of the text of later sections of
clides was greatly expanded. The fission process itself, and
this chapter.
the high neutron flux densities achieved in nuclear weapons
explosions and fission reactor cores, led to the production of
8.2.2.1. The concept of risk
large quantities of fission and activation products. Fission
products are the isotopes with atomic masses in the 70-170
There are many definitions of risk. Risk relates to quantities
range, formed by thermal fission of 235U and other heavy fis-
such as the probability that specific deleterious consequences
sile nuclei (e.g., 239Pu). High-yield fission products include
may arise and the magnitude and character of such conse-
89Sr, 90Sr, 91Y, 95Zr, 95Nb, 99Mo, 103Ru, 131I, 133Xe, 137Cs,
quences. In this assessment, the term risk is used to mean the
140Ba, 140La, 141Ce, 144Ce, 143Pr and 147Nd. However, in most
probability the likelihood that something unpleasant will
situations, the most radiologically important fission products
happen. Clearly, however, the likelihood of an adverse hap-
in the short term are 89Sr, 90Sr, 131I and 137Cs, and in the long
pening cannot be considered outside of the context of the
term, 90Sr and 137Cs, because of their yields, half-lives and
severity of the associated effect. If the consequences of hap-
chemical properties. Activation products are the isotopes
penings of equal probability are respectively fatality or minor
formed principally by the capture of neutrons by stable iso-
personal financial loss, the former is going to be respected
topes in high neutron flux environments. Typical activation
and considered far more seriously than the latter. A related
products formed in the structure of nuclear reactors include
term is that of hazard. A hazard is essentially a `set of cir-
51Cr, 54Mn, 55Fe, 59Fe, 60Co, 63Ni, 65Ni, 64Cu, 65Zn, 69Zn,
cumstances' that may result in harmful consequences. Harm
110Ag, 109Cd, 134Cs, 236U and 239U. These radionuclides are
is generally taken to include adverse effects on health or the
generally neutron-rich and decay primarily by gamma-ray
quality of life; it can also be expressed in terms of loss, in-
and beta particle emission. Some activation products are iso-
cluding loss of life, of working days, or material items, such
topes of elements of atomic number larger than uranium and
as environmental amenities or money. It is often possible,
these are referred to as `transuranic' nuclides. Prime exam-
therefore, to represent adverse effects as costs to society. Be-
ples are 239Np and 239Pu that are created in reactors as a re-
cause costs are also incurred in reducing risks, the two sets
sult of the -decay of the short lived activation product 239U.
of costs have often been used to estimate the optimum `value
As a result of the use of fission reactors for electrical
for money' in relation to measures taken to reduce risk. In
power generation, there are large quantities of fission prod-
absolute terms, no set of human circumstances is entirely
ucts in spent nuclear fuel assemblies. There is also a large in-
safe but, obviously, the lower the risk, the higher the degree
ventory of activation products in reactor assemblies and in
of safety. The two terms (risk and safety) are, therefore, in-
other materials such as 60Co radiation sources deliberately
versely related and what most people perceive as being `safe'
or inadvertently exposed to significant neutron fluxes. As a
actually corresponds to an acceptable level of risk.
result of nuclear fuel reprocessing for the recovery of pluto-
nium, substantial quantities of fission and activation prod-
8.2.2.2. Health effects and units of dose
ucts have been released to the environment in wastes.
Nuclear weapons explosions have provided the largest in-
Exposure to radiation can cause detrimental health effects.
ventory of both fission and activation products in the global
At large acute doses, radiation effects such as opacities in
environment, and many of these have been, and remain, de-
the lens of the eye sometimes leading to cataract, temporary
tectable world-wide. Indeed, nuclear explosions have pro-
or permanent sterility and, in severe cases of whole body ir-
duced the most pronounced global change in the character
radiation, acute syndromes (such as damage to bone mar-
of environmental radioactivity. Various nuclear accidents
row, gastrointestinal tract, lungs and the nervous system)
have further contributed to the inventory of radionuclides in
can lead to death within a short period of time after expo-
the environment. In recent years, the use of radiothermal
sure. Large chronic dose rates also cause clinically detectable
power generators for space vehicles has, as a result of acci-
deleterious effects. These various effects are called determin-
dents, given rise to additional isotopes detectable in the envi-
istic because they are certain to occur if the dose exceeds
ronment, most notably 238Pu.
certain threshold levels.
At low doses, radiation exposure can also plausibly in-
duce severe health effects, such as malignancies, which are
8.2.2. Effects of radionuclides
statistically detectable in a population, but cannot be un-
The main concern about radionuclides and radiation are
equivocally associated with individual exposures. Hereditary
their adverse effects on organisms, including humans. How-
effects due to radiation exposure have been statistically de-
ever, it must be remembered that ionizing radiation has med-
tected in mammals and are presumed to occur in humans as
528
AMAP Assessment Report
well. All these statistically detectable effects are called sto-
Of the large number of radionuclides produced by cosmic
chastic effects because of their random (i.e., probabilistic)
radiation only four of them (3H, 7Be, 14C and 22Na) contri-
nature. These effects are expressed after a latency period,
bute significantly to the dose to humans (NCRP 1987). The
presumably over the entire range of doses without a thresh-
most radiologically significant of these four radionuclides is
14
old level. In addition, there is a possibility of health effects
C. The annual natural production of 14C is 1 PBq and the
in children exposed to radiation in utero during certain peri-
specific activity of natural 14C in the body is 230 Bq/kg lead-
ods of pregnancy, including a greater likelihood of leukae-
ing to an annual effective dose of 12 Sv (UNSCEAR 1993).
mia and severe mental retardation.
The contributions from the ingestion of 3H, 7Be and 22Na
The fundamental dosimetric quantity in radiological pro-
are much smaller.
tection is the absorbed dose. This is the energy absorbed per
Primordial radionuclides are usually categorized as either
unit mass and is expressed in units of joules per kilogram and
`series radionuclides' which decay in a chain of radionuclides
given the name gray (Gy). The probability of stochastic ef-
to a stable isotope of lead, or `non-series radionuclides' which
fects depends not only on the absorbed dose but also on the
decay directly to stable nuclides. There are several tens of
type and energy of the radiation causing the dose. However,
non-series radionuclides in crystalline rocks and soils. How-
it is the absorbed dose averaged over a tissue or organ (rather
ever, most of the non-series radionuclides have a combina-
than at a point) and weighted for the radiation type that is
tion of half-life, isotopic abundance, and elemental abun-
pertinent. The equivalent dose is the term used in a tissue or
dance in the Earth's crust such that they have negligibly
organ when these two components have been taken into ac-
small specific activities and are not dosimetrically significant.
count through the use of appropriate weighting factors. The
The only non-series radionuclides having any dosimetric sig-
relationship between the probability of stochastic effects and
nificance are 40K and 87Rb, which are both geochemically
equivalent dose depends on the organ or tissue irradiated. It
similar alkali elements. Whilst 87Rb is a pure -emitter, 40K
is, therefore, appropriate to define a further quantity, derived
decay is accompanied by both - and -radiation. The abun-
from equivalent dose, to indicate the combination of differ-
dance of 40K in the environment makes it a major source of
ent doses to several different tissues in a way that correlates
both internal and external doses from naturally-occurring
with the total of the stochastic effects. Once a weighting fac-
radiation. 40K in rocks, soils and building materials is also a
tor is introduced to account for the relative contribution of
major contributor to external background radiation. Ac-
each organ or tissue to the total detriment resulting from uni-
cording to UNSCEAR (1988) about 40% of the average an-
form irradiation of the whole body, the term used to charac-
nual dose to humans from external radiation is due to 40K in
terize the dose is effective dose. The effective dose is then the
the surroundings. Data on 40K in the human body are well
sum of the weighted equivalent doses in all the tissues and
established, mainly from direct whole body measurements of
organs of the body. While the units of effective dose are still
persons of various ages. The average specific activity of 40K
joules per kilogram, it is given the name sievert (Sv).
in the body of adults is about 55 Bq/kg, averaged over both
sexes. The annual effective doses to adults and children are
Natural radiation and exposures
estimated to be 165 and 185 Sv, respectively. Potassium is
Cosmic radiation and ionizing radiation from radionuclides
in homeostatic control in the human body, which means
in the environment provide the major source of human radi-
that the dose from 40K is not influenced by the potassium
ation exposure.
intake with diet.
The term `cosmic radiation' refers both to the primary
The radionuclides in the decay series headed by 238U (ura-
high-energy particles of extraterrestrial origin that strike the
nium series), 232Th (thorium series) and 235U (actinium se-
Earth's atmosphere and to the secondary particles generated
ries) are called series radionuclides. The relative abundance
by their interaction with the atmosphere. The primary galac-
of 235U (0.73%) is low compared to 238U (99.2%). The de-
tic particles entering the Earth's atmosphere are high-energy
cay products in the actinium series are relatively short-lived.
protons ( 90%) and alpha-particles ( 10%).
Thus, the actinium series is of much less dosimetric impor-
Lower-energy charged particles are deflected back into
tance than the uranium and thorium series, and will not be
space by the Earth's magnetic field. This effect is latitude-
discussed further.
dependent and there is a greater flux of incident low-energy
Depending on local geology, there are large local and re-
protons at the poles than at the equator, resulting in an in-
gional variations in outdoor gamma dose rates. Outdoor
crease in the dose rate at high latitudes. Furthermore, this
gamma radiation depends mainly on 226Ra, 232Th and 40K in
latitude effect increases with altitude.
soil and rock. In certain granites and alum shales, 226Ra ac-
Buildings provide some shielding against the directly ion-
tivity concentrations of up to 500 and 5000 Bq/kg, respec-
izing component of cosmic radiation, but the magnitude of
tively, have been found. External gamma radiation indoors,
the shielding depends strongly on the structural composition
on the other hand, depends mainly on the activity concen-
and thickness of the building material. The shielding effect
tration of the building materials. The average outdoor and
of wooden houses reduces the dose rate of the direct ioniz-
indoor dose rates in air for the world population have been
ing component by less than 5% (Miller and Beck 1984),
calculated by UNSCEAR (1993). Based on an indoor occu-
whereas the reduction is between 35 and 70% for some
pancy factor of 0.8, the average annual effective dose to the
larger multi-storey concrete buildings (Miller and Beck
world population is estimated to be 0.46 mSv.
1984, Lin et al. 1986).
Exposure to 222Rn (radon), 220Rn (thoron) and their prog-
Taking into account shielding by buildings and the distri-
eny comes mainly from the inhalation of the decay products
bution of the world population with altitude and latitude,
of radon and thoron, which deposit inhomogeneously with-
the population-weighted average annual effective dose from
in the respiratory tract and irradiate the bronchial epithe-
cosmic radiation has been estimated by the United Nations
lium. The dose contribution from inhaled radon or thoron
Scientific Committee on the Effects of Atomic Radiation
gas, both which are highly soluble in body fluids and tissues,
(UNSCEAR) to be 380 Sv; the directly and indirectly ioniz-
is small ( 5%) compared with the doses from their progeny.
ing components contributing 300 Sv and 80 Sv, respec-
Outdoor radon concentrations depend on the amount of
tively. The dose is assumed to be 10-20% higher at high lati-
radon released from soil and the atmospheric factors con-
tudes (> 72°N).
trolling its upward dispersion. The annual global radon emis-
Chapter 8 · Radioactivity
529
sion has been estimated by Harley (1972) to be about 1020
Table 8·1. Global average natural radiation doses (UNSCEAR 1993).
Bq and the atmospheric inventory as 1.5
1018 Bq. This
Annual effective dose, mSv/y
would give a mean surface radon concentration of approxi-
in areas
in areas
mately 4 Bq/m3 in the northern hemisphere with higher val-
of normal
of elevated
ues of about 8 Bq/m3 over the continents.
Component of exposure
background
exposure a
The concentration of radon (and its progeny) is usually
Cosmic rays
0.380
2.01
much higher indoors than outdoors. Based on the available
Cosmogenic radionuclides
0.010
0.01
data, UNSCEAR (1993) estimated that the population-
Terrestrial radiation:
weighted world-wide average radon concentration is 40
External exposure
0.460
4.31
Bq/m3. Indoor surveys in different countries show that some
Internal exposure (excluding radon)
0.230
0.6
Internal exposure from radon
of the highest levels in the world are found in Sweden, Fin-
and its decay products
1
land and Norway. In these countries, radon activity concen-
222Rn inhalation
1.205
10.11
220
trations two to three orders of magnitude above the average
Rn inhalation
0.070
0.11
222Rn ingestion
0.005
0.11
have been reported.
Total
2.405
1
In areas of permafrost within the Arctic Circle, the radon
a. The elevated values are representative of large regions. The cosmic ray
exhalation from the ground is usually low. In addition, houses
dose rate depends on height above sea level and on latitude. Annual
are usually built without a basement and generally on piles.
doses in areas of higher exposure (locations with higher elevations) are
Consequently indoor radon concentrations are generally low.
about five times the average. The dose to a few communities living near
some types of mineral sand may be up to about 100 times the average.
The average annual effective dose to the world popula-
The dose from radon decay products depends on local geology and
tion from inhalation of radon and its progeny has been esti-
housing construction and use, with the dose in some regions being
mated by UNSCEAR (1993) to be 1.2 mSv.
about 10 times the average. Local geology and the type and ventiliation
of some individual houses may combine to give exceptionally high dose
The reported activity concentrations of series radionu-
rates from radon decay products of several hundred times the average.
clides in the body vary widely (NCRP 1987). However, ex-
cept for internal deposition of radon progeny in the respira-
8.2.3.1. Practices
tory tract, the only significant source of internally-deposited
natural radioactivity results from the ingestion of 210Pb and
In radiological protection, the primary objective is to mini-
210Po. The mean daily intake of 210Pb and 210Po through in-
mize the risks to individuals and the collective detriment to
gestion is about 0.1 Bq. For populations in the Arctic and
the exposed population. Accordingly, the focus is on both
subarctic regions, with high consumption of reindeer/cari-
individual exposures and collective exposures. Individual
bou meat, the daily intake can increase to more than ten
exposures are those to individuals and attention is given pri-
times this mean value (Kauranen and Miettinen 1969). Rein-
marily to the most (potentially) exposed group of individu-
deer breeders in northern Norway and Finland may ingest
als, referred to as the critical group. Collective exposures are
on average as much as 10 Bq/d of 210Pb and 210Po. High
individual doses integrated over the entire exposed population
consumers of seafood are also expected to receive higher
and are indicative of the overall detriment to society posed
doses than normal owing to the elevated activity concentra-
by radiation exposures from specific sources and practices.
tions of 210Pb and 210Po in marine products (UNSCEAR
The basic provisions of the `System of Radiological Pro-
1993). This exposure route is assumed to be more important
tection' in relation to proposed and continuing practices are
for some population groups living in the Arctic due to their
termed justification, compliance with exposure limits, and
high consumption of marine products. Cigarettes each con-
optimization. Simply stated, in the context of practices rele-
tain about 20 mBq and 15 mBq of 210Pb and 210Po, respec-
vant to this assessment, these are:
tively. Thus, smokers receive higher doses than non-smokers.
· Justification: Practices involving the production or use of
The world-wide average committed dose from annual in-
radionuclides should be justified as offering net benefit to
takes of natural radionuclides (excluding radon) is estimated
society before being authorized.
to be 0.23 mSv, of which 0.17 mSv is from non-series radio-
· Compliance with Exposure Limits: Limits of radiation ex-
nuclides (mainly 40K) and 0.06 mSv from radionuclides in the
238
posure to individuals (both radiation workers and mem-
U and 232Th series (mainly 210Pb and 210Po) (UNSCEAR
bers of the public) set to avoid deterministic and signifi-
1993). The annual effective dose due to the body content of
cant stochastic effects must not be exceeded.
non-series and series nuclides are estimated to be 0.18 and
· Optimization of Protection: All practices should be opti-
0.13 mSv, respectively. Table 8·1 summarizes the doses re-
mized to reduce radiation exposures to values as low as
ceived from natural exposures derived by UNSCEAR.
reasonably achievable, social and economic factors taken
into account.
8.2.3. The system of radiological protection
Many beneficial human activities involve the exposure of
Individual doses, dose limits and dose constraints
people to radiation from both natural and artificial sources.
For individual exposures, primary concern relates to ensur-
These activities, which are planned in advance, may be ex-
ing the protection of the most exposed individuals, namely
pected to increase the exposure that people already receive
members of the critical group. It is to members of this group
from natural background radiation. These activities are
that the dose limit for members of the public is applied.
termed practices. On the other hand, there are radiation ex-
Prior assessments of practices and sources are directed at en-
posures incurred de facto by people, such as those from nat-
suring that doses to members of the critical group from all
ural radionuclides and nuclear accidents. Activities aimed at
relevant practices are below this limit currently 1 mSv/y
reducing these exposures are termed interventions. The `Sys-
for members of the public. However, when dealing with in-
tem of Radiological Protection' provides the basic require-
dividual practices, only a fraction of the dose limit can be
ments for the protection of people against undue radiation
used. This is called the `dose constraint'. Dose constraints
exposures. Its aim is to prevent the occurrence of determinis-
are designed to ensure that aggregate exposures from all
tic effects due to radiation and to restrict the likelihood of
sources and practices to individuals do not exceed the dose
stochastic effects.
limit for members of the public. Dose calculations must take
530
AMAP Assessment Report
account of all pathways of exposure and all radionuclides,
`System of Radiological Protection', namely: the practice is
however, in reality, a few of these will be dominant, and
justified; protection is optimized; and there is compliance
conservative (pessimistic) consideration of these ensures that
with the individual dose/risk limits.
the dose limit/dose constraint is not exceeded.
A licence is normally required to operate a major installa-
tion. There are, however, other forms of control depending
Collective doses
of the type of the practice. For major installations, justifica-
Collective dose is the integral of dose within a population.
tion of the practice normally goes beyond the radiological
The primary application of collective dose is in relation to
protection regime. Radiation safety aspects are only one
optimization. Optimization focuses on minimizing the col-
consideration, although an important one, in justification.
lective dose as a proxy for the overall health (radiological)
In order to get a licence for a practice, the applicant needs to
detriment. This requires that estimation of collective dose be
make an assessment of the nature, magnitude and likelihood
as realistic as possible, in contrast to individual dose predic-
of the exposures attributable to the practice/source and to
tions that can be conservative to ensure compliance with
show that all reasonable measures for the protection and
dose limits and the relevant dose constraints.
safety of both workers and the public have been taken. The
following discussion is limited to the environmental aspects
Exclusion and exemption
of radiological assessments and does not address either
Any radiation exposure that is essentially beyond human
safety assessments or assessments of worker doses and
control, such as the dose from 40K in the body, is excluded
worker safety.
from radiological control. Furthermore, practices, and
The basic principles of radiological protection are in es-
sources within a practice, may be exempted if the associated
sence followed in all countries for civilian applications of
individual risks are negligible and the collective radiological
ionizing radiation. However, the practical requirements for
impact does not warrant regulatory concern. However, ex-
predictive assessments necessary for the issuing of a licence
emption is also subject to the practice or source being inher-
may vary from country to country. Likewise, the require-
ently safe in the sense of there being no significant likeli-
ments for retrospective assessments may also vary among
hood of circumstances (i.e., accidents) in which the opera-
national jurisdictions.
tional dose estimates would be exceeded.
The most common elements of assessments relating to
limiting and minimizing the environmental consequences of
a practice involving the production, use and release of ra-
8.2.3.2. Intervention
dionuclides may be grouped as follows:
In some situations, the sources, pathways, and exposed indi-
The applicant/licensee must carry out:
viduals already exist when the decisions about control mea-
· A source-related assessment, prior to licensing, covering
sures are being considered. Sometimes the new control mea-
both normal operations and accident scenarios, providing
sures can be defined as part of a review of the original practice,
input to the justification process and demonstrating com-
but, more commonly, they will constitute interventions. An
pliance with the requirements for optimisation of protec-
important group of such situations involves enhanced expo-
tion including the relevant individual dose constraints.
sures to natural sources of radiation. Accidents and emergen-
The assessment of collective dose is used to select the op-
cies will have been considered as sources of potential exposure
timum options for protection.
when dealing with practices, but if they occur, they may call
· Source-related assessments and monitoring during opera-
for intervention. In most situations, intervention cannot be
tion to confirm the validity of the `prior to licence' assess-
applied at the source and has to be applied in the environment
ment; in other words, to confirm that the conditions are
and/or to the freedom of action of individuals. The counter-
within those specified in the assessment and licence.
measures forming a program of intervention, which always
· Dose/consequence assessments in the event of an accident
have some disadvantages, should be justified in the sense that
to predict the consequences and to select necessary and
they should do more good than harm. Their form, scale and
appropriate actions.
duration should be optimized to obtain the maximum benefit.
Dose limits are only applicable to practices. The use of dose
The licensing authority carries out the following types of as-
limits established for the control of practices, or any other
sessments, independent of the applicant/licensee:
predetermined limits, as a basis for deciding on intervention
· Assessments to confirm the applicant's assessments for
might involve measures that would be out of all proportion to
items 1 and 2 above.
the benefit obtained and would therefore conflict with the prin-
· Individual-related assessments to check that the inte-
ciple of justification. Nevertheless, at some level of dose, ap-
grated dose contributions from all relevant sources/prac-
proaching that which would cause serious deterministic effects,
tices do not exceed the dose limits for individuals.
some kind of intervention will become virtually mandatory.
· Source-related assessments, in the event of an accident,
In judging the benefits and detriments of intervention
either domestic or foreign, and, where appropriate, for
aimed at reducing public exposure, the comparison should,
chronic exposure situations, for predicting consequences.
in the first place, be made for those at risk, but there will
also be an impact on the rest of society and the judgements
In other words, the licensing authority assesses individual
will have to be wide enough to also cover these impacts. The
doses for critical population groups to determine whether
application of intervention is to avert future doses. The dose
any individual doses approach thresholds for deterministic
potentially averted by the implementation of intervention
effects and whether any individual has an excessively high
measures is referred to as the avertable dose.
probability of suffering stochastic effects. For intervention,
the assessment of avertable doses by each protective action
is required to justify and optimize the protective action.
8.2.3.3. Radiological assessments
Thus, the avertable average individual and collective doses
Internationally agreed standards for radiological protection
of the affected population need to be assessed.
require that no practice involving ionizing radiation shall be
The basic structure of source-related assessment, mostly
adopted unless it accords with the basic principles of the
predictive in nature, is similar irrespective of the application
Chapter 8 · Radioactivity
531
of the assessment. Its purpose is to establish the relationship
source. Thus, whenever possible, controls applied at the
between the source (release rate) and the consequences to
source are to be preferred. Actions applied to the environ-
humans and biota (expressed in terms of dose rates). It
ment, or to individuals, are more obtrusive and may have
should be noted that it has been shown that fauna will not
social disadvantages, not all of which are foreseeable. Their
be adversely affected at the population level provided that
effectiveness will be limited because they apply only to some
the system for limiting the exposure to humans is applied.
of the pathways and individuals.
This conclusion is conditional on the proximity of the ex-
It is essential to avoid confusion between `Dose Limits
posed humans and fauna relative to the source. In cases
and Constraints' restricting releases from normal operations
where the exposed humans and fauna are at similar dis-
and the `Intervention Levels or Action Levels' for chronic
tances from a source, the conclusion is valid. However,
exposure situations or accidents, that trigger intervention or
where fauna are situated relatively close to a source com-
action. Although similar principles apply to normal opera-
pared to humans, such as in the case of sources situated on
tions of practices and to intervention in post-accident or
the deep remote ocean floor, faunal and human doses are so
chronic exposure situations (i.e., justification and optimiza-
dissimilar as to require specific and independent considera-
tion), they are applied to different quantities. For the control
tion of the dose to fauna (IAEA 1988, 1992).
of planned releases, the benefit from the source itself is com-
First, it needs to be shown that the releases will not cause
pared with the additional radiation exposures it produces. In
higher individual doses than the dose constraints assigned by
the case of intervention, the disadvantages of the interven-
the relevant national authority. Second, it needs to be dem-
tion are compared with the reduction in total radiation ex-
onstrated that the doses have been reduced by appropriate
posure (irrespective of the origin of the exposure) achiev-
measures to a level below which it is no longer reasonable to
able. Intervention levels, based on the justification and opti-
make further reductions taking into account social and econo-
mization principles, are either generic or specific, and are
mic factors (in other words that the protection is optimized).
primarily expressed in terms of avertable dose in Sv (typi-
In the optimization process, the alternative technical pro-
cally as mSv) but can also be expressed in directly measur-
tective measures can be compared with each other in their
able quantities, as dose rates or activity concentrations.
ability to reduce the collective dose in relation to the re-
For accident situations a set of `Generic Intervention Lev-
sources spent, to identify the option of optimal protection.
els' has been derived and internationally recommended.
Social factors can play an important role in the optimization
They are given in terms of avertable dose achieved by major
process, such as in the case of selecting intervention mea-
protective actions applicable in case of a nuclear accident.
sures. However, there are many decision-aiding techniques
Also, `Action Levels' (or interdiction levels) for food were
that can be used to take social factors, which are often diffi-
recommended by the FAO-WHO Codex Alimentarius Com-
cult to quantify, into account, and these are not discussed
mission and the International Atomic Energy Agency (IAEA)
further here.
(Table 8·2). They can be used if there is no shortage of food
Individual-related assessments carried out by authorities
and there are no other compelling social or economic fac-
concerning the public are basically retrospective assessments.
tors. They were originally developed for the international
These enable the authorities to ensure that: individuals do
trade in food contaminated with radionuclides but are also
not receive doses exceeding the dose limits from licensed
applied to food as consumed.
practices because of unanticipated overlapping of critical
Table 8·2. Generic action levels for foodstuffs.
groups, and, following an accident or in a chronic exposure
situation, individuals do not incur doses that would call for
Foods destined
Milk,
for general
infant foods and
consideration of protective actions.
consumption,
drinking water,
Such assessments are based on measurements of activity
Radionuclide
Bq/kg
Bq/kg
concentrations in environmental materials that can con-
134Cs, 137Cs, 103Ru, 106Ru, 89Sr 1000
1000
tribute to the internal and external doses to members of the
131I
100
public or, in some cases, on direct activity concentration
90Sr
100
measurements on humans.
241Am, 238Pu, 239Pu
10
1
If the individual exposures are excessive, intervention
should be considered. An evaluation of the individual and
Otherwise no international consensus on Action Levels in
collective doses avertable by potential intervention measures
chronic exposure situations yet exists, except for radon in
is required for justification purposes. Estimates of avertable
dwellings. Thus, the basis for intervention is to justify and
dose should be as realistic as possible to avoid overestima-
optimize the available options for protective actions in
tion of the potential benefits of protective actions.
chronic exposure situations on a case-by-case basis. How-
ever, work is underway in both the International Commis-
sion on Radiological Protection (ICRP) and the IAEA to
8.2.3.4. The basis for intervention
reach a consensus on these matters.
The measures needed to restrict the exposure of individuals,
either in the control of a practice or by intervention, can be
8.2.3.5. Other issues relevant to radiological assessment
taken by applying action at any point in the paths linking
8.2.3.5.1. Relationship between radiation exposure
the source to the individuals. The action may be applied to
and risk of adverse health effects
the source, to the environment, or to the individual, e.g.
moving people, or personal protective measures. Actions
Health protection from radiological exposures at low doses
that can be applied at the source will be the least disruptive.
(stochastic effects regime) is based on an important a priori
They can be made as effective as required, unless they fail as
assumption that the risk of adverse health effects increases
a result of an accident or for other reasons. For example,
in direct proportion to radiation exposure without thresh-
such could be the case if disposed waste is removed from
old. This permits extrapolation of the dose-response relation-
one part of the environment to another without careful as-
ship into low dose regimes from that at higher dose where
sessment of the consequences. Action at the source influ-
the relationship can be epidemiologically or experimentally
ences all the pathways and individuals associated with that
determined. There is an established relationship between
532
AMAP Assessment Report
probability of serious health defects (fatal cancer induction)
lesser extent, radioiodine, because of their source strengths,
and dose, of 0.05/Sv averaged over the population. This
mobilities or radiotoxicities.
means that a dose of 1 mSv corresponds to an increased risk
of serious health defect of 5
105.
Atmospheric transport
There is an interesting consequence of the basic assump-
Radionuclides can be released from a wide variety of differ-
tion of linear no-threshold dose-response in the low dose
ent sources and can be ejected into a variety of atmospheric
stochastic regime. If, for example, a practice results in a
layers under different conditions. Weather conditions at the
large population of people suffering increased radiation ex-
time of atmospheric release will generally determine the ex-
posure and the integrated (collective) dose in this population
tent of atmospheric dispersion. The mean residence time of
is 100 manSv then the expected number of serious additio-
radionuclides in the Arctic stratosphere is in the order of one
nal health defects in the population is 5 irrespective of the
year. The transfer of radionuclides from the stratosphere to
size of the exposed population. Obviously, the smaller the
the troposphere occurs preferentially in the spring, when the
population over which the exposure is distributed, the more
tropopause (the interface between them) is most `permeable'
seriously this collective (health) detriment might be regarded
(Brewer 1949, Dobson 1956). The mean residence time of
because of the increased individual doses.
radionuclides in the troposphere is only a few weeks. Radio-
The concept of detriment is used as a measure of the total
nuclides in the troposphere are transferred to the surface of
harm that would eventually be experienced by an exposed
the Earth as wet or dry fallout.
group and its descendants as a result of the group's exposure
Radionuclides have been introduced into the Arctic at-
to radiation. Health detriment is part of the total detriment,
mosphere from nuclear weapons testing and from accidental
however, in practice, in radiological protection, the term is
(e.g., Chernobyl) or routine (e.g., the Kola nuclear power
used solely in relation to health detriment. In optimization
plant) releases from nuclear facilities. The testing of thermo-
studies, special allowance needs to be made for other forms
nuclear weapons (in the Megatons TNT equivalent range) in
of detriment, as appropriate.
the atmosphere usually injected most of the radionuclide
yield into the stratosphere. Venting from underground nu-
clear explosions and releases from reactor accidents, atmos-
8.2.3.5.2. Transport processes and exposure pathways
pheric tests of fission weapons (in the kilotons TNT range)
Radiation exposure can be grouped into two main types:
mainly entered the troposphere.
external and internal. External exposures are those resulting
from sources outside the person or organism. Internal expo-
Marine transport
sures (comprising inhalation and ingestion) are those result-
Releases into Arctic marine ecosystems can either occur di-
ing from the incorporation of radionuclides into an organ-
rectly, through routine releases from nuclear reactors into
ism. There are a wide range of pathways, summarized be-
cooling water streams, leakage from dumped solid wastes,
low, that can lead to exposures of organisms. In construct-
direct dumping of liquid wastes, or indirectly via atmos-
ing assessments of prior exposure, the objective is to ensure
pheric deposition. In addition, radionuclides released else-
that all potential pathways of exposure are considered, al-
where may be transported into Arctic marine systems. Typi-
though in many cases there will be one or two exposure
cal examples of the latter include the releases from Sellafield
routes that will be dominant these are referred to as criti-
and Cap de La Hague reprocessing plants. Furthermore, re-
cal pathways. In Arctic ecosystems, certain critical pathways
leases into freshwater, either directly or via catchment conta-
are particularly important for this assessment. These path-
mination, may eventually be transported into the Arctic ma-
ways are discussed here in order to provide background in-
rine environment via river systems. Waterborne discharges
formation for discussions in subsequent sections. It is impor-
have occurred from a variety of different Russian nuclear es-
tant that appropriate models are available, and that they
tablishments to the Ob and Yenisey river systems. This has
adequately describe the transport of radionuclides in the en-
undoubtedly resulted in the transport of some mobile radio-
vironment. The types of models and their uses are also con-
nuclides (e.g., 90Sr) through aquatic pathways to Arctic ma-
sidered later in this section.
rine ecosystems, but it is presently difficult to quantify the
External exposure arises from radionuclides deposited
amounts of radionuclides transported in this way.
onto many different surfaces. The dose varies with the ra-
dionuclide deposited, with different exposures occurring for
Terrestrial transport
various alpha-, beta- and gamma-emitting radionuclides. In
Once radionuclides are deposited onto the Earth's surface,
addition, the dose will change with time, as radionuclides
their subsequent behavior is dependent on a number of fac-
migrate down soil profiles or are weathered from plant sur-
tors including their physico-chemical form and the type of
faces, particularly in forests or urban areas where significant
environment into which they have been released. Terrestrial
interception of radionuclides can occur above the ground.
and freshwater environments generally receive most of their
Internal exposure occurs largely through both inhalation
radioactive contamination through precipitation (wet fallout).
and ingestion. Inhalation exposure occurs when radionu-
Vegetation may be contaminated directly by deposition of
clides are breathed into the lung with air and can either im-
the radionuclides onto the surface of the plants, or indirectly
part direct exposures to the lung or be retained in lung tissue
by uptake from the soil through the roots. Further transfer
and possibly absorbed into the plasma. Ingestion exposure
of radionuclides in the food chain occurs when animals, in-
can arise through drinking or eating contaminated food-
cluding humans, consume food, drink water or breath air. A
stuffs and can therefore result from a large number of differ-
common example, with which most people are familiar, is
ent exposure pathways following releases to the atmospheric
the grass cow milk man pathway, whereby grassland
and marine/aquatic environments. The most important fac-
is contaminated through atmospheric fallout and the contami-
tors which lead to variation in rates of transfer via these
nation is transferred to humans through the consumption of
pathways differ for each radionuclide and, hence, the envi-
grass by cattle, and the subsequent production and consump-
ronmental mobility of different radionuclides also varies
tion of milk. Certain processes are of central importance in
considerably. This assessment focuses primarily on radiocae-
determining the rates of transfer and these are summarized
sium, radiostrontium, plutonium radioisotopes and, to a
below with particular reference to Arctic ecosystems:
Chapter 8 · Radioactivity
533
Interception
in the Arctic. The rate of contamination of food products
The rate of interception of aerially-deposited radionuclides
from these animals depends on three major factors:
varies with surface characteristics, meteorological conditions
· Diet selection.
and the ratios of surface area to biomass, and is particularly
In temperate regions, diet selection by food producing
high for many tree species, lichens and mosses. In addition,
agricultural animals is comparatively unimportant as the
rates of interception vary seasonally, particularly for annual
range of herbage available is highly regulated and often
crops. Intercepted fallout is gradually lost from the inter-
comprises only a few major herbage sources. In contrast,
cepting surfaces by a variety of processes, collectively termed
animals in semi-natural ecosystems ingest a wide range of
`weathering'. The initial rates of interception and subsequent
different plants and fungi at different times of the year.
rates of weathering are important factors in agricultural sys-
This leads to considerable seasonal variation in the
tems because they determine the degree of external contami-
amounts of radionuclide ingested by different species.
nation of crops and pasture grasses in the initial phase after,
A classic Arctic example is the consumption of lichen
for example, an accident. In Arctic food chains, the ability
by reindeer in winter which leads to substantially higher
of lichen to intercept, absorb and retain most of the depo-
radiocaesium activity concentrations in reindeer meat
sited radiocaesium is particularly important because of the
during the winter period. In addition, radiocaesium con-
utilization of lichen as a winter foodstuff for reindeer.
tamination of game, such as roe deer and moose often
Soil-to-plant transfer
substantially increases in autumn due to the consumption
of highly contaminated fungi.
In temperate areas, the variation in the rate of soil-to-plant
transfer of radionuclides is one of the most important fac-
· Availability for absorption in the gut.
tors influencing the extent of food contamination for both
Radionuclides are absorbed to different extents in animal
agricultural and semi-natural products. In Arctic areas, the
guts. After ingestion of contaminated vegetation the three
comparative importance of this exposure route for agricul-
most available radionuclides, in order of decreasing frac-
tural products is potentially much lower than in temperate
tions of gut absorption are: radioiodine (100%) > radio-
areas because fewer agricultural plant products are grown.
caesium (80%) > radiostrontium (ca. 20%). Most other
Plants obtain nutrients and radionuclide contaminants
radionuclides, including plutonium, are absorbed in the
from the soil solution. Thus, the rate of uptake from soil by
gut in fractions of less than 1%.
plants is determined by the rate at which the plant roots ab-
· Metabolism of the radionuclide.
sorb different elements or compounds and the activity con-
Once radionuclides have been absorbed through the gut
centrations of radionuclides in the soil solution. If a radio-
wall they are distributed within animal tissues. The tis-
nuclide has a close chemical analogue, the rate of transfer of
sues in which they accumulate and the subsequent rates
the radionuclide will be heavily dependent on its interaction
of loss, via urine and faeces, vary. The most important ra-
with the analogue, particularly any competitive effects.
dionuclides are those which contaminate parts of the ani-
When radionuclides are deposited onto the soil they are
mal which are eaten by humans, namely meat, offal and
chemically bound by different soil constituents and it is the
milk. Again, the most important radionuclides are radio-
relative strength of these associations that determines the ac-
iodine, radiocaesium and radiostrontium, all of which are
tivity concentration of the radionuclide in the soil solution.
readily transferred to milk. In addition, radiocaesium
Many radionuclides are either taken up by plant roots at
contaminates all soft tissues, and therefore ingestion of
very low rates, or form strong bonds with various soil con-
radiocaesium via meat is also important. The effective
stituents. Therefore, the rate of plant uptake of many radio-
biological half-lives of these radionuclides vary, but the
nuclides is low compared with nutrient ions. The main ex-
rates of radioiodine uptake and loss are generally faster
ceptions are radiostrontium, which has significant rates of
than those of radiocaesium and radiostrontium. Changes
uptake from many different soil types, and radiocaesium,
in radionuclide activity concentrations in ingested food
which is absorbed by plant roots much more readily from
will be reflected in milk and meat within a few days. Tar-
organic soils and, to a lesser extent sandy soils, than from
get organs for the other different radionuclides vary, but
more mineralized soils with a higher content of clay miner-
notably include bone and offal.
als which strongly bind radiocaesium.
In addition to the soil-based factors, there are marked
The effective biological half-life of a radionuclide in an or-
differences in the capacity of different plant species to ab-
ganism is a function of both the biological half-life of the el-
sorb radionuclides. However, these differences are usually
ement in the organism and the physical half-life of the radio-
smaller than those determined by the soil type.
nuclide.
For radiocaesium, a further important exposure route
1/ T
from the soil involves uptake by fungal hyphae. Many soils
1/2 eff-biol = 1/ T1/2 biol + 1/ T1/2 phy
contain large amounts of fungal hyphae that have a pro-
The effective ecological half-life of a radionuclide is a func-
nounced ability to absorb radiocaesium from the soil. When
tion of both the half-life of the element in a component of
the fruiting bodies (e.g., mushrooms) appear they often con-
an ecosystem and the physical half-life of the radionuclide.
tain much higher radiocaesium activity concentrations than
1/ T
most other food products. The extent of radiocaesium con-
1/2 eff-eco = 1/ T1/2 eco + 1/ T1/2 phy
tamination of fruit bodies is highly variable, both within
and among fungal species.
Freshwater pathways
Freshwater systems, such as lakes, rivers and groundwater,
Plant-to-animal transfer
may also be contaminated by atmospheric deposition of ra-
Animal products form an important part of the diet of
dionuclides or direct releases into rivers. The transfer of ra-
many Arctic peoples. Whilst some animal products are sim-
dionuclides from such systems occurs mainly through con-
ilar to those of temperate regions, such as milk, pork and
sumption of freshwater fish and from exploitation as drink-
lamb, a wide range of game animals and, of course, both
ing water. The mobility of a radionuclide depends on its
semi-domesticated and wild reindeer are also heavily utilized
ability to bind to river sediments and its competitive interac-
534
AMAP Assessment Report
tions with other ions. Strontium is one of the more mobile
tamination routes in semi-natural terrestrial ecosystems can
elements in aquatic systems because it does not bind strong-
usually be represented within existing model structures, with
ly to sedimentary material.
minor changes necessary to allow, for example, for surface
contamination of lichen, rather than root uptake, as a long-
Marine pathways
term exposure route via reindeer. However, transfer values
Exposure from marine pathways arises from the consump-
used in such models need to be appropriately quantified to
tion of marine food products, including fish and shellfish,
describe adequately the movement of radionuclides between
mammals such as seals and whales, and seaweed. In general,
environmental compartments in the Arctic. This cannot be
contamination of marine biota is much less than that arising
accomplished for Arctic pathways with the same degree of
from terrestrial pathways, largely because of the strong
confidence as is possible for temperate conditions because
sorption of many radionuclides by aquatic sediments and
there is much greater individual and seasonal variation in
also because of the enormous dilution which occurs in ma-
rates of transfer, associated, for example, with diet selection
rine water bodies.
by animals. Thus, parameterization is a more significant prob-
lem in the application of models to the Arctic because of the
relatively greater importance of semi-natural ecosystems.
8.2.4. Modeling
To parameterize transfer in Arctic areas, it is often neces-
sary to use both individual and aggregated transfer parame-
Measurements and models necessary for radiological assess-
ters. Individual parameters describe the direct transfer from
ments can be grouped as follows:
one environmental compartment to another. An example is
· In cases where the releases of radionuclides from the
the transfer from feed to cattle. Aggregated parameters de-
source are not known or not measured, scenarios and
scribe the transfer via a complete chain of compartments for
models for existing, projected or potential release rates
which each compartmental transfer has an individual trans-
have to be developed (e.g., releases from reactor accidents,
fer rate. An example is a transfer parameter which relates a
Komsomolets submarine, dumped packaged material,
radionuclide activity concentration in soil to that in the meat
dumped reactors, contaminated marine sediments, etc.).
of grazing animals. The diet of almost all humans comprises
· Environmental pathways from the source to humans have
food derived from diverse locations in the world. It is, there-
to be identified and the transport of radionuclides mod-
fore, not always possible to directly correlate the radionu-
eled. In box models, the pathways of radionuclides from
clide concentration in food with the contamination of the
the release point to humans are described by transfers
area where humans are living. However, for some foodstuffs
among trophic levels in the environment, such as the trans-
this is a less serious problem in the Arctic than for most
fer of airborne radionuclides from `air' to `pasture' to
other areas.
`milk'. The radionuclide transfer between compartments
Radiation dose (dose equivalent from external exposure
is commonly described by transfer parameters. In simple
plus committed effective dose from intakes of radioactive
models, these transfer parameters represent the ratio of
substances) is the relevant quantity in radiological protection
concentrations of a radionuclide in two compartments
for assessing health consequences. The procedure for calcu-
under equilibrium conditions. Changes with time are de-
lating doses to humans are based on different approaches
scribed using appropriate rate functions for decay or re-
according to the nature of the pathway. For external expo-
moval. In more complex models, an attempt is made to
sures, the dose to individuals from radionuclides in air, wa-
represent the time-dependent movement of radionuclides
ter or ground surfaces is obtained by applying the appropri-
among the various environmental compartments. These
ate dosimetric models and taking into account shielding ef-
time-dependent models are referred to as `dynamic' models.
fects, annual rates of occupancy and any other factors char-
acterizing the behavior of the individuals. For internal expo-
Simple equilibrium models have been well described and
sures, doses are calculated using metabolic models that
documented in the literature and many of the transfer para-
incorporate inhalation, food ingestion and gut absorption
meters have become virtually `standardized'. In contrast, the
rates. Such models have been internationally established and
parameters used in dynamic models tend to be both model
provide dose-intake factors for different radionuclides.
and situation specific and their values depend, amongst other
There are several ways to assess and predict individual
things, on the assumptions made in establishing the model.
dose commitments. In short, the following approaches are
Equilibrium box models are often sufficiently robust and re-
used in this assessment:
liable for radiological protection purposes.
Most predictive models, both dynamic and equilibrium,
Individual dose commitments:
have been developed to describe temperate conditions and
· Annual intakes of 137Cs and 90Sr via food and drinking
assume that external exposure arises predominantly from
water have been assessed and integrated over the time
soil, forest or urban surfaces and that internal exposure
period from 1950 to . The results have been multiplied
arises from inhalation and ingestion of contaminated food-
by relevant dose/intake factors to obtain the internal dose
stuffs produced as a result of normal agricultural or fishing
contributions from these radionuclides via consumption.
activities. It is, therefore, important to consider whether ex-
Doses via the inhalation pathway have been calculated in
isting models can be adapted to Arctic areas. For marine ex-
the same way, as appropriate. Contributions from other
posure, the transport of radionuclides in ice is an additional
radionuclides and external doses are based on ratios de-
pathway which needs to be incorporated into an assessment.
rived from UNSCEAR (1993).
The main difference with regard to external exposure is the
· The internal dose from 137Cs can also be assessed by
presence of snow and the lack of forested areas, which may
using wholebody measurement. The body burden, ex-
change exposure rates. This can be compensated for within
pressed in Bq/kg, is then multiplied by a dose factor
current models using appropriate factors. For estimating in-
given by UNSCEAR to obtain the dose commitment.
ternal exposure, the main difference lies in the origin of the
· UNSCEAR applies a compartment model (Figure 8·1) to
foodstuffs consumed in the Arctic. Agricultural production
assess the dose commitment from releases of radioactive
is limited in the Arctic, and most foods are either harvested
substances to the atmosphere from atmospheric nuclear
from semi-natural or marine ecosystems or imported. Con-
Chapter 8 · Radioactivity
535
Inhalation
consume essentially no mushrooms in poor years. This could
easily change the annual body concentrations of radiocae-
Ingestion
sium significantly in the absence of any similar changes in
the annual depositions, which are used for the annual pre-
Earth's
dictions in the UNSCEAR model. Similarly, for example, in
Input
Atmosphere
Surface
Diet
Tissue
Dose
a given year reindeer could move to an area where lichen is
0
1
2
3
4
5
less abundant, which would reduce radiocaesium body levels
in that year. Such effects would not be predictable from the
External irradiation from the ground
UNSCEAR model because changes in food consumption pat-
terns are not taken into account. However, if enough years
External irradiation from the atmosphere
of observations of environmental contamination levels in the
Arctic are available, it is possible to apply the UNSCEAR
Figure 8·1. Compartment model used to assess doses from releases of radio-
model for the calculation of the transfer coefficient P23 from
active materials to the atmosphere from nuclear testing (UNSCEAR 1982).
deposition to diet, because the temporal and local variations
are effectively smoothed out.
weapons testing. The dose commitment for a specific ra-
Integrated transfer factors provide a comprehensive as-
dionuclide Dc, due to an environmental input A0 into the
sessment of transfer over the long term. However, for many
atmosphere is given by:
products, especially in the Arctic, the data is too limited to
calculate these factors. As an alternate approach, the use of
Dc = P01(P12P23P34P45 + P14P45 + P15 + P12P25)A0
aggregated transfer coefficients can be considered.
where Pxy is the transfer coefficient from box x to box y.
P01 is the integrated concentration of a radionuclide in air
Aggregated transfer coefficients (Tag s)
at a specific location (or averaged for a broader region),
In the event of a radioactive release, there is often an urgent
divided by the amount released. The four terms in the
need for information on transfer to food products. In agri-
parentheses account for the ingestion, inhalation and ex-
cultural production systems, soil and vegetation properties
ternal exposure (cloud gamma and ground gamma, re-
are typically fairly homogeneous and models have relied on
spectively) pathways. The values of the various transfer
such parameters to define the concentration ratio for soil
coefficients adopted by UNSCEAR have, to a large ex-
plant transfer and the transfer coefficient for feedanimal
tent, been derived from observations made in northern
transfer. In contrast, semi-natural ecosystems are inherently
temperate latitudes. This assessment has considered the
more variable in both soil and vegetation characteristics,
validity of this approach for Arctic areas.
and animals graze much more selectively. Consequently, as-
sessment of transfer using established parameters may not
Integrated transfer factors
adequately account for all the pathways that contribute to
The transfer of radionuclides from deposition to diet has
an activity concentration in a given foodstuff.
been calculated by UNSCEAR (1993) according to the model:
The aggregated transfer coefficient (Tag) has, therefore,
C
been developed as a simple means of quantifying transfer be-
i = b1 Fi + b2 Fi 1 + b3 e n Fin
n =1
tween different environmental compartments in semi-natural
where Ci is the activity concentration of the radionuclide in
ecosystems after an accident (see review by Howard et al.
a food component in the year i due to the deposition density
1996). By comparing activity concentrations in a food prod-
rate in the year i, Fi, in the previous year, Fi1, and in all pre-
uct with representative measurements of total deposition, on
vious years reduced by exponential decay. This decay, with
an areal basis, aggregated values of transfer can be obtained
decay constant , reflects both radioactive decay and envi-
that can be easily applied in radiological assessment models.
ronmental loss of the radionuclide, i.e., it corresponds to the
For the current assessment, three different approaches have
effective decay constant = ln 2/Teff , where Teff is the effec-
been identified: 1) comparison of foodstuff activity concen-
tive ecological half-life (see section 8.2.3.5.2). The coeffi-
tration with total deposition, to give an estimate of transfer
cients b1, b2, b3 and the parameter are determined by re-
from all sources, 2) transfer from an individual release, Cher-
gression analysis of measured deposition and diet data. The
nobyl fallout, which has been quantified by studying changes
so-called transfer coefficient P23 (see example in section 8.4)
in the ratio between 134Cs and 137Cs, and 3) Tag values have
from deposition to diet is defined as:
been calculated from the organic layer of the soil to mush-
rooms. The second and third options may be optimal para-
P23 = b1 + b2 + b3 e n / (1 e n)
meters for modeling in Arctic ecosystems following an acci-
with the units of Bq y/kg per kBq/m2. P23 is the infinite time-
dent and have been used preferentially in this assessment
integral of the activity concentrations of a radionuclide in a
where available.
product (e.g., milk) arising from the deposition of 1 kBq/m2
Provided specific characteristics of an ecosystem are
of the radionuclide.
known, Tag values offer a convenient means of assessing
The above UNSCEAR model has been applied success-
transfer between most environmental compartments. How-
fully to nuclear weapons fallout on agricultural ecosystems
ever, like all transfer parameters they must be combined
in temperate regions, but may not be readily applicable to
with estimates of time-dependent changes in transfer, ex-
Arctic ecosystems where the pathways of radionuclide accu-
pressed as effective ecological (T1/2 eff-eco) half-lives. This is
mulation in a given product may vary locally as well as tem-
particularly important in semi-natural ecosystems where
porally. In particular, the UNSCEAR model omits the lichen-
contamination can persist for many years. Furthermore, im-
reindeer pathway, which is important in the Arctic, and is
mediately after a release Tag values are affected by the inter-
limited in its ability to take account of temporal variability
ception of fallout radionuclides by vegetation cover (this is
in production and/or harvesting which can vary consider-
further affected by season). Therefore, Tag values are of lim-
ably in semi-natural systems. For example, some terrestrial
ited value for quantifying transfer when deposition is contin-
mammals may consume large amounts of mushrooms in
uous over a long period. The rate of transfer of intercepted
years in which there are many mushrooms produced, and
deposition to the soil surface may also be subject to seasonal
536
AMAP Assessment Report
Environmental
Human
Atmospheric/deposition
Prepared food
Unprepared food
Figure 8·2. Geographical distribution of sample information in the AMAP radioactivity database.
influences. In contrast, lichens act as a surface trap for depo-
· Source-related assessments of future releases as a result of
sition, and therefore Tag values can be determined and applied
potential nuclear accidents or from contained sources in
immediately for transfer from lichen to reindeer.
the environment such as objects containing radioactive
wastes dumped at sea. These assessments are based on es-
timates of the inventories and release rates of radionu-
8.2.5. The AMAP assessment
clides from such contained sources and modeling of the
Prior assessment of proposed practices is a common require-
subsequent environmental transport and human exposure
ment imposed by legislative jurisdictions in both nuclear and
pathways (section 8.6).
non-nuclear fields. Surveillance and retrospective assessments
It should be noted that some of the sources considered under
are also employed to ensure that practices adopted by socie-
item 2 above (i.e., in section 8.5) are both current and po-
ty have consequences (detriments and benefits) that are con-
tential sources of releases to the environment. In such cases,
sistent with those outlined in the prior assessment. The latter
for example that of the sunken Komsomolets submarine, the
can be specific to particular practices or more general, such
relevant source-related assessment is presented in section
as assessments of conditions in a specific regional area like
8.5. rather than 8.6.
the AMAP Assessment.
The sections of this document dealing with radiological
assessments comprise three parts:
8.3. Past and present radioactive
· Individual-related dose assessments for individuals in
contamination of the Arctic
average Arctic populations and to individuals within se-
lected real or hypothetical population groups within the
During the AMAP assessment period, data about the past and
Arctic considered to have comparatively high radionu-
present contamination have been collected mainly through
clide intakes. These assessments are based on observed
the AMAP radioactivity data center which, together with the
activity concentrations in environmental media, dietary
AMAP radioactivity assessment group, has compiled data to
intakes and dose-intake factors and, wherever relevant,
describe levels and trends in the Arctic area. The sources of
take account of occupancy times, shielding factors and
the data include results from AMAP monitoring programs,
dosimetric models (section 8.4).
national reports compiled for the assessment group, and the
· Source-related assessments of present and future doses to
open literature. In total, measurements of more than 20 000
Arctic populations and selected population groups from
samples have been reported. There are, of course, gaps in
operational releases from nuclear power generating plants,
both temporal and spatial information. However, consider-
other civilian and military reactors, nuclear fuel repro-
able information on radioactive contamination in the Arctic
cessing installations, mining activities, nuclear explosions,
is available. An overview of the geographical distribution of
both military naval and civilian, and previous accidental
the samples in the database is given in Figure 8·2. Full refer-
releases of radionuclides to the environment (section 8.5).
ences to all data are available in the database.
Chapter 8 · Radioactivity
537
Number of Samples
1 200
1 000
800
600
400
200
0
1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Number of Samples
1 200
1 000
800
600
400
200
0
1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Atmospheric
Freshwater
Human body
Terrestrial
Marine
Figure 8·3. Available data in the AMAP radioactivity database as a function of time.
The gaps in the data are particularly notable for natural
food products, such as mushrooms, wild animals (except
8.3.1. Geographical distribution
reindeer) and freshwater fish (for some areas). In addition,
of radioactive contamination
there is a lack of data prior to 1960 for Russian rivers (espe-
cially the Ob) which most probably contained radionuclides
Radioactive contamination of the Arctic has occurred at two
in the 1950s as a result of releases from the Mayak repro-
different scales:
cessing plants to the Ob river system. Another notable lack
1. Widespread contamination, such as that associated with
of data relates to radionuclides in sea ice, for which there is
global nuclear weapons testing, Sellafield releases and the
some information from recent years, but few data prior to
Chernobyl accident.
the 1990s.
2. Localized contamination of smaller areas (e.g., resulting
As a result of atmospheric nuclear weapons testing, most
from the Thule nuclear weapons accident and radioactive
countries started monitoring of radionuclides in various
wastes dumped at sea).
samples in the late 1950s or early 1960s. Regrettably, many
of the terrestrial monitoring programs were terminated in
The following presentation focuses on 137Cs and 90Sr, since
the late 1960s or early 1970s due to the decreased fallout
these radionuclides are important for determining dose to
following the atmospheric nuclear test ban in 1963. Never-
humans, and considerable data exist on each of them.
theless, a few time series were continued to the present and
give temporal information on radionuclides in the Arctic en-
8.3.1.1. Widespread contamination of land and sea
vironment.
Marine sampling increased in the late 1970s and early
Terrestrial contamination
1980s. This is probably due to interest in the increased dis-
The two major sources of fallout in the Arctic region have
charges of radionuclides, especially 137Cs, from the British
been nuclear weapons testing and the Chernobyl accident.
Nuclear Fuel's reprocessing plant in Sellafield, UK, and
A total of 520 atmospheric nuclear weapons tests have
other European reprocessing plants. At the end of the 1980s,
been carried out (UNSCEAR 1993). Of these, 88 have taken
numerous new monitoring programs for radionuclide were
place in the Arctic on the island of Novaya Zemlya. The ma-
initiated as a response to the Chernobyl nuclear power
jority of radiocaesium deposition occurred during the period
plant accident and to the information released by the Rus-
1955-1966 as a result of (global fallout from) atmospheric
sian Federation about dumping of nuclear waste in the
nuclear weapons tests performed in the northern hemisphere.
Kara and Barents Seas by the former Soviet Union. Figure
Precipitation and latitude have been identified as the princi-
8·3 summarizes the number of available sample data as a
ple factors determining the spatial variation in global fallout
function of time.
(UNSCEAR 1993).
538
AMAP Assessment Report
The AMAP radioactivity assessment group, through the
Table 8·3. Predicted mean ground deposition of 137Cs in Arctic countries
AMAP radioactivity data center, devised a novel method of
from atmospheric nuclear weapons testing, decay corrected to 1995, using
estimating spatial variation in 137Cs deposition in the Arctic
GIS-based approach.
using Geographical Information Systems (GIS). An annual
Arctic country/
Predicted mean ground deposition,
relationship between precipitation and radiocaesium deposi-
region
Bq/m2
tion has been derived for Tromsø in Norway (Playford et al.
Canada
730
1993). The annual relationships have been combined with
Greenland a
1400
mean annual precipitation (at a resolution of 0.5°
0.5°) for
Iceland
2900
the northern hemisphere (Leemans and Cramer 1991) to
Norway
1900
Sweden
1500
calculate the input of radiocaesium within each year using
Finland
1400
GIS. A latitudinal correction based on UNSCEAR was also
Russia (east)
700
applied initially. It is possible to calculate the decay cor-
Russia (west)
1000
United States (Alaska)
1300
rected ground deposition or the total, integrated, ground
deposition from nuclear weapons testing for any spatial unit
a. Coastal inhabited areas of Greenland only, not including inland ice cap.
in any year. This approach does not consider physical losses
90
by lateral or vertical transport.
Sr in global fallout of 1.6 : 1 (UNSCEAR 1988). Estimates
The predicted values of deposition were compared to
of the mean ground deposition of 137Cs from nuclear weap-
measurements of soil inventories contained in the AMAP
ons testing for different countries, derived from the GIS, are
database, and those published for the UK by Cawse and
shown in Table 8·3. The areas with the lowest estimated
Horrill (1986). From these comparisons, it appeared that
ground deposition from nuclear weapon testing, are in the
the use of a latitudinal correction resulted in an overpredic-
Russian Far East, North Greenland and northern and central
tion of ground deposition, possibly because the latitudinal
Canada. The highest estimated ground deposition has oc-
variations in 137Cs deposition and precipitation are interre-
curred in the coastal areas of Norway and Alaska, the south-
lated. Without latitudinal correction, predictions were in
ern tip of Greenland and the southwest coast of Canada.
reasonable agreement with measured deposition. Conse-
The predictions are based on measurements of combined
quently latitudinal correction has not been used in deriving
wet and dry deposition collected at Tromsø. However, using
the fallout map but since this is a novel approach further
this approach, ground deposition in regions which receive
validation is required. An estimate of the distribution of
low annual precipitation will be underestimated. For in-
stance, at Thule, Greenland, where annual precipitation is
low (150-200 mm/y) compared with much of the Arctic, a
value of predicted decay-corrected ground deposition in
1970 of 510 Bq/m2 can be compared to a measurement of
950 Bq/m2 (Aarkrog 1978). This indicates a discrepancy of
440 Bq/m2, probably due to the higher proportion of dry
deposition at Thule, compared to Tromsø. However, as the
contribution of dry deposition will vary according to preci-
pitation rate, it is not possible to apply a simple correction
to the total predicted surface. Hence, when predicting depo-
sition from global fallout using the approach described, it is
necessary to bear in mind that ground deposition will be un-
derestimated in drier areas.
The GIS-based approach used for Figure 8·4 can also be
used to estimate the total inventory of 137Cs and 90Sr from
global fallout over land. The estimated inventory over land
north of 60°N is 35.0 PBq of 137Cs and 21.9 PBq for 90Sr.
These estimates can be compared with estimates based on
UNSCEAR data suggesting a total integrated deposition of
90Sr above 60°N of 41.8 PBq. Assuming that the total area
of land above 60°N is 50.2%, and accounting for latitudinal
variations, then the amount of deposition on land using the
UNSCEAR approach would be 25.6 PBq (16.2 PBq for the
137
50
100
250
500
1 000
2 500
5 000
10 000 25 000
Cs Bq/m2
sea). As the ratio between 137Cs and 90Sr in global fallout is
1.6:1, the present inventory of global fallout 137Cs to Arctic
land masses based on the UNSCEAR approach is 41.0 PBq
Figure 8·4. Estimated ground deposition of nuclear weapons fallout of
137
which is in reasonable agreement with the AMAP radioac-
Cs based on precipitation data, decay corrected to 1995.
tivity data center estimate given above.
137Cs ground deposition from fallout from nuclear weapons
Following the Chernobyl accident in the Ukraine in April
tests derived by the AMAP radioactivity data center is
1986, radionuclides were released to the atmosphere and
shown in Figure 8·4. Estimates of 137Cs ground deposition
fallout was observed in most European countries. The fall-
based on mean precipitation values do not take into account
out levels in the different areas were dependent on the daily
years with extreme rainfall or drought which, especially in
discharge rates, distance from the source and climate (wind
the early 1960s, could have produced local variations in
direction, wet deposition). Adjacent areas in the Ukraine,
ground deposition. Nevertheless, the validation carried out
Belarus and Russia were particularly heavily affected with
thus far suggests this approach gives reasonable estimates of
high contamination levels of up to about 37 GBq/m2 (100
137Cs ground deposition from global fallout.
Ci/km2) of 137Cs observed outside the 30 km Chernobyl ex-
The GIS approach could also be used to predict ground
clusion zone. The prevailing winds during the accident were
deposition of 90Sr by applying the ratio between 137Cs and
southeasterly. Rainfall during the passage of the contami-
Chapter 8 · Radioactivity
539
cle were considerably affected by the Chernobyl fallout with
radiocaesium deposition of up to 200 kBq/m2 .
In the northern parts of Scandinavia and Finland, ground
deposition was only about 1-2 kBq/m2, which was similar to
that due to nuclear weapons global fallout. In the Russian
Arctic, ground deposition of Chernobyl 137Cs fallout on the
Kola Peninsula was up to 1 kBq/m2. At sites further east,
137Cs deposition declined; in the region of Arkhangelsk
about 220 Bq/m2 was deposited, and in the Asian part of
Zapolyarie about 40 Bq/m2. The deposition of 90Sr after the
Chernobyl accident was an order of magnitude lower than
that of 137Cs.
Radionuclide contamination of land led to uptake in a
variety of environmental flora and fauna. At present, the
highest activity concentrations can be observed in natural
foodstuffs such as reindeer meat, mushrooms, freshwater
fish and berries (Figure 8·6). Products such as goat milk,
goat cheese and lamb meat, derived from animals ingesting
vegetation in semi-natural ecosystems, also have 137Cs activ-
ity concentrations higher than those in many agricultural
ecosystems. Contemporary 137Cs activity concentrations in
Chernobyl
reindeer meat from different parts of the Arctic can be seen
in Figure 8·25 later in this section.
137Cs kBq/m2
Marine contamination
185
The anthropogenic sources contributing to the contamina-
40
10
tion in the marine environment are mainly nuclear weapons
2
fallout and releases from Sellafield and the Chernobyl acci-
dent. Caesium-137 activity concentrations in surface seawa-
ter are shown for different years in Figures 8·7 and 8·8 (next
Figure 8·5. Ground deposition of 137Cs from the Chernobyl accident (val-
page). Relatively high values occur in the vicinity of the
ues normalized to May 10, 1986) (after EU/CIS JSP-6 1996, and data
North Pole over the Lomonosov Ridge (Figure 8·8) com-
from national sources).
pared with other areas. A slight increase in 137Cs activity
nated plume over central Norway and Sweden gave rise to
concentration can also be seen in the Laptev Sea compared
considerable fallout and many other areas in Europe were
with the Kara and Barents Seas. East of the outlets of the
also affected. The fallout was heterogeneously distributed.
large Siberian rivers, namely the Ob, Yenisey and Lena, the
Chernobyl fallout affected a smaller part of the Arctic region
137Cs activity concentrations decrease. This geographical
than nuclear weapons fallout, and the extent of deposition
trend follows a decrease in salinity and probably reflects di-
of 137Cs in 1986 after the Chernobyl fallout can be seen in
lution of 137Cs in the seawater by the input of river water.
Figure 8·5. In Scandinavia, areas just south of the Arctic Cir-
This is further confirmed by the increased 90Sr/137Cs ratio in
Mollusks
Shrimps
Potatoes
Seal
Marine fish
Whale
Hare
Cow's milk (Bq/L)
White cheese
Berries
Vegetables
Ptarmigan
Moose
Other meat
Brown cheese
Lamb
Mushrooms
Freshwater fish
Reindeer / Caribou
0.01
0.1
1
10
100
1 000
10 000
137Cs Bq/kg
Figure 8·6. Ranges and average values of 137Cs activity concentrations in food products, from data in the AMAP radioactivity database.
540
AMAP Assessment Report
137Cs Bq/m3
0 to 4
4 to 8
137Cs Bq/m3
8 to 16
0 to 4
16 to 32
32 to 64
4 to 8
> 64
8 to 16
Figure 8·7. 137Cs activity concentrations in surface seawater in 1979 and 1982.
Figure 8·8. 137Cs activity concentrations in surface seawater in 1994.
137
the low-salinity waters of the northeastern Kara Sea. In the
Cs activity concentration can be estimated to be about
vicinity of Greenland, the highest activity concentrations are
30 Bq/m3. This estimate is probably too high because cae-
found along the east coast. The lowest measured activity
sium has a higher affinity for particles than strontium, and
concentrations are found in the East Siberian Sea and over
is therefore removed more quickly from the water column.
the southeastern part of the Makarov Basin.
Consequently, the peak in 137Cs activity concentrations of
Measurements of radiocaesium in Barents Sea water are
around 50 Bq/m3 measured in the Barents Sea in the early
available since 1970. As seen in Figure 8·9, the pattern of
1980s, which was primarily due to releases from Sellafield,
releases of 137Cs from Sellafield are reflected in the levels of
is probably the highest activity concentration which has oc-
137Cs measured in the Barents Sea, with a lag time due to
curred in this Sea.
transportation of approximately four to five years. No mea-
In Figures 8·7 and 8·8, 137Cs activity concentrations in
surements of 137Cs are available in the Barents Sea before
surface seawater are shown for 1979, 1982 and 1994, re-
1970. However, earlier 137Cs activity concentrations can be
spectively. These figures also show clearly the input from
estimated from measurements of 90Sr which are available
Sellafield to the Arctic area. The distribution pattern is con-
since 1963. Before the increase in liquid discharges from Sel-
sistent with the transport of 137Cs from Sellafield, which in-
lafield in the late 1960s/early 1970s, the only major source
dicates that the maximum releases in the 1970s are now re-
contributing 137Cs to the Barents Sea was fallout from nu-
flected in the peaks found in the vicinity of the North Pole
clear weapons testing, in which the ratio of 137Cs/90Sr was
and in the Laptev Sea. In 1994, the rate of input of 137Cs
1.6. In the years of maximum global fallout (1963-1964),
from Sellafield to the Barents Sea was estimated to be 200-
the measured activity concentration of 90Sr in surface water
300 TBq/y (Kershaw and Baxter 1993a) with a total inven-
in the Barents Sea was about 20 Bq/m3 (RCRA 1997). Ap-
tory of 10-15 PBq. Recent measurements of the distribution
plying the 137Cs/90Sr fallout ratio of 1.6 to this value, the
of 137Cs in the Arctic show increased activity concentrations
in waters east of Greenland which are of polar origin. These
137Cs releases from Sellafield (TBq)
137Cs in seawater (Bq/m )
3
are attributable to radiocaesium from earlier Sellafield dis-
6 000
50
charges that have entered the Arctic Ocean circulation and
East Greenland Current
been transported back into the Atlantic through the East
Greenland Current. However, in addition to the contribu-
5 000
Barents Sea
40
tion from atmospheric fallout, the Arctic Sea may also be
Sellafield
contaminated by marine transport from the North Sea and
4 000
the Baltic Sea, the catchments of which have both received
30
considerably more fallout from Chernobyl (by combined
fallout and runoff) than the Arctic Ocean. Contamination of
3 000
the Baltic Sea, as a result of atmospheric fallout following
20
the Chernobyl accident, was very heterogeneous. The high-
2 000
est 137Cs activity concentrations occurred in the Gulf of
Bothnia and the Gulf of Finland where they were about two
10
1 000
orders of magnitude higher than before the Chernobyl acci-
dent. 90Sr activity concentrations in the waters of the Gulfs
of Finland and Riga in May 1986 were increased by 20%
0
0
compared with pre-accident levels of about 20-25 Bq/m3 .
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
Thus, the 137Cs inventory in the Baltic Sea increased about
Figure 8·9. Seawater concentrations of 137Cs in the Barents and East Green-
tenfold as a result of the Chernobyl accident. This is further
land Seas compared to the yearly releases from Sellafield.
discussed in section 8.5.
Chapter 8 · Radioactivity
541
Test site C
Matoskhin Shar
Test site B
137Cs Bq/kg
< 1
C
1 to 10
hern
10 to 100
aya
Test site A
> 100
Bay
Figure 8·10. Average 137Cs activity concentrations in surface sediments of
some Arctic seas sampled from 1992 to 1995.
An overview of average levels of 137Cs in surface sedi-
ments from some Arctic seas sampled between 1992 and
8£10d01
Figure 8·11. Nuclear weapons test sites on Novaya Zemlya.
1995 is given in Figure 8·10. For most locations, average
137Cs activity concentrations are below 10 Bq/kg. In the
Norwegian Sea and in the open Kara Sea, some locations
had samples with activity concentrations up to 100 Bq/kg.
This may reflect unusual sedimentation rates or sediment
Novaya
characteristics. The only locations where samples with activ-
Zemlya
ity concentrations over 100 Bq/kg were found were in Cher-
naya Bay, on the southern coast of Novaya Zemlya, and
close to dumped nuclear wastes adjacent to Novaya Zemlya,
4.7
where levels up to several thousand Bq/kg are found. Events
involving leakage from the Russian Northern Fleet's storage
Mt.
Lazareva
sites for spent nuclear fuel on the Kola Peninsula have oc-
curred, and have probably contaminated sediments adjacent
27.7
to the stores. However, it has not been possible to obtain
3.6
quantitative data on the contamination of these areas.
Mat
S
oskhin Shar
4.9 h
Mt.
o
Radionuclides can be taken up from seawater and sedi-
u
5.7 Chernaya
Mt.
mi
ments by marine plants and animals. Current activity con-
li
Moiseev
kha
centrations in selected marine fish from Arctic and adjacent
Val ey
marine areas are shown later in this section, together with
137Cs activity concentrations in marine mammals such as
whales and seals. Marine food products are generally much
Total ground contamination
less contaminated by radionuclides than food products from
137Cs (kBq/m2)
the terrestrial environment.
90
0
10
20km
Sr (kBq/m2)
Figure 8·12. Local radionuclide contamination on Novaya Zemlya.
8.3.1.2. Localized contamination
8.3.1.2.1. Short-range fallout from Novaya Zemlya tests
contaminants (137Cs, 90Sr and 60Co) are largely present in
the top 6-10 cm of the sediments. At the present time,
There have been some 130 tests at Novaya Zemlya, 88 in
ground contamination by 137Cs and 90Sr is in the range
the atmosphere, 3 underwater and 39 underground (Mikhai-
(30-480 kBq/m2 0.8-13 Ci/km2) with maximum dose
lov et al. 1996). These tests have been mainly confined to
rates of 30 R/h.
three areas, identified as A, B and C, with local radionuclide
· A radioactive trace from the explosion conducted on the
contamination, as shown in Figures 8·11 and 8·12.
surface in 1957 extended from the southern shore of the
Radioactive contamination in area A of the island can be
Chernaya Bay to the eastern coast of Novaya Zemlya,
associated with five different explosions.
crossing the whole of southern Novaya Zemlya. The re-
· A radioactive trace which crossed the Koushny Peninsula
sulting plume extended approximately 1500 km from the
in a southerly direction following the underwater nuclear
site of the explosion. Radioactive contamination within a
explosion in 1955. This trace has a width of approxi-
400 m radius around the crater comprised 90Sr, 137Cs,
mately 2 km and an area of several km2. The radioactive
60Co, 152Eu and 239Pu and is now characterized by dose
542
AMAP Assessment Report
rates of 10-150 Gy/h (1000-15 000 R/h). The current
sedimentation basins within the Pechora Sea that contain in-
ground contamination of 137Cs is 9.3-1110 kBq/m2 (0.25-
creased proportions of fine material. Underwater nuclear ex-
30 Ci/km2), and of 90Sr, 1.5-555 kBq/m2 (0.04-15 Ci/km2).
plosions took place within the Bay in October 1955, and
Thirty km from the test crater current contamination by
September 1957, and in the vicinity of the Bay in 1961. The
137Cs plus 90Sr is 26 kBq/m2 (0.7 Ci/km2).
distribution of several radionuclides in the sediments of the
· A radioactive trace from a low level atmospheric explo-
Pechora Sea, including Chernaya Bay, has examined by
sion in September 1957. At the present time, an area with
Smith et al. (1995a). The surface sediments of Chernaya Bay
a diameter of approximately 0.5 km has a dose rate of
contain elevated 239,240Pu ( 8500 Bq/kg), 241Am ( 430
approximately 0.30 Gy/h (30 R/h) and ground conta-
Bq/kg) and 137Cs ( 160 Bq/kg) activity concentrations com-
mination of 152Eu of 2-130 kBq/m2 (0.05-3.5 Ci/km2),
pared to external areas. The activation product 60Co also oc-
60Co of 22 kBq/m2 (0.6 Ci/km2), and 90Sr plus 137Cs of
curs in measurable amounts (90 Bq/kg). Smith et al. (1995a)
approximately 2 kBq/m2 (0.05 Ci/km2).
have closely examined the ratios among the plutonium and
· A radioactive trace from an explosion that took place on
americium isotopes and made comparisons with ratios in
the surface of the sea in 1961 extends from Chernaya Bay
global fallout and with areas contaminated by other nuclear
to the northeastern part of Novaya Zemlya and is charac-
explosions, including underwater devices. Whilst some con-
terized by current dose rates of 0.20-0.25 Gy/h (20-25
tribution from radionuclides contained in wastes dumped in
R/h) and contamination values for 90Sr plus 137Cs of
Chernaya Bay in 1991 (OPRF 1993) cannot be ruled out,
4-40 kBq/m2 (0.1-1.2 Ci/km2).
the distribution and relationships among the transuranic iso-
· In 1973, there was an underground explosion following
topes suggest that the underwater explosions were the pri-
which a vented release of gas products of 20 minutes du-
mary source of the enhanced radionuclide activity concen-
ration produced a plume extending in a southeasterly di-
trations in the sediments.
rection. Current dose rates near the detonation site are
The 239,240Pu inventory in central Chernaya Bay is ap-
0.25 Gy/h (25 R/h), part of which arises from approx-
proximately 300 kBq/m2. This is similar to other sites of
imately 37 GBq (1 Ci) of 137Cs created from disintegra-
major plutonium contamination, such as the vicinity of the
tion of the vented 137Xe.
test explosion sites at Enewetak Lagoon (Nelson and Nosh-
kin 1973), the most contaminated area of Bylot Sound (Aar-
In area B, maximum external dose rates in several places,
krog et al. 1987) and the Irish Sea in the vicinity (i.e., within
close to the epicenters of the explosions, reach 1 Gy/h (100
an adjacent 100 km2 area) of the Sellafield nuclear fuel re-
R/h), but in the rest of the area, the values decline to 0.10-
processing plant (Pentreath et al. 1986) where sediment in-
0.20 Gy/h (10-20 R/h).
ventories are of comparable magnitude but exceed 300 kBq/
Table 8·4. Radioactive contamination in four regions of area C of Novaya
m2. The integrated Chernaya Bay sediment inventory of 3
Zemlya.
TBq estimated by Smith et al. (1995a) is comparable with
that of Bylot Sound (1 TBq) resulting from the weapons plu-
Dose level,
137Cs
152Eu
60Co
Gy/h
kBq/m2
kBq/m2
kBq/m2
tonium spill at Thule (Aarkrog et al. 1987) and of the same
Region
Area, km2
( R/h)
(Ci/km2)
(Ci/km2)
(Ci/km2)
order as that of Enewetak Lagoon (8.5 TBq). It is, however,
dwarfed by the inventory in the upper 30 cm of Irish Sea
Eastern part
0.4
0.30-0.40
3
5.6-20
2
(30-40)
(0.07)
(0.15-0.6)
(0.05)
sediments of 280 TBq.
Central part
0.3
0.25-0.35
2.2
17
(25-35)
(0.06) (0.45)
Western part
0.5
0.25-0.30
19
8.3.1.2.3. The Thule accident
(25-30)
(0.5)
Northern part
0.3
0.20-0.25
2.0
In January 1968, an American B-52 aircraft carrying four
(20-25)
(0.05)
nuclear weapons crashed on the ice in Bylot Sound near
Thule, Greenland. The impact triggered the conventional
In area C, four regions of radioactive contamination were
explosive, fragmenting the weapons, and resulting in the
measured, and results are given in Table 8·4.
release of plutonium onto the ice. Debris and the upper-
The highest doses and deposition levels occur in area A,
layer of contaminated snow was removed, but the ice had
where levels are 10-100 higher than in areas B and C. In ad-
been broken around the point of impact and some pluto-
dition, there is a larger range of radionuclides present at the
nium was deposited on the underlying sediments (see also
A area site, including 239Pu.
section 8.5.3.2).
To summarize, the current average ground contamination
of 137Cs at the Novaya Zemlya test site is 3 kBq/m2 (0.09
Plutonium in Bylot Sound seawater
Ci/km2) of 137Cs and 2 kBq/m2 (0.06 Ci/km2) of 90Sr. During
In the summer of 1968, half-a-year after the accident, the
the underground nuclear explosions carried out from 1964
seawater in Bylot Sound contained about 0.2 Bq 239,240Pu/
to 1990, most of the radionuclides were retained under-
m3. This was higher than measured at several other loca-
ground at the location of the explosions. Only a small pro-
tions along the Greenland coast (Qaanaaq, Qeqertarsuaq
portion of the activity (1-10%) escaped into the atmosphere
(Godhavn), Nuuk (Godthåb), Ammassalik and Danmark-
leading to localized radionuclide contamination of the No-
shavn) (Figure 8·13) which are assumed to be contaminated
vaya Zemlya test site territory.
by global fallout only. As the total volume of water in Bylot
Sound is 50 km3, the amount of 239,240Pu in the Thule sea-
water in 1968 was estimated to be 1010 Bq, of which half (5
8.3.1.2.2. Chernaya Bay
GBq or 2 g plutonium) was due to the accident.
Chernaya Bay is a 15 km long fjordic inlet on the southwest-
In 1970, seawater activity concentrations (0.02-0.1 Bq/m3)
ern coast of Novaya Zemlya (Figure 8·11). It has a variable
of 239,240Pu did not differ significantly from the global fallout
width of 1-6 km and is connected to the Pechora Sea which
background. This has been verified by subsequent sampling.
occupies the southeastern extremity of the Barents Sea. The
The only increase found in seawater was in particle-bound
239,240
Bay contains finer, organic rich sediments compared with
Pu in near-bottom water at the point of impact, prob-
those of the Pechora Sea, although there also exist deeper
ably due to resuspension of contaminated sediments.
Chapter 8 · Radioactivity
543
Pu left on the ice after the decontamination effort in 1968
was also estimated to be 1 TBq ( 50%). It therefore seems
likely that a substantial part of the plutonium in the sedi-
ments is derived from the melting of sea ice. On the other
hand, it is evident that the highest levels are found beneath
the point of impact and, as some of the contaminated ice
drifted away before it melted, it seems likely that some debris
entered the sea directly through the impact hole in the ice.
The low solubility of PuO2 (high Kd value) makes it prob-
118
152
able that most of the plutonium that might have entered the
sea from the Thule accident would be bound to marine sedi-
ments on the bottom of Bylot Sound. The extensive sam-
pling has shown that the Pu from the accident has only been
minimally dispersed from the crash site (Figure 8·14). Thus,
the site currently does not constitute a significant source of
plutonium contamination to the surrounding environment.
8.3.1.2.4. Contamination at sea dumping sites
141
Between 1960 and 1991, the former Soviet Union carried
out dumping of radioactive waste in the Kara and Barents
Seas. The wastes dumped at sea included liquid and solid
waste, the latter including reactor compartments and entire
37
submarines. Some of the reactors contained spent nuclear
fuel. In addition to the official information provided by the
93
government of the Russian Federation (OPRF 1993), the In-
ternational Arctic Sea Assessment Project (IASAP) of the IAEA
239, 240
has produced revised inventories (see also section 8.6.4).
Pu mBq/m3
During joint Norwegian-Russian expeditions from 1992-
150
1994 (Strand et al. 1997) to the dumping sites, the dumped
50
material was visually inspected and samples collected both
beside and further away from the dumped material.
Figure 8·13. Concentrations of 239,240Pu in seawater around Greenland,
On the east coast of Novaya Zemlya, in the Abrosimov
1968.
and Stepovogo Fjords, enhanced 137Cs, 90Sr, 60Co and Pu-
isotope activity concentrations were measured in sediments
collected in the immediate vicinity of the dumped nuclear
Plutonium in Bylot Sound sediments
waste (Table 8·5). In addition, 152Eu and 154Eu were identi-
From measurements of plutonium in marine sediments col-
fied in one sample collected close to the hull of the dumped
lected during expeditions to Thule in 1968, 1970, 1974,
Table 8·5. Range of radionuclide activity concentrations (Bq/kg dw) in sed-
1979 and 1984 (Aarkrog 1971, 1977, Aarkrog et al. 1984,
iments near to the dumped objects in Abrosimov Fjord and Stepovogo
1987, Smith et al. 1994) it was calculated that about 1 TBq,
Fjord (Strand et al. 1997)
or half a kilogram of plutonium, was deposited on the bot-
Radionuclide activity concentrations
tom of Bylot Sound from the Thule accident. The amount of
Location /
in sediments, Bq/kg dw
object(s)
137Cs
90Sr
60Co
239,240Pu
Abrosimov Fjord
Containers
23-31000
4-8850
0.4-180
1-18
Bylot
Vessels
38-196
0.3-3
0.5-53
0.7-2.6
Sound
Submarine reactor compartments 33-8445 0.4-3250
1-61
1-5
Stepovogo Fjord
Containers
14-109000
1-310 < 0.3-3150 0.2-28
Submarine
4-1670 a
0.4-6
< 0.1-6
< 0.1-6
a. Result was not confirmed by additional sampling.
Bylot
Sound
nuclear submarine in the Stepovogo Fjord. In Abrosimov
Fjord, radionuclide contamination was measurable in the
Saunders
Island
upper 5 cm (90Sr) and upper 10 cm (137Cs, 60Co) sediment
Thule Airbase
layer. Figure 8·15 shows the 137Cs activity concentrations in
the sediments. In Stepovogo Fjord, enhanced 137Cs, 90Sr, 60Co
and Pu-isotope activity concentrations were observed in the
upper 5 cm of sediments collected in the close vicinity of the
dumped containers (Figures 8·16 and 8·17). Substantially
239, 240Pu kBq/m2
lower 137Cs and 90Sr activity concentrations were observed
0.1
0.4
1.6
6.4
25
100
in sediments at greater distances from the localized objects.
0
10 km
Activity concentrations at these latter sites were similar to
values in the open Kara Sea.
In the Tsivolky Fjord, traces of 60Co found in the sedi-
Figure 8·14. Activity concentrations of 239,240Pu in sediments near Thule,
Greenland.
ments close to the dumped vessel could indicate leakage
544
AMAP Assessment Report
Table 8·6. Range of radionuclide activity concentrations, Bq/m3 in seawater
in 1993-94.
Activity concentrations in seawater, Bq/m3
Novaya
Zemlya
Abrosimov
Stepovogo
Tsivolky
Open Kara
Fjord
Fjord
Fjord
Sea
137Cs 90Sr
137Cs 90Sr
137Cs 90Sr
137Cs 90Sr
Surface water
4-7 2-4
3-9 2-7
4-6 4-6
3-8 3-11
Near-bottom water
4-9 2-4
6-31 3-26
6-14 3-4
8-20 4-6
activity concentrations in fish from the area showed no simi-
137Cs Bq/kg dw
lar enhancement. Activity concentrations in Arctic cod (Bo-
< 20
reogadus saida) and Arctic char (Salvelinus alpinus) ranged
20 to 35
from 0.5-2.2 Bq/kg 137Cs and 0.9-1.6 Bq/kg 90Sr. The activ-
35 to 45
ity concentration of 238Pu and 239,240Pu were below the de-
45 to 70
0
1
2
3 km
tection limits of 0.01 Bq/kg and 0.008 Bq/kg, respectively.
> 70
Swamp
8.3.1.2.5. Sunken Komsomolets submarine
8£15d01
Figure 8·15. Activity concentrations of 137Cs in sediments of Abrosimov Bay.
On April 7, 1989, the Soviet nuclear submarine Komsomo-
lets caught fire and sank southwest of Bear Island in the Nor-
wegian Sea. This submarine contains two torpedoes with nu-
clear warheads and a nuclear reactor. The reactor was shut
down prior to sinking. An estimate of the inventory of the
Novaya
Zemlya
main long-lived radionuclide constituents of the reactor and
nuclear weapons of the Komsomolets is given in Table 8·7
(Høibråten and Thoresen 1995) (see also section 8.5.3.5.1).
Scientific studies in the vicinity of the site where the Kom-
somolets is located indicate that only minor contamination
can be attributed to the submarine (Table 8·8) (Kolstad
137
1995, Strand et al. 1996).
Cs Bq/kg dw
5 - 7
Table 8·7. Inventory of selected radionuclides in Komsomolets, decay cor-
7 - 9
rected to 1 January, 1995 (Høibråten and Thoresen 1995).
9 - 14
Radionuclide
TBq
14 - 40
40 - 800
137Cs
2700
90Sr
2400
239Pu
0022
0
1
2
3 km
Table 8·8. Radionuclide activity concentration in sediments near to and ap-
Figure 8·16. Activity concentrations of 137Cs in sediments of Stepovogo Bay.
proximately 1 nautical mile in different directions from Komsomolets com-
pared to average values for the North European Seas in 1995.
Radionuclide activity concentrations, Bq/kg dw
Sample site
239, 240Pu
238Pu
241Am
137Cs
134Cs
Close to
Komsomolets,
1993a
0.3 0.2
90b
0.2 0.1
7 4
1.4 0.8
1994a
0.4 0.4
8 4
0.5 0.3
1 nm south (1994)
0.85
0.04
< 0.7
5.1
n.d.
1 nm west (1994)
0.65
6.6
n.d.
1 nm north (1994)
0.95
0.03
< 0.96
5.4
0.3
1 nm east (1994)
0.96
0.13
0.57
9.7
n.d.
Close to
Komsomolets,
1995a
1.16 0.08 0.04 0.01 0.86 0.066 7.1 0.4 0.6 0.3
Other North
European seas a
1.4 0.8
0.1 0.1
0.8 0.5
a. These numbers are averages of many measurements, the uncertainty
given is the standard deviation of the measurements.
b. Plutonium-238 was detectable in only one sample.
n.d.: not detected.
Figure 8·17. Some of the dumped containers in Stepovogo Bay.
8.3.2. Time dependence of radioactive
from dumped waste. Water and sediment samples obtained
contamination
from the Novaya Zemlya Trough showed no indication of
leakage from dumped wastes. The only location of enhanced
A number of examples of the changes with time in radionu-
levels of radionuclide contamination in water was the inner
clide activity concentrations in Arctic environmental samples
Stepovogo Fjord with an increased 90Sr activity concentra-
have been collated in this section. The purpose is to give a
tion in the bottom water (Table 8·6). However, radionuclide
selected overview of radionuclide contamination of the Arc-
Chapter 8 · Radioactivity
545
Table 8·9. Available sample data for showing time trends in radionuclide activity concentrations.
Sample type
90Sr
137Cs
Air
Finland, Norway, Russia
Deposition
Finland, Greenland
Finland, Russia
Lichens
Finland, Greenland
Finland, Greenland, Russia
Reindeer meat
Finland, Greenland, Norway, Russia
Freshwater
Finland, Greenland, Russia
Finland
Freshwater fish
Finland
Seawater
Greenland waters, Barents Sea, Kara Sea
Greenland waters, Barents Sea
Marine fish
Greenland waters
Marine mammals
Greenland waters
Whole body measurements
Finland, Norway, Sweden, Russia
tic, both with regard to local and temporal variations. As
Within the larger Arctic regions (northern Russia, north-
137Cs and 90Sr are important in determining dose to man, the
ern Canada, Greenland and Alaska) local variations in ra-
presentation is focused on these radionuclides and is limited
dionuclide activity concentrations in environmental samples
to the most comprehensive radionuclide-specific data sets.
may be an order of magnitude, mainly due to heterogeneity
This assessment focuses on specific environmental data sets,
in the deposition of nuclear weapon global fallout. For cer-
as identified in Table 8·9, from particular areas where it has
tain types of sample (e.g., 137Cs in freshwater fish and mush-
been possible to show trends in the selected sample types.
rooms), variations may be even higher due to ecological dif-
In a few cases, logarithms of activity concentration have
ferences. In comparison, the variation in radionuclide activ-
been used due to the high between-year variations. The exam-
ity concentrations within the Arctic areas of Finland, Norway
ples given below, showing trends with time in radionuclide
and Sweden is usually less than for the larger Arctic coun-
activity concentrations, are deliberately not reported with spe-
tries. Because of the similar dietary habits, this is reflected in
cific uncertainties. This is because such uncertainties would
the estimated individual doses from 137Cs received by popu-
not have been comparable and, thus, might be misinterpreted.
lations in the Arctic parts of the above Nordic countries,
For example, if we consider the 137Cs activity concentrations
which appear to be similar (see section 8.4.2.).
in lichen, as shown later in Figure 8·24, the data are given as
5-year means. These means may, for some periods and coun-
8.3.2.1. Air and deposition
tries, have been based on all years and sampling locations,
while in other cases they may be based on less complete data.
Measurements of radionuclide activity concentrations in air
Furthermore, some countries have had a more comprehen-
provide indications of recent releases. Figure 8·18 shows the
sive net of sampling locations than others. If, for example,
137Cs activity concentration in air samples collected since the
the data arise from only a limited part of an Arctic country,
early seventies in the Tromsø area (Tromsø and Skibotn)
the sampling errors may be lower than those for a country
(Norway), in the Helsinki area of Finland (outside the Arc-
with a more widespread sampling net, but the representative-
tic), and in Russia (Naryan Mar and Norilsk). Since 1986,
ness of the data may be better in the latter case than in the
measurements have been made in Arctic Finland at Rovani-
former. Hence, error indications would have been misleading.
emi (Sinkko et al. 1987, Aaltonen et al. 19901991, Toivo-
137Cs µBq/m3
1 000 000
100 000
Norilsk
Naryan Mar
Norway
10 000
Finland
1 000
100
10
1
0.1
0.01
1960
1965
1970
1975
1980
1985
1990
1995
Finland
Norway
Naryan Mar
Norilsk
Finland
Norway
Naryan Mar
Norilsk
Figure 8·18. Changes with time in 137Cs activity concentration in air in Norway, Finland and Russia.
546
AMAP Assessment Report
131I µBq/m3
10 000
1 000
Finland
100
Ivalo
10
Nurmijärvi
Helsinki
1
0.1
0.01
1975
1980
1985
1990
1995
2000
Helsinki (southern Finland)
Ivalo (Arctic Finland)
Nurmijärvi (southern Finland)
Helsinki (southern Finland)
Ivalo (Arctic Finland)
Nurmijärvi (southern Finland)
Figure 8·19. Changes with time in 131I activity concentration in Finnish air samples.
nen et al. 1992). Activity concentrations decreased rapidly
ucts, such as leakage from underground nuclear test explo-
after 1981, which was the year after the last atmospheric
sions or accidental atmospheric releases from nuclear instal-
nuclear test carried out in China. In 1986, the Chernobyl
lations. The peak shown in 1987 was due to a Soviet under-
accident increased the air concentrations of 137Cs by several
ground nuclear weapons test carried out in August that year
orders of magnitude in certain Arctic areas. Thus, in 1986,
at Novaya Zemlya (Bjurman et al. 1990).
the annual mean air activity concentration measured at Ski-
Wet deposition, in rain or snow, is the main mechanism
botn was about 350 Bq 137Cs/m3, and that at Rovaniemi,
for the transfer of radionuclides from the atmosphere to ter-
Finland, approximately 1200 Bq 137Cs/m3. Since 1986, air
restrial and aquatic ecosystems. Three of the Arctic areas are
concentrations of 137Cs have decreased to 0.5 Bq/m3 at
represented by a nearly complete time series of measurements
Skibotn and about 2.5 Bq/m3 at Rovaniemi. In recent
for wet + dry deposition: Finland, Russia and Greenland.
years, the decrease has been less rapid than that observed in
Data are available for both 90Sr (Figure 8·20) and 137Cs
the first few years following the Chernobyl accident, prob-
(Figure 8·21) in Arctic Finland (Salo et al. 1966-1996a,
ably due to the resuspension of deposited 137Cs.
1966-1996b), for 90Sr in Greenland (Figure 8·20) and for
137
Iodine-131 activity concentrations in Finnish air samples
Cs in northwest Russia (Figure 8·21).
are given in Figure 8·19. Due to its high fission yield, vol-
All data show the same trend with time. Deposition of
atility and short physical half-life (8 days), 131I may be used
fallout from atmospheric nuclear weapons testing peaked in
as a short-term indicator of recent releases of fission prod-
1963. The deposition rate of 90Sr and 137Cs in global fallout
8£20d01
90Sr Bq/m2
600
137
Finland
Cs Bq/m2
500
Greenland
Finland
500
Russia
400
400
300
300
200
200
100
100
0
0
1960
1965
1970
1975
1980
1985
1990
1995
1960
1965
1970
1975
1980
1985
1990
1995
Figure 8·20. Changes with time in wet and dry deposition of 90Sr in Arctic
Figure 8·21. Changes with time in wet and dry deposition of 137Cs in
Finland and Greenland.
Arctic Finland and north west Russia (Nenets Autonomous Okrug).
Chapter 8 · Radioactivity
547
has declined with an effective half-life of 3.6
0.3 years
137Cs Bq/kg
Finland
since the peak in 1963. The decay in the rate of deposition
2 500
Russia
to the ground was rapid from 1963 to 1966 with a half-life
Greenland
of a little more than one year, followed by a period of slower
decay until 1981 with an effective half-life of 4-5 years.
2 000
From 1981 until the Chernobyl accident, the decrease in air
concentrations of 137Cs again became more rapid, corre-
1 500
sponding to an effective half-life of about two years. These ob-
served variations in the effective decay rate of 90Sr and 137Cs
deposition may be explained as follows: from 1963 to 1966
1 000
atmospheric radioactive contamination was predominantly
in the stratosphere and the decay in the atmospheric levels
depended mostly on the stratospheric mean residence time.
500
From the middle of the sixties until 1980, China and France
performed thermonuclear tests in the atmosphere and this
contributed more radionuclides to the atmosphere, hence
0
1960
1965
1970
1975
1980
1985
1990
1995
the effective decay was reduced. In 1986, the Chernobyl
accident increased 137Cs and 90Sr annual deposition rates.
Figure 8·22. Changes with time in 137Cs activity concentration in lichen
from Arctic Finland, north west Russia, and Greenland.
Table 8·10. Estimated integrated ground deposition densities of 90Sr and
137
90
Cs in Arctic countries from global fallout, kBq/m2 since 1950 without
Sr Bq/kg
Greenland
decay correction (decay corrected values are given in Table 8·3).
300
Russia
Integrated ground deposition
Mean GIS-based
using data compiled
integrated ground
250
Region
in AMAP database, kBq/m2
deposition, kBq/m2
137Cs
90Sr
137Cs
200
Arctic Canada
1.4
Arctic Finland
1.7
2.5
2.8
Greenland
2.7
4.3 a
2.7 c
150
Iceland
5.8
Arctic Norway
2.5 b
4.4
3.7
Arctic Russia
1.7 b
3.1
100
Arctic Russia, west
2.1
Arctic Russia, east
1.4
Arctic Sweden
1.6 b
2.9
3.0
50
United States, Alaska
2.6
a. Value estimated from 137Cs.
0
b. Value estimated from 90Sr.
1960-1964 1965-1969 1970-1974 1975-1979 1980-1984 1985-1989 1990-1995
c. Includes the land area covered by the ice cap.
Figure 8·23. Changes with time in 90Sr activity concentration (5-year
means) in lichens in Greenland and Russia.
The estimated integrated ground deposition of 90Sr and
137Cs, for various Arctic regions since 1950 (without decay-
137
correction), based on measurements or calculated from the
Cs Bq/kg
2 000
137Cs : 90Sr ratio of 1.6, are given in Table 8·10. For the
Finland
Greenland
Russia
years prior to measurements in the various countries, depo-
1 500
sition densities were estimated from measurements of 90Sr
deposition carried out by USAEC in New York (HASL 1958-
1 000
1978). These values have been compared with (non-decay
corrected) estimates from GIS-based analysis of the integrated
500
ground deposition by the AMAP radioactivity assessment
0
group. GIS-based estimates decay-corrected to 1995, shown
1960-1964 1965-1969 1970-1974 1975-1979 1980-1984 1985-1989 1990-1994
in Table 8·3, are ca. 50% lower (see also section 8.3.1.1).
Figure 8·24. Changes with time in 137Cs activity concentrations (5-year
means) in lichens in Arctic Finland, Greenland, and Russia.
8.3.2.2. Terrestrial and freshwater ecosystems
The observed effective ecological half-life of 137Cs in
8.3.2.2.1. Lichen
lichens from Arctic Finland and Arctic Russia since the mid-
Lichen is an efficient collector of atmospheric contamination
1960s and until the Chernobyl accident was 5-6 years. The
due to its large surface area, nutritional uptake characteris-
real effective ecological half-life should be shorter but the
tics and its slow growth. Lichen is the winter fodder of rein-
true value is masked by fresh global fallout added during the
deer and is thus an important determinant of 137Cs activity
period of observation from atmospheric nuclear tests con-
concentrations in most populations of reindeer during winter.
ducted by China and France between 1974 and 1980.
Nearly complete time series data are available for 90Sr
The 90Sr activity concentrations in lichen from Greenland
and 137Cs in lichen from Arctic Finland, Greenland and Arc-
and Arctic Russia were similar until the Chernobyl accident
tic Russia. Mean 137Cs activity concentrations in lichen from
when lichen in Russia became significantly more contami-
Arctic Finland (Rissanen and Rahola 1996, Rahola and Ris-
nated (Figure 8·23). Compared with 137Cs (Figure 8·24), 90Sr
sanen 1996), Arctic Russia and Greenland are shown in Fig-
activity concentrations in lichen were significantly lower, by
ure 8·22. Activity concentrations peaked in 1965-1969. The
almost a factor of 3-5. Although the Chernobyl accident in-
Chernobyl accident was clearly reflected in 137Cs activity
creased 137Cs activity concentrations in some Arctic lichen,
concentrations in Finnish lichen, whilst only a modest signal
contamination was still lower than in 1965-1969 in most of
was observed in lichen from Greenland and Arctic Russia.
these Arctic areas.
548
AMAP Assessment Report
137
Arctic Norway
Cs Bq/kg
8.3.2.2.2. Reindeer meat
Arctic Finland
3 500
Greenland
Reindeer meat is consumed by some Arctic populations in
Western Russia
substantial amounts. Due to the relatively high 137Cs conta-
3 000
Eastern Russia
mination of reindeer meat, this important component of diet
is a major contributor to intake of anthropogenic radionu-
2 500
clides by some Arctic populations.
Radionuclide activity concentrations in reindeer meat are
2 000
determined by those in fodder. In summer time, reindeer
mainly eat herbaceous vegetation, whereas in winter they
1 500
eat lichen. Because of the higher 137Cs activity concentra-
tions in lichen than in other vegetation, reindeer meat is
1 000
more highly contaminated in winter and spring than in sum-
mer. Maximum 137Cs activity concentrations in reindeer in
500
all Arctic countries were reached in the middle of the 1960s
(Figure 8·25). Since then, 137Cs activity concentrations have
0
decreased gradually with an observed effective ecological
1960
1965
1970
1975
1980
1985
1990
1995
half-life of 4.9
1.1 years in western Arctic Russia, 8.3
4.7
Figure 8·25. Changes with time in activity concentration of 137Cs in rein-
years in eastern Arctic Russia, 5.4
1.1 years in Arctic Fin-
deer meat in Arctic Norway, Arctic Finland, Greenland, and Arctic Russia.
land (Rissanen and Rahola 1996, Rahola and Rissanen
137Cs Bq/kg
1 500 to 2 500
600 to 1 500
200 to 600
0 to 200
Figure 8·26. Average activity concentrations of 137Cs in reindeer meat after 1990.
Chapter 8 · Radioactivity
549
1996), and 7.2
1.3 years in Arctic Norway. The decrease
90Sr Bq/m3
was slowed by the atmospheric tests carried out by France
250
and China during the late 1970s. In 1986, a clear increase in
reindeer meat contamination was observed in Arctic Russia,
Sweden, Norway and Finland due to the Chernobyl acci-
200
Danmarkshavn
dent. Five to ten years after the Chernobyl accident, 137Cs
Upernavik
activity concentrations in reindeer have stabilized and are
Scoresbysund
expected to decrease more slowly in the future. In recent
150
Qeqertarsuaq
years, the highest 137Cs activity concentrations in reindeer
Nuuk
meat have been observed in the western part of Russia and
in Norway, Sweden and Finland (Rissanen and Rahola
100
Prince Christians Sund
1996). In contrast, 137Cs activity concentrations are lowest
in the eastern part of Russia, in Canada, Greenland and Ice-
land (Figure 8·26). In Canada, after 1986, 137Cs activity
50
concentrations were generally about 100 Bq/kg. For Iceland,
data from the 1990s are available showing 137Cs activity
concentrations in the order of 10 Bq/kg; such low levels are
0
due to reindeer feeding almost entirely on herbaceous vege-
1960
1965
1970
1975
1980
1985
1990
1995
tation which is less highly contaminated than Cladonia
Figure 8·29. Changes with time in average activity concentration of 137Cs
lichen species (Palsson et al. 1994).
in drinking water in Greenland.
90Sr Bq/m3
8.3.2.2.3. Freshwater ecosystems
200
One of the most mobile radionuclides from global fallout en-
tering freshwater systems is 90Sr. Unlike many other radionu-
Danmarkshavn
clides, including 137Cs, it is not significantly retained by soils.
150
Measurements of 90Sr in Russian river water have been
Upernavik
Scoresbysund
carried out since the beginning of the 1960s. The levels
Qeqertarsuaq
peaked around 1964 and have generally declined since then.
100
Nuuk
90Sr Bq/m3
60
Finland
Prince Christians Sund
Tomijoki
Kemijoki
50
50
Russia
Eastern Russia
40
Mid Russia
0
Western Russia
1960 - 1969
1970 - 1979
1980 - 1989
1990 - 1995*
30
Precipitation
Drinking water
* five year period
Precipitation
Drinking water
* five year period
Figure 8·30. Ten-year averages of 90Sr activity concentrations in drinking
water and precipitation in Greenland.
20
The sources of the contamination of Russian river water is
10
partly direct deposition of global fallout from the atmos-
phere and partly runoff of previously deposited global fall-
out from catchments. A third source is waterborne discharges
0
from nuclear facilities, such as Mayak in the Urals. The ma-
1960
1965
1970
1975
1980
1985
1990
1995
jor discharges of 90Sr from Mayak to the Ob river system oc-
8£36d01
Figure 8·27. Changes with time in activity concentration of 90Sr in Rus-
sian and Finnish rivers.
curred around 1950. However, data on the levels of 90Sr in
the Arctic parts of the Ob river from that time are not avail-
137Cs Bq/m3
able. From 1964 to 1994, 90Sr activity concentrations in both
50
Tomijoki
Russian and Finnish river water decreased by a factor of ten
(Figure 8·27). By comparison, 137Cs activity concentrations
Kemijoki
in Finnish river water (Figure 8·28) (Salo et al. 1966-1996a,
40
1966-1996b) decreased from 20 Bq/m3 in 1964 to 1 Bq/m3
in 1985. The Chernobyl accident increased 137Cs activity
30
concentrations to 40 Bq/m3 in 1986 but, by 1993, they had
decreased to 3.4 Bq/m3. This clearly demonstrates that 137Cs
activity declines more rapidly in river water than 90Sr be-
20
cause of the higher particle reactivity of Cs.
Drinking water has been analysed in Greenland for 90Sr
since 1962 (Figure 8·29). The drinking water has been col-
10
lected from six locations: Danmarkshavn, Scoresbysund,
Prins Christiansund, Nuuk (Godthåb), Qeqertarsuaq (God-
HERT
0
havn) and Upernavik. The drinking water in Greenland is
1960
1965
1970
1975
1980
1985
1990
1995
mostly derived from ice and snow for the northern localities.
Figure 8·28. Changes with time in activity concentration of 137Cs in
At the southern locations, surface water plays a greater role.
Finnish rivers.
Figure 8·30 shows 10-year mean values of 90Sr in precipita-
550
AMAP Assessment Report
tion and drinking water collected at these six locations in
90Sr Bq/m3
Greenland. Local variations in the concentrations exist.
40
Barents Sea
The highest drinking water activity concentrations of 90Sr
Greenland Water
were found in the south at Prins Christiansund and the low-
Kara Sea
est in the northwest at Qeqertarsuaq and Upernavik. It is
interesting to note that the ratio between 90Sr in drinking
30
water and precipitation has been increasing with time, illu-
strating the contribution of 90Sr from previous precipitation
(in ice) to the drinking water. This contribution is, as would
be expected, highest at the northern stations.
Data for a few fish species have been selected to illustrate
20
the levels and trends in 137Cs activity concentrations in Arc-
tic freshwater fish. The species from lakes in Finnish Lap-
land (Jokelainen 1965, Kolehmainen et al. 1966, Rissanen
unpubl.) selected as examples were pike, whitefish, perch
10
and trout (Figure 8·31). The first species represents preda-
tory fish, and the last a non-predatory species consuming
bottom fauna, plankton or both. The two intermediate spe-
cies are partially predatory. The uptake of radionuclides by
lake fish depends on hydrology and lake type. Cs-137 up-
0
take, particularly for the non-fish-eating species, increases
1960
1965
1970
1975
1980
1985
1990
1995
from eutrophic to oligotrophic lakes. The lakes in Finnish
Figure 8·32. Changes with time in 90Sr activity concentrations in surface
Lapland and in Arctic Scandinavia are dysoligotrophic and
seawater from Greenland waters and the Barents and Kara Seas.
oligotrophic and differences in the measured 137Cs activity
concentrations in fish among the lakes were not substantial.
from European reprocessing plants, particularly from Sel-
lafield. These discharges peaked in the mid-1970s and were
137Cs Bq/kg
evident in the Barents Sea around five years later as shown
600
in Figure 8·9.
Perch
Pike
Strontium-90 activity concentrations in surface seawater
Trout
from Greenland during the last 35 years have been decreas-
500
Whitefish
ing with an observed effective half-life of about 13.5 years
(Aarkrog 1995). This decrease is probably representative of
400
the Arctic Ocean as a whole (Figure 8·32). The highest ac-
tivity concentrations around Greenland occur in the East
300
Greenland Current reflecting surface seawater concentra-
tions in the Arctic Ocean that are higher than those in the
North Atlantic.
200
Since the early 1970s, 137Cs has been measured in surface
seawater collected around Greenland. In contrast to 90Sr, no
100
significant decrease in 137Cs activity concentrations has been
observed. This is due to the input of 137Cs from the (1974-
0
1982) liquid discharges to the Irish Sea from Sellafield in the
1960
1965
1970
1975
1980
1985
1990
1995
UK. The Chernobyl accident in 1986 also added 137Cs to the
Arctic Ocean. As is the case of 90Sr, the highest 137Cs activity
Figure 8·31. Changes with time in activity concentration of 137Cs in fresh-
concentrations have been observed in the East Greenland
water fish in Arctic Finland.
Current. The 137Cs activity concentrations in surface seawa-
In several studies in other areas in Finland that received
ter along the East Greenland coast are about two times high-
higher deposition from the Chernobyl accident, the maxi-
er than those measured along the west coast of Greenland.
mum transfer of Chernobyl radionuclides to fish occurred
within the first three years for most species with 137Cs peaks
8.3.2.3.2. Fish and marine mammals
in the plankton-feeding fish occurring before those in preda-
tory fish such as pike.
Samples of marine fish, seals and whales have been collected
in Greenland and Icelandic waters since the early 1960s and
have been analysed for 137Cs (Figures 8·33 - 8·35). A slight
8.3.2.3. Marine ecosystems
decline has been observed in 137Cs activity concentrations in
8.3.2.3.1. Seawater
all such samples. Contamination of fish and marine mam-
Monitoring of radioactive contamination of the western
mals was similar. Present 137Cs activity concentrations are
Arctic seas of Russia has been carried out since the early
about 0.2-0.5 Bq/kg.
1960s. The highest 90Sr activity concentrations in the seas
were reported in 1963-1964 and were: 85 Bq/m3 in the Kara
8.3.3. Human wholebody measurements
Sea, 52 Bq/m3 in the Laptev Sea, 22 Bq/m3 in the East Sibe-
rian Sea and 26 Bq/m3 in the Chukchi Sea. 90Sr activity con-
The 137Cs content of Arctic human populations is signifi-
centrations in Arctic Seas have mainly been influenced by
cantly influenced by their consumption of locally produced
global fallout from the testing of nuclear weapons in the at-
food products, particularly intake of reindeer meat. Signifi-
mosphere.
cant amounts of 137Cs may, however, also be contributed by
Cs-137 activity concentrations, on the other hand, have
consumption of mushrooms and freshwater fish from, e.g.,
largely been determined by discharges of this radionuclide
oligotrophic lakes.
Chapter 8 · Radioactivity
551
137Cs Bq/kg
Maximum 137Cs activity concentrations in reindeer
7
meat (Figure 8·25) and in reindeer herders from various
Arctic regions (Figures 8·36 and 8·37) were reached in the
middle of the 1960s. Average wholebody concentrations
6
have decreased gradually over the following 20 years by
a factor of 3 to 7. The peak in 137Cs wholebody concen-
trations after 1986 is due to fallout from the Chernobyl
5
accident. The highest body burdens of 137Cs were ob-
served in reindeer herders living on the Kola Peninsula in
western Russia. Except where affected by Chernobyl fall-
4
out, lower body burdens were found in Finnish (Rahola et
al. 1993) and Norwegian reindeer herders (by a factor of
2-3) and in the far eastern part of the Russian Arctic (by a
3
factor of 1.5-10). The difference between Nordic and west-
ern Russian herdsmen is probably due to the greater con-
sumption of less-contaminated, imported food by the for-
2
mer group. In eastern Russia, the lower levels are due to
lower 137Cs contamination of this area by global nuclear
weapons fallout.
1
Cs bodyburden, Bq
137Cs bodyburden, Bq
0
50 000
1960
1965
1970
1975
1980
1985
1990
1995
Figure 8·33. Changes with time in 137Cs activity concentration in marine
fish from Greenland waters.
40 000
137Cs Bq/kg
20
30 000
15
20 000
10
10 000
5
0
1965
1970
1975
1980
1985
1990
1995
2000
0
Arctic Finland
Kautekeino (northern Norway)
1960
1965
1970
1975
1980
1985
1990
1995
Arctic Finland
Kautekeino (northern Norway)
Sweden
Snåsa (southern Norway)
Figure 8·34. Changes with time in 137Cs activity concentration in whales
Sweden
Snåsa (southern Norway)
from Greenland waters.
Figure 8·36. Changes with time in 137Cs wholebody measurements of rein-
deer herders in northern and central Norway, Arctic Finland, and Sweden.
137Cs Bq/kg
137Cs bodyburden, Bq
Eastern Russia
10
120 000
Mid Russia
Western Russia
8
100 000
80 000
6
60 000
4
40 000
2
20 000
0
0
1960
1965
1970
1975
1980
1985
1990
1995
1960
1965
1970
1975
1980
1985
1990
1995
Figure 8·35. Changes with time in 137Cs activity concentration in seals
Figure 8·37. Changes with time in 137Cs wholebody measurements of Rus-
from Greenland waters.
sian reindeer herders.
552
AMAP Assessment Report
No measurements of 137Cs in seawater from that period are
8.3.4. Summary
available, but by applying the established ratio between
137
Radionuclide activity concentrations in air and precipitation
Cs and 90Sr in nuclear weapons fallout, activity concen-
have closely reflected the rates of emission of radionuclides
trations of 137Cs in seawater can be estimated to have been
into the atmosphere from above-ground nuclear weapons
up to 50 Bq/m3. Equivalent and higher values were also
tests, with identifiable peaks in the period when most tests
found in the Barents Sea in the early 1980s due to Sellafield
were conducted, or associated with specific events, such as
releases. Radionuclide activity concentrations in marine bio-
vented underground nuclear tests or accidental releases. Ra-
ta have consistently been low in the Arctic compared to the
dionuclides are accumulated in terrestrial ecosystems and
levels found in terrestrial and freshwater biota. Even during
water bodies, and the rates of decline in contamination lev-
the heaviest fallout period in the mid-1960s, average 137Cs
els in biota in both types of ecosystem are slower than those
activity concentrations in marine fish or marine mammals in
in the atmosphere.
the North European seas did not exceed 10 Bq/kg.
In general, radionuclide contamination levels in terrestrial
In addition to contamination from the sources affecting
biota have consistently been higher than those in marine
large areas of the Arctic, some places have locally enhanced
biota. Within the terrestrial environment, the highest activity
contamination due to specific sources or events. The most
concentrations have been found in products harvested from
significant of these are at Thule, Greenland, following a nu-
natural or semi-natural ecosystems, followed by products
clear accident, and Novaya Zemlya in northwest Russia
produced by extensive farming in semi-natural ecosystems.
where terrestrial or underwater nuclear testing occurred and
For representative foodstuffs, the order of contamination
where solid nuclear waste has been dumped along the east-
generally decreases as follows:
ern coast. At these latter sites, contamination levels may be
orders of magnitude above the average for the Arctic. It
reindeer, mushrooms, freshwater fish > lamb meat, goat
should be stressed that such elevated contamination only
cheese > potatoes, vegetables >> marine fish, whale and
occurs within a few kilometers of these sources.
seal meat.
In most Arctic areas, levels of radionuclide contamination in
terrestrial biota reached a maximum in the second half of
8.4. Individual doses to man estimated from
the 1960s due to global fallout from nuclear weapons tests.
environmental measurements
The geographical distribution of fallout reflects the patterns
of precipitation for much of the Arctic. Fennoscandia and
Arctic populations are exposed to ionizing radiation from
western Russia were also affected by fallout from the Cher-
three major sources. Exposures from natural sources deliver
nobyl accident. In parts of Norway and Sweden, peak radio-
the main part of the dose. Medical exposures are a second
caesium activity concentrations in terrestrial biota due to
potentially important source but are not dealt with here.
Chernobyl fallout attained values similar to those during the
Exposures arising from the exploitation of nuclear energy,
period of atmospheric nuclear weapons testing.
for military as well as peaceful purposes, is the third major
The highest radiocaesium activity concentrations in the
source. Some of these exposures show considerable variation
terrestrial environment have usually been found in compo-
with time and location, whereas others, such as those from
nents of natural or semi-natural ecosystems, especially lichen
natural sources, are generally less variable. Exposures to an-
and mushrooms, due to the high rate of interception or up-
thropogenic sources depend on source characteristics, eco-
take of radiocaesium by these organisms. These high conta-
logical factors, which influence rates of transfer, and on the
mination levels are then transferred up the food chain and
living habits of the exposed population.
are especially reflected in the meat of Arctic reindeer or cari-
bou, which largely depend on lichen for winter fodder. Simi-
8.4.1. Natural radiation
larly, in animals which consume mushrooms, such as moose,
lamb or cattle, high levels of radiocaesium can be found in
External exposures from natural sources
the autumn. During the late 1960s, 137Cs activity concentra-
(see also section 8.2.1.1)
tions in reindeer/caribou meat varied over the Arctic. In
According to UNSCEAR (1993) (Table 8·1) the typical indi-
Alaska, Canada, Greenland and the Asian part of Russia,
vidual doses received from external exposures from natural
such activity concentrations were up to 1000 Bq/kg, whereas
sources, including cosmogenic radionuclides, are 0.39 mSv/y
in the northern part of the European continent, they were up
from cosmic rays and 0.46 mSv/y from terrestrial gamma
to 2000-3000 Bq/kg. After the mid-1960s, 137Cs activity
rays, giving an aggregate individual dose rate of 0.85 mSv/y.
concentrations in reindeer meat decreased with an observed
In regions with elevated natural radiation, these values can
effective ecological half-life of about 5-10 years until the
reach 2.0 mSv/y and 4.3 mSv/y, respectively, giving a total of
Chernobyl accident in 1986.
6.3 mSv/y. The AMAP radioactivity assessment group has
In all Arctic countries, population groups with high in-
assumed that members of the Arctic population receive, on
takes of reindeer meat exist. Wholebody measurements on
average, an external exposure from natural sources equal to
some of these groups have shown average body burdens of
that considered typical by UNSCEAR of 0.85 mSv/y. A Nor-
137Cs up to 50 000 Bq in the late 1960s. The observed half-
dic study (Christensen et al. 1990) describing the variation
lives of wholebody 137Cs have been shorter than for reindeer
in exposures from natural sources among the Nordic coun-
meat since the food consumption patterns of some of these
tries, estimated that it varied between 0.5 and 1.0 mSv/y.
population groups have changed since the 1960s with an in-
creasing portion of their food originating from agricultural
Internal exposures from natural sources
ecosystems.
According to UNSCEAR (1993), the average annual effec-
In the marine environment, the highest levels of radiocae-
tive internal dose from natural radionuclides (mainly 40K
sium contamination were found in the North European seas
and radionuclides from the 238U and 232Th series) is 0.23
in the late 1970s and early 1980s due to releases from the
mSv/y. In addition, there is an average dose of 1.3 mSv/y re-
Sellafield reprocessing plant. During the late 1960s, 90Sr ac-
ceived from radon, thoron and their decay products. In re-
tivity concentrations in seawater were about 20-30 Bq/m3.
gions with elevated natural radiation, these annual internal
Chapter 8 · Radioactivity
553
doses may be as high as 0.6 and 10 mSv, respectively. For
Of the Lapland population, 20.8% live in large apartment
the Arctic regions, the assessment group has used the aver-
houses and 79.2% in single-family-houses or attached
age internal dose estimated by UNSCEAR of 0.23 + 1.3 =
houses. The reindeer herders all live in single-family houses.
1.53 mSv/y. Christensen et al. (1990) have reported the vari-
The adult Saami reindeer herders (males and females) in
ation in internal exposure in the Nordic countries to be 0.5-
the Saami district form the selected group for the Finnish
4 mSv/y, due mainly to differences in radon exposures in
Arctic area for the discussion of dietary intake of radionu-
dwellings. Consumption of reindeer meat and fish contri-
clides. The dietary habits of both the average population
butes to the internal exposure from natural sources because
and the selected group have changed during the past 40
they contain enhanced 210Po concentrations (Tracy et al.
years. In particular, milk and grain consumption has de-
1995, Woodhead 1982, Pentreath 1988). Consumers of
clined and vegetables and fruit consumption has increased
large amounts of reindeer meat may receive a dose from
in both groups. The consumption of reindeer meat by the
210Po in the order of 10 mSv/y.
selected group has remained fairly constant during the past
three decades but is lower than in the early 1960s. Reindeer
meat consumption among the average population is small.
8.4.2. Radionuclide contamination
The consumption of freshwater fish by the selected group
To estimate the doses from past and present radioactive con-
has decreased since the 1960s whilst the consumption of
tamination in the Arctic, information on population charac-
marine fish (e.g., salmon from the rivers discharging into the
teristics, including living habits such as occupation, housing
Arctic Ocean) increased during the 1990s (Jokelainen 1965,
and food consumption, for the average populations of the
Hasunen and Möttönen 1976, Hasunen et al. 1976, Hälinen
eight Arctic countries have been collated. Furthermore, `se-
and Sikkilä 1993, Laitinen et al. 1996, Rissanen unpubl.).
lected' groups have been defined, denoting groups of people
in specific Arctic areas expected to receive higher doses from
Greenland (Kalaallit Nunaat)
the intake of radiocaesium. It should be stressed here that the
Greenland is one of the three countries within the Kingdom
various selected groups are not necessarily comparable, some
of Denmark. The total number of people in Greenland on 1
are relatively large whilst others are small, some are based
January, 1994, amounted to 55 419. Of this population, 87%
on actual populations whilst that for Greenland is hypothet-
were born in Greenland and 13% outside, mainly in Den-
ical. Estimates of external dose have, where appropriate,
mark. The mean life-expectancy for women is 68 years and
taken account of the shielding effect of different types of
61 for men. About 80% of the population live in the towns,
dwellings.
where large apartment houses were built in the 1960s and
1970s. In the villages, where about 20% of the population
live, most residences are small single-family houses.
8.4.2.1. Information base for individual dose estimates
Approximately 20% of the population is dependent on
The following sections provide information on which the
hunting activities, primarily for seal. There are wild as well
dose estimates for average and selected groups within the
as domestic reindeer in Greenland (P. Hansen, Grønlands
Arctic are based. In some instances, the population charac-
Hjemmestyre, pers. comm. 1996). They are mainly found
teristics differ from those given in the chapter 5. This results
between 64°N and 69°N on the west coast of Greenland.
from the need to estimate individual doses at an early stage
The number of wild reindeer varies considerably and is pre-
in the preparation of the radioactivity assessment chapter.
sently (1996) very low. The exact number is not known, but
It should, however, be noted that much of the information
it is estimated to be around 15 000-20 000. In earlier years
used here was specifically collated for the purposes of radio-
the number approached 100 000 and the annual hunting
logical dose estimation and were reviewed at an earlier stage
was 5000-6000 animals. In 1995, only 2000 wild reindeer
in the assessment process by the AMAP assessment groups
were shot, with a live weight of about 300 000 kg. Hunting
responsible for chapters 5 (Peoples of the Arctic) and 12
takes place mainly during August and September. The total
(Pollution and Human Health).
number of domestic reindeer is about 5000-6000 and ap-
proximately 30% of the stock are slaughtered annually, cor-
Finnish Lapland
responding to about 200 000-300 000 kg live weight. Do-
The total number of inhabitants in Finnish Lapland was
mestic reindeer are mostly slaughtered from the middle of
202 400 in 1992, and they inhabit an area of 93 057 km2.
August to the middle of September, however, winter slaugh-
Of these 94 700 live in towns and 107 700 in rural areas.
tering also occurs.
Of the rural population, 10.8% depend on agriculture for
Sheep farming is carried out in southwestern Greenland
their livelihood, 23.7% on industry and 62.7% on services.
between 60°N and 62°N. The number of sheep is about
The reindeer-herding area is approximately 114 000 km2
17 900, and about 13 300 lambs (live weight ca. 485 000 kg)
and there are 7100 reindeer owners. Approximately 4000 of
are slaughtered each year in September-October. The de-
the reindeer owners are Saami people who depend mainly
mand for lamb exceeds local production and lamb is im-
on reindeer husbandry. The total Finnish reindeer herding
ported from Iceland and New Zealand.
area is divided into 56 herding co-operatives, 40 of which
The selected group for Greenland is a hypothetical group
are in Lapland. Annually 130 000-150 000 reindeer are
that is assumed to consume only reindeer meat instead of
slaughtered producing 3 million kg of meat. Reindeer herd-
imported meat and lamb. Furthermore, it is assumed that
ers and their families consume 130 000 kg of meat them-
the group consumes freshwater fish rather than marine fish
selves. Reindeer herding is economically most significant in
and locally collected berries rather than imported fruit.
the Saami district, where over one third of the herd is man-
aged by about 1500 of the owners.
Northern Canada
About 21% of the population are younger than 15 years,
The total population of the Canadian Arctic is ca. 70 000
67% are between 15 and 64 years and the remaining 12%
(Yukon 23 075; Northwest Territories 46 000, plus a small
are 65 years and older. In rural areas, the houses are mainly
number in northern Quebec) (Canadian Encyclopaedia 1985).
single family-houses, largely constructed of wood with some
The indigenous population of the Canadian Arctic is approx-
of brick. In urban areas there are also large apartment houses.
imately 36 000 (cf. chapter 5). This latter estimate compares
554
AMAP Assessment Report
well with the sum of the estimated rural (non-urban) popu-
tants of villages and small settlements who are not involved
lation of the Yukon (8300) and that of the Northwest Terri-
in reindeer breeding . The urban population consumes main-
tories (24 000) (Canadian Encyclopaedia 1985) allowing for
ly imported food products. The dietary habits of this seg-
some of the urban population comprising indigenous people.
ment of the population are, therefore, similar to those of in-
The dietary intake for the average Canadian Arctic resi-
habitants of other large Russian cities. The dietary habits of
dent is estimated on the basis of a weighted combination of
the rural population is more diverse but still includes a rela-
the average (dominantly southern) Canadian consumption
tively large proportion of imported products.
pattern obtained to represent the non-indigenous population
Food habits of the average and selected groups of the Rus-
of the Arctic and the estimated dietary intake for indigenous
sian Arctic population have been carefully studied since the
residents after Coad (1994). The average Canadian con-
1960s to allow internal dose estimation. These studies in-
sumption figures are based on a `Nutrition Canada' survey
cluded specially-devised population surveys combined with
of food consumption patterns (B. Tracy, pers. comm.). The
wholebody counting (Ramzaev et al. 1993). The results of
different consumption patterns for individuals in distinct age
these studies have been used for the calculation of radionu-
groups (in the range 12-64) have been averaged to derive
clide intake by inhabitants and subsequent internal dose esti-
these values. The consumption of beverages have been in-
mation.
cluded with drinking water and the consumption of miscel-
laneous foodstuffs has been ignored. This has been done in
Northern Norway
full recognition that this community does not represent a
The total number of inhabitants in Arctic Norway was
group of individuals with common habits analogous to a
379 461 in 1990, and they inhabit an area of 95 489 km2.
critical group that might be selected for assessing the accept-
At that time, the number of Saami was about 50 000, con-
ability of individual doses arising from a practice. The diet-
stituting 13.2% of the total population.
ary characteristics of the Old Crow selected group have been
The Saami People in Norway live along the Norwegian
determined from recent estimates described in Coad (1994).
Swedish border from as far south as Engerdal municipality
There is insufficient historical information to estimate any
in Hedmark county to the border with Russia in the north.
changes in dietary patterns that might influence the retro-
The largest Saami population is found in Troms and Finn-
spective individual dose reconstruction attempted here.
mark counties, especially in the municipalities of Karasjok,
Kautokeino, Tana, Nesseby and Porsanger. Like other peo-
Northern Russia
ple, the Saami have gradually migrated to the villages and
The total population of the Russian Arctic in the AMAP
cities but many people still live in the countryside. The num-
area numbers about 2 million persons; 1.7 million living in
bers of Saami in the three Arctic counties of Norway was
cities, towns and settlements (the urban population) and 0.3
recently estimated to be 21 689 in Finnmark, 12 457 in
million living in villages and small settlements (the rural
Troms, and 3239 in Nordland.
population) (see chapter 5). For the purposes of average ef-
The general population of Arctic Norway does not differ
fective dose estimation, the population of the Russian Arctic
very much from the population of the rest of Norway. The
is divided into two groups defined on the basis of geographi-
selected group for Arctic Norway comprises males and fe-
cal and social criteria, including dietary habits.
males associated with reindeer-breeding.
The selected group comprises reindeer-breeders and their
families. This group is represented mainly by indigenous
Alaska
peoples and has a population size of about 100 000.
The USA definition of Arctic Alaska includes approximately
In winter, reindeer are pastured in the forest-tundra zone,
51 930 people (Boedeker 1991) and an area of approximate-
where there is an opportunity to find shelter from bad
ly 700 000 km2.
weather and pastures are rich with lichens. In spring, herds
For rural Alaska as a whole, fish are 59% by weight of
move to the northern meadow pastures in coastal areas. In
the total subsistence harvest; for certain regions, fish com-
autumn, the reindeer again return to the forest tundra to the
prise over three-quarters of the harvest. Except for the north-
south. Along with reindeer-breeding, the indigenous people
ern and northwestern regions of Alaska, fish represent the
of the North hunt game and fur-bearing animals, gather ber-
majority of the subsistence harvest by weight. Salmon are
ries, mushrooms and fish. The majority of reindeer-breeders
the most important species, but whitefish, burbot, and trout
live with their families in settlements and take turns in tend-
species are significant as well. Several species of shellfish, in-
ing the herds. In summer, they move from place to place, to-
cluding clams and crab, are also important to subsistence
gether with their families, on the tundra.
harvests.
About 70% of the reindeer-breeders have permanent win-
For coastal communities in Arctic Alaska, marine mam-
ter dwellings which are standard wooden houses. In perma-
mals are a critical and highly valued resource. They are also
frost areas, houses are mounted on piles with boarding
the reason that many communities are located on the coast,
along their perimeter.
as migrating marine mammals pass within close range. In
In settlements on the coast, indigenous people undertake
northern and northwest Alaska (the Arctic Slope, the NANA
hunting of sea animals and fish. However, meat producing
and Bering Straits regions, i.e., the areas adjacent to the
reindeer-breeding farms also exist. No more than 10-15%
Beaufort, Chukchi and northern Bering Seas) marine mam-
of the indigenous people of northern Russia live directly on
mals account for 42% of the subsistence harvest, or 99 kg/y/
the coast.
cap. In these regions, the primary species taken are bowhead
Because of different levels of radioactive contamination
whales, beluga whales, walrus, bearded seals, ringed seals,
of the western and eastern parts of the Russian Arctic, the
other species of seals, and polar bear.
data on 137Cs and 90Sr activity concentrations in food prod-
For both coastal and inland communities, terrestrial mam-
ucts, estimates of intake, and internal doses of inhabitants
mals form a significant part of the subsistence harvest. Cari-
are considered separately for these two regions.
bou are the primary species hunted, although moose, Dall
The average population comprises 1700 000 inhabitants
sheep, muskox, brown and black bear, and a variety of smal-
of large ports and industrial cities including Murmansk,
ler mammals are also taken. Reindeer herding, introduced in
Archangelsk, and Norilsk, and about 200 000 rural inhabi-
the early 20th century, continues in some areas of Alaska. At
Chapter 8 · Radioactivity
555
present, caribou populations throughout the Arctic region
The critical group in northern Sweden belongs to the
are high, as are harvest levels. Due to their migrations, cari-
reindeer-herding population with relatively high consump-
bou are hunted throughout the year in different communi-
tion of reindeer meat and freshwater fish from the region.
ties, depending upon their local availability. Moose, sheep,
muskox, and smaller mammals are available more consis-
Diet intakes by Arctic populations
tently, although local preferences and government hunting
Tables 8·11 and 8·12 (next page) show the annual mean
regulations may restrict harvests.
consumption rates for the average populations and some
selected groups within some of the Arctic countries. The
Iceland
selected groups are generally assumed to be those with the
Iceland is the second largest island in Europe, located in the
highest consumption of reindeer meat.
North Atlantic just south of the Arctic Circle. The total area
of the country is 103 000 km2 and the coastline, including
8.4.2.2. External and internal doses to humans
fjords and inlets, is about 5960 km long.
Iceland is the most sparsely populated country in Europe
External exposure from anthropogenic sources
with an average of 2.4 inhabitants per km2. On 1 December,
The effective dose commitment due to radionuclides pro-
1994, the total population of Iceland was 266 783, compris-
duced in atmospheric nuclear testing is 1 mSv for exter-
ing 133 781 males and 133 002 females. Compared with
nal exposure in the northern temperate zone (40-50°N)
neighboring countries, the Icelandic age distribution is rela-
(UNSCEAR 1993). Half of this is due to 137Cs, the other
tively young. Thus, in 1994, around 10.8% of the popula-
half comes from short lived radionuclides such as 95Zr, 106Ru,
tion was aged 65 years and older, and 24.8% was aged 15
54Mn and 95Nb. The assessment group have assumed that
years or younger.
the total external dose is proportional to 137Cs deposition
The Icelandic diet is typically western European in most
from nuclear weapon testing, which UNSCEAR estimates to
respects. Nevertheless, it retains some characteristics of a
be 5.2 kBq 137Cs/m2 in the 40-50°N latitude belt. The effec-
subarctic region, making it somewhat unique among Euro-
tive dose commitment in the 60-70°N latitude belt, where
pean nations. Fish, meat and milk are traditionally the main
the majority of the Arctic population lives, assuming that an
foods produced in Iceland and this local production affects
integrated deposition density of 137Cs in that latitude belt is
what people consume. Icelanders consume more fish than
3 kBq/m2 (UNSCEAR 1993), can then be calculated as
any other nation in Europe (73 g/d/cap) and, in general,
3 / 5.2
1 mSv 0.6 mSv. External dose will decrease from
foods of animal origin constitute a large proportion of the
south to north due to decreasing amounts of precipitation
Icelandic diet. Young people consume the least amount of
and thus of global fallout. Furthermore, the regions which
fish while people over fifty consume the most (Steingrims-
received high amounts of Chernobyl fallout will show en-
dottir et al. 1991). Grains (with the exception of small
hanced external doses. It has been estimated that the exter-
amounts of barley and oats) and fruit are not grown in Ice-
nal individual dose commitments from Chernobyl to the
land and vegetable production is mostly limited to potatoes
Norwegian, Swedish and Finnish average populations were
and greenhouse plants. Consequently, Icelanders eat less
1.0, 0.6 and 1.7 mSv, respectively (Strand et al. 1987, Mo-
vegetables than inhabitants of most European countries,
berg 1991, STUK-A74 1991). The external doses received
even through the economy allows substantial import of such
by the Arctic populations are, in general, less than the nation-
foods. The diet of Icelanders is unusually rich in protein and
al means. The external doses from the Chernobyl accident
fat (protein: 17.4%; fat: 41%) but unlike the situation in
should be added to the 0.6 mSv from global fallout to ob-
many other western countries, this is not entirely a modern
tain the total external exposure from anthropogenic sources.
development since the traditional diet of the nation had
It is assumed that external exposures to members of the Arc-
some of the same characteristics in earlier times.
tic population will be in the range 0.6-1 mSv from all an-
thropogenic sources.
Arctic Sweden
The total number of inhabitants in the two northernmost
Internal doses from anthropogenic sources
counties in Sweden, Västerbotten (55 401 km2) and Norr-
Internal doses arise from ingestion and inhalation of radio-
botten (98 911 km2), was 259 775 and 267 648, respectively,
nuclides. The inhalation pathway is of minor importance for
in 1994. The rural population constitutes 26% and 19% of
most radionuclides and will not be considered here. Accord-
the total in each county, respectively.
ing to UNSCEAR (1993), the most important contributor to
The total Saami population in the main reindeer-herding
internal doses from anthropogenic sources is 14C from inges-
area of Sweden is about 17 000 (10 000 in Norrbotten, 5000
tion that will deliver an individual effective dose commitment
in Västerbotten, and 2000 in Jämtland). Of these, about
of 2.6 mSv to the average member of the world population
2500 are reindeer owners.
over the next many thousands of years. This dose will be the
From the mid-1980s, about 80 000 to 100 000 reindeer
same irrespective of where people live. However, only 10%
have been slaughtered annually, producing 2 million kg of
of the dose will be delivered before the year 2200. For cur-
meat per year. About 15-20% of the meat production is
rent generations, the most important anthropogenic contribu-
derived from the slaughter in September, 40-50% during
tors to internal dose are 90Sr and 137Cs. These radionuclides
November-December, and 30-40% during January-April.
have been the most intensively studied in the environment.
During other periods of the year the slaughter is relatively
From 1995 onward, it was assumed that activity concen-
small.
trations of both radionuclides in the Arctic diet would de-
Cattle, primarily for milk production, constitutes a signi-
crease with an effective ecological half-life of ten years, and
ficant source for the regional supply in both Västerbotten
the future time integrals were calculated according to this
and Norrbotten counties. Beef production, and in Västerbot-
model. However, some measurements (in Canada for in-
ten also pork production, is substantial. Although lambs are
stance) suggest that the effective ecological half-life of 137Cs
found in these areas, meat production is small and only cor-
in the human Arctic population may be significantly lower
responds to a few percent of the beef production. The con-
than ten years (Walton 1995). In contrast, Scandinavian
tribution from other livestock is of even less importance.
studies indicate effective ecological half-lives in species of
556
AMAP Assessment Report
Table 8·11. Contemporary annual mean consumption of foodstuffs (kg/y/cap) for average populations of the Arctic.
Finnish
Northern
Northern
Norwegian
Swedish
Alaska
Lapland
Greenland
Iceland
Canada
Russia
Lapland
Lapland
Diet Mainly imported products
Cow milka
80
230
92
176
86
70b
206
158
Other milka
Cow cheesea
12
5.8
14
5b
9.5
12.7
Other cheesea
6.6
0.4
Grain products as flour
60
72
57
57
51
100b
59
60
Potatoes
40
60
28
50
30
70b
60
84
Leafy vegetables (e.g. cabbage)
10
20b
Root vegetables (e.g. carrots)
50
30
11.6
26
61
10b
39
Fruit (imported)
60
36
57
30b
65
55
Pork
60
15
15b
Beef
40
36
41
15b
40
55
Poultry and eggs
10
15b
Diet Mainly local products
Lamb
9.1
24
6
1
Marine fish (incl. fish from fish farms)
14
23
27
10b
20
15
Deer
1.5
0.2
Elk or moose
1.5
28
3c
1.6
Reindeer or caribou
10
1.0
3.8
0.04
0.7
0.2
Freshwater fish (wild)
10
4
0.6
12
5b
1.2
1
Berries (wild)
10
10b
2.6
4.5
Mushrooms (wild)
1.4
5b
0.2
2.8
Various game (e.g. birds, hare, etc.)
5
0.3
1.1
4.7
0.05
Seal
5
30
6.2
Whale
2.3
Drinking water
700
700
550
600
700
600b
600
550
a. Milk and milk products are locally produced in some countries.
b. For 90Sr, intake calculated as 0.25 kg bone.
c. For 90Sr, intake calculated as 0.05 kg bone.
STUK-94A -A62 -A78, Risø Reports 1986-1993, Steingrimsdottir et al. 1991, Coad 1994, Tracy et al. 1995, Strand et al. 1987, Moberg 1991, Johanson
and Bergström 1993.
Table 8·12. Contemporary annual mean consumption of foodstuff (kg/y/cap) for selected groupsa.
Northern
Northern
Finnish
Northern
Russia,
Russia,
Norwegian
Swedish
Alaska
Lapland
Greenland
Iceland
Canada
West
East
Lapland
Lapland
Diet Mainly imported products
Cow milkb
150
92
213
158
Other milkb
10c
10c
Cow cheeseb
15
5.8
5
Other cheeseb
Grain products as flour
80
57
180c
180c
50
Potatoes
50
28
15c
15c
55
Leafy vegetables (e.g. cabbage)
55
Root vegetables (e.g. carrots)
20
5c
5c
5
Fruit (imported)
10
20c
20
55
Pork
Beef
9
11
3
Poultry and eggs
5
5c
5c
Diet Mainly local products
Lamb
2
Marine fish (incl. fish from fish farms)
5
40c
40c
14
Deer
Elk or moose
0.1
364
91c
91c
6
Reindeer or caribou
300
70
82
120
60
Freshwater fish (wild)
10
20
23
12
70c
70c
5
15
Berries (wild)
20
10
15c
15c
12
3
Mushrooms (wild)
1
20c
20c
Various game (e.g. birds, hare, etc.)
10
1
12
10c
10c
2
Seal
30
Whale
0
Drinking water
700
700
550
700
700c
700c
550
a. These groups are not necessarily directly comparable (see text).
b. Milk and milk products are locally produced in some countries.
c. For 90Sr, intake calculated as 1.5 kg bone.
STUK-94A -A62 -A78, Risø Reports 1986-1993, Steingrimsdottir et al. 1991, Coad 1994, Tracy et al. 1995, Strand et al. 1987, Moberg 1991, Johanson
and Bergström 1993.
mushrooms and for moose which approach the physical
in Arctic diet. Where data were missing, they were estimated
half-life of 137Cs of 30 years. Although the effective ecologi-
from neighboring periods by interpolation or extrapolation.
cal half-life obviously represents an important uncertainty it
The composition of the diet may have changed throughout
does not greatly influence the overall dose commitment be-
the years and this was taken into account in Arctic Finland
cause the major part of the dose was received between 1960
and Greenland. Otherwise, a constant composition of the
and 1994.
diet throughout the period was assumed. The data used in
Future doses, i.e. doses from 1995 and onward, contri-
these calculations are based upon national survey results re-
bute less than 10% to the total dose commitment from 137Cs
ferred to in the references. However, the national data used
Chapter 8 · Radioactivity
557
Table 8·13. Estimated dietary intakes of 137Cs by the average population (kBq over period).
Table 8·15. Estimated dietary intakes of 90Sr by the
average population (kBq over period).
Arctic Northern
Arctic
Arctic
Arctic
Period
Alaska Finland Greenland Iceland Canada
Russia
Norway
Sweden
Period
Greenland
Arctic Russia
1950-59
35
10
158
18
38
18
1950-1959
1.91
2.41
1960-64
39
11.1
176
20
42
20
1960-1964
2.11
2.61
1965-69
42
13.5
175
25
42
20
1965-1969
1.31
3.61
1970-74
22
4.5
112
15
19
11
1970-1974
0.66
1.51
1975-79
11.1
3.7
91
10.3
34
7
1975-1979
0.41
1.01
1980-84
8.2
4.1
63
7.7
5.6
6.5
1980-1984
0.28
0.54
1985-89
10.9
4.0
32
10.3
16.3
31
1985-1989
0.20
0.39
1990-94
5.9
1.8
21
6.5
13.3
22
1990-1994
0.14
0.20
1995
17
4.3
~ 45a
61
19
40
64
1995
0.39
0.58
1950
191
57
890
132
247
199
1950
7.41
12.81
mSv
2.5
0.74
~ 3.3b
11.6
1.7
3.2
2.6
mSv
0.21
0.36
a. From Dahlgaard (1994a).
b. Estimated from Greenland dose assuming proportionality between doses in Iceland and Greenland.
Table 8·16. Estimated dietary intakes of 90Sr
Table 8·14. Estimated dietary intakes of 137Cs by selected groups (kBq over period).
by selected groups (kBq over period).
Arctic
Arctic
Arctic
Arctic
Arctic Green-
Northern Russia,
Russia,
Arctic Arctic
Green- Russia, Russia,
Period
Alaska
Finland
land
Iceland Canada
West
East
Norway Sweden
Period
land
West
East
Alaska
1950-1959
330
560
142
2100a
860
240
580
300
1950-1959
3.2
31
20
0.5
1960-1964
370
620
158
2300a
960
270
640
330
1960-1964
3.5
35
23
0.6
1965-1969
420
1000
192
2300a
1050
330
1050
630
1965-1969
2.2
47
26
0.6
1970-1974
240
490
64
1500a
630
240
700
145
1970-1974
0.8
18.8
14.5
0.2
1975-1979
58
210
53
1200a
340
155
460
57
1975-1979
0.5
12.3
8.9
0.2
1980-1984
30
149
58
820a
240
133
260
59
1980-1984
0.3
4.6
3.5
0.1
1985-1989
15
195
57
420a
250
154
560
310
1985-1989
0.2
2.6
2.2
0.04
1990-1994
8
146
26
270a
160
151
270
250
1990-1994
0.2
1.0
0.9
0.02
1995
23
420
61
790a
450
440
800
720
1995
0.5
2.8
2.5
0.07
1950
1390
3800
810
11700a
4900
2100
5300
2800
1950
11
155
102
2.3
mSv
18
49
10.5
152a
64
27
69
36
mSv
0.3
4.4
2.8
0.06
a. Individuals in the selected group in Canada eat 13 times more reindeer meat than the
average Arctic residents. Hence, the dose is estimated to be 13 times higher than that of the
average, because reindeer is the dominating factor for 137Cs in the diet for Arctic Canada.
to calculate the values shown in the tables have been criti-
selected group from Old Crow, where people are assumed to
cally evaluated within the AMAP consultation process. Fur-
consume 1 kg reindeer/d/cap did not validate the high dose cal-
thermore, it should be noted that, due to changes in diets in
culated from the dietary estimate. The wholebody dose was,
Finland and Greenland, it is not possible to achieve the total
in this case, an order of magnitude lower than the dietary
intakes of radionuclides shown in Tables 8·13 to 8·16 from
dose estimate. However, if the Canadian wholebody measure-
simple multiplication of the amounts of food (Tables 8·11
ments were considered representative for the average Cana-
and 8·12) by the activity concentrations in the diet (tables in
dian Arctic population, the agreement between wholebody
Annex). In some cases, activity concentrations in the various
measurements and dietary estimates of dose was satisfactory.
diet components differ between the average population and
The relatively good agreement achieved for some popula-
the selected group. In such cases, a separate activity con-
tions between assessments of dose based on diet and whole-
centration table has been provided for the relevant selected
body measurements may, in some cases, be fortuitous. The
group in the annexed tables.
deposition of 137Cs associated with the accident at Cherno-
The estimated dose from 137Cs in the diet to the average
byl was small in Arctic Finland, Arctic Norway and Arctic
population varies between 0.74 mSv (Greenland) and 11.6
Russia. In areas of Norway and Sweden which were more
mSv (Canada), and for the selected groups range from 10.5
highly contaminated by fallout from the Chernobyl accident,
mSv (Greenland) to 152 mSv (Canada). The time-integrated
the comparison between wholebody 137Cs content and depo-
concentrations of 137Cs in the human body given in Table
sition is influenced by countermeasures. Thus, the weighted
8·53 may be used to estimate doses to members of the se-
averages based on dietary estimates are higher than those
lected groups in three of the Arctic countries. According to
obtained from wholebody measurements. Accordingly, con-
UNSCEAR (1993) a wholebody time integral of 1 Bq y/kg
siderable overestimation of doses to reindeer herding groups
for 137Cs delivers a dose of 2.4 nSv. Hence, the doses to the
of the Saami population occur in some areas of high deposi-
selected groups in Arctic Finland, Arctic Norway and Arctic
tion of Chernobyl caesium if estimated on the basis of diet.
Russia become 21, 32 and 56 mSv, respectively. These doses
However, the dose estimates match reasonably well in areas
may be compared with those calculated from diet intakes in
where Chernobyl deposition was relatively low. Assessments
Table 8·14. It appears that the doses based on wholebody
based on dietary data must, therefore, be made with caution
measurements are about half of those derived from dietary
and should consider possible influences of the provision of
data for Finland and Norway but, in the case of Russia, the
information and the adoption of countermeasures. Any
two estimates are similar.
changes in the use of local food products or changes in the
Wholebody measurements on Arctic population groups
activity concentrations in actually consumed foodstuffs need
have also been carried out in Sweden and Canada. In Sweden,
to be taken into account in dose estimation based on dietary
the dose estimate based on wholebody measurements was a
habits. Further procedures are needed to provide a basis for
factor of two lower than that obtained from dietary studies,
taking account of such effects in dose assessments.
i.e., in agreement with the observations in Finland and Nor-
A tendency to overestimate doses when using dietary cal-
way. In the case of Canada, wholebody measurements on the
culations is often observed. This may be attributed to several
558
AMAP Assessment Report
Table 8·17. Estimated dose commitments to the average Arctic resident
Doses to average members of Arctic populations from an-
from anthropogenic releases of radionuclides.
thropogenic and natural sources are summarized in Tables
8·17 and 8·18. The doses are given as the estimated range
Dose
Radiation type
Dose commitment, mSv
between minimum and maximum average values for the
External
137Cs
0.3-1
eight Arctic countries. The lowest anthropogenic doses are
Other radionuclidesb
0.3-1
those for Greenland and the highest are for Canada. The
Internal
90Sr
0.1-0.4
137
lowest doses from natural sources occur in Iceland. Com-
Cs
0.7-12
pared with UNSCEAR's data for doses from natural sources,
Ingestion and inhalation
14C
2.6a
the Arctic anthropogenic doses correspond to 2-7 years ad-
Other radionuclidesb
0.2-0.6
ditional background radiation.
Total dose commitment
4-18
The dose estimated for the selected groups in the Arctic
a. Infinite dose commitment, which is delivered over several thousands of
are shown in Tables 8·19 and 8·20. The highest dose com-
a. years. The dose from 14C is minor during the period in which the dose a.
mitments are those received by the selected group in Canada,
commitment from 90Sr and 137Cs is received.
which was assumed to have an extremely high consumption
b. The contribution from other radionuclides were estimated from
UNSCEAR (1993).
of reindeer meat (1 kg/d/cap). The Canadian group receives
a dose from 210Po in reindeer meat of about 10 mSv/y. The
dose commitment from 137Cs is calculated to be approximate-
Table 8·18. Estimated annual dose rate to the average Arctic resident from
natural sources.
ly 150 mSv for the same selected group, which corresponds
to the dose received in 15 years from naturally occurring
Source
Annual dose rate, mSv/y a
210Po in the consumed reindeer meat (Beak 1995). However,
External
0.5-1 (0.85)
it is conceivable that there are other population groups in
Internal
0.5-4 (1.53)
Arctic areas of other countries consuming similarly large
Total annual dose rate
1-5 (2.4)
amounts of reindeer/caribou meat and having correspond-
a. Values in parentheses are the global average annual dose rates according ingly high exposures to natural and anthropogenic radionu-
to UNSCEAR (1993).
clides. Compared to the global average doses received from
natural sources, the selected Canadian group receives a dose
Table 8·19. Estimated dose commitments to members of selected Arctic
from natural sources which is four times higher. The dose
population groups associated with previous anthropogenic releases of ra-
commitment received from 137Cs by this selected group is
dionuclides in the environment.
40-50 times higher than that of the average Arctic resident.
Dose
Radiation type
Dose commitment, mSv
The individual mean doses for the average populations
(Tables 8·13 and 8·15) in the various Arctic countries have
External
137Cs
0.3-1
been multiplied by the corresponding population numbers to
Other radionuclides
0.3-1
00
obtain the collective doses for 137Cs and 90Sr from dietary in-
Internal
90Sr
0.1-4
137Cs
10-150
take. The estimated collective dose from 137Cs intake is 9000
manSv, and that from 90Sr is estimated, from the Russian
Ingestion and inhalation
14C
-2.6a
Other radionuclides
- 1~
and Greenland data, to be 25% of the 137Cs dose, i.e. (2000
Total dose commitment
14-160
manSv. The remaining collective dose is estimated to be
6000 manSv, assuming that the anthropogenic mean exter-
a. Infinite dose commitment.
nal dose and the internal doses from radionuclides other
than 90Sr and 137Cs (also excluding 14C) is 1.5 mSv (Table
Table 8·20. Estimated annual dose rate to members of selected Arctic pop-
8·19) and that the total Arctic population is 3.8
106.
ulation groups from natural sources.
Hence the total collective dose is estimated to be 17 000
Source
Annual dose rate, mSv/y a
manSv. This value may be compared with an estimate for
global fallout only of 15 000 manSv (see section 8.5.1.1).
External
0.5-1
These two estimates are in fairly good agreement particu-
Internal (incl. 210Po)
0.5-10
Total annual dose rate
1-11
larly considering that the dose from Chernobyl is included
in the 17 000 manSv. In this calculation, it was assumed that
the individual dose commitment to the average population
factors. Probably the most important of these is an overesti-
in Alaska from 137Cs is 2 mSv and to the Faeroese popula-
mate of the amounts of food actually consumed. Other fac-
tion 3.3 mSv. The range of individual dose commitment
tors are loss of 137Cs during food preparation and a tendency
(4-17 mSv) received by the average Arctic inhabitant ex-
for measurements to represent maximum rather than average
ceeds the average dose commitment for individuals living in
activity concentrations. It should also be noted that whole-
the north temperate zone (40-50°N) which according to
body measurements can be biased because the individuals
UNSCEAR (1993) is 4.4 mSv.
participating in wholebody measurements may not be repre-
sentative of the population group within which they reside.
8.4.3. Intakes of 137Cs through various
Doses from 90Sr in diet were calculated for a few coun-
dietary components
tries only. The effective dose commitment from 90Sr varied
from 0.21 mSv to 0.36 mSv for the average population, and
Figures 8·38, 8·40, 8·42, 8·44, 8·47, and 8·49 depict the
for selected groups between 0.06 and 4.4 mSv. The doses
yearly intakes of 137Cs in various dietary components of the
from 90Sr are thus significantly lower than those from 137Cs,
average members of the populations of Arctic Finland,
particularly for the selected groups. Reindeer meat is the
Greenland, Canada, Russia, Norway and Sweden, respec-
predominant source of 137Cs in the Arctic diet. 90Sr activity
tively. Figures 8·39, 8·41, 8·43, 8·45, 8·46, 8·48, and 8·50
concentrations in reindeer meat are typically two orders of
depict the yearly intakes of 137Cs in dietary components of
magnitude lower than 137Cs activity concentrations. For in-
the selected populations of Arctic Finland, Greenland, Can-
takes of 137Cs by dietary component, see the annex to this
ada, eastern Russia, western Russia, Norway and Sweden,
chapter.
respectively.
Chapter 8 · Radioactivity
559
137Cs intake
137Cs intake
Bq/person/y
Bq/person/y
10 000
5 000
9 000
4 000
8 000
3 000
7 000
2 000
6 000
1 000
5 000
0
1960-1964
1965-1969
1970-1974
1975-1979 1980-1984
1985-1989
4 000
Freshwater fish
Mushrooms/Berries
Other
3 000
Sheep and goat milk
Reindeer/Game
Figure 8·40. Y
8£54d01
early intake of 137Cs from various dietary components by
2 000
the average population of Greenland.
137Cs intake
1 000
Bq/person/y
50 000
0
1960-1964
1965-1969
1970-1974
1975-1979 1980-1984
1985-1989
45 000
Freshwater fish
Mushrooms/Berries
Other
40 000
Sheep and goat milk
Reindeer/Game
Figure 8·38. Y
8£52d01
early intake of 137Cs from various dietary components by
35 000
the average population of Arctic Finland.
30 000
137Cs intake
Bq/person/y
200 000
25 000
180 000
20 000
160 000
15 000
140 000
10 000
120 000
5 000
100 000
0
1960-1964
1965-1969 1970-1974 1975-1979
1980-1984 1985-1989
80 000
Freshwater fish
Mushrooms/Berries
Other
Figure 8·41. Yearly intake of 137Cs from various dietary components by
R i d
/G
60 000
the `selected' population of Greenland.
40 000
Variations in sources of 137Cs intake
Figures 8·39, 8·41, 8·43, 8·45, 8·46, 8·48 and 8·50 for the
20 000
selected Arctic groups show, in all cases, the same feature:
reindeer/caribou meat is, for these groups, the dominating
0
source of dietary 137Cs. In general, less than 10% of the
1960-1964
1965-1969 1970-1974 1975-1979
1980-1984 1985-1989
137Cs comes from other dietary components, mostly from
freshwater fish.
Freshwater fish
Mushrooms/Berries
Other
For average populations, the various dietary compo-
nents contributing to 137Cs intake vary considerably among
Sheep and goat milk
Reindeer/Game
Arctic countries. In Arctic Canada (Figure 8·42), rein-
deer/caribou has been the dominant source of 137Cs in the
Figure 8·39. Yearly intake of 137Cs from various dietary components by
the `selected' population of Arctic Finland.
diet. This is because the percentage of indigenous people
560
AMAP Assessment Report
137Cs intake
137Cs intake
Bq/person/y
Bq/person/y
50 000
7 000
6 000
5 000
40 000
4 000
3 000
30 000
2 000
20 000
1 000
0
1960-1964
1965-1969
1970-1974
1975-1979 1980-1984
1985-1989
10 000
Freshwater fish
Mushrooms/Berries
Other
Sheep and goat milk
Reindeer/Game
Figure 8·44. Yearly intake of 137Cs from various dietary components by
the average population of Arctic Russia.
0
1960-1964 1965-1969 1970-1974 1975-1979
1980-1984 1985-1989
137Cs intake
Reindeer/Game
Bq/person/y
70 000
Figure 8·42. Yearly intake of 137Cs from various dietary components by
the average population of Arctic Canada.
60 000
137Cs intake
Bq/person/y
50 000
500 000
40 000
400 000
30 000
300 000
20 000
10 000
200 000
0
1960-1964 1965-1969 1970-1974 1975-1979
1980-1984 1985-1989
100 000
Freshwater fish
Mushrooms/Berries
Other
0
Sheep and goat milk
Reindeer/Game
1960-1964
1965-1969 1970-1974 1975-1979 1980-1984 1985-1989
Figure 8·45. Yearly intake of 137Cs from various dietary components by
Reindeer/Game
the `selected' population of eastern Arctic Russia.
Figure 8·43. Yearly intake of 137Cs from various dietary components by
the `selected' population of Arctic Canada.
products such as seals and fish. Contributions of 137Cs
from freshwater fish are most important in Arctic Finland
(with high caribou consumption) in Arctic Canada is
(Figure 8·38).
higher than in the other Arctic countries. For Arctic Fin-
land (Figure 8·38), Arctic Russia (Figure 8·44) and Arctic
Temporal variations in 137Cs intake
Norway (Figure 8·47) agricultural products (i.e., milk,
All the calculations give a maximum intake of 137Cs with
cereals, beef and pork) are important contributors to diet-
diet in the 1960s due to the peak in global fallout around
ary 137Cs. In Arctic Sweden (Figure 8·49), mushrooms and
1962-1964. In some cases, the maximum occurred in the
berries dominate for most years. In Greenland (Figure 8·40),
first half of the decade, in others in the second half. The
lamb and reindeer are the most important sources of 137Cs
Chernobyl signal was evident in 1985-1989 in all Arctic
in the diet of the average population. Although 87% of the
areas included in these calculations except Canada (Figures
Greenland population is indigenous, reindeer is not as im-
8·42 and 8·43). In Greenland (Figures. 8·40 and 8·41)
portant to 137Cs intake for the average population as it is
(and probably also Iceland) the signal was weak, but none-
in Arctic Canada, because Greenlanders prefer marine
theless present. The strongest response on the Chernobyl
Chapter 8 · Radioactivity
561
137Cs intake
137Cs intake
Bq/person/y
Bq/person/y
250 000
250 000
200 000
200 000
150 000
150 000
100 000
100 000
50 000
50 000
0
0
1960-1964 1965-1969 1970-1974 1975-1979 1980-1984
1985-1989
1960-1964
1965-1969 1970-1974 1975-1979 1980-1984
1985-1989
Freshwater fish
Mushrooms/Berries
Other
Freshwater fish
Mushrooms/Berries
Other
Sheep and goat milk
Reindeer/Game
Sheep and goat milk
Reindeer/Game
Figure 8·46. Y
8£60d01
early intake of 137Cs from various dietary components by
Figure 8·48. Yearly intake of 137Cs from various dietary components by
the `selected' population of western Arctic Russia.
the `selected' population of Arctic Norway.
137
137Cs intake
Cs intake
Bq/person/y
Bq/person/y
6 000
10 000
5 000
9 000
4 000
8 000
3 000
7 000
2 000
6 000
1 000
5 000
0
4 000
1960-1964
1965-1969
1970-1974
1975-1979 1980-1984
1985-1989
3 000
Freshwater fish
Mushrooms/Berries
Other
Sheep and goat milk
Reindeer/Game
2 000
Figure 8·49. Yearly intake of 137Cs from various dietary components by
the average population of Arctic Sweden.
1 000
120
137 000
Cs intake
Bq/person/y
0
1960-1964
1965-1969
1970-1974
1975-1979 1980-1984 1985-1989
100 000
Freshwater fish
Mushrooms/Berries
Other
80 000
Sheep and goat milk
Reindeer/Game
60 000
Figure 8·47. Yearly intake of 137Cs from various dietary components by
the average population of Arctic Norway.
40 000
fallout was seen in Arctic Sweden (Figures 8·49 and 8·50)
where the estimated 137Cs intake increased by a factor of 5
20 000
from 1980-1984 to 1985-1989.
Changes in the relative importance of
0
1960-1964 1965-1969
1970-1974 1975-1979 1980-1984 1985-1989
dietary components with time
The relative importance of the various diet sources to 137Cs
Freshwater fish
Mushrooms/Berries
Other
intake by the Arctic average populations showed only minor
changes throughout the years. In Arctic Sweden (Figure 8·49),
Sheep and goat milk
Reindeer/Game
however, the relative importance of mushrooms has been in-
Figure 8·50. Yearly intake of 137Cs from various dietary components by
creasing since the 1960s.
the `selected' population of Arctic Sweden.
562
AMAP Assessment Report
8.4.4. Summary
8.5.1.1. Atmospheric nuclear weapons tests
The doses to Man in the Arctic from natural and anthro-
The three major sites for atmospheric testing of thermonu-
pogenic radiation derive from both external and internal
clear weapons have been Novaya Zemlya in the Arctic re-
sources. The major contribution to the average population
gion of the former Soviet Union (FSU), Bikini and Eniwetok
is the dose from inhalation of naturally-occurring radon; the
Islands (USA) in the Pacific Ocean and the Nevada test site
annual dose from radon is 0.5-4 mSv/y, corresponding to a
(Figure 8·51). In addition, the FSU conducted tests at Semi-
lifetime dose of 30-300 mSv. Lifetime doses to present gen-
palatinsk in Soviet Central Asia, China at Lop Nor in west-
erations of the Arctic average population due to anthro-
ern China, France at Mururoa in the Pacific Ocean and the
pogenic radionuclides vary between 2 and 15 mSv or about
United Kingdom in Maralinga and the Monte Bello Islands,
5% of the dose from natural sources. The most important
Australia, and the Christmas Islands in the Pacific. The USA,
contribution to enhanced doses from anthropogenic radio-
UK and FSU all discontinued atmospheric testing by 1962.
nuclides in the Arctic is the consumption of reindeer/caribou
Since 1980, there has been no atmospheric testing carried
meat. Certain specific population groups, for instance those
out by any country.
in Arctic Canada having extreme consumption rates of rein-
A total of 520 atmospheric explosions took place up to
deer meat, i.e., of the order of 1 kg/d/cap, may get lifetime
1980 (UNSCEAR 1993). The total explosive yield amounted
doses from 137Cs of the order of 150 mSv. In addition, indi-
to 545 Mt (TNT equivalents): 217 Mt from fission explo-
viduals in such population groups may receive annual doses
sions and 328 Mt from thermonuclear (fusion) explosions.
from naturally-occurring 210Po in reindeer meat of 10 mSv/y.
According to UNSCEAR (1993), of the aggregate releases of
It cannot be ruled out that there are small numbers of indi-
radionuclides to the environment, fallout activity deposited
viduals within the other Arctic countries having similar diet-
close to the test sites accounts for 12%, tropospheric fallout,
ary habits as the selected Canadian community. Accordingly,
which is deposited in a band around the Earth at the latitude
comparable or higher doses than those calculated for the
of the test site, for 10%, and global fallout, which is mainly
Canadian selected group may exist within the Arctic.
deposited in the same hemisphere as the test site, for 78%.
This study shows that people eating marine products
As most test explosions have been carried out in the north-
only, for instance fish and marine mammals, receive doses
ern hemisphere, (see Figure 8·51), most of the radioactive
from anthropogenic radionuclides that are at least an order
contamination is found there. Contamination of Arctic re-
of magnitude lower than those to people consuming terres-
gions is, in general, less than that of temperate areas. The
trial products such as reindeer/caribou, freshwater fish and
AMAP GIS-based estimate of the total integrated deposition
mushrooms. For this reason, individuals predominantly con-
of 137Cs on land north of 60°N (section 8.3), derived during
suming seafood have lower dose rates and lifetime doses
the course of this assessment, yields a value of 35 PBq (cor-
than average members of both the Arctic and northern hemi-
responding to 18 PBq in 1995). The average ratio between
137
spheric populations.
Cs and 90Sr in global fallout is 1.6. Hence, the present in-
ventory of global fallout of 90Sr in the Arctic (60-90°N), es-
timated using the GIS approach, is approximately 11 PBq.
Local fallout from thermonuclear tests tends to be minor
8.5. Source-related assessments of
compared with the global fallout because these tests were
past and present releases
frequently conducted at high altitude in which much of the
As discussed in section 8.2.3.7, two categories of source-
debris is injected into the stratosphere and the fireballs cre-
related assessments need to be discussed in this document.
ated by the explosions did not reach the ground. In the pop-
These are:
ulated part of the Arctic, where the major food production
occurs (i.e., the 60-70° N latitude band), the integrated de-
· Previous and continuing releases of radionuclides to the
position density of 90Sr (Figure 8·52) is 1.7 kBq/m2 and that
environment from human activities.
of 137Cs is 2.6 kBq/m2 (UNSCEAR 1982). The equivalent
· Potential, or possible future, releases of radionuclides to
value for deposition on land of 137Cs using the GIS-based
the environment resulting from human activities.
approach is 2.2 kBq/m2. The UNSCEAR estimates of depo-
In this section of the document, previous and continuing re-
sition densities have been used for dose estimation for the
leases from human activities are discussed. Such releases are
Arctic population from global fallout.
of two distinct types: operational and accidental. Operatio-
The transfer coefficients of UNSCEAR (see section 8.2.4)
nal releases are those authorized in the licensing of practices
predict the doses to individuals at temperate latitudes from
and are generally routine releases from nuclear fuel cycle op-
ingestion, inhalation and external exposure pathways, re-
erations. Nuclear explosions intended both for the testing of
spectively. If the transfer coefficients in Table 8·21 are as-
nuclear weapons and for peaceful purposes are also classi-
sumed to be also valid for the Arctic population, and if the
fied for the purposes of this review as operational releases.
integrated deposition densities given above for the Arctic
Accidental releases are unintentional releases of radionu-
are applied, average individual doses in the Arctic are, by
clides to the environment.
simple multiplication, 0.1 mSv for 90Sr and 0.4 mSv for 137Cs.
Taking the total population of the Arctic to be 3.8
106,
the collective dose from 90Sr and 137Cs is, according to the
8.5.1. Nuclear explosions
UNSCEAR model, estimated to be approximately 2000
The predominant releases of artificial radioactivity on a glo-
manSv. This dose would be very unevenly distributed. Popu-
bal scale have been derived from atmospheric testing of nu-
lation groups consuming local terrestrial products, such as
clear weapons. Thermonuclear weapons tests in the atmos-
reindeer/caribou meat, would receive significantly higher
phere have contributed the most to global fallout. Contami-
doses than those people living on imported products from
nation resulting from underground nuclear weapons testing
temperate regions. In contrast, people consuming locally-
and underground civilian nuclear explosions is, from a global
produced marine products, such as fish and sea mammals,
perspective, negligible. However, some underground explo-
would receive lower doses than those consuming largely im-
sions have resulted in local environmental contamination.
ported products. For instance, dietary studies in Greenland
Chapter 8 · Radioactivity
563
Novaya Zemlya
Semipalatinsk
Nevada
Test Site
Reggane
Lop Nor
Bikini Is.
Eniwetok Is.
Christmas Is.
Monte Bello
Islands
Mururoa
Maralinga
8£65d01
Figure 8·51. Sites where atmospheric testing of nuclear weapons has taken place since 1945 (Zander and Araskog 1973).
90
reindeer meat is the main contributor to the 137Cs dose.
3 500
0
500
1 000
1 500
2 000
2 500
3 000
Sr Bq/m2
The total annual production of reindeer meat in the Arctic
Latitude
has been estimated by the assessment group to be 2.8
107
90°N
kg (Table 8·54). If the integrated transfer coefficient for
80°N
reindeer meat is equal to the mean of that observed in this
Northern
70°N
Hemisphere
study for the Arctic countries of 10 kBq 137Cs kg1 y per
60°N
kBq 137Cs/m2 (Table 8·51), the total intake of 137Cs from
50°N
reindeer becomes 2.8
107
104
2.6 7.3
1011 Bq 137Cs
40°N
(assuming an integrated deposition density of 2.6 kBq/m2 of
137
30
Cs from global fallout in the Arctic). The dose intake fac-
°N
tor is 13 nSv per Bq 137Cs and the collective dose thus be-
20°N
comes 9500 man Sv. In other words, if we conservatively
10°N
assume that all reindeer meat produced in the Arctic is con-
0°
sumed by Arctic residents (i.e., a consumption rate of 7.4
10°S
kg/y/cap), this would increase the estimated dose by 9500
20°S
manSv (i.e., to 12 000 manSv). The individual mean dose
30°S
commitment from 137Cs then becomes 3.2 mSv, which is
40°S
Southern
eight-fold greater than that (0.4 mSv) estimated using the
Hemisphere
50
UNSCEAR methodology.
°S
It is unlikely that all reindeer meat produced in the Arctic
60°S
is consumed by the Arctic people. For instance, if the indi-
70°S
vidual consumption data given in 8.4.2 are multiplied by the
80°S
respective populations of the Arctic countries, the estimated
90°S
consumption of reindeer meat by Arctic populations is only
Figure 8·52. Integrated deposition density of 90Sr (UNSCEAR 1993).
1.5
107 kg/y which is approximately half the estimated an-
nual production of reindeer meat of 2.8
107 kg/y. How-
have shown that 137Cs intake from the three types of diet:
ever, other terrestrial food products with enhanced 137Cs lev-
local terrestrial, imported, and local marine, vary in the
els are also consumed, such as freshwater fish, game, mush-
ratio 20 : 2 : 1 (Aarkrog et al. 1963-1995).
rooms and berries, and these will also contribute significant-
The transfer coefficients estimated by UNSCEAR in
ly to the dose. On these grounds, the collective dose from
Table 8·21 do not consider the enhanced intakes of 137Cs
90Sr and 137Cs in global fallout from atmospheric nuclear
from Arctic semi-natural and natural ecosystems. There-
weapons testing to the Arctic population would more realis-
fore, the estimated doses calculated above are lower than
tically be of the order of 10 000 manSv, i.e. five-fold greater
the actual doses received by Arctic populations because
than the UNSCEAR-based estimate.
Table 8·21. Transfer coefficients adopted by UNSCEAR (1993) for global fallout (see Figure 8·1).
Ingestion
Inhalation
External exposure
Transfer coefficient
Unit
90Sr
137Cs
90Sr
137Cs
137Cs
Remarks
P01
Bq m3 y (PBq) 1
9.3 106
40-60°N
P12
Bq m2 (Bq m3 y) 1
5.56 105
Equivalent to 1.8 cm/s
P23 (total diet)
Bq kg1 y (kBq m2) 1
3.8
8.4
00
0
Mean of Argentina,
}
P34 (total diet)
kg y1 cap1
500
.6
0
Denmark and USA
P45
nSv Bq1
28.6
13.6
350.6
8.5
ICRP
P14
m3 y1 cap1
.6
.6
7300
Inhalation rate : 20 m3/d
P24
Bq cap1 (kBq m2) 1
1900.6
4200.6
13
.6
P23 P34
P25
mSv (kBq m2) 1
52.6
55.6
4.6
0.11
97
P24 P45 ingest
564
AMAP Assessment Report
The contribution from other shorter-lived radionuclides
along fissures, and takes a considerable time. Only inert
present in global fallout, such as 3H, 54Mn, 95Zr, 95Nb, 106Ru,
gases and, exceptionally, isotopes of highly volatile ele-
131I, and 144Ce, will increase the dose to present generations
ments (e.g., halogens), are released into the atmosphere.
in the Arctic by about 1000 manSv (UNSCEAR 1993). Fur-
Releases of the isotopes 133Xe, 135Xe, 137Xe and 138Xe are
thermore, if the doses to be received within the next 50 years
the most common.
from very long-lived radionuclides such as 14C and the trans-
With shallow ventilated underground explosions, only
uranic elements (Pu and Am) are included, the dose to Arctic
isotopes of iodine provide unambiguous indicators of vent-
populations from global fallout radionuclides other than
ing. However, only insignificant proportions of these iso-
90Sr and 137Cs is estimated to be in the order of 2000 manSv.
topes are normally released into the atmosphere. If the re-
Hence, the total collective dose commitment over the next
lease of radioactive products into the atmosphere in case of
50 years from nuclear weapons testing in the atmosphere to
a camouflet explosion (one that does not result in immediate
the Arctic population of 3.8
106 will be of the order of
venting to the atmosphere) commences a few minutes after
13 000 manSv (see section 8.4.3).
the explosion, the isotopes of inert gases, whose decay prod-
ucts form aerosols, may enter the atmosphere. Such isotopes
of inert gases (89,90Kr and 137Xe) have rather short half-lives
8.5.1.2. Underground nuclear explosions
and, therefore, their decay products (89,90Sr and 137Cs) are
8.5.1.2.1. Underground explosions carried out in the Arctic
more important. These radionuclides, together with tritium
by the former Soviet Union
and long-lived isotopes of induced activity (e.g., 60Co), are
Military nuclear explosions carried out by the FSU at under-
the dominant contaminants of the environment in cases
ground locations on Novaya Zemlya are described in section
where releases of radioactive products into the atmosphere
8.3.1.2.1 in the context of localized contamination. The de-
actually occur. These radionuclides will be released to the
scription of these weapons tests is not repeated here.
atmosphere in the radioactive cloud in the case of camouflet
During the period 1965-1988, a total of 116 peaceful un-
or excavating explosions, either when the cavity breaks or
derground nuclear explosions (PUNEs) have been carried out
the rock roof fails due to gas pressure.
in the FSU (Figure 8·53). Of these, 17 were conducted in re-
A short-range surface plume in the vicinity of the explo-
gions near the Arctic Circle from 1971 to 1988. Specifically,
sion results from the deposition of most of the larger parti-
one was carried out in the Komi Republic, two in the Mur-
cles from the radioactive cloud and may extend for several
mansk region, two in the Tjumen region, four in the Krasno-
hundred meters. The short-range plume may be defined by
yarsk region, four in the Sakha-Yakutia Republic, and four in
fallout from the cloud during the first 24-hours following
the Archangelsk region (RCRA 1997, Lystsov 1995). A recent
releases from the explosion.
Russian report compiling information on nuclear explosions
At three PUNE sites there has been significant local con-
by the FSU (Mikhailov et al. 1996) has been used as an addi-
tamination:
tional source of information in this assessment.
· At the first site, located 100 km to the north from Kras-
novieshersk, an underground explosion took place on 23
March, 1971. Three 15 kt devices were exploded simulta-
neously at a rather shallow depth of 128 m. The purpose
of the explosion was to construct a canal, as part of a
major project to alter the direction of some northward
flowing rivers to flow to the south. This explosion was
the first of 250 planned PUNEs intended to excavate this
canal. The 700 m long trench formed by the three devices
was significantly contaminated. Information is not avail-
able on the types or amounts of this contamination. How-
Krasnovieshersk
ever, radiation levels in the area were up to 1 mrem/h (10
Aykal
mSv/h) 15 years after the explosion.
· At the second site, located 90 km to the northeast of Ay-
kal, an explosion took place on 2 October, 1974 at a
depth of 98 m. The explosive yield was 1.7 kt and its pur-
pose was the construction of a dam. The explosion re-
Civilian nuclear explosions
sulted in an accidental release of radionuclides, mainly
8·53. Location of PUNE's in Arctic Russia.
137Cs, 90Sr, 239,240Pu, 60Co, 125Sb and 241Am. A total of
eight such explosions were planned, but the program was
The main application of PUNEs was for mining and con-
abandoned after this accident. In situ measurements were
struction purposes. PUNEs were also used for emergency ex-
not performed until 1990-1993. These showed contami-
tinction of gas-gushers, oil and gas production, creating un-
nation in soil samples of almost 20 kBq/kg of 137Cs and
derground storage cavities, disposal of toxic liquid waste,
more than 35 kBq/kg of 239,240Pu. Unfortunately, no infor-
ore crushing, extinguishing gas fires in coal pits, rock exca-
mation is given on soil sample depth and, therefore, these
vation, rock loosening and cratering.
values cannot be transformed into estimates of areal con-
These explosions had yields ranging from less than 0.5-40
tamination in Bq/m2. Neither is it possible, from the in-
kt of TNT equivalent and the total yield was somewhat less
formation given, to estimate the amounts of specific radio-
than 550 kt of TNT. They were carried out at depths of be-
nuclides released. However, there is some information
tween 100 and 2860 m. It is difficult to estimate the scales
about environmental consequences, such as the death of
of surface radioactive contamination resulting from these
trees (Lystsov 1995).
nuclear explosions as information on the condition of the
· At the third site, located 120 km to the southeast of Ay-
surrounding regions has only recently been made available.
kal, an explosion took place on 24 August, 1978, at a
With deep underground explosions, the release of ra-
depth of 577 m. The explosive yield was 19 kt. Its pur-
dioactive products is possible only through soil layers or
Chapter 8 · Radioactivity
565
pose was seismic sounding of the Earth's crust. After the
Environmental contamination that has occurred as a re-
fifth second following the explosion, a radioactive release
sult of these tests is specified as:
was recorded. It is believed that the release took place via
· Subsurface contamination surrounding each shot cavity.
an incompletely sealed well. The contaminated cloud
· Several seeps containing trace amounts of radionuclides
moved over the workers camp and the site and about 80
(principally tritium) have been found at, or near, the mud
persons were exposed. The dominant radionuclides dur-
disposal pits at the Long Shot ground zero.
ing the first days following the explosion were 131I, 140Ba
· Erosion of buried waste disposal pits has the potential for
and 140La. Currently, the dominant radionuclides are
137
leakage into the ocean.
Cs, 90Sr, 238Pu, 239,240Pu, 60Co and 125Sb. Large-scale de-
contamination was carried out in the summer of 1981.
An environmental organization issued a report on October
From the information provided, it is not possible to esti-
30, 1996, detailing the results of radionuclide sampling con-
mate the magnitude of the contamination. Soil samples
ducted on Amchitka Island during the summer of that year.
were taken in 1990-1992 and showed concentrations of
This organization alleges, on the basis of these data, that
less than 1 kBq/kg of 90Sr, less than 2 kBq/kg of 239,240Pu
samples of moss indicate leakage of 239,240,241Pu and 241Am
and less than 10 kBq/kg of 137Cs. Concentrations of 90Sr
from the cavity created by the Cannikin explosion. While it
in reindeer moss (lichen) and shrubs of up to 44 kBq/kg
does not claim that these releases constitute a significant risk
have been found. At this site, a `dead forest' (100 hect-
to human health, it contends that future releases could con-
ares) has been reported (Lystsov 1995).
stitute a significant risk, especially if bioaccumulation occurs.
The USEPA will assess the samples collected by the en-
vironmental organization in 1996 to validate the results
8.5.1.2.2. Underground explosions carried out in the Arctic
and USEPA samples to be collected in 1997 will be pro-
by the United States
vided to the environmental organization and other inter-
Amchitka Island, Alaska, the southernmost island of the Rat
ested groups for cross-checking. Results of the most recent
Island Group in the Aleutian Chain, was the site of three
sampling will be made available by September 30, 1997.
high-yield underground nuclear explosions for seismic stud-
The biennial sampling program at Amchitka will be ex-
ies, calibration and warhead development between 1965 and
panded to include appropriate biota to address concerns
1971. Amchitka lies approximately 2100 km southwest of
about bioaccumulation of radionuclides in the food chain.
Anchorage. The island is about 64 km long and 1.6-6.4 km
The State of Alaska, native groups (including the Aleutian
wide with an area of approximately 30 000 ha. The surface
and Pribiloff Islands Association) and other interested par-
elevation of all three test sites was 41 m above sea level.
ties will participate in the formulation of objectives and
Unlike the underground explosions in the Russian Arctic,
protocols for the sampling program. The USDOE will re-
which had explosive yields of less than 50 kilotons, the USA
view all available data to determine whether there are any
explosions in Amchitka were of much larger yields as item-
monitoring or other data which would indicate previously
ized below:
unreported releases of radionuclides from the three shot
cavities. Finally, relevant classified materials will be reviewed
· Long Shot, October 29, 1965, was detonated at 716 m
with a view to de-classification.
depth in basalt as part of the Vela Uniform program to
obtain event measurements relating to the detection of
underground nuclear explosions. The yield of Long Shot
8.5.2. Operational releases from
was 80 kt.
the nuclear fuel cycle
· Milrow, October 21969, was detonated at 1220 m depth
The term `nuclear fuel cycle' is used to delineate activities
in pillow lava as a seismic calibration test. The yield of
associated with the production of energy from fission reac-
Milrow was approximately 1 Mt.
tors encompassing uranium mining, uranium enrichment,
· Cannikin, November 6, 1971, was detonated at 1790 m
fuel fabrication, reactor fuel insertion, nuclear reactor oper-
depth in basalt as a test of the proposed warhead for the
ation, spent fuel removal, reprocessing and the ultimate dis-
Spartan missile. The yield of Cannikin was 5 Mt.
posal of wastes from the nuclear power industry. This sub-
Site deactivation was started by the US Department of En-
section deals with operational releases from nuclear power
ergy in February, 1972, and completed in September, 1973.
plants and nuclear fuel reprocessing plants.
No known escape of radionuclides to the surface has oc-
curred as a result of these tests with the exception of trace
8.5.2.1. Nuclear power plants
quantities of tritium which can be detected in surface water
and shallow groundwater above the Long Shot detonation
There are two nuclear power plants (NPPs) within the Arc-
point. A preliminary assessment conducted by the US De-
tic, the Kola NPP near Polyarny Zori on the Kola Peninsula
partment of Energy (USDOE) under the terms of the Com-
and the Bilibino NPP which is in the Chukchi region of east-
prehensive Environmental Response, Compensation and Li-
ern Russia. In addition, there are NPPs in Sweden, Finland
ability Act was submitted to the US Environmental Protec-
and Russia that are within 1000 km of the Arctic Circle.
tion Agency (USEPA) in 1988. This assessment recommen-
ded that sampling in the vicinity of the test sites be contin-
8.5.2.1.1. Nuclear power plants in the Arctic
ued and that the results of migration studies carried out at
the Nevada test sites be evaluated for estimating the poten-
The two NPPs operating in the Arctic are: one plant with
tial for radionuclide migration at Amchitka.
four VVER (water-cooled and water-moderated energy reac-
The three test sites continue to be monitored as part of
tor) units at Kola and one plant with four EGP-6 (light wa-
the Long-Term Hydrological Monitoring Program (LTHMP)
ter cooled graphite-moderated) units at Bilibino (Table 8·22,
carried out by the USEPA. Under this program, samples are
next page) (RCRA 1997).
collected biennially. Samples of groundwater (23), surface
The criteria (Tsaturov 1996) for critical group exposure
water (19), spring water (1) and rainwater (1) were last col-
used to calculate release limits for normal operation of nu-
lected in 1995.
clear power plants in Russia are shown in Table 8·23.
566
AMAP Assessment Report
Table 8·22. Russian NPPs in the Arctic.
Atmospheric releases
The limits for releases of radionuclides in airborne effluents
Installed
Commercial
are calculated from the above criteria and are shown in
Unit
Type/model
capacity, MW(e)
start-up date
Table 8·24.
Kola 1
VVER - 440/230
440
1973
Actual releases into the atmosphere from the Kola NPP
Kola 2
VVER - 440/230
440
1974
between 1985 and 1994 are shown in Table 8·25.
Kola 3
VVER - 440/213
440
1981
Kola 4
VVER - 440/213
440
1984
Actual radionuclide releases into the atmosphere from the
Bilibino 1
EGP-6
12
1974
Bilibino NPP for the period 1991-1994 are shown in Table
Bilibino 2
EGP-6
12
1974
Bilibino 3
EGP-6
12
1975
8·26.
Bilibino 4
EGP-6
12
1976
Liquid releases
The liquid release rates of some radionuclides from the Kola
Table 8·23. Dose criteria for releases from NPPs in Russia.
NPP for the period 1988 -1993 are shown in Table 8·27.
Criteria for
Criteria for
airborne
liquid
Other wastes
Critical organ
effluents, mSv/y effluents, mSv/y
Disposal sites at the Kola NPP contain 5000 m3 of solid ra-
Whole body and bone marrow
0.2
0.05
dioactive wastes having 155
1010 Bq ( 42 Ci) of activity
Other organs (except skin and extremities)
0.6
0.15
and 65 000 m3 of liquid wastes having 7
1014 Bq ( 1.9
Skin and extremities
1.2
0.30
104 Ci) of activity.
The associated collective and critical group doses associ-
ated with releases from these NPPs were not available to the
Table 8·24. Release limits of radionuclides in airborne effluents for Kola
assessment group.
and Bilibino NPPs.
Release limits, GBq/y (Ci/y)
8.5.2.1.2. Nuclear power plants in the vicinity of the Arctic
NPP
Inert gases
Radioiodine
Long-lived nuclides
There are additional NPPs in Russia, Finland and Sweden sit-
Kola
27000000
540
810
(730000)
(15.0)
(22.0)
uated within 1000 km of the Arctic Circle. These may have
Bilibino
6700000
140
2040
relevance to Arctic populations, particularly in the event of
(180000)
(3.7)
(55.0)
accidents, and have been included here for completeness.
Russian NPPs
Table 8·25. Annual radioactive releases into the atmosphere from the Kola
NPP, 1985-1994.
The Leningrad (now St. Petersburg) NPP is situated about
1000 km from the Arctic Circle. It has four units with graphite
Releases, GBq
moderated pressure tube boiling water reactors, a type of reac-
Year
Inert gases
131I
Long-lived nuclides
tor which has only been constructed in the former Soviet Union.
The graphite consists of blocks that are arranged in the form
1985
1040000
0.54
0.22
1986
540000
0.77
0.43
of columns penetrated by vertical channels that provide loca-
1987
560000
1.60
0.91
tions for the fuel rods, control rods, graphite reflector coolant
1988
420000
1.10
0.89
tubes and instrumentation (NKS 1994). The electrical output
1989
420000
64
113
1990
270000
3.5
85
of each unit is 1000 MW(e). The plant has been built in two
1991
000
.
.
versions. The main differences between these versions are the
1992
280000
1.20
2.55
nature of the emergency cooling systems and containment sys-
1993
180000
5.6
3.20
1994
82000
3.1
3.0
tems. The four units of the Leningrad NPP were put into oper-
ation in the years 1973, 1975, 1979 and 1981, respectively.
The releases as gas and aerosols into air from the Lenin-
Table 8·26. Annual radionuclide releases into the atmosphere from the
grad power plant are given in Table 8·28.
Bilibino NPP, 1991-1994.
Table 8·28. Annual releases to the atmosphere of radionuclides as gas and
Releases, GBq
aerosols from the Leningrad NPP, 1992-1995, GBq (Ci).
Short-lived
Long-lived
radionuclides,
radionuclides,
Releases in gases and aerosols to the atmosphere, GBq
Year
Inert gases
half-life < 24 h
half-life > 24 h
131I
(Ci, values in parentheses).
Radio-
1991
274100
1.71
Background
n.d.
nuclide
1992
1993
1994
1995
1992
353800
4.41
Background
n.d.
1993
326300
0.89
Background
n.d.
Inert radio-
1390000
1610000
1790000
1070000
1994
417100
1.81
Background
n.d.
active gases
(37625)
(43629)
(48360)
(29000)
Long-lived 81
20.9
59.9
35
nuclides
(2.2)
(0.564)
(1.62)
(0.95)
131I
89
20
50
20
Table 8·27. Annual liquid releases of some radionuclides from the Kola
(2.4)
(0.53)
(1.36)
(0.53)
NPP, 1988-1993.
51Cr
14
13.6
5.9
(0.37)
(0.367)
(0.16)
Releases, GBq
54Mn
0.22
0.31
0.1
Radionuclide
1988
1989
1990
1991
1992
1993
(6 103)
(8.43 103)
(2.6 103)
60Co
0.67
0.41
0.14
Chromium-51
0.037
0.310
0.540
0.093
0.19
(18 103)
(11.2 103)
(3.7 103)
Manganese-54
0.110
0.630
0.790
0.630
0.82
137Cs
2.6
0.95
1.88
1.2
Cobalt-58
0.100
1.070
0.950
0
0.27
(69 103)
(25.6 103)
(50.9 103)
(33 103)
Cobalt-60
0.280
1.310
1.140
0.990
1.07
89Sr
0.13
0.10
0.28
0.08
Strontium-90
0.014
0.005
0.005
0
(3.4 103)
(2.63 103)
(7.6 103)
(2.1 103)
Caesium-134
0.020
0.035
0.072
0.160
0.14
90Sr
0.03
0.10
0.04
0.02
Caesium-137
0.083
0.078
0.240
0.24
(0.8 103)
(2.63 103)
(1.1 103)
(0.54 103)
Chapter 8 · Radioactivity
567
The collective and critical group doses associated with
release rate limits for radionuclides during normal operation
releases from the Leningrad NPP were not available to the
are defined and specified in the authorization. Adherence to
AMAP radioactivity assessment group and, thus, could not
these limits alone does not suffice; in addition, radionuclide
be included for comparative purposes.
releases shall be kept as low as reasonably achievable
(ALARA). To limit the overall exposure, the global collective
Finnish NPPs
dose commitment to the population, truncated at 500 years,
Finland has two nuclear power plants both situated on the
arising from normal operation of a nuclear power plant for
Baltic Sea coast: Loviisa on the Gulf of Finland; and Olkilu-
any period of one year, is limited to 5 manSv/GW(e) in-
oto on the Gulf of Bothnia. Two units are in operation at
stalled net electrical capacity (STUK 1992).
both sites (Table 8·29).
Figure 8·54 shows that the annual radiation doses to crit-
ical groups associated with releases from Finnish NPPs are
Table 8·29. Finnish NPPs.
very much lower than the limit of 0.1 mSv/y.
Installed capa- Commercial
Table 8·30 tabulates the radionuclide releases (GBq/y) for
Reactor
city, gross/
start-up
the six radionuclides in liquid effluent contributing most to
Unit (Company)
type/model
net, MW(e)
date
the individual dose, for each of the two NPPs, for the period
Loviisa 1 (IVO)
VVER 213 (PWR)
465 / 445
1977
1990-1994. The corresponding releases in gaseous effluent
Loviisa 2 (IVO)
VVER 213 (PWR)
465 / 445
1981
are presented in Table 8·31 (STUK 1994).
Olkiluoto 1 (TVO)
BWR
735 / 710
1979
Olkiluoto 2 (TVO)
BWR
735 / 710
1982
Table 8·31. Annual airborne releases from Loviisa and Olkiluoto NPPs
(GBq), 1990-1994.
PWR: Pressurized water reactor; BWR: Boiling water reactor.
Operation of nuclear power plants is regulated by the
Releases, GBq
national Nuclear Energy Act of 1987 (NEA 1987) and the
Site/radionuclide
1990
1991
1992
1993
1994
Nuclear Energy Decree of 1988 (NED 1988). The Council
Loviisa
of State has issued General Regulations for the Safety of
Noble gases
1000
1000
1800
1600
1500
Nuclear Power Plants (DCS 1991). The Radiation Act (RA
Tritium
740
480
230
210
210
Carbon-14
310
320
150
190
180
1991) and the Radiation Decree (RD 1991) set forth the
Aerosols
0.2
0.2
0.3
0.08
0.2
general regulations for the limitation of radiation exposure.
Iodines
0.02
0.2
0.03
0.03
0.0002
The Decision of the Council of State stipulates that the limit
Olkiluoto
for the dose commitment to an individual of the population
Noble gases
22000
43000
29000
9500
3500
Tritium
100
130
350
430
240
associated with the normal operation of a nuclear power
Carbon-14
640
640
640
650
470
plant in any one year period is 0.1 mSv. Based on this limit,
Aerosols
0.2
0.7
0.3
0.1
0.1
Iodines
0.06
0.3
0.2
0.08
1.1
Dose µSv/y
6
Loviisa
The annual release limits for liquid effluents from the
Olkiluoto
Loviisa power plant are:
5
Tritium
150 000 GBq/y
4
Other nuclides
890 GBq/y
The release limits for liquid effluents from the Olkiluoto
3
power plant are:
Tritium
18 000 GBq/y
2
Other nuclides
300 GBq/y
1
The group of noble gases has been calculated as 87Kr-equi-
valents. The principal noble gas released from Loviisa is
0
41Ar, and for Olkiluoto xenon isotopes. Among the aerosols,
1975
1980
1985
1990
1995
the main releases from Loviisa comprise 54Mn, 60Co, 76As,
Figure 8·54. Annual radiation doses to critical groups from the Loviisa
110 mAg and 124Sb, and those from Olkiluoto 51Cr, 54Mn,
and Olkiluoto NPP's.
58Co, 60Co and 99mTc.
Table 8·30. Annual releases of the most abundant radionuclides in liquid
Dose mmanSv
effluents from the Loviisa and Olkiluoto power plants in Finland (GBq),
35
Total
1990-1994.
Loviisa
30
Releases, GBq
Olkiluoto
Site/radionuclide
1990
1991
1992
1993
1994
25
Loviisa
Tritium
12000
14000
10000
12000
6600
20
Cobalt-60
2.0
1.3
1.4
0.8
0.102
Silver-110m
3.1
1.6
0.9
0.5
0.020
Antimony-124
7.2
0.1
0.2
0.1
0.202
15
Caesium-134
1.7
0.4
0.2
0.5
0.001
Caesium-137
3.3
1.5
0.5
0.2
0.002
10
Olkiluoto
Tritium
1300
1900
1800
840
2800
5
Chromium-51
1.5
4.3
3.6
2.1
0.202
Manganese-54
9.0
4.6
2.9
2.9
3.102
Cobalt-58
2.5
1.2
0.9
0.3
0.302
0
Cobalt-60
16.0
8.4
7.3
3.6
5.902
1975
1980
1985
1990
1995
Caesium-137
0.6
0.5
0.2
0.04
0.502
Figure 8·55. Collective doses from the Loviisa and Olkiluoto NPP's.
568
AMAP Assessment Report
The annual release limits for gaseous effluents from the
Table 8·33. Annual doses (mSv/y) to critical groups from Swedish
Loviisa power plant are:
NPPs, 1989-1994.
Noble gases (in 87Kr-equivalents) 22 000 000 GBq/y
Annual doses, Sv/y
Iodines 220
GBq/y
NPP
1989
1990
1991
1992
1993
1994
The annual release limits for gaseous effluents from the Ol-
Barseback
1.86
0.29
0.86
0.56
0.15
0.09
(14C) a
(0.92)
(0.92)
(0.96)
(0.30)
(0.32)
(0.6)
kiluoto power plant are:
Forsmark
0.87
0.48
0.50
0.26
0.35
0.10
(14C) a
(0.57)
(0.57)
(0.62)
(0.28)
(0.29)
(0.4)
Noble gases 18 000 000 GBq/y
Oskarshamn
2.90
2.24
1.72
0.90
0.75
0.44
Iodines
110 GBq/y
(14C) a
(0.45)
(0.45)
(0.49)
(0.17)
(0.15)
(0.2)
Ringhals
1.59
2.38
1.39
3.4
19
36
Collective doses calculated on the basis of measured release
(14C) a
(11.0)
(11.0)
(11.4)
(8.5)
(8.6)
(6.7)
rates, excluding 14C, from Loviisa and Olkiluoto power
a. Calculated 14C contribution.
plants are depicted in Figure 8·55.
Dose estimates for both individual and collective expo-
The purpose of a reference value expressed as collective
sures do not include 14C because 14C releases are estimated
dose per GW(e) is partly to limit future individual doses from
from energy output figures and only occasionally checked by
several power plants, all of which may contribute to dose to
measurement. However, since 1992, the 14C release in gaseous
the same individual, and partly to determine the total risk
effluents from the Loviisa NPP has been measured. The an-
associated with releases. As can be calculated from the val-
nual effective dose equivalent to the most exposed individual
ues given in Table 8·34 and the installed electrical capacities
is estimated to be of the order of a few µSv. The estimated
of the power plants in Table 8·32, the reference value for
global collective doses from annual 14C releases are 4 manSv
collective dose has been exceeded several times. However, it
for Loviisa and 7 manSv for Olkiluoto (assuming 1010 global
should be noted that the assumed extent of investment in
population and 500 years integration time) (STUK 1991).
nuclear power used when deriving the reference value was
much greater than that actually constructed. Accordingly,
Swedish NPPs
there is assurance that the highest future dose to individuals
Sweden has 12 nuclear power units at four sites. Table 8·32
will not exceed 0.1 mSv/y which was the main purpose of
portrays the types, electrical installed capacity and the year
placing limits on collective dose per unit installed capacity.
of commercial start-up.
Table 8·34. Collective doses (mmanSv) associated with releases from Swedish
NPPs, 1989-1994.
Table 8·32. Swedish NPPs.
Collective dose, mmanSv
Installed capacity,
Year of commer-
NPP
1989
1990
1991
1992
1993
1994
Plant and unit
Type
gross / net, MW(e)
cial start-up
Barsebäck
90
10
30
10
3
0.24
Barsebäck 1
BWR
615 / 600
1975
(14C) a
(7100)
(7100)
(7400)
(4000)
(4300)
(5700)
Barsebäck 2
BWR
615 / 600
1977
Forsmark
140
20
20
10
10
6.1
Forsmark 1
BWR
1006 / 968
1981
(14C) a
(18000)
(18000)
(19300) (14900) (15400) (17600)
Forsmark 2
BWR
1006 / 969
1981
Oskarshamn
40
30
20
20
17
13
Forsmark 3
BWR
1197 / 1155
1985
(14C) a
(13000)
(13000)
(13800)
(8300)
(7500)
(9200)
Oskarshamn 1
BWR
462 / 442
1972
Oskarshamn 2
BWR
630 / 605
1974
Ringhals
10
10
10
50
293
510
(14C) a
(9800)
(9800)
(10600)
(7000)
(6800)
(8500)
Oskarshamn 3
BWR
1205 / 1160
1985
Ringhals 1
BWR
825 / 795
1973
a. Calculated 14C contribution.
Ringhals 2
PWR
905 / 875
1974
Ringhals 3
PWR
960 / 915
1980
Ringhals 4
PWR
960 / 915
1982
As in the case of Finnish NPPs, the radionuclides in liquid
PWR: Pressurized water reactor; BWR: Boiling water reactor.
effluents contributing most to critical group dose are 3H,
51Cr, 54Mn, 58Co, 60Co, 134Cs, 137Cs, 65Zn and 124Sb. The or-
Based on the national Radiation Protection Act (RPA
ders of magnitude of corresponding releases are GBq/y ex-
1988) and Ordinance (RPO 1988) on radiation protection,
cept for tritium which is of the order of TBq/y. Airborne re-
the Swedish Radiation Protection Institute has stipulated
leases comprise mainly krypton and xenon isotopes and
reference values for release limits in terms of radiation dose
these occur at rates in the order of TBq/y.
equivalent to the critical group and global collective dose
It is noteworthy that airborne effluent releases of 131I
per GW(e) installed net electrical capacity. The values are
from the Leningrad NPP are reported to vary between 20
0.1 mSv/y and 5 manSv/GW(e) respectively for one year's
and 89 GBq/y. This is a substantial release rate compared
releases using a truncation time for the collective dose of
with the Finnish effluent releases presented in Table 8·30,
500 years. Reference releases were calculated on the basis
but, nevertheless, lower than the Finnish release rate limits.
of a dose of 0.1 mSv/y to the critical group taking into ac-
Airborne releases of 131I from Swedish NPPs have varied
count all radionuclides and all pathways of exposure. In
between 0.006 and 38 GBq/y during the last decade.
addition, the ALARA principle is applicable. There are de-
Overall, routine discharges from nuclear power plants are
tailed requirements for reporting of situations in which the
small and contribute little to the contamination of the Arctic
reference releases are expected to be exceeded and the upper
environment or to the doses of Arctic residents.
limits for operational releases are defined on the basis of
ICRP Dose Limits for members of the public (SSI FS 1991).
8.5.2.2. Russian civilian nuclear fleet
Table 8·33 depicts doses to individuals in critical groups
based on measured releases. Carbon-14 is listed separately
Russia currently has seven nuclear-powered civilian vessels
as the values for exposures to this nuclide are based on theo-
in operation of which six are nuclear-powered icebreakers
retical calculations and test measurements. Table 8·34 de-
and one is a nuclear-powered container ship. These are:
picts corresponding collective doses (SSI 1994).
· Four icebreakers with two reactors each of capacity 171
Table 8·33 clearly shows that doses to critical group mem-
MW(t) (Arctica, Rossiya, Sovietskiy Soyuz and Yamal).
bers are lower than the reference dose value of 0.1 mSv/y.
Chapter 8 · Radioactivity
569
· Two icebreakers with one reactor each of capacity 171
disposal and storage vessel decommissioning is currently
MW(t) (Taimyr and Vaigach).
being addressed by French, British and Russian experts
· A lighter-container carrier Sevmorput with one reactor of
(Filippov 1996).
135 MW(t) capacity.
In general, the operations and equipment of civilian
atomic fleet vessels accord with relevant international and
The vessels, all operated by the Murmansk Shipping Com-
national regulations. This should ensure appropriate levels
pany (MSC), operate from their base Atomflot located 2 km
of safety in the processes of handling and storage of nuclear
north of the town of Murmansk on the Kola Peninsula. The
wastes, including activities on the storage vessels. During the
nuclear-powered icebreakers have been constructed since
many years of operation of auxiliary ships, there has not
the 1950s for facilitating increased shipping along the north-
been a recorded case of an emergency having negative effects
ern coast of Siberia. They have also been used for scientific
on the environment.
expeditions in the Arctic and, since 1989, the carriage of
The nuclear fleet vessels are currently stationed, repaired
tourists to the North Pole (Filippov 1996).
and maintained, along with reactor refueling, at the techno-
Spent nuclear fuel was unloaded from two old, partially-
logic enterprise Atomflot located in the vicinity of Murmansk.
decommissioned, icebreakers Lenin and Sibir. These vessels
Radionuclides can be released to the environment both from
are anchored and are manned to maintain safety. During the
ships and the Atomflot base during the following operations:
early operational period of the first nuclear-powered ice-
breaker Lenin there were some reactor accidents. As a result
· Refueling.
of one serious accident, the reactor unit with three reactors
· Decontamination and repair.
was cut out and dumped in 1967 in the Kara Sea near No-
· Handling radioactive waste including transportation
vaya Zemlya. One of the reactors contained part of the
and processing.
damaged spent fuel with an estimated activity at the time of
· Storage of spent nuclear fuel (SNF) on ships.
disposal of about 3.7 PBq (OPRF 1993). Potential radiolo-
The main environmental concern relates to releases of radio-
gical dangers associated with this operation are considered
active wastes generated as a result of the operation of nu-
in section 8.6.
clear-powered ships.
The Murmansk Shipping Company also operates auxil-
The gas-aerosol releases on nuclear icebreakers comprise
iary ships for the collection and storage of spent fuel and ra-
mostly radioactive inert gases and vary in the range 40-400
dioactive waste:
GBq/y, which is two to three orders of magnitude less than
· Two floating technical bases for reloading fuel assemblies
the actual releases of these radionuclides from land-based
and the storage of spent fuel (with 1530 assemblies on
nuclear power plants. During storage of SNF on the service
the Imandra and 4080 assemblies on the Lotta).
ship Imandra, the gas release consisted primarily of 85Kr.
· A floating technical base Volodorasky for the storage of
During the storage period, not more than 400 GBq is re-
solid radioactive waste of 300 m3 volume.
leased from a spent core, over 90% of it during the first
· A special tanker Serebryanka for the collection and stor-
three months of storage. The air concentration of radionu-
age of liquid radioactive waste of 1000 m3 volume.
clides outside the reactors, nuclear ships and service ship
· Vessels for sanitary treatment of personnel and dosimetric
Imandra during gas releases in all refueling and reactor re-
control (e.g., Rosta 1).
pair operations does not exceed the derived air concentra-
Four auxiliary ships were out of service and two of them
tions prescribed by the Russian norms for radiation safety.
need comprehensive remediation:
When burning solid wastes, a mixture of radionuclides is
released to the atmosphere containing 137Cs, 60Co, 90Sr, 54Mn,
· A floating technical base/storage facility for spent nuclear
152Eu and 154Eu. The actual releases of these radionuclides
fuel Lepse.
amount to 3
106 to 1
102 percent of the permissible level.
· A floating dosimetric control unit (FDCU-5).
In total, during four years of operation of a single reactor
A total of about 6000 spent fuel assemblies from the Mur-
core, 130 m3 of liquid radioactive waste are generated, of
mansk Shipping Company are stored on the floating bases
which 100 m3 arises on the nuclear vessel and 30 m3 on a
Lepse, Imandra and Lotta. Auxiliary ships of the atomic
floating refueling base. The total annual activity of the liquid
fleet are ice-class vessels with double hulls. The atomic ice-
waste amounts to 4-300 GBq with an average of 20 GBq.
breakers and floating bases were designed to withstand col-
Since 1989, a pilot unit for the decontamination of liquid ra-
lisions with other ships and to remain afloat in the case of
dioactive wastes of 1200 m3/y capacity has been operating
flooding of two adjacent compartments. The tanks contain-
on RTP Atomflot. Following decontamination to permissible
ing spent nuclear fuel have biological shielding made of steel
concentrations and analysis of the remaining artificial radio-
of thickness 380-450 mm. Prevention of criticality is achieved
nuclide content, the waste liquid is discharged to Kola Bay.
through the use of a specific geometric arrangement of spent
Altogether, during the operational life of one core, about
fuel assemblies. The storage tanks have special water-cooling
32 m3 of solid radioactive wastes of total activity 2 TBq are
systems independent of those of the vessels. Special tanks for
generated. Most of the solid radioactive wastes ( 90%) are
liquid radioactive wastes are made of stainless steel and are
low-level, of which 50-70% is combustible. Solid radioac-
placed in the vessel's hull. The tanks are provided with cor-
tive wastes from ships are passed to a special complex of
rosion and biological protection (Filippov 1996).
RTP Atomflot for compaction and storage. Since 1989, in-
Spent nuclear fuel is offloaded from ship reactors and
cineration of combustible solid radioactive wastes has been
stored on the Imandra for 0.5-1 year, and then transferred
performed in a burning unit having a capacity of 40 kg/h.
to the Lotta where it is stored for a further period of about
The actual levels of radionuclides in air samples over
three years. Subsequently, spent fuel is transported by rail to
water areas adjacent to RTP Atomflot from 1985-1994 are
the reprocessing plant Mayak in the Urals.
below control levels. Increased levels detected in air and
The storage facility on the floating base Lepse is in a poor
rainfall in May 1986, were due to the Chernobyl accident.
state. It contains 28 PBq in 642 damaged fuel assemblies
Environmental monitoring during 1992-94, when five re-
within two tanks. Long-lived transuranic radionuclides com-
fueling operations were carried out, indicates that inputs
prise 0.6 PBq of this total. The problems of damaged fuel
from refueling constitute 3-32% of the annual release limits
570
AMAP Assessment Report
for specific monitoring points. Wholebody measurements
during the period of operation of naval nuclear-powered
on working personnel show that the levels of 134Cs, 137Cs
ships. These cases are summarized in section 8.5.3.5.
and 95Zr are about 0.01% of permissible levels.
In 1993-94, studies of bottom sediments in Kola Bay in
8.5.2.3.2. Decommissioning
the vicinity of RTP Atomflot were conducted. Gamma-spec-
trometry revealed the presence of 137Cs, 60Co, 152Eu and
As a result of disarmament and other technological reasons,
154Eu. However, the measured radionuclide concentrations
nuclear submarines of the Northern Fleet are being decom-
do not provide reason for undue concern about the impact
missioned. By the beginning of 1996, the number of sub-
of discharges from this enterprise on the environment (Kolo-
marines taken out of service was about 90 and is likely to
miets et al. 1992, Filippov 1996).
exceed 100 by the year 2000. The procedure for decommis-
sioning of submarines involves unloading of SNF from reac-
tors on the ships, after which decontamination is performed,
8.5.2.3. The Russian Northern Fleet
equipment is removed for further use, and the reactor com-
8.5.2.3.1. Nuclear-powered vessel operations
partment is cut out and placed in a prepared and environ-
Since the 1950s, nuclear-powered submarines and surface
mentally-safe storage or disposal site. However, at present,
ships have been operating in the Northern Fleet of the USSR,
practices do not fully conform to this scheme due to the lack
now that of the Russian Federation. The Russian Northern
of technical infrastructure and resources. The most urgent
Fleet currently includes several tens of nuclear-powered sub-
task is unloading of SNF from reactors. By the beginning of
marines and two nuclear-powered cruisers, operating mainly
1996, SNF was unloaded from only 25% of the submarines
from nine major bases on the Kola Peninsula extending from
taken out of service in the Northern Fleet. In all, six reactor
Gremikha in the east to the Litsa Fjord in the west. Most of
compartments have been prepared for long-term storage.
these vessels are equipped with two nuclear reactors of pres-
Other submarines that have been decommissioned, but still
surized, light water moderated design with capacity of 70-
contain nuclear fuel, are now moored in harbors on the
300 MW(t) (CCMS/CDSM/NATO 1995).
Kola Peninsula awaiting completion of decommissioning
Radiation monitoring during the period 1968-1992 has
(Petrov 1995, Ecological Safety 1996).
demonstrated that the main sources of radioactive contami-
According to one scenario, long-term storage of the un-
nation of base locations of the Northern Fleet were ships
loaded reactor compartments is planned to take place in
with nuclear reactors, depot ships, coastal technical bases
large man-made rock cavities constructed on the shores of
and supply ships carrying liquid radioactive wastes. The
the Kola Peninsula about two decades ago. Unloading of
main sources of radioactive waste generation in the North-
SNF from submarines with damaged cores remains an un-
ern Fleet are: bases at Zapadnaya Litsa Inlet, Olenya Bay,
solved problem as there are currently inadequate facilities
Saida Bay, Ara Bay, Pala Bay, and Iokan'ga; sites of nuclear-
for removal of fuel rods from damaged reactors. Further-
powered military vessel deployment (i.e., Polarny and Seve-
more, in the next few years, those facilities that are coming
rodvinsk); sites of interim storage of SNF (i.e., Zapadnaya
to the end of their planned operational lifetime are due to be
Litsa and Iokan'ga); floating naval bases for refueling of sub-
scrapped, in particular the tenders used for refueling and
marines; shipyards; and the repair plants at Polyarny, Vyu-
spent nuclear fuel storage. Some of these do not meet con-
zhny, Severodvinsk. The maximum radioactive releases have
temporary radiation safety requirements.
occurred in liquid form through disposal of decontamina-
tion water, leakages of liquid radioactive wastes during the
8.5.2.3.3. Storage of the spent nuclear fuel and
process of transfer from ships to storage bases and through
radioactive waste
unauthorized releases. Radioactive contamination of the en-
vironment (water and air) in such cases is generally of local
In recent years, the problem of spent nuclear fuel and radio-
scale and decays to pre-existing levels relatively quickly (i.e.,
active wastes disposal has become more acute for the Rus-
within a few hours) (OPRF 1993, Lisovsky et al. 1996).
sian Northern Fleet. Transport of spent nuclear fuel to re-
Numerous observations have shown that the radiation
processing facilities in the Urals has been slowed and the
situation in nuclear-powered ship bases of the Northern
Russian Navy lacks the necessary processing plants. Further-
Fleet is only abnormal in the cases of accidents. The concen-
more, since 1992, radioactive wastes are no longer dumped
trations of radionuclides in the environment are normally no
at sea. All this has led to a build-up of hazardous materials
higher than background levels (see Table 8·35) (Lisovsky et
at locations where the nuclear-powered ships are based and
al. 1996). Only in exceptional cases, are traces of radionu-
overhauled, with associated negative effects on the radiolo-
clides derived from reactors detected in places where moni-
gical and environmental conditions. Also, it is difficult to en-
toring of reactor servicing activities and radioactive waste
sure radiation and environmental safety because of the diffi-
storage facilities is conducted. There have been several cases
culties associated with implementation of the program for
of accidental radioactive contamination of the environment
the decommissioning of submarines (OPRF 1993, Petrov
1995).
Table 8·35. Average radionuclide content in the environment of base
At present, more than 3000 cells with spent fuel assem-
locations of the Russian North Fleet in 1987 (Lisovsky et al. 1996).
blies are stored at facilities of the Northern Fleet. As there
Concentration of radionuclides
are seven assemblies in each cell, the total number of assem-
Medium
Unit
90Sr
137Cs
144Ce
60Co
210Po
blies is over 21 000. The majority of these spent fuel assem-
blies are in storage on the shore of Andreev Bay (Fjord of
Seawater
Bq/L
9 103 9 103 9 103 5 103 2 103
Zapadnaya Litsa). Because of accidents at previous `wet'
Drinking water
Bq/L
6 103 1 102 6 103 7 103 9 104
Atmospheric aerosols Bq/m3 2 105 3 105 2 105 7 105 1 104
spent nuclear fuel storage facilities located in the same area
Atmosph. depositions Bq/m2 1 108
1 108
7 107
1 107
in the 1980s, spent nuclear fuel is now kept `dry' in contain-
Sea algae
Bq/kg
4
3
4
1.5
0.4
ers intended primarily for liquid radioactive wastes. Tempo-
Sea bottom sediments Bq/kg
10
11
8
11
4
Benthic sea organisms Bq/kg
2
4
2.5
0.7
2
rary storage facilities were built on a crash basis and do not
Soil
Bq/kg
10
7
3
4
9
fully comply with radiation safety and environmental pro-
Surface vegetation
Bq/kg
6
4
4
3
2
tection requirements. Spent nuclear fuel is stored in coastal
Chapter 8 · Radioactivity
571
facilities in Andreev Bay (80%), Iokan'ga and on tenders
tive waste storage and processing have not yet been com-
(5%). Part of the spent fuel is situated at the shipyard Atom-
pleted and/or implemented because of limited financial re-
flot of the Murmansk Shipping Company because naval
sources (Petrov 1995, NEFCO 1996).
facilities for temporary storage of spent nuclear fuel have
The chemical and radionuclide composition of the radio-
been exhausted (OPRF 1993, Petrov 1995).
active wastes is complex and changes with time. It depends
The technical support system for nuclear reactors on
upon design features of the nuclear propulsion units, leakage
ships that was created in the late 1950s and early 1960s was
rates from fuel rods and the age of the wastes. Isotopes with-
not designed for decommissioning of a large number of nu-
in the wastes include both fission and activation products,
clear submarines in a short period. This is the primary rea-
mainly 54Mn, 60Co, 90Sr, and 137Cs (RCRA 1997).
son for the unsatisfactory situation regarding SNF and ra-
A Russian Federal Program, approved by the government
dioactive wastes in the Russian Navy. There are several fac-
in 1995, will significantly improve the handling and disposal
tors aggravating the problem: the unloading of spent nuclear
of radioactive waste and spent nuclear fuel including decom-
fuel cannot keep pace with the decommissioning of nuclear
missioning of naval nuclear reactors between 1996 and 2005.
submarines; some storage facilities are in a dangerous con-
However, measures are being implemented slowly because of
dition; and there are no available storage containers.
contemporary economic problems in Russia (Ecological Safe-
At present, there are two options for the transport of
ty 1996 ).
spent nuclear fuel from the Northern Fleet, in newly-de-
signed TUK-18 transport flasks with improved safety fea-
8.5.2.3.4. Shipyards
tures, from Severodvinsk and from Atomflot. This, however,
does not resolve the problem. To enable transport of SNF
The largest Russian shipyard Sevmash is located in Severod-
from Andreev Bay, 30 km of railway line needs to be con-
vinsk on the coast of the White Sea. The ship repair plant
structed and/or a special transport ship built. The spent nu-
Zvezdochka is located in the same area. Nuclear submarines
clear fuel on submarines which suffered accidents is still an
have been built, serviced, repaired and refitted at these ship-
unresolved issue. A technology is being developed for the
yards for 35 years and they are now also decommissioned at
transport of spent fuel from liquid metal cooled reactors.
these locations. During these operations large amount of
This is planned to be implemented at Iokan'ga in 1998 (Pe-
solid and liquid radioactive wastes are generated which are
trov 1995, NEFCO 1996).
stored in stationary and floating storage facilities. In the ship-
As a result of the accident at a spent nuclear fuel storage
yard, there are six floating hull sections containing reactors
site in Andreev Inlet, water of the cooling ponds continued
following spent fuel removal, 3370 m3 of solid radioactive
to be released into the soil of the adjacent territory and en-
waste in the stationary storage facility and 1250 m3 in an in-
tered a nearby stream for several years. The associated ra-
terim storage site. Some 950 m3 liquid radioactive wastes
dioactive contamination affected an area of 1300 m2 to-
are stored in on-land and floating containers (Koupri 1995).
gether with waters of the bay adjacent to the location where
Also at these shipyards, 15 nuclear submarines with 29
the stream discharge occurs. The storage of spent nuclear
reactors, from which the fuel has been unloaded (more than
fuel on the coast of Andreev Bay continues to be a source of
7000 fuel assemblies), are being repaired. After unloading,
current and potential radioactive contamination of the local
the spent fuel is sent to Mayak for reprocessing. However,
environment (Petrov 1995, Lisovsky et al. 1996).
the rate of spent fuel accumulation is higher than the rate of
Liquid radioactive wastes generated during the operation
reprocessing. As a result, the amount of spent fuel in storage
of nuclear-powered vessels are stored in both coastal and
continues to increase.
floating containers. The volume of all available containers
Monitoring around the facilities in Severodvinsk in the 1990s
for the Northern Fleet is 10 000 m3 but 30% of these are un-
(Table 8·36) showed that there is no cause for any major con-
fit for use. The amount of liquid radioactive wastes from the
cern about environmental contamination (Koupri 1995).
Northern Fleet is estimated to be 7000 m3 containing a total
activity not exceeding 3.7 TBq. About 2000-2500 m3 of liq-
Table 8·36. Radionuclides in environmental media in the vicinity of
Severodvinsk shipyards.
uid wastes are generated annually but there are essentially
no empty containers in which to store them. Liquid wastes
Environmental medium
Radionuclide concentration (range)
of the Northern Fleet are partly transported for processing
Atmospheric aerosols
2 104-11 104 Bq/m3
at Atomflot, Murmansk (Petrov 1995). In 1994-96, the
Atmospheric deposition
2-15 MBq/m2 per month
amount of liquid radioactive wastes processed was 1500 m3.
Seawater in shipyard areas
4-9 Bq/m3
Solid radioactive wastes are stored in storage facilities
Bottom sediments
2-11 Bq/kg dw
and on open temporary storage sites which were largely
built in the 1960s and 1970s and do not fully meet the re-
The progressively increasing accumulation of spent nu-
quirements for environmental protection. High-level wastes
clear fuel and radioactive wastes stored partly in interim
are stored in special facilities only. The amount of solid ra-
storage facilities poses a threat of contamination of both ter-
dioactive wastes stored in facilities of the Northern Fleet is
restrial and marine environments in the case of accidents
estimated to be 8000 m3 with a total activity not exceeding
(Koupri 1995, NEFCO 1996).
37 TBq. On average, 1000 m3 of solid radioactive wastes
Operational releases from these civilian and military ves-
are generated annually. Considering the increasing rate of
sels do not appear to have been documented, but are prob-
submarine decommissioning, the rate of generation of ra-
ably relatively small. Finally, it should be noted that military
dioactive waste may increase by at least a factor of two. The
nuclear-powered vessels from other countries such as USA,
major storage sites for solid radioactive wastes in the North-
France, Canada and UK can transit Arctic waters.
ern Fleet are Andreev Inlet, Iokan'ga and Polyarny. By vol-
ume, 50% of the solid wastes are combustible, 15% are
8.5.2.4. European nuclear fuel reprocessing plants
compressible, 35% are non-compressible and 1% are spent
ion-exchange filter resins. No processing of solid radioactive
While there are no nuclear fuel reprocessing plants in the Arc-
wastes currently takes place by the Navy. Numerous plans
tic, radionuclide releases to the marine environment of west-
and projects to build and reconstruct facilities for radioac-
ern Europe can be transported to the Arctic by ocean currents.
572
AMAP Assessment Report
Table 8·37. Contributions (%) made by European nuclear fuel reproces-
dose reconstruction. This provides a basis for assessing the
sing plants to total discharges within the European Community (Bq), for
degree to which these operations might be of concern rela-
discharges up to the end of 1984.
tive to their effects on the Arctic and its residents. In this dis-
% of total discharge
cussion, only exposures to humans have been considered.
Facility
-emitters
-emitters
Tritium
This is because no evidence exists to suggest that doses to
organisms resulting from nuclear fuel reprocessing activities
Sellafield
95.2
86.9
52.9
in western Europe would be of any concern beyond the vi-
La Hague
0.52
5.3
16.4
Dounreay
1.8
6.7
0.33
cinity (i.e., the receiving area of releases) of particular repro-
Total (%)
97.5
98.9
69.6
cessing operations.
Total EC discharges, Bq
1.4 1015
1.5 1017
4.6 1016
8.5.2.4.1. British nuclear fuels plant at Sellafield, UK
The principal sources of nuclear waste discharges to the
marine environment from western Europe are nuclear fuel
The nuclear fuel reprocessing plant at Sellafield (formerly
reprocessing operations at Sellafield in Cumbria on the west
Windscale) on the eastern side of the Irish Sea is the largest
coast of England, at La Hague near Cherbourg, France, and
nuclear complex in the United Kingdom and is operated by
at Dounreay in the northeast of Scotland (see Table 8·37)
British Nuclear Fuels plc (BNFL). Discharges from this site
(after CEC 1990). These plants are all involved in the recov-
started in 1951 when the facility was initially put into oper-
ery of plutonium from irradiated nuclear fuel for further use
ation by the UK Atomic Energy Authority. The UK govern-
in the electrical generating industry. While there are also re-
ment has been assiduous both in maintaining records of the
processing operations at Marcoule, France, which discharges
basis of its discharge authorizations (e.g., HMSO 1959) and
to the Rhone, and at Karlsruhe in Germany, these are of
providing, for public information, the results of its monitor-
minimal significance in relation to the impact of European
ing associated with civil nuclear site activities through the
reprocessing operations on the North Atlantic, northern Eu-
medium of reports issued by the UK Atomic Energy Author-
rope and the adjacent Arctic. Accordingly, the latter two re-
ity and the Ministry of Agriculture, Fisheries and Food
processing plants have not been considered in this summary.
(MAFF 1967). The first of this latter series of reports reflects
Authorization for releases from such operations are based
the commitment to protecting the most exposed individual
on limiting radiation exposures to the most exposed individ-
from exposures exceeding the ICRP dose limits and the com-
uals for a particular plant. The potentially exposed individu-
mitment to what became known subsequently as the princi-
als of primary concern are those in the vicinity of the plant
ple of optimisation. These same reports also contain results
and the area of its releases to the environment. Thus, expo-
relevant to estimating potential exposures of residents of the
sures to distant individuals, such as residents of the Arctic,
Channel Islands as a result of releases from the La Hague re-
are not directly relevant to the authorization process for Eu-
processing plant.
ropean nuclear fuel reprocessing operations.
Until 1954, when formal authorisation of such releases
Although collective doses to larger populations have not
occurred under the Atomic Energy Authority Act (UK), dis-
been used in the context of authorizing nuclear fuel repro-
charges took place under controls implemented by the site
cessing plant operations and releases, collective dose rates
operators in consultation with the Government Departments
and commitments associated with such plants within west-
(primarily the Ministry of Housing and Local Government).
ern Europe are useful for assessing the large-scale radiolo-
Operation of the facility was largely transferred from the
gical impact of such activities. Thus, collective dose esti-
UK Atomic Energy Authority to British Nuclear Fuels in
mates provide a useful framework for considering the ex-
April, 1971. Throughout, authorizations have been based on
tent of radiological consequences and associated risks to
limiting exposures to individuals in the most exposed group
Arctic residents resulting from European nuclear fuel re-
to the prevailing values recommended by the ICRP (Wix et
processing operations.
al. 1960, Hunt 1995). Discharges from the site have fluctu-
It is worth noting, in an Arctic context, that the radio-
ated considerably over time with a maximum for most radio-
nuclide compositions of the discharges from European re-
nuclides occurring in the mid- to late 1970s as a consequence
processing plants have altered significantly during their per-
of changes in practice and the introduction of additional
iod of operation. Although Sellafield remains the primary
measures to reduce discharges introduced predominantly in
contributor to activity releases among the three plants, the
the mid-1970s (Gray et al. 1995, Kershaw and Baxter 1993a,
dominant sources of individual radionuclides have, in some
1993b). The nuclides representing the major proportion of
cases, altered dramatically. This is particularly well illu-
the total releases from Sellafield (ca. 130 PBq excluding 3H
strated by the relative contributions made to releases of the
up to 1986) are the beta-/gamma-emitters 137Cs (30%), 106Ru
long-lived radionuclide 129I by Sellafield and La Hague.
(21%), 241Pu (16%) and 95Zr/95Nb (18%) (CEC 1990); CEC
While the aggregate discharge of 129I remained relatively
1990 reported the 95Zr/95Nb value as 41%, but this number
constant from 1975-1990, and remains so for Sellafield, the
appears inconsistent with the record of discharges from Sel-
increased fuel reprocessing throughput at La Hague has re-
lafield, cf. Figure 8·56. Total alpha-emitter releases in the
sulted in substantially increased discharges of this radionu-
same period were about 1% of the total and are dominated
clide making La Hague currently the dominant contributor
by 239Pu (0.0053%) and 241Am (0.0041%) (CEC 1990). The
(90%) to the aggregate 129I discharge (Yiou et al. 1995).
chronology of releases, both to sea and to the atmosphere,
From a scientific perspective, this is an important observa-
from the Sellafield reprocessing operation has been most re-
tion because of the relatively conservative nature of iodine in
cently and comprehensively documented by Gray et al.
seawater and the important contribution of this isotope to
(1995). Figures 8·56 and 8·57 show the rates of different liq-
Arctic waters through transport into and through the Nor-
uid discharges from 1952-1992, reproduced from Gray et al.
wegian Coastal Current into the Arctic making it, in combi-
(1995).
nation with other isotopes, a valuable oceanographic tracer.
Environmental measurements and modeling have been
This section deals with the history of releases from Euro-
used to confirm the chronological record of liquid discharges
pean nuclear fuel reprocessing plants, the basis under which
from Sellafield with emphasis given to measurements in com-
such releases are authorized, and individual and collective
ponents of critical pathways of exposure (Gray et al. 1995).
Chapter 8 · Radioactivity
573
TBq/y
characterization of the local critical group, the rates of con-
6 000
106
sumption of seafood and changes to the dosimetric data for
Ru
actinides have given rise to changes in the estimates of criti-
241Pu
cal group exposures. Nevertheless, it is clear that the recent
5 000
90Sr
doses to individuals within the critical group arising from
137Cs
liquid releases from Sellafield are relatively small (of the or-
4 000
der of 200 µSv/y) and well below the currently recommended
ICRP individual dose limit of 1 µSv/y and the constraint for
3 000
doses arising from this category of source. Furthermore, as
all of the most exposed individuals are in population group-
ings situated in the general vicinity of the site, individual
2 000
doses to more distant members of the public, such as resi-
dents of the Arctic, will be much lower.
1 000
8.5.2.4.2. La Hague, France
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
The reprocessing facility at Cap de la Hague was brought
Figure 8·56. Discharges of principle beta-gamma emitters in liquid effluent
into operation in 1965. Major modifications have been made
from Sellafield, 1952-1992 (after Gray et al. 1995).
to process additional types and amounts of spent fuel on
two subsequent occasions. The total discharges from this
TBq/y
site were, excluding 3H, about 8 PBq up to 1986; much less
120
241
than those from Sellafield. Releases of individual radionu-
Am
238
clides up to 1982 were given by Calmet and Guegueniat
Pu
100
239Pu
(1985) and these were cited and extended to 1986 in the
`Project Marina' report (CEC 1990). In contrast to the con-
80
tributions made to activity discharges from Sellafield by in-
dividual radionuclides, in which 137Cs has been the domi-
60
nant contributor up to the mid-1980s, for La Hague the
dominant contributor to activity releases (55%) is the beta-
40
emitter 106Ru followed by 90Sr rather than 137Cs.
As in the case of Sellafield, authorization of discharges
20
from La Hague is based on limiting the doses to individuals
within potentially critically exposed groups to the dose lim-
0
its for members of the public recommended by the ICRP. A
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
discussion of the identities and habits of the critical groups
Figure 8·57. Discharges of principle alpha emitters in liquid effluent from
for doses arising from discharges from La Hague is given in
Sellafield, 1952-1992 (after Gray et al. 1995).
Calmet and Guegueniat (1985), together with calculated
proportions of the dose limit for fishermen via consumption
Initial assessments showed that the pathways likely to give
and external exposure pathways. It is difficult to assess,
rise to the largest exposures were internal exposures associ-
from this reference, the quantitative importance of these
ated with the consumption of fish and seaweed and external
pathways of individual exposure. However, if it is assumed
exposures associated with shoreline occupation. Figure 8·58
that, consistent with international recommendations at the
presents a comparison of doses to individuals within critical
time, the individual dose limit for members of the public
groups over the period of releases from Sellafield up to 1993
used in France was 5 mSv/y, the dose to individuals within
after Hunt (1995). It should be appreciated that the applica-
the critical group resulting from sea discharges was of the
tion of, and values for, the dose limits applicable to the au-
order of 0.2 mSv/y. Critical group doses associated with liq-
thorization of practices, as recommended by the ICRP, have
uid releases from La Hague were reassessed for the period
changed during the period concerned. Also, changes in the
1982-1986 in `Project Marina' (CEC 1990). This reassess-
ment yields individual dose rates of 0.03-0.04 mSv/y for the
Committed effective dose rate µSv/y
seafood consumption pathway and 0.06-0.27 mSv/y for ex-
2 000
ternal exposures over the five-year period.
1 500
8.5.2.4.3. Dounreay, UK
Releases from Dounreay are much smaller than those from
1 000
Sellafield. The aggregate activity discharged up to 1986 was
about 10 PBq with 95Zr/95Nb representing 55% of this ac-
tivity; 144Ce, 17%; and 106Ru, 10% (CEC 1990). Discharges
500
of beta-activity was at its highest during the 1960s and early
1970s with small peaks in 1968 and 1973 resulting from
plant washout and decontamination procedures. In the early
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
years of operation of the Dounreay reprocessing plant, some
of the irradiated fuel, which was derived from the Dounreay
Dose to laverbread consumers
Dose to fish/shellfish consumers
Fast Reactor, was of higher specific activity than that proces-
External exposures
Measured exposure to houseboat dwellers
sed by any similar plant. In 1980, fuel from the Prototype
Fast Reactor began to be reprocessed, and this gave rise to
Figure 8·58. Comparison of doses to critical groups for Sellafield dis-
charges from 1952 to 1993 (after Hunt 1995).
increased discharges of alpha-emitters, mainly 238Pu, 239Pu
574
AMAP Assessment Report
and 241Am. The critical groups for individual exposures
Integrated collective dose manSv
from Dounreay discharges are fishermen handling fixed
103
salmon nets contaminated by sludge, where dose to hands
from beta-emitters is the primary route of exposure, al-
though attention was also given to seafood consumption by
102
fishermen and their families, crustacean consumption by
workers in a local canning factory and, potentially, occa-
sional occupants of inlets who might receive external expo-
101
sures from contaminated sediments (Freke et al. 1969).
100
8.5.2.4.4. Dose reconstruction for releases from
Western European reprocessing plants
10-1
For the purposes of uniform and consistent comparison, the
predominant source used for dose reconstruction informa-
tion has been `Project Marina' (CEC 1990). This latter re-
10-2
view deals with all sources of radionuclides that give rise to
exposures to the European Community population and this
further justifies the use of this source for broader consistency
10-3
in assessing both individual and collective doses within a
larger context. `Project Marina' concluded that `the quality
and detail of environmental monitoring data from nuclear
10-4
site operators, governments and other sources were more
1950
1960
1970
1980
1990
2000
than adequate for assessing critical group doses.' In this
context, discharges from reprocessing plants in general, and
Total
Sellafield
La Hague
Sellafield in particular, gave rise to the highest critical group
Dounreay
Winfrith
doses arising from nuclear site discharges. During the period
EC integrated collective dose by discharge site
1977-1986, critical group doses, albeit delivered to only a
Figure 8·59. Collective dose commitment from nuclear fuel reprocessing
few individuals consuming above average amounts of sea-
activities (after CEC 1990).
food harvested near to the discharge point, were up to 3.5
mSv/y. Nevertheless, ICRP dose limits have been met as ex-
(peak in the late 1980s), 40 manSv/y (peak in the mid-1980s)
posures were generally less than 1 mSv/y and exposures ex-
and 7 manSv/y (peak at the beginning of the 1970s), respec-
ceeding this value did not occur for long enough for lifetime
tively. Collective dose rates have declined substantially since
exposure to have exceeded 1 mSv/y on average. Discharges
then and can be projected (based on aggregate discharges up
from Sellafield have declined since the mid-1970s and this
to the end of 1986) to be of the order of 20 manSv/y, 0.6
has led to a decline in critical group doses. In 1986, critical
manSv/y and 0.3 manSv/y, respectively, in the year 2000
group exposures from Sellafield discharges were about 0.3
mSv. The corresponding doses from the other two European
Collective dose rate
reprocessing plants at La Hague and Dounreay have been
manSv/y
substantially less than those arising from Sellafield. The indi-
103
vidual effective dose equivalents for La Hague are less than
0.3 mSv/y, and for Dounreay less than 0.05 mSv/y. It is
102
again stressed that these exposures are to members of criti-
cal groups resident in the immediate vicinity of the respec-
tive reprocessing plant and not to Arctic residents.
101
Collective dose estimation allows an improved basis to
assess the consequences of nuclear fuel reprocessing opera-
tions on a larger scale and in an appropriate context to the
100
doses arising from other human activities. The collective
dose commitment to the population of the European Com-
munity, truncated in the year 2500, resulting from total dis-
10-1
charges up to the end of 1984 from nuclear fuel reprocessing
operations in western Europe is estimated to be 5150 manSv
with the dominant contribution (90% or 4600 manSv) from
10-2
Sellafield and progressively smaller contributions from La
Hague (8% or 430 manSv) and Dounreay (2% or 120 manSv)
(CEC 1990) (Figure 8·59). The predominant radionuclide
10-3
contribution to the collective exposure is 137Cs. About 80%
of the collective dose commitment to the European Commu-
10-4
nity population had already been delivered by the mid-1980s.
It is also worth noting that the aggregate collective dose
1950
1960
1970
1980
1990
2000
from European reprocessing operations essentially accounts
Total
Sellafield
La Hague
for all the collective dose commitment from civilian nuclear
activities in the European Community that is estimated, on
Dounreay
Winfrith
the same basis, to be 5300 manSv.
EC collective dose rate by discharge site
The peak collective dose rates for Sellafield, La Hague
Figure 8·60. Collective dose rates from European nuclear fuel reprocessing
and Dounreay discharges were of the order of 300 manSv/y
plants based on releases up to 1986 (after CEC 1990).
Chapter 8 · Radioactivity
575
(Figure 8·60) (CEC 1990). Recognising that the major ma-
evacuated in 1953-1960, inhabitants of its upper part re-
rine contributor to collective dose for the European Com-
ceived average doses to the wholebody of over 1 Gy during
munity population arises for nuclear fuel reprocessing activ-
25 years, and some individuals received extreme doses of
ities, particularly at Sellafield, and that the seafood con-
more than 2 Gy. Specific features of the exposure of inhabi-
sumption pathway is the dominant exposure pathway for
tants of the River Techa are the enhanced doses to red bone
the general population, it is worth noting that the collective
marrow and bone surface that exceed the wholebody dose
dose rate to the non-EC population arising from the con-
by a factor of 1.5-15. High levels of exposure of bone mar-
sumption of seafood caught in EC waters is about an order
row caused tens of cases of chronic radiation sickness in in-
of magnitude below that to the EC population (CEC 1990).
habitants of the upper Techa River in the 1950s and a high-
Given the European Community population of 3.2
108
er incidence of leukaemia over longer time periods (Degteva
at the time of `Project Marina' (CEC 1990), the contempo-
et al. 1992, Kozheurov et al. 1994, UNSCEAR 1993).
rary per capita representation of the total collective dose
rate from nuclear fuel reprocessing operations is approxi-
mately 1.1 µSv/y. Although this could result in approximate-
ly 15 additional cancer deaths within the European Commu-
Krasnoyarsk
nity per year of such exposure, such deaths would not be
Kasli
Tomsk
epidemiologically detected against the normal death rate in
Mayak
the Community. Furthermore, the associated incremental
risk of 5
108 per year of serious health defect (i.e., fatal
Ozyorsk
cancer induction) to an average individual in the European
Community per year of practice is relatively trivial (IAEA
R-3
R-2
1993). The representation of the collective dose rate arising
R-4
R-9 R-17
R-10
from European nuclear fuel reprocessing operations on a
Mayak
Left bank
site
Right bank
canal
per capita basis for the entire global population yields an in-
canal
R-11
As
cremental individual risk of fatality of about 7
109 per
anov
Muslyumovo
swamps
year (IAEA 1993). This may be a reasonable representation
of the average risk to members of the Arctic community,
spread as they are throughout northern Europe, Asia and
Techa
North America, even allowing for the contributions from
other, non-European, nuclear fuel reprocessing waste releases.
Figure 8·61. The system of dams and drainage channels at Mayak.
Between 1951 and 1966, a system of dams was constructed
8.5.2.5. Russian nuclear fuel reprocessing plants
along the upper part of Techa River, creating several artificial
In Russia, fuel reprocessing takes place at the Mayak Pro-
water reservoirs along the old river bed to retain most of the
duction Association, Chelyabinsk; and at plants in Krasno-
radioactivity (Malyshev 1995). The river water, originating
yarsk and Tomsk. Although these plants are not situated
from Lake Irtysh, is today led outside the reservoirs via a
in the Arctic, radioactive releases from Mayak and Tomsk
canal on the northern side of the dam system (left bank can-
enter the drainage area of the Ob River and those from
al). Similarly, the water of the Mishelyak River flows in a
Krasnoyarsk directly into the Yenisey River. Releases to
right bank canal on the southern side. The two canals join
these rivers can be transported downstream, ultimately into
the Techa River about 3 km downstream of dam no. 11, as
the Kara Sea.
depicted in Figure 8·61.
It is estimated that the accumulated activity in reservoirs
nos 3, 4, 10 and 11 amounts to about 9.8 PBq (0.26 MCi),
8.5.2.5.1. Mayak
mainly in the sediments (see Table 8·38). The activity con-
The nuclear enterprise Mayak Production Association (Ma-
tained in the several reservoirs, including Lake Karachay
yak PA) is located outside the town Ozyorsk (until recently
(reservoir no. 9), are presented in Table 8·38 (Mayak 1993).
called Chelyabinsk-65) between the cities of Yekaterinburg
In the floodplain of the upper Techa River (the Asanov
and Chelyabinsk, just east of the Ural mountains. Mayak PA
Swamp) about 37 TBq (1000 Ci) of 90Sr and 185 TBq (5000
was the first plant in the former Soviet Union established for
Ci) of 137Cs is retained in the surface soil (Malyshev 1995,
the production of nuclear weapons material and began oper-
Romanov 1995). The maximum contamination of the flood-
ation in 1948. The plant had five special nuclear reactors for
plain (8 TBq 90Sr/km2 and 185 TBq 137Cs/km2) is compara-
the production of 239Pu and a facility for the separation of
ble to the maximum contamination following the Kyshtym
plutonium as weapons material.
accident (75 TBq 90Sr/km2).
Very large quantities of radioactive waste resulted from
Since 1951, radioactive waste has mainly been discharged
this operation. The liquid radioactive wastes were released to
into Lake Karachay, a small lake of about 150 000 m3 vol-
the River Techa. In all, about 100 PBq were released, includ-
ume with no outlet, at a total amount of about 4400 PBq
ing 95 PBq in 1950/51. 90Sr and 137Cs constituted 12% of the
(120 MCi) of which about 80% is 137Cs. The annual dis-
activity of the radioactive mixture which included significant
charge has now been reduced to about 10 PBq (0.3 MCi)
amounts of 95Zr/Nb, ruthenium isotopes, and other fission
Table 8·38. Inventories of activity contained in some Mayak storage
products. The concentration of radionuclides in water of the
reservoirs, decay corrected to 1994.
upper Techa River (reservoir 5) in 1950-1951 reached an av-
Reservoir no.
Activity contained in reservoir, PBq (kCi)
erage of 1 MBq/L. The population of the villages along the
river, numbering 28 000, was subjected to external exposures
3
1.6 (44)
from water, water-meadows and irrigated kitchen-gardens.
4
0.27 (7.3)
10
6.7 (180)
The water of the River Techa was used by the population for
11
1.2 (32)
drinking, watering cattle and fishing, etc. Although a consid-
9 (Karachay)
4400 (1.19 105)
erable portion of the population along the River Techa was
17
74 (2000)
576
AMAP Assessment Report
(Glagolenko 1995, Malyshev 1995, Romanov 1995). Spring
sists of three RBMK-type graphite-moderated reactors, a re-
flooding has contaminated large parts of the Asanov Swamp
processing plant for the production of weapons-grade pluto-
along the river banks of upper Techa River. As described
nium, and storage facilities for radioactive wastes. Two of
above, the Asanov Swamp has retained much of the dis-
the reactors have been shut down since 1992; the third reac-
charged activity. However, some of the discharged radionu-
tor is operated as a dual-purpose reactor, supplying heat and
clides, especially mobile radionuclides such as 90Sr, have
electricity to the region. Construction of a new reprocessing
probably been carried in aqueous phase more than 2000 km
facility was initiated in 1983, but suspended in 1989 as a re-
downstream to the Kara Sea.
sult of public opposition and economic problems. The Presi-
Within the Asanov Swamp, activity decreases with dis-
dent of Russia has since issued a decree calling for the con-
tance from the dam of reservoir no. 11. Whereas, during the
tinuation of construction of this plant which, when com-
period of direct discharges to the Techa River, the swamp
pleted, could treat both domestic and foreign spent fuel.
acted as a filter retaining radionuclides, it now acts as a
Most of the radioactive waste originating from the Com-
source for slow release. Radiostrontium (90Sr) is remobilized
bine reprocessing activities has been stored in large stainless
more readily than 137Cs, with a peak concentration in river
steel tanks (ca. 8000 PBq) (Lebedev et al. 1996) or injected
water during the period of spring flooding. However, the an-
into the ground within the site boundary. A total of between
nual release of 90Sr as a proportion of the total inventory in
7
108 and 1
109 Ci (26 000-37 000 PBq) at time of dis-
the swamp has been decreasing with time. The annual 90Sr
posal have been injected. The current activity of this waste
discharge has fallen from 10-15% of the total inventory of
is estimated at 4000 PBq (RCRA 1997). The activity has
the swamp in the early 1960s to 3-4% in the early 1990s.
been transported to the injection site via a reportedly leaky
It has been estimated that current releases of 90Sr from the
pipeline that has spilled an unknown amount of radioactive
Asanov Swamp are about 20-30 Ci/y ( 0.7-1.1 TBq/y) and
waste along its path to the injection site (Bradley and Jen-
future releases are predicted to be 15-30 Ci/y ( 0.5-1.1
quin 1995).
TBq/y) (NRPA 1997).
Current operational releases are considerable lower fol-
8.5.2.5.4. Assessment of river transport and associated doses
lowing the shut-down, in 1990, of the last of the five ura-
nium-graphite reactors that produced weapons grade pluto-
River transport of radionuclides occurs through runoff from
nium. Current operational releases are about 20 000 m3/y of
catchment areas contaminated by global fallout, discharges
waste corresponding to an activity of 370-746 MBq/y enter-
from nuclear installations and accidental releases. In the
ing Lake Karachay.
Yenisey drainage basin, the main source, apart from global
Large amounts of high-level radioactive waste have been
fallout, is previous releases from nuclear installations at
vitrified at the special facility at Mayak. The stored vitrified
Krasnoyarsk. For the Ob, the main sources originate within
wastes at Mayak currently contain about 8000 PBq of long-
the following tributary river systems: Techa Iset Tobol
lived radionuclides (Russian Federal Program 1995).
Irtysh (contaminated by discharges from Mayak since 1948,
the Kyshtym accident in 1957, and airborne release from
Lake Karachay in 1967); Karabolka Sinara Iset Tobol
8.5.2.5.2. Tomsk-7
Irtysh (contaminated by the Kyshtym accident in 1957); and
The Siberian Chemical Combine at Tomsk-7 or Seversk, is
Romashka Tom (contaminated by discharges and acciden-
one of the largest nuclear weapons production facilities in the
tal releases from the Siberian Chemical Plant, Tomsk-7).
world. The site contains five graphite-uranium plutonium pro-
Total fluxes of radionuclides to the Arctic seas through
duction reactors, a uranium enrichment plant, a reprocessing
river discharges are difficult to estimate, especially prior to
plant and other plants engaged in the military nuclear materi-
1961. Data in the open literature is still inconsistent with
als cycle. Three of the reactors have been shut down. The re-
respect to direct discharges from Mayak to the Techa River
maining two are now dual-purpose plants that also provide
during the period 1949-1952. Recent investigations of the
heating and electricity for the towns of Tomsk and Seversk.
vertical distribution of 137Cs and plutonium isotopes in
Tomsk-7 came to international attention in April, 1993, when
dated sediment profiles from the Ob estuary reflect signals
a chemical reaction caused an explosion in a tank containing
of 137Cs from global fallout only (Panteleyev et al. 1995).
uranium nitrate solution. The plant adjoins the River Tom
However, other radionuclides, such as 90Sr, may have been
that ultimately drains into the River Ob. Since 1956, contami-
transported from Mayak to the Kara Sea because of their
nated cooling water has been discharged to the river. It has
higher aquatic mobility.
been estimated that the inventory of radioactivity remaining
During recent years, substantial new information has be-
in the River Tom in 1995 was about 3.7
104 Ci ( 1.4 PBq).
come available on river transported radionuclides, as well as
A recent report of the Russian Federation Security Coun-
on past and present sources, especially Mayak PA. Based on
cil states that the total inventory of radioactive wastes with-
annual mean concentrations (Chumichev 1995) about 1.5
in the industrial zone of the site is estimated to be 44 000
PBq 90Sr were estimated to have been transported to the
PBq. The majority of this waste is in the form of liquid ra-
Kara Sea during 1961-1990 by the Ob (0.65 PBq), the Yeni-
dioactive waste, part of which was discharged into several
sey (0.37 PBq), the Severnaya Dvina (0.10 PBq), the Pechora
reservoirs (estimate 5000 PBq). In addition to surface dis-
(0.08 PBq), the Lena (0.29 PBq) and the Indigirka (0.02
charges, Tomsk-7 is one of two sites in Russia where under-
PBq) Rivers. Several recent investigations in the tributary
ground injection has been used as a means of disposal for
system of the Ob River confirm that the highest activity in
large volumes of waste, estimated in the early 1990s to be
sediments is found in the upper Techa River; in reservoirs,
about 15 000 PBq (RCRA 1997).
in the Asanov Swamp and in areas contaminated from the
Kyshtym accident (Trapeznikov et al. 1995, Christensen et
al. 1995, Romanov 1995). Radioisotopes of caesium are
8.5.2.5.3. Krasnoyarsk-26
strongly associated with sediment and soil components,
The Krasnoyarsk Mining and Chemical Combine, formerly
while 90Sr is comparatively mobile (Tronstad et al. 1995).
known as Krasnoyarsk-26 and now renamed Zhelezno-
There are no available data defining how much radioactive
gorsk, is situated on the Yenisey River. This combine con-
contamination has been transported by rivers into the Arctic
Chapter 8 · Radioactivity
577
marine environment in the period up to 1961. As a result, it
Dose (mSv)
is not possible to reconstruct doses for river-transported ra-
102
dionuclides.
8.5.2.6. Mining activities
The only country known to have mining activities in the
Arctic of potential radiological significance is Canada that
101
has a single uranium mining operation at Baker Lake north
GIT, int
of 60°N. This is an entirely exploratory venture and no ap-
plication for a license to operate a mine has yet been submit-
ted. Any approval would be contingent on the provision of
E
an Environmental Impact Assessment. The licensing require-
ments of the Atomic Energy Control Board of Canada im-
, ext
pose a dose limit of 5 mSv/y for members of the public.
100
There is no national policy of reducing doses below the dose
RBM, int
limit (i.e., the implementation of optimisation) as there is for
nuclear power plants.
8.5.3. Accidental releases
10-1
There have been several historical radiation accidents that
101
102
103
104
constituted sources of contamination of the Arctic environ-
Time (days)
ment with artificial radionuclides. These events began with
the Kyshtym accident in 1957, followed by the Lake
GIT
: gastro-intestinal tract
E
: effective dose of internal (int) and external (ext) exposure
Karachay accident in 1967, the spillage of plutonium from
RBM : red bone marrow
nuclear weapons at Thule in 1968, the re-entry of the Soviet
Union's Cosmos-954 satellite over Canada in 1978, the
Figure 8·62. Dose accumulation in adult rural residents after the Kyshtym
accident in 1957.
Chernobyl accident in 1986 and the accident at Tomsk-7 in
1993. There were also a series of accidents involving Soviet
dent. The average effective dose in the most exposed group
nuclear submarines in 1961, 1968, 1978 and 1989. These
of inhabitants reached 0.5 Sv. A significant portion of 90Sr
accidents and their consequences are discussed in the follow-
could have been transported to the Arctic Ocean via the Ob
ing sections.
River drainage system but this flux could not be distin-
guished from other Mayak releases. The present level of ra-
dioactivity from this atmospheric release is estimated at
8.5.3.1. The accidents at the Mayak weapons production
44 000 Ci (
plant in 1957 and at Lake Karachay in 1967
1.6 PBq) with the main constituent being 90Sr.
In addition to intentional discharges into the River Techa
Lake Karachay, 1967
during the period 1949-1951 described in section 8.5.2.5.1,
In 1967, ten years after the storage tank explosion, Lake
two major accidents have contaminated the areas surround-
Karachay partially dried out during a long summer period
ing the Mayak reprocessing plant.
without precipitation and about 16 TBq of radioactive dust,
mainly contaminated with 137Cs and 90Sr, was spread with
The Kyshtym accident, 1957
the wind, contaminating parts of the same area as the Kysh-
In the Kyshtym accident in 1957, a storage tank of highly
tim accident. In this case, however, levels of environmental
radioactive material exploded. Of the 740 PBq (20 MCi) in
contamination were lower: the area within the isopleth of
the waste tank, about 90% settled in the immediate vicinity
4 kBq/m2 of 90Sr was 1800 km2 (Malyshev 1995, Romanov
of the site of the explosion. The remaining 74 PBq (2 MCi)
1995).
was dispersed by the wind and subjected to deposition.
The explosion of the tank containing highly-active wastes
8.5.3.2. The Thule nuclear weapons accident in 1968
at the Mayak plant in the Urals resulted in release of about
74 PBq of a radioactive mixture to the atmosphere to a
On 21 January, 1968, a B-52 aircraft from the US Strategic
height of up to 1 km. 144Ce/Pr and 95Zr/Nb were the domi-
Air Command (SAC) attempted an emergency landing at
nant short-lived radionuclides while the dominant long-lived
Thule Air Base following a fire in the on-board electrical sys-
radionuclide, 90Sr, represented 5.4% of the total activity.
tem (Risø-R 213, 1970). All aircraft electrical power was
The radioactive footprint of the 90Sr distribution in soil had
lost and the aircraft crashed on the sea ice of Bylot Sound 11
a scale length of 300 km. A population numbering 270 000
km west of the Thule Air Base. The aircraft disintegrated on
was subjected to external exposure by gamma radiation
impact and an explosion and fire ensued. The crew bailed
from 95Zr/Nb, internal exposure of the intestinal tract from
out before impact. Six survived but one died. The aircraft
144Ce/Pr and of red bone marrow and bone surfaces by
carried four nuclear weapons all of which were destroyed by
90Sr/Y. Figure 8·62 shows the dynamics of accumulation of
the impact when the conventional high explosive charges in
the effective dose and of its main components in adult in-
the unarmed bombs detonated. This resulted in dispersion of
habitants, normalized to the density of soil contamination
fissionable material viz. plutonium and, presumably, urani-
with 90Sr of 1 Ci/km2 (37 GBq/km2). The contribution of
um. Only the aircraft engines were not totally destroyed.
90Sr to the effective dose became dominant 1-2 years after
There was speculation that parts of the aircraft might have
the decay of other radionuclides.
plunged through the ice but this was shown later not to have
About 10 000 persons who received a collective dose of
been the case. Ice cores from an area of approximately 100
1300 manSv were evacuated 7-670 days following the acci-
m in diameter were contaminated. Outside this zone, conta-
578
AMAP Assessment Report
mination was found on the surface only. Contamination
Table 8·39. Dose commitments from annual intakes of 1 kg of marine
from the accident can be divided in four categories:
product contaminated by 239,240Pu from the Thule accident (Aarkrog 1995).
· Plutonium carried aloft in the cloud from the explosion
Sample
Dose commitments 1968-1995, Sv
and fire and dispersed regionally or even globally by the
Bivalves
7.1
prevailing meteorological conditions.
Shrimps
15.11
· Plutonium deposited on the ice and snow surfaces locally.
Fish
1.1
Seabirds
0.2
· Plutonium deposited on aircraft and weapons debris.
Seal and walrus
0.4
· Plutonium in and beneath the ice at the point of impact.
It has been impossible to determine the exact amounts of
The number of Bq was then multiplied by the dose factor
plutonium in these four categories. However, from a local
0.95 µSv/Bq 239,240Pu and the dose commitment for con-
point of view, it was important to know the amount, form
sumption of the product over the given period was calcu-
and association of plutonium (and tritium) on the surface
lated (Table 8·39).
in the immediate vicinity of the crash site especially at the
Assuming an annual individual consumption of 5 kg bi-
point of impact where decontamination operations were
valves, 5 kg shrimps, 25 kg fish, 5 kg seabirds, and 30 kg
technically feasible. Most of the estimated amount of pluto-
seal and walrus from Bylot Sound, the dose commitment to
nium (3.5
0.7 kg or 8.8 TBq) on and in the ice at the crash
1995 becomes 143 µSv, corresponding to the dose received
site was found within a black teardrop-shaped area of 700
in three weeks from local background radiation, i.e., an in-
150 m2 (0.11 km2). The amounts of plutonium recovered
significant dose from a human health point of view. There
with aircraft and weapons debris is not included in this 3.5
appears to be no actual consumption of local marine prod-
kg. Recent information provided to the Government of Den-
ucts from the contaminated area. The doses calculated are
mark by the U.S Department of Energy disclosed that the
therefore entirely hypothetical.
Thule weapons contained a total of 6 kg of plutonium. The
In 1968, the isotopic ratio 241Am/ 239,240Pu in various
plutonium contamination was in the form of oxide particles
media was 0.05 (Aarkrog et al. 1984). Since then, the ratio
with a very wide size distribution, the median diameter be-
has increased to about 0.15 but the 239,240Pu activity con-
ing 2 m. The plutonium particles were often associated
centrations in biota have typically decreased since 1968 by
with larger particles of low density inert material.
an order of magnitude. The dose factor for 241Am is 0.98
It was decided to remove as much of the contamination
µSv/Bq (i.e., nearly the same as for 239,240Pu). In bivalves
from the sea-ice as possible. First, the weapons and aircraft
and shrimps, the 241Am/ 239,240Pu ratio is about twice that
debris were recovered. A month after the accident, this
in the marine sediments. Hence, if the annual individual
phase was completed and the weapons parts were returned
consumptions of the various products are the same as sug-
to the United States while aircraft wreckage was stored for
gested above, the dose commitment from 241Am in marine
future disposal.
products from Bylot Sound consumed 1968-1995 is of the
The next decontamination step involved scraping the
order of 10 µSv, i.e., an order of magnitude less than the
contaminated area of snow and ice with heavy road machin-
dose from 239,240Pu.
ery. `Hot spots' were monitored and removed by shoveling.
The contaminated snow and ice was placed in sixty-seven
8.5.3.3. The Cosmos-954 satellite re-entry in 1978
25 000-gallon tanks (total volume: 6300 m3). In September,
1968, these tanks were shipped to the Savannah River Plant
The disintegration of the Cosmos-954 nuclear-powered
in the USA. The clean-up was very efficient and it was esti-
satellite containing a fission reactor occurred on 24 January
mated that only approximately 1 TBq ( 50%) of plutonium
1978, when the satellite re-entered the Earth's atmosphere
remained on the ice. During the summer months (June-July)
over Canada's Northwest Territories. Early search and re-
the sea ice broke up and started to melt. It was observed
covery operations showed that significant quantities of radio-
that the ice from the crash site drifted northward. However,
active debris had spread out over a 1000 km path stretching
when and where the ice contamination was released to the
north-east from Great Slave Lake. About 65 kg of large ob-
sea was not observed.
jects were recovered including steel plates, beryllium rods,
From measurements of plutonium in marine sediments
etc., exhibiting high dose rates. This material represented
collected during expeditions to Thule in 1968, 1970, 1974,
only a small fraction of the total satellite mass, assumed to
1979 and 1984 (Aarkrog 1971, 1977, Aarkrog et al. 1984,
be several tonnes. Much material was never recovered.
1987, Smith et al. 1994) it was calculated that about 1 TBq
Air sampling and analyses have revealed radioactive par-
or 0.5 kg of plutonium was deposited on the bottom of
ticles consisting mainly of isotopes of Zr, Nb, Ru and Ce.
Bylot Sound from the Thule accident. The amount of pluto-
The absence of the volatile fission products 137Cs and 131I on
nium left on the ice after the decontamination effort in 1968
particles showed there was some melting and reformation of
was estimated to be 1 TBq ( 50%). It therefore seems likely
core material during re-entry. Environmental measurements
that the majority of the plutonium in the sediments comes
did not detect contamination of air, drinking water, soil or
from the melting of the sea-ice. On the other hand, it is also
food products. Only snow samples taken from the vicinity
evident that the highest levels are found beneath the point of
of the debris path indicated contamination with a mixture of
impact and, because some of the contaminated ice drifted
fission and activation products. Increased concentrations of
137
away before it melted, it seems likely that some debris en-
Cs (500-800 Bq/kg) in caribou meat were attributable to
tered the sea directly through the impact hole in the ice.
residual weapons test fallout. Intensive aerial surveys indi-
The doses from possible consumption of marine biota
cated that about 25% of the estimated radionuclide inven-
from Thule were calculated from environmental analysis of
tory in the satellite reactor was deposited in the form of mil-
plutonium in bivalves, shrimps, fish, seabirds, seals and wal-
limeter-particles over an area of 124 000 km2. The remain-
rus collected at Thule following the accident. To obtain the
ing 75% is suspected to have been volatilized and dispersed
integrated 239,240Pu intake over a given period, the time-inte-
as fine dust in the upper atmosphere. This material, contain-
grated activity of plutonium (Bq y/kg) was multiplied by the
ing long-lived 90Sr and 137Cs, descended to the surface over
annual individual consumption rate (kg/y) of the product.
several years. The deposited activity of 90Sr in the northern
Chapter 8 · Radioactivity
579
hemisphere is estimated to be about 3
1012 Bq compared
was contaminated with 137Cs at levels over 40 kBq/m2 and
with the total deposition of 1.1
1015 Bq in 1980 (UNSCEAR
more than 7000 km2 with levels higher than 600 kBq/m2.
1982).
The surface contamination of pasture vegetation caused
The personnel involved in debris recovery obtained indi-
rapid and high contamination of local milk products with
vidual effective doses up to 5 mSv and a collective dose of
131I, 89,90Sr and 134,137Cs. After the decay of short and medi-
about 0.1 manSv among 145 persons. Regarding public ex-
um-lived radionuclides and the removal of the surface conta-
posures, a person spending several hours a day near an unre-
mination in the autumn of 1986, the root transfer of 134,137Cs
covered core fragment could receive an effective external
from soil to vegetation and further along the food chain be-
dose of about 5 mSv. Handling a millimeter-size particle for
came the main route of internal exposure of populations.
several hours would give a skin dose of about 1 mSv. If in-
From 1987 to 1991/1993, the effective ecological half-life
gestion of typical core particle occurred one month after re-
was 1-2 years for most local agricultural foodstuffs from con-
entry, the effective dose due to GIT (gastro-intestinal tract)
taminated regions. However, the high level of contamination
irradiation is estimated to be between 4 and 12 mSv. Radio-
of mushrooms and berries in forests of these regions declined
active decay would cause the dose rate to decrease rapidly
much more slowly. Natural food products have, therefore,
with time. UNSCEAR (1993) estimated the collective dose
made a significant contribution to internal exposures of pop-
to the population of the northern hemisphere from the Cos-
ulations that consume them in the last 3-5 years (UNSCEAR
mos-954 accident to be about 16 manSv mainly from 137Cs
1993, Chernobyl Papers VI 1993, Strand et al. 1996).
and 90Sr and, over the longer term, from 239Pu.
Following the Chernobyl accident, countermeasures of
unprecedented scale, particularly in the context of FSU agri-
culture, were introduced for the protection of population
8.5.3.4. The Chernobyl accident in 1986
from long-term internal exposure (UNSCEAR 1993, Cher-
8.5.3.4.1. The accident and associated source term
nobyl Papers VI 1993). The average body burden in some
After nuclear tests in the atmosphere, the accident at Cher-
villages of the FSU was 0.4-0.6 MBq with individuals having
nobyl in the Ukraine on April 26, 1986, was the most signi-
burdens up to 4 MBq. The average internal dose in affected
ficant large scale source of environmental radioactive conta-
areas during the first year reached 5 mSv, the total dose be-
mination. At about 01: 23 hours Moscow time, two explo-
ing 10-30 mSv with individual extremes exceeding 50 mSv.
sions in quick succession blew the roof off the Unit 4 reactor
Where countermeasures were not undertaken immediately,
building of the Chernobyl plant, ejecting concrete, graphite
the content of 134,137Cs in the bodies of inhabitants of conta-
and debris and leaving a gaping hole exposing the reactor
minated areas reached its maximum in the summer of 1986.
core to the outside air. During a ten-day fire, smoke and
A number of European countries introduced timely counter-
gases rose to a height of over 1 km into the atmosphere, fol-
measures. Accordingly, in these countries, early incorpora-
lowed by fragments of uranium fuel. Transuranics and fis-
tion was prevented and the activity peak in body content oc-
sion product radionuclides from the reactor core, plus acti-
curred in the summer of 1987. Decreases in the 137Cs con-
vation products and essentially all the noble gases were re-
tent in the body reflected an effective ecological half-life of
leased to the atmosphere during the first ten days after the
0.3-2 years initially. Compared with the long-term external
accident before the releases could be contained. The heat
exposure, the internal dose is insignificant in black-earth re-
from the fire increased the release rates of radioiodine, a
gions, is similar in regions of turf-podzolic soils, and pre-
substantial fraction of the volatile metallic elements, includ-
dominates in peaty soils areas in Polesye and the Arctic. The
ing radiocaesium, and somewhat lesser amounts of other ra-
contributions of 90Sr, plutonium isotopes and 241Am to inter-
dionuclides normally found in the fuel of a reactor that has
nal dose are insignificant.
been operating for about three years.
According to the best current information regarding the
8.5.3.4.3. Transport and deposition in the Arctic
source characteristics and measurements on environmental
samples, the total activity of radionuclide mixture released
Due to the intensity of the fire, which continued for ten
in this accident was about 8000 PBq (200 MCi) (as of 26
days, and the unique meteorological conditions at the time,
April, 1986) or 1900 PBq (50 MCi) (decay-corrected to 6
parts of the initial radioactive cloud reached altitudes as
May,1986). These values do not include the noble gases
high as 5-10 km. Thus, although the surface winds were to
xenon and krypton which were almost totally released into
the west and northwest, parts of the radioactive cloud were
the atmosphere in an amount of about 7000 PBq (as of 26
more widely dispersed. However, the prevailing transport of
April, 1986) (IAEA 1996, NEA 1995). Total releases into
radioactive materials after the explosion was westward and
the environment of the most radiologically-significant vol-
then, as the wind gradually turned toward the north, north-
atile radionuclides 131I, 137Cs and 134Cs have been recently
east and south. Part of the initial radioactive releases lofted
estimated at about 1500 PBq, 85 PBq and 46 PBq, respec-
to a height of 1.5-3 km moved first over Sweden, then over
tively (NEA 1995). About 8 PBq of 90Sr and 0.1 PBq of
Norway and southern Finland, they then continued east-
alpha-emitting plutonium nuclides were also released and
ward over the Archangelsk region and the southern part of
mostly deposited in the vicinity of Chernobyl.
the Kola Peninsula (28 April), and subsequently over
Salekhard in the estuary of the Ob river (28-29 April). In
late April and early May, Chernobyl fallout was essentially
8.5.3.4.2. Radiological consequences at temperate latitudes
not observable in the far north of Russia (e.g., on the Franz
The explosion of a high-power operating reactor and subse-
Joseph Land Archipelago or Cape Zhelanie at the northern
quent 10-day fire resulted in radionuclide contamination of
end of Novaya Zemlya). In the north of eastern Siberia, in
Eurasia, especially within the territory of the former Soviet
contrast to the European Russian Arctic, radioactive deposi-
Union. Short-lived radionuclides, including 131I; soil conta-
tion was relatively insignificant.
mination with 134,137Cs and, to a lesser extent with 90Sr; and
As the radioactive material from the reactor explosion re-
plutonium isotopes and, later, 241Am, near to the source,
mained largely in the troposphere, from which aerosols are
were the factors affecting population exposure. More than
scavenged fairly quickly, the 137Cs concentration in Arctic
140 000 km2 of the territory of Ukraine, Belarus and Russia
air was reduced by nearly one half as early as June, 1986,
580
AMAP Assessment Report
Table 8·40. Changes in 137Cs deposition from the atmosphere in different
Nielsen (1996) have used the extensive MORS data to calcu-
regions of the Russian Arctic in 1986, following the Chernobyl accident.
late the inventory of 137Cs in Baltic seawater for the years 1985-
137
1990. These Baltic Sea inventories are shown in Table 8·41.
Cs deposition by region, Bq/m2/month
The concentrations in water declined by a half within the
Month, 1986
Murmansk
Arkhangelsk
Asian North
first year following the Chernobyl accident due to sedimen-
May
553.01
187.3
27.5
tation of 137Cs. Since 1987, the inventory of 137Cs has de-
June
512.31
15.1
8.21
cayed with an effective environmental half-life in the range
July
3.14
6.85
1.98
August
1.64
4.4
0.91
10-15 years compared with a mean residence time of Baltic
September
1.27
3.74
0.63
Sea water of 20- 45 years. From these data, it can be esti-
October
0.70
3.11
0.65
mated that the present (1996) input of 137Cs to the Arctic
November
0.61
2.58
0.50
December
0.46
0.68
0.13
from the Baltic Sea is of the order of 50 TBq/y and that the
total transport of 137Cs derived from the Chernobyl accident
and continued to decrease in the following months. During
to the Arctic has been in the range 0.5-1 PBq.
subsequent years, the 137Cs concentration in the atmosphere
The ratio of 134Cs/137Cs in Kara Sea water, determined in
in the northern European part of Russia decreased exponen-
1992, was relatively constant (in the range 0.019-0.034)
tially but 1.5-2 times more slowly than in the mid-latitudes
(Strand et al. 1993) and systematic geographical variations
where the effective ecological half-life was 17 months. Table
could not be distinguished. The primary known sources of
134
8·40 shows changes with time in the monthly atmospheric
Cs are the discharges from the Sellafield reprocessing
deposition of 137Cs in various regions of the Russian Arctic
plant and Chernobyl fallout. Assuming that the ratio of
134
during 1986. It can be seen that rather high levels of 137Cs in
Cs /137Cs in Sellafield discharges has been consistently less
deposition were exhibited during May-June on the Kola
than 0.1 since the mid-1970s, the 134Cs /137Cs ratio in the
Peninsula (the Murmansk region) and the lowest values oc-
Kara Sea, if Sellafield were the sole source, would be insig-
curred in the Asian part of the Arctic. High levels of 137Cs
nificant because of the 2-year physical half-life of 134Cs. It is,
deposition in June in the western part of the Russian Arctic
therefore, unlikely that the 134Cs observed in the Kara Sea in
are explained by wind resuspension and migration of radio-
1992 originates from Sellafield. Accordingly, assuming that
active dust particles.
Chernobyl is the only source of 134Cs, approximately 30%
Large-scale environmental monitoring programs, started
of the 137Cs in the Kara Sea is derived from the Chernobyl
soon after the accident both in the former Soviet Union and
accident (Strand et al. 1993).
other European countries, enabled the production of maps
of aerial contamination of Europe with long-lived 137Cs (EU/
8.5.3.4.4. Food chain and human contamination
CIS JSP-6 1996). Figure 8·5 presents such a map covering
the northern part of Europe. It can be seen that closest to the
As the ecological food chain lichen reindeer humans is
Arctic, the significantly contaminated area extends in an east-
the most important in the exposure of members of the Arctic
west direction across the Leningrad region of Russia, south-
population to environmental contamination with radionu-
ern Finland and Scandinavia. There are no large areas above
clides, this is considered in relation to the dynamics of 137Cs
the Arctic Circle contaminated by more than 10 kBq/m2. Con-
following the Chernobyl accident. Measurements of the dis-
tamination of northern Fennoscandia is mainly < 2 kBq/m2.
tribution of 137Cs in the environment following the Cherno-
In contrast, most of the European part of the Russian Arctic
byl accident are presented in section 8.3.3.
is more contaminated, but still at a level of less than 10 kBq/
m2, reflecting some additional radionuclide deposition of
Lichen
Chernobyl origin. This observation is based on measure-
A clear increase in 137Cs concentrations in Arctic lichen was
ments of Chernobyl-derived 134Cs in deposition. More de-
observed following the Chernobyl accident. In Arctic Fin-
tailed mapping by Roshydromet has identified some areas
land, the peak concentration of 137Cs in dried lichens in
with deposition of between 4 and 18 kBq/m2 on the coast
1986-1987 was up to 1300 Bq/kg above previous levels. The
of the White and Pechora Seas (Izrael et al. 1990).
concentrations decrease with an effective ecological half-life
of about 3-4 years. This is faster than that occurring after
Marine transport to the Arctic seas
global fallout because the Chernobyl accident was a single
In addition to the direct atmospheric fallout from Cherno-
contamination event in contrast to the sustained period of
byl, the Arctic seas may also be contaminated by marine
nuclear weapons testing in the 1950s and 1960s. Less pro-
transport from the North Sea and the Baltic Sea, the catch-
nounced was the 137Cs concentration increase in Arctic lichens
ments of both of which received considerably more radionu-
collected in the Murmansk region and the Nenets area of
clides from Chernobyl (through direct fallout and runoff)
northern Russia and in Greenland. This geographical differ-
than the Arctic.
ence can be explained by the heterogeneity of ground depo-
Studies of radionuclides in the Baltic Sea have been carried
sition, location or sampling sites, and/or variations in snow
out under the aegis of the Monitoring of Radioactivity in the
cover conditions during May-June 1986.
Sea (MORS) group under the Helsinki Commission. As a re-
sult of the atmospheric fallout following the accident, conta-
Reindeer
mination of the Baltic Sea was highly heterogeneous. Nies and
There was an increase in radiocaesium in reindeer meat pro-
duced in Finland, Norway, Sweden and Western Russia in
Table 8·41. Inventory of 137Cs in Baltic seawater (TBq).
1986. Because the 137Cs content in live reindeer in winter
Year
Inventory, TBq
follows the concentration in lichen, it decreased with an ef-
fective ecological half-life of 3-4 years following the peak in
1985
325
1986-1987.
1986
4260-5000
1987
2700
1988
1790
Human body
1989
2320
Long-term radioactive contamination of the Arctic from
1990
2060
Chernobyl, mainly with caesium radioisotopes, 134Cs and
Chapter 8 · Radioactivity
581
137Cs, in 1986 and subsequent years caused significant in-
8.5.3.4.5. Countermeasures
creases in the wholebody content of these radionuclides in
Arctic indigenous people consuming local natural and semi-
A wide range of countermeasures were radiologically justi-
natural foods (reindeer meat, freshwater fish, mushrooms
fied and introduced, especially in Norway and Sweden, fol-
and berries) that concentrate certain radionuclides. This in-
lowing the Chernobyl accident, and some of these are still
crease occurred in all European Arctic countries except Ice-
being applied. Such countermeasures included: ploughing
land and Greenland which were not so significantly conta-
and fertilizing of natural pasture to reduce uptake by plants
minated by Chernobyl fallout, see section 8.3.3. A typical
from soil; feeding of animals with fodder containing lower
wholebody value for 137Cs in Finnish Saami living in the
levels of contamination; use of caesium-binders (saltlick,
Inari region in the early 1980s was 4 kBq. This had risen
mixed in concentrate or boli) for freely grazing animals;
to about 9 kBq in 1986-1988 and decreased to the original
changing the slaughtertime of wild or semi-wild animals;
value during the subsequent seven years (Rahola et al.
and dietary advice to consumers. Countermeasures consider-
1993). In both northern and southern Saami areas of Nor-
ably reduced the doses to the population in the affected
way, Saami people were contaminated by Chernobyl fallout.
areas, especially for reindeer-breeding Saami. The dose re-
Deposition levels varied considerably. The pre-accident body
duction achieved for this latter group was up to 90% (Strand
burden of 3 kBq reached 40 kBq in areas most affected by
et al. 1990). The countermeasures also had positive effects
Chernobyl fallout and about 7 kBq in areas which were
on social well-being (Strand 1994, Brynhildsen et al. 1996,
least affected. For both groups, the consumption of reindeer
Strand et al. 1990).
meat dominated the intake of radiocaesium. Figure 8·63
presents dose estimates based on wholebody measurements
8.5.3.4.6. Human dose estimation
for the two groups of Norwegian reindeer-breeding Saami,
Internal dose estimation was made on the basis of whole-
Dose
body measurements conducted prior to, and after, the Cher-
mSv
nobyl accident. The 137Cs wholebody content was integrated
2.0
over a period of 5-8 years after 1986 and combined with the
Northern Saami, Norway
appropriate dose conversion factors. It was assumed that
Southern Saami, Norway
50-70% of the effective internal dose was delivered during
1.5
the observational period. A further 30-50% is expected to
be delivered in the future. For the groups described in the
previous paragraph, the committed effective internal dose
from caesium radioisotopes of Chernobyl origin ranges
1.0
from 0.5-10 mSv, depending mainly on 137Cs soil contami-
nation density and food habits. Given an initial deposition
of 1 kBq/m2 of 137Cs (and 0.5 kBq/m2 of 134Cs), the dose
0.5
commitment will be in a narrower range of about 0.5-1.5
mSv, which is an order of magnitude more than in temperate
latitudes. For comparison, at the same level of initial deposi-
tion, the external dose commitment for Nordic indigenous
0
people is about 0.1 mSv. Thus, for people consuming Arctic
1965
1970
1975
1980
1985
1990
1995
natural and semi-natural foods, internal exposure from Cher-
Figure 8·63. Dose estimates based on wholebody measurements (see Fig-
nobyl contamination of the environment contributes 80-
ure 8·36) for two groups (northern Saami and southern Saami) of Norwe-
90% of the total (i.e., external + internal) dose. The collec-
gian reindeer-breeding Saami.
tive dose for about 2000 000 persons in the average popula-
tion and about 100 000 persons in the indigenous popula-
northern Saami and southern Saami. After 1986, there was
tion living in the European Arctic, contaminated after the
an increase in both groups but of smaller magnitude in the
Chernobyl accident at an average level of about 1 kBq/m2 of
northern population. The increase in the southern Saami
137Cs in Fennoscandia and 2 kBq/m2 in western Russia, has
would have been considerably larger but for the implemen-
been preliminarily estimated to be about 500 manSv.
tation of countermeasures. Reindeer-breeding Saami people
living in the north of Sweden accumulated an average of
8.5.3.5. Accidents involving nuclear-powered vessels
about 40 kBq of 137Cs in 1987-88 compared to 5 kBq pre-
viously (Johanson and Bergstrom 1993). In Russia, the most
There have been several cases of radioactive contamination
detailed information is available for indigenous inhabitants
of the environment during the period of operation of nuclear-
of the Murmansk region (Kola Peninsula). Before the Cher-
powered ships. In 1961, a submarine with a damaged reactor
nobyl accident, the average 137Cs wholebody content of
returned to its base on the Kola Peninsula and local contami-
adults was about 20-30 kBq, the highest values within the
nation occurred. In 1965, a local release of radioactive mate-
Eurasian Arctic (Ramzaev et al. 1993). In June, 1986, the
rials was reported during an accident with a submarine reac-
average wholebody content remained at a level of 26 kBq,
tor in the Severodvinsk shipyard. The largest release of liquid
but by July, 1991 it had increased to 33 kBq.
radioactive waste (74 TBq) occurred in 1989 from a North-
The peak in the 137Cs wholebody content of indigenous
ern Fleet submarine in the Ara inlet. This accident led to ra-
people living in different areas and countries depends on the
dioactive contamination of an area of about 1.0 km2. Two
extent of deposition, meteorological conditions, the duration
other nuclear submarine accidents occurred in the Norwe-
of snow cover in May-June, 1986, and individual food hab-
gian Sea in 1989; one involving failure of the cooling circuit
its. Contamination with both 137Cs and 134Cs of Chernobyl
in a Soviet (NATO Echo-class) submarine and the other the
origin occurred. The mean ratio of 134Cs/137Cs, which was
loss of the (NATO Mike-class) submarine Komsomolets.
0.54 at the time of the accident, has decreased over time due
During accidents involving submarine nuclear reactors,
to the faster decay of 134Cs (physical half-life 2.05 y).
personnel have been affected by gamma-radiation which can
582
AMAP Assessment Report
result in high doses causing development of acute radiation
for the people onboard neighboring Norwegian surveillance
sickness to its highest degree. In addition, beta-radiation
and rescue vessels were in the range 0-43 000 Bq 131I. The
from radioactive gases entering the hull of the submarine
doses to thyroid in the most exposed personnel were 1.6
during major accidents can cause acute radiation damage to
mGy. Doses to the Norwegian population, assessed on the
skin. Collective doses of gamma-radiation obtained by per-
basis of analysis of air and milk samples, were shown to be
sonnel of nuclear submarines during several previous acci-
insignificant (NRPA 1989).
dents have been between 17 and 74 manSv, and the average
individual doses between 0.2 and 0.6 Sv. The average indi-
vidual dose to skin was between 2 and 6 Gy and the average
individual dose to thyroid, associated with the inhalation of
radioiodine, was between 2 and 10 Gy.
In 1961, a submarine, which had suffered an accident,
returned to its base. Samples of seawater were subsequently
taken at the distances of 5, 50, 100, and 300 m from the
submarine hull at the surface and at depths of 5, 10, and 20
m. The highest specific activity in seawater, of 37 Bq/L, was
found at a distance of 5 m from the hull at a depth of 5 m.
Radioactivity in samples of plants, fish, and sediment taken
during ventilation of submarine compartments had back-
ground levels. Three months later, the measurements of ra-
dionuclides in seawater in the vicinity of the submarine
mooring were repeated. The activity concentrations in sam-
ples of seawater, taken predominantly in the region of the
reactor power compartments, were between 0.1 and 27
Bq/L, which corresponds to background.
In 1968, radioactive contamination of a naval base oc-
curred during decontamination work on a submarine. In-
creased alpha- and beta-activity was detected in seawater
near the mooring location. However, monitoring of sedi-
ments, algae and fish at all sites gave values corresponding
to background levels.
In 1979, during work on remediating the consequences
of another submarine accident, the gamma-radiation dose
rate in the sanitary zone of the naval base reached 1.3 µGy/h.
The sources of contamination were the submarine which
had suffered the accident, the stored radioactive waste re-
moved from it, and releases to the atmosphere during the
VEIGÅRD
ventilation of contaminated compartments. The average
concentrations of fission products in seawater in the bay
ERIK
adjacent to the base during the period of remediation of
Figure 8·64. Burning nuclear submarine.
consequences varied within the range of background associ-
ated with global fallout. Short-term contamination of sea-
8.5.3.5.1. Sunken Komsomolets submarine
water was observed at the mooring location of the subma-
8.5.3.5.1.1. Accident and source term
rine, and was caused by products of decontamination of the
submarine hull and by the release of low-level waste from
On 7 April, 1989, the Soviet nuclear submarine Komsomo-
the coastal reservoir to the bay. No significant increases of
lets (K278, NATO Mike-class) caught fire and sank 180 km
radioactivity of algae in the region of the submarine moor-
southwest of Bear Island (Bjørnøya) in the Norwegian Sea.
ing, bottom sediments or benthic organisms (starfish) in the
Of the 69 crew members on board, 42 were killed in the ac-
bay were detected. Any increased levels could not be attrib-
cident. The submarine was designed to operate at depths of
uted to the submarine because similar values were measured
up to about 1000 m, and its double hull was made of a tita-
prior to the accident. During the period of work on the sub-
nium alloy. At present, the wrecked submarine rests at a depth
marine, the concentrations of short-lived radionuclides in at-
of ca. 1650 m. The single reactor was shut down in an or-
mospheric air at the location of its mooring did not increase
derly manner prior to sinking. To estimate causes and conse-
significantly. No contamination of soil and vegetation at the
quences of the accident and to develop countermeasures, a
base location was observed during this period.
number of Russian expeditions to the Komsomolets have
A Soviet nuclear submarine (NATO Echo-class) with nu-
taken place in recent years.
clear weapons onboard had problems in the Norwegian Sea
The sunken submarine contains one nuclear reactor
in June, 1989. The prime cooling circuit failed and a supply
with an inventory of long-lived radionuclides comprising
ship had to deliver cooling water (see Figure 8·64). The sub-
2.8
1015 Bq of 90Sr and 3.1
10l5 Bq of 137Cs along with
marine subsequently sailed back to its base on the Kola Penin-
other fission and neutron activation products. The change in
sula. During its time off the Norwegian coast there were mea-
the reactor nuclide inventory with time is depicted in Figure
surable amounts of radioiodine in the surrounding water and
8·65 derived from Gladkov et al. (1994) and reproduced
in air samples in the north of Norway. Radioiodine in milk
from the NATO-CCMS Report (CCMS/CDSM/NATO
from animals grazing freely in the north of Norway ranged
1995). It can be seen that a decade after the accident long-
from 0-10 Bq/L. Based on air measurements in Norway and
lived 137Cs and 90Sr will dominate among the fission prod-
information from the former Soviet Union, the total releases
ucts. Most activation products are likely to have essentially
of 131I were of the order of 0.1-10 TBq. The inhaled activity
decayed before major releases through corrosion are liable
Chapter 8 · Radioactivity
583
Activity
from the Russian Navy (Lisovsky et al. 1996). In addition, a
TBq
study of the release and transport of radionuclides from the
10 000
wreck has been carried out by scientists at the Norwegian
147
Institute of Marine Research, Bergen (Blindheim 1994).
Pm
The NATO study was intentionally based on a relatively
137 Cs
simple model that estimates the corrosion rate of the reactor
144 Ce
and torpedoes, considers transport of water-soluble radionu-
90 Sr
clides by ocean currents but takes no account of radionu-
1 000
clide partitioning between water and particles. The study
draws conclusions primarily from the results of the modeled
134 Cs
dispersion of soluble 137Cs, the predominant long-lived fis-
sion product. Because the modeling is based on an assump-
241
85
Pu
Kr
tion that all radionuclides are completely water soluble, it
106 Ru
overestimates the dispersion of particle-reactive nuclides
55 Fe
100
such as 60Co, 241Am and plutonium isotopes.
Barriers to the release of radionuclides within the reactor
60 Co
fuel rods comprise cladding of 5 mm thick stainless steel, the
reactor vessel of 150 mm thick carbon steel and the subma-
rine hull of 100 mm titanium alloy. These are expected to
prevent corrosion of the reactor fuel for about two thousand
years. By that time, only plutonium isotopes and americium
10
will be present in the reactor in significant activities. In the
63 Ni
intervening period, the only pathway of radionuclide release
from the reactor will be water exchange through the reactor
compartment ventilation tube. The rate of 137Cs release can,
therefore, be assumed to be comparable with the present
241 Am
rate of about 0.5 TBq/y with appropriate correction for ra-
1
dioactive decay.
1
10
100
500
Plutonium in the torpedoes is not protected from sea wa-
Time (y)
ter to the same degree and is expected to be open to corro-
Figure 8·65. The changing inventory of major radionuclides in the Kom-
sion much more quickly than the reactor fuel. Plutonium is
somolets reactor with time (after Gladkov et al. 1994 and CCMS/CDSM/
NATO 1995).
highly particle-reactive and insoluble. Accordingly, any plu-
tonium released is likely to be retained in marine sediments
to occur. Two nuclear torpedoes with mixed uranium/pluto-
close to the point of release as occurred following the Thule
nium warheads, situated in the forepart of the hull contain
nuclear weapons accident.
about 1.6
1013 Bq of weapons-grade plutonium.
The transport of radionuclides released from the Komso-
Minor releases of radionuclides from the reactor com-
molets has been modeled by Blindheim (1994) and NATO
partment have already been detected in the close vicinity of
(CCMS/CDSM/NATO 1995). Both studies have, however,
the submarine wreck during Russian expeditions. These
used conservative models that will be more-or-less applica-
surveys indicate radionuclide releases through a reactor
ble for nuclides having low particle reactivity (low kds) but
ventilation tube, the 137Cs activity concentration in the wa-
be overly conservative (i.e., pessimistic) for reactive nuclides
ter inside the tube being of the order 1 MBq/m3. The annual
such as 60Co, plutonium and americium. The models predict
release of 137Cs from the submarine was estimated to be no
highest concentrations of conservative radionuclides north-
more than 0.5 TBq. The likelihood of large-scale releases of
west of Spitsbergen but the contamination is distributed
radionuclides from the Komsomolets submarine in the near
over a large area both in the Arctic and Atlantic Oceans.
future is small. As the containment barriers in the submarine
The Blindheim (1994) study also considers the importance
are breached by corrosion, further gradual releases may oc-
of a `thermal plume' arising from residual heat in the Kom-
cur and these will increasingly comprise long-lived fission
somolets reactor but concludes that the energy is insufficient
products from the reactor and uranium and plutonium from
to create a plume rising more than 500 m from the bottom
the nuclear-tipped weapons. While uranium is relatively sol-
of the Norwegian Sea.
uble and will be mobilized as the structural integrity of the
Assuming that the releases from the submarine occur as
torpedo and warhead casings is breached, the environmental
suggested above and fish permanently inhabit contaminated
contribution will be essentially insignificant in the context of
water, the contribution of Komsomolets releases to radionu-
the natural uranium content of the surrounding environ-
clide contamination of the Arctic waters and to the most se-
ment. Plutonium has limited solubility and a high affinity
verely affected fish can be estimated. The NATO study esti-
for particles. Accordingly, most of the plutonium released
mate that the increase in 137Cs in seawater was about 0.02
from the warheads is likely to be retained in sediments with-
Bq/m3 in 1991, 0.01 Bq/m3 in 1995 and is expected to de-
in the immediate vicinity of the wreck.
crease to about 0.001 Bq/m3 in 2089. These levels should be
compared with the concentrations of 137Cs attributable to
other sources in surface waters of the open Barents and Kara
8.5.3.5.1.2. Radiological assessments of
Seas of between 1 and 10 Bq/m3, respectively.
the Komsomolets accident
The radionuclide concentration in fish is assumed to be
There have been two major assessments of the radiological
proportional to the concentration in surrounding water.
threat posed by the Komsomolets. The first of these was
Komsomolets releases are predicted to contribute concentra-
carried out by Norwegian experts under the auspices of the
tions of about 0.005 Bq/kg in 1991, 0.002 Bq/kg in 1995
NATO sub-Committee on Challenges to Modern Society
and 0.00005 Bq/kg in 2089 in the most significantly affected
(CCMS/CDSM/NATO 1995) and the second by experts
fish. The dominating radionuclides in 1995 are 147Pm, 137Cs
584
AMAP Assessment Report
and 55Fe. For comparison, typical Arctic fish generally have
Table 8·42. Plutonium-239 distribution in the aqueous marine environ-
a level of about 1 Bq/kg of 137Cs.
ment (half-period of release is assumed to be 10 000 hours).
This conservative modeling indicates that radionuclide
Time, hours
concentrations in seawater and fish caused by past, present
100
500
1000
5000
10000
50000
and future releases from Komsomolets are at least two to
three orders of magnitude lower than current concentrations
Activity
concentration,
of 137Cs in the same media. Human intake of released ra-
Bq/L
Water, km2
dionuclides through seafood consumption should represent
an even lower proportion of current exposures because
0.27
2
0
0
0
0
0
0.027
51
48
80
9
0
0
seafood is generally of minor importance as a source of in-
0.0027
460
1200
3000
1400
200
0
ternal dose to humans. Thus, the Komsomolets submarine
0.00027
5400 30000 40000
76000 40000
0
constitutes an insignificant source of existing and projected
Activity
marine contamination and radiation exposure. The NATO
concentration,
study (CCMS/CDSM/NATO 1995) concludes that the re-
Bq/m2
Sediments, km2
sults `clearly indicate that the sunken submarine represents
2.7
0
0
5
5
5
5
no significant hazard to man, today or in the future'.
0.27
0
80
100
1500
5800
9900
For completeness, a brief presentation of the Russian
Navy assessment (Lisovsky et al. 1996) of the risks posed
background contamination by an order of magnitude) has a
by releases from the Komsomolets submarine reactor and
form close to a symmetrical ellipse with its axis elongated in
weapons is given below.
a northwesterly direction. Under slow release conditions, the
A generalized computer model was constructed for esti-
contamination zone dimensions do not exceed 200
200 km.
mating the radioecological consequences of the release of ra-
The water contamination zone under slow release conditions
dionuclides to the marine environment. The model embodies
can persist for about 4-5 years. The bottom contamination
the main processes of radionuclide migration and accumula-
zone continuously increases but stabilizes in response to the
tion in environmental compartments and the radiological
decline in the level and extent of water contamination. In
consequences for humans and marine biota. For estimating
the slow release scenario (half-period of the release = 10 000
the radionuclide releases from the active zone of the reactor,
hours), water contamination during the initial period reaches
the corrosion rate was assumed to be linearly dependent on
0.027 Bq/L and a zone with such a 239Pu activity concentra-
temperature. The corrosion rate was chosen on the basis of
tion will persist for up to 5000 hours.
engineering data on the corrosion resistance of the various
The international expedition conducted in 1994 suggested
barrier materials. The model for mass and heat transfer was
that some kinds of zoocenosis in the region of the Komso-
based on ordinary differential equations.
molets wreck could result in the transfer of radionuclides
Of the number of radionuclides that are formed in the ac-
from deep ocean layers to surface layers and serve as food
tive zone during operation of a reactor, three long-lived ra-
for fishes feeding on plankton. In this respect, Calanus hy-
dionuclides 137Cs, 90Sr and 239Pu are of primary concern.
perboreus and Themisto olyssorum were regarded as of pri-
Calculations were performed only for caesium. The content
mary importance, although whether these organisms feed at
of plutonium in the active zone of the reactor of the given
depth has been questioned. In the absence of data on the dy-
type is extremely small. For the calculations, it was assumed
namics of plutonium accumulation by these organisms, it
that the reactor compartment and the active zone of the re-
was conservatively assumed that body content reaches equi-
actor were damaged at the moment of the vessel's impact on
librium in 100-150 hours. The most unfavourable situation
the sea bottom. It was further assumed that the open area of
will be one in which the release of plutonium from the sunken
the active zone is 5 cm2, the area of holes in the guard of the
submarine occurs a short time before the seasonal migration
reactor compartment is 2 m2 and the total area of defects in
of this species of zooplankton to the upper ocean layers. In
fuel cladding at the time of the accident was 80 cm2.
such a case, the incorporation of plutonium into fish and
The predicted maximum release rate of 137Cs to the envi-
consequent exposures to human seafood consumers will be a
ronment is calculated to be 0.05 TBq/y. This is within an or-
maximum. Assuming the accumulation factor to be 2600,
der of magnitude of the release rate estimated on the basis of
for the scenario of slow plutonium release, the maximum
measurements during sea expeditions to the Komsomolets
possible concentration in plankton rising to the upper layers
wreck site during the period 1992-1994 which is 0.004-0.4
will be 70-700 Bq/kg and the area of contamination will not
TBq/y. These are relatively small release rates and would not
exceed 80 km2. The zooplankton species migrating from the
be expected to entail any significant radiological consequences.
near-bottom layers comprise 2-17% of the ration of market-
About 1.5
1013 Bq of weapon 239Pu in the torpedoes of
able fish (i.e., the specific content of 239Pu in the biomass that
the sunken submarine is assumed to come into direct con-
serves as food for marketable fish will range from 1 to 120
tact with seawater, resulting in corrosion and plutonium re-
Bq/kg). After one to three months, zooplankton dispersion
lease to the environment. Two release rate scenarios were
will result in a decreasing plutonium specific activity.
used in the modeling. The first is based on a release rate that
According to conservative modeling of the radionuclide
is governed by an exponential law with a half-period of 24
accumulation by hydrobionts, fish having a maximum pluto-
hours; the second is based on an exponential release of half-
nium content of 0.1-6 Bq/kg in edible parts may appear for
time of 10 000 hours (i.e., ca. 1 year). The latter variant cor-
some months in an area of about 80 km2. When individuals
responds to a case of a release rate governed by the corro-
leave the contaminated zone as a result of natural migration,
sion rate alone and yields a constant rate of release.
the plutonium content in fish will decrease according to a line-
The results of generalized calculations of plutonium con-
ar or exponential law. Thus, within 25-30 days of the cessa-
centration fields are presented in Table 8·42 for the case of
tion of contaminated plankton consumption, the contamina-
the longer-term plutonium release. These indicate that the
tion level in edible tissues will decline to less than 0.1 Bq/kg.
maximum level of contamination does not exceed 0.27 Bq/L
On the basis of the standard maximum consumption of
for water and 2.7 Bq/m2 for sediments. The zone of contam-
sea products, the hypothetical dose to members of the criti-
ination (i.e., the zone in which concentrations exceed the
cal group can be no more than 0.03 mSv/y based on conser-
Chapter 8 · Radioactivity
585
vative assumptions. Realistically, it should actually be at
8.6.1. Nuclear power plant reactor accidents
least an order of magnitude less, namely of the order of
1 Sv/y. The estimated dose rate in zooplankton perma-
Prevention of nuclear reactor accidents has been the main
nently inhabiting the most contaminated sea area may reach
objective of nuclear safety since the beginning of the nuclear
about 1 mGy/h for some months. Fish consuming this zoo-
era. International and, in most countries, national nuclear
plankton would receive about 1 µGy/h. This exposure level
safety efforts were improved following the Chernobyl reac-
cannot cause any significant biological effects in marine bio-
tor accident. The first step in national nuclear safety admin-
ta populations. For migrating hydrobionts, the levels of ex-
istration is the development of a regulatory infrastructure,
posure will be one or two orders of magnitude lower.
laws and regulations to specify the criteria against which
Thus, based on the results of the studies carried out by
safety is judged.
NATO and Russian Navy experts, it can be concluded that
To achieve and maintain a high level of nuclear safety
the threats posed by radionuclides in the wreckage of the
worldwide through the enhancement of national measures,
Komsomolets submarine are minor.
an International Convention on Nuclear Safety was signed
under IAEA auspices in 1994 (IAEA 1994). National regula-
tions understandably vary in detail from country to country.
8.5.4. Summary
The Nuclear Safety Convention stipulates the minimum re-
The major contribution ( 15 000 manSv) to the collective
quirements for the operation of a nuclear installation; for
dose to Arctic populations results from fallout from nuclear
example, it prohibits the operation of a nuclear installation
weapons testing. The range of individual dose commitments
without a license and it requires comprehensive and system-
to members of the Arctic population resulting from fallout is
atic safety assessments to be carried out prior to licensing
1-150 mSv. The next most important contribution ( 500
and throughout the lifetime of the installation.
manSv) to collective dose within the Arctic derives from the
The Convention also requires that, when necessary, a
Chernobyl accident. Individual dose commitments associ-
Contracting Party shall ensure that all reasonably practica-
ated with releases from Chernobyl were in the range 1-50
ble improvements are made as a matter of urgency to up-
mSv. Individual annual doses to the most exposed residents
grade the safety of an existing nuclear installation. If such
of the Arctic from Chernobyl releases were, however, about
upgrading cannot be achieved, plans should be implemented
10-20 mSv/y. Countermeasures introduced by several coun-
to shut down the installation as soon as practically feasible.
tries following the Chernobyl accident and, in some cases,
The primary risk posed by a nuclear reactor relates to the
being maintained to the present day, resulted in individual
large amount of radioactive material, primarily fission prod-
doses and dose commitments being much smaller than they
ucts, that are generated during its operation. The release of
would have been if no intervention measures had been in-
only a small fraction of these to the environment could cause
troduced. The effectiveness of these measures can be judged
severe harm to humans and to the environment. Therefore,
by the 90% reductions in dose to native people that were
one of the main aims of nuclear safety is to prevent the re-
achieved in Fennoscandia. The next most important collec-
lease of radioactive fission products into the environment.
tive dose contribution ( 50 manSv) results from releases
To prevent such releases, a number of barriers are placed
from the Sellafield fuel reprocessing plant but the contribu-
between the primary risk source (i.e., the fuel) and the envi-
tion of this source to individual dose has been relatively
ronment. In addition, the nuclear chain-reaction in the reac-
small (i.e., in the range 0-0.05 mSv) (UNSCEAR 1993).
tor core should be self-controlling or inherently stable, so
Smaller-scale releases from accidents in military opera-
that small perturbations in operating conditions always
tions, such those in northern Russia, the plutonium spill at
cause the reactor to return to normal conditions by itself.
Thule and the loss of the Komsomolets submarine in the
To cope with abnormal conditions, the reactor should pos-
Norwegian Sea, have resulted in no significant increases in
sess effective and expedient shut-down capabilities.
radiation exposures within Arctic populations. However, a
limited number of military personnel have been exposed to
Safety criteria
significant doses in connection with accidents on nuclear
Comparisons between some Western and Eastern safety cri-
vessels. It has been estimated that the collective dose to per-
teria have been made for example, by the Nordic Nuclear
sonnel operating Russian nuclear vessels was in the range
Safety Research (NKS) program (NKS 1994). The following
17-74 manSv with the highest individual doses in the range
observations have resulted from these studies:
0.2-0.6 Sv.
· In the West, safety design has often been demonstrated
For some previous releases to the environment, it has
through tests and experiments in pilot plants. These de-
been difficult to carry out an assessment of the collective or
monstrations have shown the functioning of the system in
individual doses specifically to Arctic populations. Such
question and helped to verify computer codes developed
cases include releases from the Mayak plant and releases
for analysing the safety of nuclear power plants.
from Sellafield for which an assessment in relation to the
· In contrast, in the former Soviet Union, safety design was
Arctic area was based on a global assessment.
often based on calculations rather than experiments. How-
ever, the systems were often designed conservatively, such
that pipe dimensions, the number of pumps, the size of
8.6. Source-related assessments
vessels, etc., were larger than necessary. This compen-
of potential releases
sated for some of the uncertainties in the calculations and
the lack of experiments and verification of codes.
This section discusses potential releases that may occur in
the future: as a result of accidents within the nuclear power
Differences between Eastern and Western practices re-
and weapons industries; from contained sources of radionu-
garding some basic safety criteria are described in more de-
clides within the marine and terrestrial environments; as a
tail below.
result of failures of containment structures for radionuclides
Western safety concepts give priority to measures for acci-
stored in, or previously released to, restricted areas of the
dent mitigation and accident management as well as auto-
environment; and from accidents involving nuclear weapons.
matic actions of safety systems. To relieve operators and to
586
AMAP Assessment Report
reduce the response time of protection systems, a progressive
The number of failed fuel rods resulting in direct con-
concept of protection by automatic control is applied.
tact between fuel and coolant must be less than 0.1%
The barriers present between the primary risk source and
of the total number of fuel rods (West and East).
the environment differ in the West and the East as follows:
· During accident conditions the following limits are applied:
West
East
The maximum cladding temperature must be less than
Fuel matrix
Fuel matrix
1200°C (West and East).
Fuel cladding
Fuel cladding
Local depth of oxidation of fuel cladding must be less
Pressure boundary of
Pressure boundary of
than 17% (West) or 18% (East) of the original cladding
primary coolant system
primary coolant system
thickness.
including reactor vessel
including reactor vessel
The NKS comparison among the respective western and east-
Reactor containment
Confinement
ern safety criteria indicates that little difference exists between
Filter --
them. Overall, the largest difference is the lack of full con-
According to the relevant safety criteria, protective mea-
tainment in many of the reactors of the former Soviet Union.
sures are realized in the West at four, and in East at three,
The values stipulated in Finland are presented as exam-
different safety levels:
ples of the dose and activity levels related to the technical re-
quirements at various safety levels. In a design basis accident,
West
East
the limit for the dose to an individual in the population aris-
Normal operation
Normal operation
ing from external radiation in any period of one year and as-
Transient conditions
Upset conditions
sociated radioactive materials intake is 5 mSv. In the case of
Design basis accident
Design basis accident
a beyond-design basis accident, the limit for the release of
Incidents beyond
radioactive materials arising from such an accident is that
design basis accidents
--
which causes neither acute harmful health effects to the pop-
The principal aim of all western safety considerations is
ulation in the vicinity of the NPP nor any long-term restric-
to ensure that the radioactive materials present in a nuclear
tions on the use of extensive areas of land and water. For
power plant are confined at all times. In other words, a nu-
satisfying the requirement relating to long-term effects, the
clear power plant must be designed and operated in such a
limit for atmospheric releases of 137Cs is 100 TBq. The com-
way that, at all times, during specified normal and upset op-
bined fallout comprising nuclides other than caesium-isotopes
eration and during the so-called design basis accidents, the
shall not cause, in the long-term (starting three months from
following design goals must be fulfilled:
the accident), a hazard (dose equivalent) greater than that
arising from a radiocaesium release corresponding to the
· The reactor can be safely shut down and be kept shut
specified limit. The probability that, as a result of a severe
down;
accident, the above mentioned requirement is not met, shall
· The residual heat can be removed;
be `extremely small' (STUK 1992).
· The radiation exposure of personnel and radioactive re-
After the Chernobyl accident, most countries introduced
leases to the environment must be kept as low as possible.
enhanced safety measures, such as the introduction of filters
To achieve these design goals, the safety precaution princi-
to reduce accidental releases of radionuclides in particulate/
ples were set up with a multiple level safety concept as fol-
aerosol form. In Sweden, requirements for such additional
lows:
safety measures were formulated by limiting releases, even in
very severe accidents, to noble gases and to less than 0.1%
· Assurance of normal operation with least possible
of the core inventory of long-lived radionuclides such as
occurrence of abnormal operating conditions;
137Cs (Sweden 1982). The implementation of such measures
· Control of abnormal operating conditions that might
took place between 1985 and 1989.
occur through the application of engineered safety
features; and
· Assurance that design basis accidents stay within given
8.6.1.2. Probabilistic safety assessment (PSA)
limits with assurance of dose minimization by means
of engineered safety features.
Previous safety analyses that needed to be performed before
licensing and throughout the life of a NPP were mostly de-
Furthermore, the so-called `single failure criteria' must be
terministic and did not address the probabilities of events.
fulfilled; that is, the safety systems must comply with the de-
Compliance with safety requirements was checked by deter-
sign criteria even under the assumption of a single compo-
ministic analyses that used pessimistic assumptions to ensure
nent failure in one of the safety systems.
that the results of the assessments were conservative (i.e.,
When analysing emergency conditions, the following cri-
`on the safe side'). However, the lack of events of safety sig-
teria are applied, with the differences between Western and
nificance in the operating experience of NPPs does not pre-
Eastern practice indicated in parentheses:
clude the existence of underlying safety deficiencies. Safety
· With the reactor at rated power, (a maximum diameter
experts considered that a gap existed between the bulk of
pipe break (West); a break of a pipe with a diameter of
operating experience and events of safety significance that
500 mm (East)) with a two-way free outflow of coolant
could only be bridged by theoretical analyses. PSA is a sys-
(a so-called guillotine break) is postulated to be the de-
tematic approach to performing such analyses (IAEA 1992).
sign basis accident. (Note: for some of the oldest Russian
Thus, currently, both deterministic and probabilistic safety
reactors, the pipe break diameter was limited to 32 mm.
analyses are used to complement each other in nuclear safety.
The pipes then had flow-reducing orifices).
In practice, PSA aims at:
· Regarding fuel, the following design limits are applied to
· Identifying and delineating the combinations of events
normal operation:
that may lead to a severe accident.
The number of failed fuel rods with gas leakage must
· Assessing the expected probability of occurrence for
be less than 1.0% of the total number of fuel rods
each combination.
(West and East).
· Evaluating the consequences.
Chapter 8 · Radioactivity
587
There are three levels of PSA:
· Fire risks are relatively high.
· Level 1 provides a review of plant design and operation,
· Redundancy of the safety systems is lower as they are
focusing on sequences that could lead to core damage.
not designed for large leaks or severe accidents.
· Level 2 addresses, in addition to the analyses on Level 1,
· During an accident, the high ventilation stack will be
the phenomenon of a core damage accident, the response
closed and the radionuclides will be blown, via openings,
of the containment to the expected loads, and the trans-
to the inner yard of the plant site.
port of radioactive material from the damaged core to the
· In accidents where the radionuclides are released straight
environment.
above the reactor, no system has been designed to control
· Level 3 addresses, in addition to the requirements of lev-
releases or to mitigate the consequences.
els 1 and 2, the dispersion of radionuclides in the sur-
· In a core melt accident, there is apparently a higher prob-
rounding environment and potential environmental and
ability of larger releases than in units fitted with contain-
health effects.
ment.
Levels 1 and 2 PSAs are fairly well-established tools for nu-
Because of the similar thermal-hydraulic features in the Lo-
clear safety, although not carried out for all NPPs relevant
viisa and Kola NPPs, the accident analyses made for the Lo-
to this assessment. PSA Level 1 studies are available for all
viisa plant can also be used to illustrate the behavior of the
Finnish and Swedish reactors. Level 2 studies for these reac-
Kola NPP. However, the probability of a core damage acci-
tors are underway and have been partially reviewed by safe-
dent in the Kola NPP cannot be assessed from the results of
ty authorities. Level 3 PSAs were not available to the assess-
the PSA made for Loviisa NPP. A systematic PSA for Kola
ment group for any of the NPPs relevant to effects in the
NPP would be the only way to estimate the core damage risk
Arctic. Typical official or unofficial criteria used in some of
for the plant and such an assessment has not been made avail-
the western Arctic countries are 1.0
104 to 1.0
105
able to the AMAP radioactivity assessment group. However,
(core damage probability/year) for Level 1 and 1.0
106 to
the assessment group was informed that the present techni-
1.0
107 (for large off-site releases/year) for Level 2 (NEA
cal and protection devices are not adequate to retain the ra-
1994). The Russian legal criterion for `equipment failure or
dioactive products inside the plant in cases of severe low-
active reactor zone melting' is 1.0
105 per reactor year
probability accidents. Accordingly, the consequences would
(Tsaturov 1996).
exceed those of the design basis accident (Tsaturov 1996).
Most of the studies in the literature dealing with the con-
The core damage frequency for the Loviisa plant is of the
sequences of nuclear accidents do not take into account the
order of 1.0
104 per year, which is slightly higher than the
probabilities of abnormal events. They also use varying as-
value for the other Finnish plant at Olkiluoto comprising
sumptions of source terms, meteorological conditions, sea-
BWR-type reactors (STUK 1991). The present PSA Level 1
sons, combinations of extreme conditions, etc., and are
studies for all Swedish reactors result in an estimated core
therefore difficult to compare. However, they are still useful
damage frequency of 1.0
105 per year (Swedish Institute
for emergency preparedness planning. Emergency response
for Radiation Protection and Swedish Nuclear Power In-
preparedness is, in practice, also an important consideration
spectorate 1995).
when considering the ultimate consequences of an accident.
Some studies take the efficiency of protective measures into
account when predicting accident consequences.
8.6.1.3. Studies to assess the consequences of major
As an example, we can qualitatively compare the Loviisa
reactor accidents
NPP in Finland and the Kola NPP in Russia as they are both
There are considerable shortcomings in the analyses avail-
of the VVER-440 design (Rantalainen 1995). Kola NPP
able to the AMAP radioactivity assessment group that allow
Units 1 and 2 are of model 230 which does not have an
conclusions to be drawn about the probability and conse-
accident localization system. The design basis accident is a
quences of potential accidents in nuclear power plants in the
break of a primary circuit pipe with a diameter of 200 mm.
Arctic. Despite these limitations, some general conclusions
Kola NPP Units 3 and 4 are of model 213 having accident
can be drawn from consequence analyses which focus on
localization systems. The design basis accident is a break of
short-period consequences and external doses. It should be
a pipe having a diameter of 500 mm. The accident localiza-
stressed, however, that without associated probability val-
tion system is not leak-tight like a containment, but consists
ues, consequence analyses are a less than satisfactory basis
of rooms, suppression pools and sprays to delay and reduce
for such an assessment. Consequence analyses have been
any radioactive releases. The Loviisa NPP has two units of
made for most NPPs of relevance to the Arctic and provide
VVER design model 213. The design basis accident is a
some idea of the order of magnitude of consequences that
double-ended break of a 500 mm diameter primary circuit
can be expected if a severe accident occurs.
pipe. Both units have free-standing steel containments sur-
Studies on two different types of reactor will be given as
rounded by concrete walls with internal and external spray
examples. In both cases, the scenario can be considered to
systems. The Kola and Loviisa NPPs have several positive
represent a severe, but not worst-case, accident. Neither sce-
common features, e.g., the water volumes in the primary
nario, however, represents an extremely unlikely combina-
and secondary circuits are large compared with those in
tion of situations although no calculated probability for
typical western PWRs making energy densities low and fuel
these cases can be given. The first example is a study of one
cooling effective. Despite its positive features, the Kola plant
1000 MW RBMK unit of the four at the St. Petersburg (Le-
has some special risk characteristics, namely:
ningrad) NPP (Ilvonen et al. 1994). This is the same type of
· Two reactors are in the same building.
reactor as the one at Chernobyl that suffered an extremely
· There is no leak-tight full-scale containment.
serious accident in 1986.
· Emergency cooling systems are more limited.
Improvements have been made at the St. Petersburg plant
· The axes of the turbo-generators are parallel to the reac-
since the Chernobyl accident. Therefore, the release for the
tors, causing a higher missile risk to the control room
study is assumed to be smaller. It corresponds to 100% of
and reactor cooling systems in the event of turbine disin-
the noble gases, 10% of the more volatile radionuclides,
tegration.
such as radioiodine, radiocaesium, etc., and 1% of the other
588
AMAP Assessment Report
fission product radionuclides. A 3-hour release at heights of
20 and 100 m is assumed.
The calculations give the following orders of magnitude
for dose rates and doses for outdoors external and inhala-
tion pathways north of the Arctic Circle:
Norway
Malaya
· External dose rate from cloud gamma is 0.01-0.1 Sv/h
Linakhamari
Lopatka Guba
Ara
Vidyaevo
30 hours after the release.
Andreyeva Bay
Guba
Chan-Ruchei
· Dose received from cloud gamma within 96 hours to a
Nerpichya Gula
Gadzhiyevo Guba
person outdoors is 0.1 to 10 Sv.
Bolshaya
Ura Guba
Sayda Guba
Lopatka Bay
· Dose commitment from radionuclides inhaled outdoors
Gremhika
Olenya
Nerpa
Murmansk
Pala Guba
Polyarnyy
during 96 hours is 10-100 Sv.
Arkhangelsk
· Dose from deposited gamma in the first year is 0.01-1 mSv.
Atomflot
Severomorsk
Sevmorput
Severodvinsk
Rosta Shipyard
These values are averages of about 3000 different dispersion
Russia
Radon Murmansk
situations, representing expectation values with no presump-
Russia
tions regarding the meteorological conditions at the time of
Finland
the release.
Figure 8·66. Major Russian naval bases along the Kola Peninsula and
The food pathways are not described because they are
White Sea.
highly dependent on the season in which the accident occurs
and on local habits, which were not taken into account suffi-
breaker fleet of the Murmansk Shipping Company). The ex-
ciently in the study. However, the doses from local foodstuffs
perience of previous radiation accidents on Russian nuclear-
in the first year would be higher than from surface deposited
powered vessels (see section 8.5) indicates that accidents in
gamma-emitting radionuclides (assuming both pathways are
the open sea, both underwater and on the surface, pose nei-
without protective measures), with the magnitude depending
ther threats of serious radioactive contamination of the ter-
on food habits.
restrial environment nor significant public exposures. In
It was concluded by the AMAP radioactivity assessment
these events, however, vessel crews are usually subjected to
group that there is no risk to Arctic residents of determinis-
significant radiation exposure. In contrast, accidents involv-
tic health effects from releases at NPPs situated further than
ing nuclear-powered vessels located in harbors and coastal
about 1000 km from the Arctic Circle. Nevertheless, because
bases, which may occur during fueling operations (refueling,
of the particular ecological conditions in the Arctic (cf. sec-
unloading of nuclear reactors, etc.), can pose significant ra-
tion 8.7), it is likely that contamination of some food path-
diation risks to the population of surrounding territories.
ways could impose a requirement for protective measures.
The following discussion of potential accidents occurring
The second study given as an example deals with the
with nuclear-powered vessel reactors is based entirely on in-
Kola NPP (Rantalainen 1995) which is situated within the
formation provided by the Russian Federation.
Arctic. It assumes a hypothetical accidental release, at a
Because of the presence of the Russian Northern Fleet
height of 100 m, of 100% of the noble gases, 10% of the
bases (Figure 8·66), northwest Russia contains the highest
easily volatile radionuclides, and 1% of the less volatile ra-
concentration of nuclear-powered vessels and nuclear reac-
dionuclides of the radionuclide inventory of the core of a
tors in the world. In addition, the civilian Russian icebreaker
445 MW NPP after a long period of steady operation at full
fleet is based in Murmansk. For comparative purposes, the
power. The calculations show that, with high likelihood, the
number of nuclear reactors operating in northern Russia in
doses are less than 1 Sv at distances greater than 5 km from
military and civilian categories with their aggregate invento-
the NPP during the first 24 h, and smaller than 0.1 Sv at dis-
ries of 90Sr plus 137Cs are shown in Table 8·43.
tances greater than 30 km. This means that, even without
Table 8·43. Nuclear reactors operating in northwest Russia.
protective measures, acute health effects are not expected at
distances greater than 5-10 km. A similar result was ob-
Number of
Approx. aggregate inventory,
tained by Amosov et al. (1995).
Type of reactor
reactors
(137Cs + 90Sr), PBq
The magnitude of external and inhalation doses in the
Military naval
120 a
2000
Kola area would be similar to those at any other site but
Civilian naval
11
200
contamination has more severe consequences in terms of
Nuclear power plant
4
300
dose commitment via food pathways than at lower lati-
a. Approximate number (NEFCO 1996).
tudes because of the ecological characteristics of the region.
There is no information available of the probability of acci-
Disposal of decommissioned nuclear-powered submarines
dents of this magnitude. Levels 1 and 2 PSAs would be re-
represents a complex technical, scientific, ecological and eco-
quired to obtain such information. If such an accident were
nomic challenge. About 99% of the total radionuclide in-
to occur, measures would be needed to ensure adequate pro-
ventory in submarine reactors is located in the spent nuclear
tection of the local population close to the NPP against acute
fuel (SNF). Accordingly, it is important to focus on safe de-
health (i.e., deterministic) effects and in the population in
fueling procedures and on the short and long term storage
the large scale (over thousands of km2) against late health
and disposal of spent nuclear fuel. At present, spent nuclear
(i.e., stochastic) effects
fuel with an activity of about 2000 PBq (137Cs + 90Sr) is
stored on land (40%) and on submarines taken out of ser-
vice (60%). The Murmansk Shipping Company also has an
8.6.2. Potential accidental releases from nuclear
estimated inventory of 500 PBq (137Cs + 90Sr) in a range of
vessels and nuclear storage sites
wastes stored on ships at its base.
There are significant risks of accidents during the routine
At the moment, a complete assessment of the risks and
operation of nuclear-powered vessels both military (i.e.,
impacts in relation to possible accidents with Russian spent
those in the navies of China, France, Russia, the United
nuclear fuel does not exist. However, some information is
Kingdom and the United States) and civilian (i.e., the ice-
available, the most relevant of which pertains to potential
Chapter 8 · Radioactivity
589
accidents associated with nuclear submarine decommission-
The individual doses can be segregated into three categories:
ing which has been provided to the AMAP radioactivity as-
· Emergency dose in the first few hours after an accident.
sessment group by the Russian Federation (RCRA 1997).
· Short-term dose up to ten days after an accident.
The results of radiation and safety analysis for submarine
· Medium-term dose for a period of up to a year after
refueling operations at the Sevmash enterprise in Severod-
an accident.
vinsk (on the coast of the White Sea) are presented in this
report. These analyses demonstrate that, in the case of de-
In these contexts, particular significance is given to the thy-
sign basis accidents, no inadmissible concentrations occur
roid dose.
outside the sanitary zone designated for refueling opera-
Under the basic criteria and requirements for ensuring
tions. Radiation risks to the public are associated with be-
safety in the siting of nuclear power plants set down by
yond-design basis accidents and the report presents the re-
the State Atomic Inspection Authority (Gosatomnadzor),
sults of the analysis of such accident scenarios involving
total and thyroid doses of less than 5 mSv and 50 mSv,
chain reactions.
respectively, do not require protective measures following
The most serious design-basis accident considered was a
an accident. The foregoing analysis showed that, while
loss of integrity of the primary cooling circuit as a subma-
the public beyond the shipyard would not be at great risk,
rine approaches the shipyard location for refueling. An as-
protective measures would clearly be required in relation
sumption was made of a rupture of a primary circuit pipe
to radioiodine and inhalation doses for those within the
with an aperture of 10 mm. About 90% of the gaseous fis-
sanitary zone. The cited document then deals with the re-
sion products and 8% of the other radionuclides in the
quirements for protective actions for this latter group of
coolant would be released to the reactor compartment. It
individuals.
was further assumed that the reactor compartment loses its
It still remains for the probabilities and doses associated
integrity and all such releases enter the atmosphere. How-
with accidents during the transport, storage and disposal of
ever, wind transport is not predicted to result in radionuclide
spent fuel from submarine reactors to be evaluated.
activity concentrations outside the restricted or `sanitary'
The summary conclusions of the Russian assessment are
zone exceeding the values prescribed for areas around nu-
as follows:
clear power stations. On the basis of this analysis, it is con-
· The beyond-design accident during dockside trials with a
cluded that design-basis accidents do not result in undue ra-
newly loaded core does not lead to significant radiologi-
diological dangers to the surrounding population in the
cal consequences to the near-field population. Exposures
urban area of Severodvinsk.
to individuals in the population outside the shipyard do
The largest beyond-design accident scenario that has
not exceed 5 mSv and no special protective measures are
been considered during submarine reactor refueling opera-
required. Within the shipyard, protective measures against
tions is the ejection of two or more control rods from
radioiodine ingestion and protection of respiratory or-
the core of the reactor when either the primary circuit is
gans and skin may be required.
pressure tested without the control rod retainers being in
· The major hazard in the cases of accidents involving a
place (as a result of an operator error) or through capsizing
chain reaction in spent cores is associated with the release
of the vessel without the control rod retainers in place.
of 137Cs and subsequent surface contamination. In such
Such an accident would result in a criticality event with
cases, decontamination of the shipyard and, possibly,
0.5
1020-1.5
1020 fissions. This is sufficient to melt the
parts of the territory beyond the shipyard, would be re-
reactor core. The core would achieve a temperature suffi-
quired.
cient to melt the fuel rods within three seconds following
· The dose to workers in the shipyard (excluding the per-
the ejection of the control rods. It is assumed that all of the
sonnel servicing the nuclear propulsion plant at the time
fuel rods lose integrity as a result of such an accident. Four
of the accident) within two hours of a reactivity accident
other specific scenarios for beyond-design basis accidents
in a spent core can be as high as several sieverts which
were considered:
can result in some deaths. The implementation of an
Type 1: Chain reaction in the fresh core to be loaded
emergency contingency plan could ameliorate these doses
(not relevant to decommissioning).
significantly.
Type 2: Chain reaction in the spent core with a cooling
· In the case of an accident near to the slip docks under ad-
time of 30 days.
verse wind conditions, a short-term dose to about 5000
Type 3: Chain reaction in the spent core with an cooling
city dwellers in Yagry could be higher than 0.5 Sv which
time of 90 days.
would require immediate evacuation of this portion of
Type 4: Chain reaction in the new core following dock-
the population.
side trials (the reactor is assumed to have worked
· The short-term dose to another 30 000 to 35 000 of the
for a month at 30% of the nominal power rating
city's population would range between 0.05 and 0.5 Sv.
with a subsequent cooling time of 1 day) (not rel-
· If a criticality accident in a new core during dockside tri-
evant to decommissioning).
als were to occur, total and thyroid doses could lead to
deaths among workers at the shipyard. However, the
It was assumed that the radionuclides are released into the
much reduced radiocaesium isotope release, compared to
atmosphere at a height of 20 m and that the contaminant
accidents involving spent cores, means that the conse-
cloud is transported and dispersed as a result of turbulent
quences for the external population are much less severe.
diffusion and advection. The exposure of individuals after
the accident occurs through four pathways:
This is the only study of its type made available to the
AMAP radioactivity assessment group and, although its
· External exposure from the cloud.
validity cannot be judged in isolation, it appears to be the
· Internal exposure due to inhalation.
kind of safety assessment that would be warranted for
· External exposures from contaminated surfaces.
evaluating the probability and consequences of reactor
· Internal exposures from the long-term consumption of
refueling accidents.
contaminated agricultural and fisheries products.
590
AMAP Assessment Report
into the river through groundwater. They calculate that
8.6.3. Potential releases from reprocessing plants
failure of ponds 10 and 11, the most likely ponds to be
The primary issue of concern here is the possibility of acci-
subject to such an event, could raise 90Sr activity concen-
dents in nuclear fuel reprocessing plants that could result
trations at Salekhard to 1000-2500 Bq/m3. The scenario of
in major releases of radionuclides to the environment.
release into groundwater has also assumed release of the
In an Arctic context, accidents in western European (i.e.,
complete inventory of ponds over a time determined from
at Sellafield and Cap de la Hague) and Russian (e.g., Ma-
information on current fluxes in groundwater. The esti-
yak) reprocessing plants would be of primary concern.
mates obtained, which represent a highly conservative
Unfortunately, no evaluations of the potential for accidents
worst-case scenario, suggest a flux at Salekhard of 11 000
and associated consequences have been provided to the
Bq/m3 from pond 9 (Karachay), and 10-20 Bq/m3 from
AMAP radioactivity assessment group on which to base
ponds 3 and 10.
an analysis.
The final scenario considered is that of release of activity
Also of relevance here are accidental releases from envi-
from the Asanov Swamps. Remobilization of the total in-
ronmental reservoirs into which discharges from nuclear
ventory (a worst-case scenario) would involve drying out of
fuel reprocessing plants have previously taken place. These
the marsh and flora and subsequent flood events washing
reservoirs include those associated with Russian nuclear fuel
organic matter down the River Ob. This scenario would in-
reprocessing operations in the drainage basins of the Ob and
volve the release of an additional 327 TBq of 90Sr within
Yenisey Rivers and coastal marine sediment areas close to
one year.
the discharges from nuclear fuel processing operations in
The movement of radioactive contaminants in the Yenisey
western Europe. Considering that the remobilization of ra-
River system has been studied by Vakulovsky et al. (1993).
dionuclides from western European coastal reservoirs is
They found that, of the principal contaminants, 24Na and
51
likely to be of limited importance to the Arctic, emphasis
Cr were transported in dissolved form, whereas 46Sc, 54Mn,
58/60
here has been devoted to the remobilization of radionuclides
Co, 59Fe and 65Zn were transported predominantly by
derived from Russian reprocessing operations in the Ob and
suspended particles. Transport of 32P and 137Cs was more
Yenisey drainage basins.
evenly distributed among solution and sediment phases.
There has been widespread recognition that nuclear facili-
These differences were reflected in the activity concentra-
ties in central Siberia have the potential to release significant
tions of the radionuclides in water samples. The authors
quantities of radionuclides to tributaries of the Ob and Ye-
concluded that, of the long-lived radionuclides, only 137Cs
nisey drainage basins, which ultimately discharge into the
could reach the Kara Sea, with an average discharge of be-
Arctic Basin. Several open reservoirs of radionuclides from
tween 0.8 and 2.8 TBq/y. Panteleyev et al. (1995) concluded
previous releases from nuclear fuel reprocessing activities
that most of the 137Cs and 239,240Pu in sediments of the Ob
exist in the drainage basins of these major rivers. Concerns
delta was derived from nuclear weapons testing and that the
regarding potential releases in the present context relate to
contribution from central-Siberian reprocessing plants was
the behavior, movement and effects of radionuclides cur-
small by comparison.
rently retained within fluvial systems and the consequences
The model estimates for the discharge of 90Sr assume
of possible future accidental releases from fuel reprocessing
complete mobility. As the transport of radionuclides will
operations. In view of the considerable distance of central
involve both the dissolved fraction and transport of conta-
Siberia from the Arctic, consideration is given here only to
minated sediments, in reality the rate of diffusion toward
the future mobilisation and the effects of radionuclides cur-
the Kara Sea will depend on interactions with soils, sedi-
rently retained in environmental reservoirs as a result of pre-
ments and biota. For instance, Trapeznikov et al. (1995)
vious operational releases.
found that, in the River Techa, 250 miles downstream of
Mayak, the activity concentration of 90Sr in water was
roughly halved, 239,240Pu decreased fourfold and 137Cs de-
8.6.3.1. Mobilisation of radionuclides released to
creased by an order of magnitude, reflecting the relative
the terrestrial environment
particle reactivities of these radionuclides. The rivers carry
Modeling has focused particularly on 90Sr because of its rel-
the greatest sediment load at times of greatest flow during
atively high aqueous mobility (i.e., low particle reactivity).
the late spring-melt.
The models have conservatively assumed no interaction with
The rate of transport downriver will therefore be influ-
sediments or biota. At Salekhard (the Ob river/estuary inter-
enced both by the distribution of a contaminant among sol-
face), the current flux of 90Sr is assumed to be dominated by
uble and particulate phases and its affinity for different par-
surface runoff of atmospheric fallout with, on average, an
ticle size fractions. Tronstad et al. (1995) found that 90Sr
annual discharge from the Ob River of around 10-40 TBq.
was remobilized from Asanov Swamp sediments more read-
This flux has been used as a benchmark against which the
ily than 137Cs or 239,240Pu. They concluded that the Asanov
effects of scenarios of additional release upriver can be com-
Swamp area could act as a long-term source of contamina-
pared using fluvial transport models.
tion of the River Techa.
A model analysing available data relating to the move-
The sorption of a radionuclide is often modeled using an
ment of radiostrontium, radiocaesium and plutonium
equilibrium distribution or partition coefficient, Kd. A lower
through the Ob River system has been developed by Palu-
Kd indicates enhanced mobility, i.e., a greater proportion of
szkiewicz et al. (1995). Transport of radioactive contami-
the radionuclide is in the soluble phase. In an open system,
nation downstream is governed by meteorological/hydro-
such as a river, changes in physical, chemical and biological
logical forcing, decay, sediment partitioning and radionu-
parameters can influence the Kd value. Radionuclides can be
clide release conditions. Based on an estimate of the current
transferred between sediment and water phases by chemical,
activity concentration of 90Sr in the River Ob at Salekhard
physical or biological processes. Physical-chemical changes
of 25-100 Bq/m3, these authors have calculated scenarios
in water characteristics can change the equilibrium parti-
for releases from the storage reservoirs at Mayak, both on
tioning conditions resulting in sorption or release of radio-
a scenario of dam failure, assuming total release of the in-
nuclides. Chemical mobilisation includes ion-exchange,
ventory within a year, and on scenarios of a steady flux
leaching and dissolution; biological processes can effect both
Chapter 8 · Radioactivity
591
chemical and physical mobilisation. In addition, physical
was recorded in 1993. This was due to an increase in re-
transport can occur due to natural or anthropogenic resus-
servoir volume after an unusually wet period. Similarly,
pension of sediments.
discharge of 90Sr through the right- and left-bank con-
Changes in ionic composition of water can affect Kd
duits reached 14 GBq and 100 GBq, respectively, in 1993
significantly. Kd values for both 90Sr and 137Cs in river wa-
(NRPA 1997).
ter and ice are higher than those in seawater and sea ice.
The risk of resuspension of contaminated spray and se-
The mobilisation of sediment-associated 137Cs and 90Sr
diments has been mitigated by a project to fill in the lake
may be caused by changes in pH, ionic strength, salinity
by depositing concrete sections on the lakebed. However,
and/or concentrations of exchangeable elements.
the potential remains for contamination of the water table
Ice may be an important mechanism for the transport of
through expansion of a lens of contaminated groundwater
sediment-bound radionuclides in the Arctic Sea. The extent
resulting from diffusion of radionuclides through the lake-
to which sea ice could have a role in transporting sedi-
bed. Currently, the lens of contaminated groundwater ex-
ments deposited in estuaries is not well known. It appears
tends over an area of about 10 km2 beneath Lake Kara-
that contaminated sediments can be transported consider-
chay to a depth of around 100 m. The lens straddles a sub-
able distances in the sea ice mass. The direction of trans-
terranean watershed with complex hydrogeology. Depend-
port is much dependent on the drift of ice within the polar
ing on the nature, capacity and replenishment of the aqui-
pack. Meese et al. (1995) suggest that sea ice is a primary
fers, expansion of the lens can be northward toward the
transport mechanism by which contaminated sediments
area of the storage ponds or southward toward the River
are redistributed throughout the Arctic Ocean.
Mishelyak, a minor tributary of the Techa River. Lateral
expansion of the lens in this area may be enhanced by the
use of groundwater as a source of potable water by some
8.6.3.2. Mayak
local villages. The primary radiological concern arises be-
The Mayak facility is situated around the headwaters of
cause of the mobility of 90Sr, which may be enhanced by
the River Techa, which ultimately drains into the Kara Sea
the presence of dissolved organic compounds in the dis-
via the Ob River. The system of waste management has
charged waste. Caesium-137 is less mobile and is largely
been dependent on a series of natural and artificial reser-
sorbed to lake sediments.
voirs and drainage canals. A total of 4000 PBq (decay cor-
Severe flooding of the Asanov Swamps, perhaps in con-
rected to 1994) comprising mainly 137Cs and 90Sr has been
nection with failure of dams in the pond chain above, could
released to this restricted system (NRPA 1997). The princi-
result in substantial releases of 90Sr and 137Cs to the Techa
pal risks of release are from leakage through the walls of
River system.
dams and conduits and from drainage through lakebeds
into groundwater. This latter scenario is relevant to the
8.6.3.3. Tomsk
situation around Lake Karachay. A further source is exist-
ing contamination in the Asanov Swamps that has accu-
Storage ponds at the site are believed to be contaminated
mulated from accidental and routine discharges during the
to a similar degree as Lake Karachay. They contain an esti-
period of operation of the plant.
mated 1.3
108 Ci ( 4800 PBq). As at Mayak, there are
Failures in the waste containment in this system could
concerns about potential contamination of groundwater,
result in discharges of radionuclides to the Techa River.
compounded by the proximity of the Tom River and the
Such failures include partial or total dam failure, leakage
relatively high groundwater level, the water table of which
from the reservoirs into the neighboring diversionary con-
is only 20 m beneath the surface.
duits that discharge into the Techa River below the lowest
reservoir (reservoir no. 11), and overflow of reservoirs as
8.6.3.4. Krasnoyarsk
they approach capacity. Total failure of any of the dams,
but particularly of dam no. 11, would potentially release
The storage ponds at Krasnoyarsk are believed to contain
a great volume of contaminated water and sediment into
an inventory of about 5
104 Ci ( 2 PBq). As at the other
the Techa River and Asanov Swamps below the dam. Re-
sites, there is a risk of contaminated groundwater migrat-
servoir no. 11 has a volume of 216
106 m3.
ing into rivers, in this case the Yenisey River.
Further release scenarios identified include filtration of
contaminated water through the dams of reservoirs nos 10
8.6.4. Radioactive wastes dumped at sea
and 11, and leakage through the beds of the reservoirs into
the aquifers beneath. Leakage is monitored through a sys-
The former Soviet Union dumped high, intermediate and
tem of boreholes. The potential importance of the former
low level radioactive waste in the Arctic seas during the
pathway may increase as it is thought that the sorption ca-
years 1959-1991. In spring 1993, the Russian Federation
pacity of the rock strata in the vicinity of reservoir no. 11
published a report, the so-called `White Book' (Office of
is almost exhausted and contamination will enter the River
the President of the Russian Federation, OPRF 1993) that
Techa more readily than previously. Contamination of
included information on sea dumping operations. Accord-
groundwater from the smaller storage reservoirs nos 3 and
ing to this report, the total amount of radioactivity dumped
4 has not been determined.
in the Arctic Seas was approximately 90 PBq at the time of
Leakage of waste from reservoirs nos 10 and 11 has
dumping. The items dumped included six nuclear subma-
been identified in the diversionary conduits that discharge
rine reactors and a shielding assembly from an icebreaker
into the Techa River below the reservoirs as well as in wa-
reactor each containing spent fuel with an aggregate inven-
ter seeping through the dam of reservoir no. 11. The rate
tory of 85 PBq; ten reactors (without fuel) containing 3.7
of seepage increases with increasing volume in the reser-
PBq; liquid low-level waste containing 0.9 PBq; and solid
voir. The discharge of 90Sr into the Techa River after seep-
intermediate and low-level waste containing 0.6 PBq. The
age through the dam of reservoir no. 11 has increased
packaged and unpackaged solid waste and the nuclear re-
steadily during the last 15 years. Despite some remedial ac-
actors were dumped in the fjords of Novaya Zemlya at
tion in the period 1990-1993, a peak discharge of 27 GBq
depths of between 12 and 135 m, and in the Novaya Zem-
592
AMAP Assessment Report
8.6.4.1. Surveys of dumped objects
A Joint Norwegian-Russian Expert Group was established
in 1992 to investigate radioactive contamination of northern
areas as a result of the dumping of nuclear waste in the Bar-
ents and Kara Seas. Exploratory cruises to the dumping
areas were conducted in 1992, 1993 and 1994. Using high-
frequency side-scan sonar and a remotely operated vehicle
(ROV) equipped with an underwater video camera, a NaI-
Blagopoluchiye Fjord
Techeniy Fjord
detector and a sediment sampler, attempts were made to
Sedov Fjord
identify and examine the dumped wastes. All of the four
Oga Fjord
Tsikolka Fjord
sites where spent nuclear fuel was dumped were visited, but
Novaya Zemlya Trough
only some of the objects containing high level waste were
Stepovogo Bay
Abrosimov Bay
successfully located. However, dumped vessels and numer-
ous containers of low level waste were also located.
The Norwegian-Russian Expert Group also took sediment,
Kolguyev
seawater and biota samples, both in the immediately vicinity
Island
of the dumped objects and in the surrounding area. The lev-
els of radionuclides in waters, sediments and biota in the
Kara Sea are very low compared to other marine systems,
Liquid waste
e.g., the Irish, Baltic and North Seas (NRPA 1996). Never-
theless, levels in the immediate vicinity of dumped low level
Solid waste
waste containers indicate that some leakage has occurred.
Figure 8·67. Locations of sea dumping of radioactive waste in the Russian
Arctic.
8.6.4.2. International Arctic Seas Assessment Project
(IASAP)
lya Trough at a depth of 380 m. The liquid low-level waste
was discharged in the open Barents and Kara Seas. Figure
While it appears that no significant global or regional effects
8·67 shows the locations of these dumping sites. Tables 8·44,
have yet resulted from the dumping of radionuclide waste in
8·45 and 8·46 list types, locations and radioactivity of ob-
the Arctic, there is concern about the gradual deterioration
jects containing spent nuclear fuel (SNF), objects devoid of
of the waste containments that could lead to releases of ra-
SNF, and solid low and intermediate level wastes, respective-
dionuclides in the future. This could result in contamination
ly, dumped in Arctic marine areas as given in the `White
of the marine food chain and increased radiation exposures
Book' (OPRF 1993).
of human consumers of fish and other marine foodstuffs.
Table 8·44. Dumped objects containing spent nuclear fuel in Arctic seas (OPRF 1993).
Total
Dumped objects
Year
Location
Depth, m
activity, PBq a
Radionuclide content
Compartment of nuclear submarine no. 285
1965
71°56'2"N 55°18'5"E
20
29.6
Fission products
with two reactors, one containing SNF
Abrosimov Fjord
Compartment of nuclear submarine no. 91
1965
71°56'2"N 55°18'9"E
20
14.8
Fission products
with two reactors containing SNF
Abrosimov Fjord
Shielding assembly of reactor from
1967
74°22'1"N 58°42'2"E
49
3.7
137Cs (1.8 GBq), 90Sr (1.8 GBq),
OK-150 unit of icebreaker Lenin contain-
Tsivolka Fjord
{ 238Pu, 241Am, 244Cm} ( 70 TBq)
ing residual SNF (60% of fuel complement
Reactor from nuclear submarine
1972
72°40'N 58°10'E
0300
29.6
Fission products
no. 421 containing SNF
Novaya Zemlya Trough
Nuclear submarine no. 601 with
1981
72°31'15"N 55°30'15"E
50
7.4
Fission products
two reactors containing SNF
Stepovogo Fjord
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total (5 objects with 7 reactors containing SNF) 1965-1981
85.1
a. Expert estimates were made at the time of dumping.
Table 8·45. Objects devoid of spent nuclear fuel dumped in Arctic seas (OPRF 1993).
Total
Dumped objects
Year
Location
Depth, m
activity
Radionuclide content
Reactor compartment of nuclear submarine
1965
71°55'13"N 55°32'32"E
20
Requires
no. 254, containing two reactor assemblies
Abrosimov Fjord
special analysis
Reactor compartment of nuclear submarine
1966
72°56'2"N 55°8'5"E
20
Requires
no. 260, containing two reactor assemblies
Abrosimov Fjord
special analysis
Nuclear power plant of icebreaker Lenin containing 1967
74°26'4"N 58°37'3"E
50
1.9 PBq
Mainly 60Co
three OK-150 reactors with primary coolant lines
Tsivolka Fjord
Two reactors from nuclear submarine no. 538
73°59'N 66°18'E
35- 40
Requires
1972
Techneniye Fjord
35- 40
special analysis
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total (5 objects with 9 reactors without SNF)
1965-1972
< 3.7 PBq at dumping
Chapter 8 · Radioactivity
593
Table 8·46. Low and intermediate level radioactive waste dumped in the
reactor operating history, detailed inventories of the radio-
Kara and Barents Seas (OPRF 1993).
nuclide composition of the wastes and the likely behavior of
Number
Years
Activity,
protective barriers with time. Attention was focused on the
of disposals
dumped
TBq.
high level waste, i.e., the dumped reactors and the container
of spent fuel from the icebreaker Lenin which obviously pose
Solid wastes
Novaya Zemlya Trough
22
1967-1991
123.0
the highest potential risks.
Sedov Fjord
8
1982-1984
126.0
During the IASAP project, general information about the
Oga Fjord
8
1968-1983
75.0
actual dumping operations was obtained (Yefimov 1994,
Tsivolka Fjord
8
1964-1978
99.0
Stepovoy Fjord
7
1968-1975
47.0
IAEA 1996). Fuel had been removed from ten of the reac-
Abrosimov Fjord
7
1966-1981
25.0
tors prior to dumping. Those dumped with spent fuel (six
Blagopoluchiye Fjord
1
-19811972
8.0
reactors) had usually suffered an accident prior to dumping,
Techeniye Fjord
3
1982-1988
68.0
Off Kolguyev Island
1
-19811978
1.5
in which the fuel had been damaged. The dumping of the re-
Zornaya Bay (Novaya Zemlya)
1
-19811991
11.0
actors took place by four methods: 1) Most of the subma-
Barents Sea
1
?
> 4.0
rine reactors were dumped within their respective reactor
Liquid wastes
compartments. 2) In some cases, the reactors were taken out
(Volume >189634 m3)
About 100 operations
900.0
of the compartment and placed in a special metal box prior
at 5 sites plus 6 accidental
releases elsewhere
to dumping. 3) In the case of the lead-bismuth cooled reac-
------------------------------------------------------------------------------------------------
tors, the submarine compartment was filled with bitumen
Total
1500.0
and the entire submarine was dumped (in this case, the so-
lidified liquid metal coolant forms an additional protective
Because the wastes are lying in shallow waters, the possi-
barrier). 4) The dumped components of the nuclear ice-
bility of radiation exposure by other routes, such as direct
breaker include a reactor compartment with three reactor
exposures following the movement and transport of the
vessels from which the fuel was removed. About 60% of the
waste packages by natural events (ice or storm action) or
fuel from one of the reactors was dumped in a separate metal
human actions cannot be totally ruled out. The International
lined concrete box. All of the reactors containing nuclear
Arctic Seas Assessment Project (IASAP) was established by
fuel and the icebreaker fuel box were filled with a special
the IAEA in 1993 to address these and other related issues
polymer mixture, FurfurolTM.
partly at the request of the London Convention 1972.
The total activity of the dumped reactors (with and with-
The objectives of the IASAP study were:
out nuclear fuel) at the time of dumping was re-estimated by
IASAP to be 37 PBq. This may be compared with the first
· To assess the risks to human health and to the environ-
estimate of ca. 89 PBq provided in spring 1993 by the Rus-
ment associated with solid radioactive waste dumped
sian Federation (see Tables 8·44 and 8·45). The total inven-
in the Kara and Barents Seas; and
tory of the dumped reactors had, by 1994, declined through
· To examine possible remedial actions related to the
radioactive decay to an estimated 4.7 PBq (Sivintsev 1994a,
dumped solid waste and to advise on whether they are
1994b, Yefimov 1994, IAEA 1996). The peak of radioactiv-
necessary and justified.
ity of the dumped material, 25 PBq, was reached in 1967,
The IASAP project was carried out by a multidisciplinary
when spent fuel from one of the reactors of the icebreaker
team of scientists from several countries within the normal
was dumped (Figure 8·68) (IAEA 1996, Sivintsev 1995,
procedures of the IAEA. The following approach was
NRPA 1996).
adopted:
Construction details of the steam generation installations
were analysed to determine the likely ingress routes for sea
· Examination of the current radiological situation in
water and the time after dumping at which this occurs. The
Arctic waters to assess evidence for releases from the
evaluation of radionuclide release rates was based on an
dumped waste;
analysis of the weak points of the protective barriers. Re-
· Prediction of potential releases from the dumped wastes
lease was assumed to be controlled by corrosion of the ma-
concentrating on the solid high-level waste objects con-
taining the majority of the radionuclide inventory;
Activity, kCi
· Modeling of the environmental transport of released
700
radionuclides and assessing the associated radiological
impact on humans and biota;
· Examination of the feasibility, costs and benefits of pos-
600
sible remedial measures applied to a selected high-level
waste object.
500
The study was divided into a series of five working areas:
1) source term reconstruction, 2) existing environmental
400
concentrations, 3) transfer mechanisms and modeling,
4) radiological impact assessment, and 5) assessment of
remedial measures.
300
8.6.4.2.1. Source term reconstruction
200
The information needed about the dumped radioactive
wastes for the purposes of assessing the radiological impact
100
and evaluating the need for remedial measures was acquired
1960
1970
1980
1990
2000
and evaluated by a special working group of IASAP. The
Figure 8·68. Predicted release rates of different radionuclide groups from
work involved obtaining knowledge of the characteristics of
the submarine reactor dumped in the Novaya Zemlya Trough (best esti-
the steam generating installations and nuclear fuel, data on
mate scenario).
594
AMAP Assessment Report
terials forming the reactor structure and nuclear fuel. The
The IASAP models differ, inter alia, in their spatial and
best available predictions for corrosion rates in an Arctic
temporal resolutions. Thus, the results from different models
environment were derived from simple computer models of
were used for different specific endpoints. For example, the
containment failure.
results from models with greater spatial resolution were used
It was assumed that all corroded material is immediately
for critical group calculations. For global collective dose cal-
released to the environment. This is a highly conservative
culations, which involve long time scales, radiological com-
assumption, as most of the corroded matter will be both
partment models were used.
heavy and insoluble and will be retained in the hull or reac-
For individual dose estimation, three critical groups were
tor pressure vessel until other barrier corrosion is well ad-
considered:
vanced.
· A group living in the Ob and Yenisey estuaries and on
The following release scenarios were used for impact as-
the Taimyr and Yamal Peninsulas whose subsistence is
sessment calculations:
heavily dependent on the consumption of locally-caught
(A) A best estimate scenario release occurs through grad-
Kara Sea fish, marine mammals, and seabirds and their
ual corrosion of the barriers, waste containers and the
eggs, and who spend 250 hours/y on the seashore. These
fuel itself.
habits are also typical of subsistence fishing communities
(B) A plausible worst case scenario normal gradual cor-
in other countries bordering the Arctic.
rosion followed by catastrophic disruption of two
· A hypothetical group of military personnel patrolling
sources at a single dump site (the fuel container and
the foreshore of the fjords of Novaya Zemlya containing
the reactor compartment of the icebreaker) in the year
dumped radioactive wastes, for assumed periods of 100
2050 followed by accelerated release of the remaining
hours/y. The exposure pathways considered include exter-
radionuclide inventory of these sources.
nal radiation and the inhalation of seaspray and resus-
(C) A climate change scenario corrosion up to the year
pended sediment.
3000 followed by instantaneous release, due to glacial
· A group of seafood consumers considered representative
scouring, of the radionuclide inventory remaining in all
of the northern Russian population situated on the Kola
sources.
Peninsula eating fish, mollusks and crustaceans harvested
from the Barents Sea. No consideration was given to the
The release rates with time resulting from each of these three
consumption of seaweed or marine mammals, nor to ex-
scenarios constitute the respective input terms for the model-
ternal radiation.
ing of environmental transport and exposure pathways.
The total annual individual doses in the critical groups of
seafood consumers (Groups 1 and 3) for all three scenarios
8.6.4.2.2. Consideration of possible criticality
are small and very much less than variations in natural
A preliminary analysis of the possibilities of criticality in the
background doses. For scenario A, the maximum individ-
dumped reactor assemblies concludes that run-away critical-
ual dose rates to members of the critical groups in the Ob
ity (k >> 1) is so extremely unlikely, given the moderating
and Yenisey estuaries, the Yamal and Taimyr Peninsulas
conditions, as to be ruled out. Nevertheless, the probability
and the Kola Peninsula are less than 107 Sv/y with the
of some criticality in some of the reactor assemblies is some-
doses on the Kola Peninsula being the lowest. These low
what higher and this might accelerate corrosion of barriers
doses were mainly delivered through fish consumption,
and lead to enhanced release of radionuclides. If this occur-
with 137Cs, 239Pu and 90Sr being the dominant radionu-
red in situ, it is unlikely to be of major concern, especially
clides. The maximum dose rates for the other two release
as the inventory of the reactor assemblies declines with time.
scenarios (B and C) are about one order of magnitude high-
If, however, it occurred during recovery of one of the assem-
er than those of scenario A. These maximum doses to the
blies, it could involve significant risks of direct radiation ex-
general public are essentially trivial. For members of the
posure for those involved in recovery operations. It is there-
hypothetical critical group 2, comprising military personnel
fore suggested that, before any decision to undertake reme-
on Novaya Zemlya, the maximum dose is estimated to be
dial action is taken, a thorough criticality study should be
3 mSv/y (for the plausible accident scenario B) derived
conducted.
primarily from external exposure and inhalation. This lat-
ter dose is high enough to warrant continued restrictions
on the occupancy of the foreshores of the fjords of Novaya
8.6.4.2.3. Pathway modeling and radiological assessment
Zemlya in which wastes have been dumped.
When the IASAP project was launched, there were prac-
Collective doses were estimated only for the best esti-
tically no radiological assessment models dealing with Arc-
mate release rate scenario (A). The collective dose to the
tic marine areas. In addition, very little information was
world population arising from the dispersion of radionu-
available on the oceanographic, sedimentological and bio-
clides in the world's oceans (for nuclides other than 14C and
129
logical conditions in the Kara Sea. Thus, the first task to-
I) were calculated for two time periods: 1) up to the year
ward the radiological assessment of the impact of the
2050 to provide information on the collective dose to the
dumped waste to human health and environment was to
current generation, and 2) over the next 1000 years, a time
develop realistic marine environmental transport models
period which covers the estimated peak releases. The esti-
for the region. The approach taken within the IASAP pro-
mated collective doses are 0.01 and 1 manSv, respectively.
ject was to involve several national modeling groups, each
Appropriate global models were used to calculate the col-
of them extending their models to the target area or cre-
lective doses associated with 14C and 129I, which are very
ating completely new models. At the same time, through
long-lived nuclides and circulate globally in the aquatic, at-
the work of the Norwegian-Russian Expert Group and
mospheric and terrestrial environments. Assuming that the
several Russian and international institutes, improved en-
entire 14C inventory of the wastes is released around the
vironmental information was gradually acquired (Strand
year 2000, integrating the dose to the world's population
et al. 1997, Pavlov 1994, Ivanov 1994, Sazykina and Kry-
over 1000 years into the future (i.e., to the year 3000)
shev 1994, IAEA 1996).
yields a collective dose of about 8 manSv. The correspond-
Chapter 8 · Radioactivity
595
ing value for 129I is much lower at 0.0001 manSv. Thus,
8.6.4.2.6. Conclusions of IASAP
the total collective dose over the next 1000 years to the
world's population from all radionuclides in the dumped
The IASAP study concludes that the radiological risks
waste is of the order of 10 manSv. In contrast, the annual
posed to human health and the environment by the ra-
collective dose to the world's population from naturally-
dioactive wastes dumped in the western bays of Novaya
occurring 210Po in the ocean is estimated to be about three
Zemlya and the Kara Sea are minor. It further concludes
orders of magnitude higher and the collective dose from
that remedial actions are not warranted on radiological
previous sea dumping of low-level radioactive waste in the
grounds. The existing restrictions on the occupations and
Northeast Atlantic Ocean is 1 manSv over 50 years and
habits of military personnel on Novaya Zemlya are suf-
3000 manSv over 1000 years.
ficient to prevent significant radiation exposures to humans
occurring. It is, however, recommended that attempts be
made to locate and identify all dumped high level waste
8.6.4.2.4. Effects on marine organisms
objects.
The IASAP study also included an evaluation of doses to
marine organisms arising from the dumped radioactive
8.6.5. Nuclear weapons
waste objects and the likelihood of their having effects on
populations of such organisms. The doses to marine organ-
History confirms that measurable risks exist of releases of
isms are orders of magnitude below those at which detri-
radioactive material following accidents involving nuclear
mental effects on populations might be expected to occur.
weapons (see, for example, section 8.5.3.2.). Platforms
Furthermore, these doses are delivered to only small pro-
(ships, aircraft, ground vehicles) carrying nuclear weapons
portions of the resident populations.
can be involved in accidents such as aircraft crashes, fires
onboard vessels and loss of vessels or aircraft at sea. There
are two main categories of concern in this context: the risk
8.6.4.2.5. Remediation
of a nuclear explosion; and the risk of releases of radioac-
The Contracting Parties to the London Convention 1972
tive constituents of weapons to the environment.
specifically requested the IAEA to study possible remedial
Weapons designers have used various features to mini-
actions and to advise on whether they are necessary and
mize the risks of nuclear explosion and release of fissile ma-
justified.
terial from weapons deployed in the field. These include:
A preliminary engineering feasibility and cost study was
· Permissive Action Links (PAL), believed to be used on all
conducted as a case study for the container of spent fuel
weapons. PAL is a mechanism that prevents a weapon
from the icebreaker. It was chosen because it has the high-
being armed before an authorized code is introduced to
est inventory of any of the dumped containers (2200 TBq)
an electronic arming system. PAL is intended to prevent
and is the best documented in terms of construction and
unauthorized ignition outside the chain of command.
mode of disposal. Furthermore, there is little risk of criti-
· The effectiveness of the PAL system can be easily evalu-
cality occurring during remedial measures.
ated in the laboratory.
A group of salvage experts defined two broad categories
· Enhanced Nuclear Detonation Safety (ENDS), a mecha-
of option for remedial measures that would warrant initial
nism to reduce the probability of a warhead detonation
evaluation from the perspectives of engineering feasibility
in an accident. However, ENDS is probably not employed
and cost:
in all nuclear weapons.
· Capping in situ with concrete or other suitable material
· One Point Safety, intended to eliminate the possibility of
to encapsulate the material.
criticality being attained if the conventional explosive ig-
· Recovery for land disposal.
nites at a single point. Implosion weapons can only be
properly initiated if a precise sequence of a series of con-
Both options were deemed to be technically feasible but
ventional explosive detonations occurs. One Point Safety
the costs (US$ 6-10 million) would be very high relative
prevents the correct detonation sequence being initiated
to the potentially avertable dose. The radioactive waste
if a single conventional explosive component is deto-
sources in the Barents and Kara Seas are predicted to give
nated. One Point Safety is believed to have been intro-
rise to future annual doses of less than 1 µSv to individuals
duced in all weapons.
in population groups bordering these seas. The risk of
· Insensitive High Explosives (IHE), these are chemical
fatal cancer induction from a dose of this magnitude is
high explosives that are less easily detonated by impact
about 5
108 which is a wholly trivial risk. The collec-
or fire than are conventional high explosives. A large
tive dose to the world's population over the next 1000
proportion of the USA stockpile of nuclear weapons
years from the dumped wastes in the Barents and Kara Seas
does not employ conventional charges containing IHE
is of the order of 10 manSv. A simplified approach to con-
and it is not known whether similarly insensitive chemi-
sidering collective dose in a decision-making framework is
cal explosives are used in weapons manufactured by
to assign a monetary value to the health detriment that
other nuclear countries.
would be prevented if remedial action was implemented.
· Fire-Resistant Pits (FRP), these are pits (i.e., Uranium-
Such a scoping approach indicates that remedial measures
Plutonium weapon cores) that are covered with high-
applied to the single largest source (the dumped spent fuel
melting-point metal shells that reduce the possibility of
package from the nuclear icebreaker) costing in excess of
dispersing plutonium through fire following an accident.
US $ 200 000 would not appear to offer sufficient benefit
FRP has only been introduced in a few USA weapon de-
to be warranted. If, on the other hand, remedial actions
signs and is not known to be used by other weapons
were to be taken for reasons other than radiological ones,
manufacturing countries.
it appears that doses to those involved in remedial actions
would not be large.
Thus, PAL is intended to prevent unauthorized detonation.
One Point Safety, IHE and ENDS are used to reduce the
probability of an unintended nuclear explosion by ensuring
596
AMAP Assessment Report
that the conventional charges do not ignite in a manner
sile material in such operations. There is a need for en-
that initiates a nuclear explosion in the event the weapon
hanced information on such activities and assurance that
is exposed to fire, mechanical shocks or fragment/bullet
adequate surveillance and safety measures are being im-
penetration. FRP is used to reduce the possibility of the
plemented.
dispersion of nuclear material from the core of a weapon
The dangers associated with weapons are clearly related
in the event of fire.
to concerns about nuclear proliferation. New nuclear na-
The problem remains that not all weapons employ all
tions may not be able to make safe weapons and any use of
these safety features. Modern weapons design is extremely
nuclear material by irresponsible factions or organizations
complicated and no precautions can entirely eliminate the
to construct weapons, even crude ones having virtually no
possibility of an unintended nuclear detonation. Neverthe-
incorporated safety features or deployment controls, would
less, an inadvertent nuclear weapons detonation has never
immeasurably enhance concerns about nuclear weapons
been known to occur and the risk is considered low.
accidents.
In contrast, there have been several cases of releases of
radioactive material from nuclear weapons as a conse-
8.6.6. Radionuclide thermoelectric generators
quence of accidents involving weapons platforms. The
Thule B-52 aircraft crash (see section 8.5.3.2.) and an inci-
Radionuclide Thermoelectric Generators (RTGs) have been
dent at Palomares, Spain, in 1966 are merely documented
used as power sources in remote areas because of their rug-
examples of such events. The greatest radiological risks
gedness and reliability. RTGs are deployed in both Alaska
posed by such accidents are the ejection of fine particulate
and Russia. The first USA RTG was installed in 1973, and
plutonium into air that may be inhaled. By comparison, the
nine additional units were installed in 1985. They were es-
radiological risks associated with marine-deposited weap-
tablished at remote sites to provide reliable electricity for
ons plutonium and uranium following accidents are rela-
the seismic stations established for nuclear treaty verifica-
tively minor.
tion. Electricity is generated by the Seebeck thermoelectric
On January 17, 1966, four nuclear weapons were inad-
effect in which an electrical potential is generated across a
vertently dropped from a USA aircraft in the vicinity of the
thermocouple exposed to a temperature gradient. The tem-
small village of Palomares in Spain. The conventional ex-
perature gradient is provided by the radioactive decay of a
90
plosive in one of the bombs detonated and scattered pluto-
Sr source. The 90Sr capsule is made of strontium titanate,
nium but no nuclear explosion occurred. A recently-re-
a ceramic material. Depending on the unit, this may weigh
leased report from Los Alamos, however, revealed that `by
between 0.5 and 1.8 kg (i.e., between 1.2 and 3.9 pounds).
good fortune' the weapons involved had just previously
This material is fire resistant with a melting point above
been modified after small-scale nuclear tests had uncovered
1100°C (2000°F). It also has a very low solubility in water.
a safety problem. If the weapons had not been modified to
There are 10 RTGs remaining in Alaska, located at the US
reduce the risk, the latter report states that `the chance of a
Air Force seismic observatory at Burnt Mountain, Alaska
significant nuclear explosion would have been more than a
(67°25'N, 144°36'W). As of April, 1994, the total inven-
thousand times greater'. This was at a time when the USA
tory for the 10 RTGs at Burnt Mountain was approximate-
had experience of more than 430 weapon tests. To date,
ly 26 PBq (700 000 Ci). Each RTG weighs between one and
France has only conducted half as many tests and China
two tonnes including radiation shielding, insulation, ther-
about 10%. There may, therefore, be a risk that some of
mocouples and housing. The units are designed to ensure
the weapons in the stocks of these countries have less de-
that external exposure at a distance of 1 m does not exceed
veloped safety systems and may pose a greater risk of nu-
0.1 mSv/h.
clear detonation in the case of accidents. Altogether, the
No accidental releases of radioactive material have been
USA has conducted 1032 nuclear tests and the experience
associated with the units installed at Burnt Mountain, Alas-
has largely benefited both the United States and the United
ka. A 1994 evaluation of these units by the US Office of
Kingdom. The former USSR has conducted 715 nuclear
Technology Assessment (OTA 1995) found that, even in the
tests and this should imply that a relatively sophisticated
event of a forest fire impacting the observatory site and
level of safety design features has been incorporated into
burning one of the buildings housing a RTG, there is little
Russian weapons.
risk of a release of 90Sr . In the event of a release from the
The most serious concern relates to a possible critical
radioactive capsule, the main risk would be associated with
event in a weapon i.e., an accidental nuclear explosion.
radiation exposures to people cleaning up the site. These
This has been prevented to date but, clearly, the fail-safety
radiological risks would most likely be posed by external
of nuclear weapons is an area in which public reassurance
exposure to 90Sr because the strontium titanate is not read-
is warranted. If some accidental ignition of a part of the
ily converted into fine particles that are likely to be inhaled.
conventional explosive within a weapon occurred, a full-
In this form, 90Sr is not readily biologically available to ei-
scale nuclear detonation is unlikely. However, partial crit-
ther plants or animals. These RTGs, therefore, pose little
icality from a very small yield to almost full yield cannot
risk to the general public and worker exposures are mini-
be ruled out. The most serious such accident would be one
mized by the use of appropriate safety procedures.
occurring at ground level. The neutron flux from a partial
In 1994, there were approximately 155 Russian RTG-
criticality at ground level would also result in the forma-
powered lighthouses in service in northern areas. Each of
tion of a larger quantity of activation products in the vicin-
these RTGs contains up to 13 PBq of 90Sr (i.e., a total of
ity of the detonation.
(2000 PBq). Only one accident involving RTGs has been
Clearly, the risk of releases of radioactive material fol-
reported. A helicopter carrying a RTG source crashed off
lowing accidents involving nuclear weapons is high. Al-
the east coast of Sakhalin Island in 1987 and the source
though little information has been made available regard-
was lost. No enhanced levels of radioactivity have been
ing safety precautions during the handling, storage and de-
detected in the area.
ployment of nuclear weapons by either the Russian Federa-
Some satellites also employ RTGs as power sources. A
tion or the United States, there are reasons for concern
US Navy navigation satellite (Transit 5BN) containing a
about possible accidents involving fires and spillage of fis-
SNAP 9A RTG source fueled by 238Pu, burned up in the
Chapter 8 · Radioactivity
597
atmosphere over the Indian Ocean in April 1964. As a re-
in ecological, social and other associated factors (e.g., food
sult, a considerable quantity of 238Pu was released to the
production) can lead to spatial differences in: 1) radiocae-
upper atmosphere and this has resulted in altered relation-
sium intake by humans, and 2) total radiocaesium output
ships between 238Pu and other plutonium isotopes in at-
(or net flux) from different Arctic areas. This section con-
mospheric fallout.
siders the extent to which contamination of Arctic food
products may vary spatially considering, primarily, radio-
caesium, but also radiostrontium and radioiodine.
8.6.7. Summary
During the initial phase (days to weeks) external radia-
The greatest threats to human health and the environment
tion, short-lived radionuclides and direct contamination of
posed by human and industrial activities in the Arctic are
foodstuffs are of major concern. Fluxes, on the other hand,
associated with the potential for accidents in the civilian
are mainly relevant to long-term considerations. By using
and military nuclear sectors. Of most concern is the poten-
fluxes, it is possible, beforehand, to identify areas and food-
tial for accidents in nuclear power plant reactors, during
stuffs most vulnerable to radioactive contamination. After
the handling and storage of nuclear weapons, in the de-
a nuclear accident, there will probably be an uneven distri-
commissioning of nuclear submarines and the disposal of
bution of radioactive contamination and, accordingly, there
spent nuclear fuel from vessels. The risks posed by radio-
will be a need to map and characterize the fallout distribu-
active wastes dumped in the marine environment of the
tion. The priority should be given to areas with the greatest
Russian Arctic and by radioelectric thermal generators de-
vulnerability, identified from proximity to the source and
ployed in the Arctic environment are relatively minor.
previously calculated fluxes.
It is not possible to judge the risks posed by the remobi-
lization of radionuclides previously released from nuclear
8.7.1. Sources of radionuclide intake by humans
reprocessing activities currently residing in storage basins
of northern Russian river systems because of the limited
The radionuclide intake of an individual depends on both
understanding of the rates and modes of transport of ra-
the rate of consumption and the contamination level of each
dionuclides within terrestrial environments, especially the
dietary component. Information on dietary composition is
effects of episodic events. Unfortunately, neither has it been
available for the average population and, to some extent,
possible to assess with confidence and quantitatively the
for indigenous groups, the latter groups constituting, in
risks posed by potential reactor accidents in the Arctic,
varying ways, the specially selected groups for each country
both military and civil, potential accidents in nuclear re-
given in the individual dose assessment in section 8.4.
processing operations and potential accidents in the han-
The accuracy of such estimates of intake for different
dling and storage of nuclear weapons, either because of the
population groups is dependent on the accuracy or repre-
limited relevant information made available to the AMAP
sentativeness of the information on consumption and con-
radioactivity assessment group or because of shortcomings
tamination of foodstuffs provided by participating coun-
in contemporary safety assessments of these practices.
tries (cf. also discussions in chapter 5). Furthermore, the
It is concluded that there is a need for more detailed
approach does not intrinsically consider the proportions of
probabilistic safety assessments of civilian nuclear power
the foodstuffs produced locally or imported from outside
plant installations. It is also essential that account is taken
the study area, but does allocate different foodstuffs as ei-
of medium and long-term internal doses via terrestrial Arc-
ther imported or local. Additionally, the respondents to di-
tic pathways. In addition, continuing attention should be
etary surveys may have a tendency to overestimate their
paid to the modes and rates of radionuclide mobilisation in
consumption of some foodstuffs and there is also likely to
major river catchments of Siberia, including groundwater.
be considerable variation within the groups surveyed. Nev-
ertheless, this analysis provides an indication of which
products are most important in determining 137Cs intake
8.7. Spatial analysis of vulnerability
and identifies the foodstuffs for which the collection of spa-
of Arctic ecosystems
tial data would be most appropriate, such as reindeer meat,
lamb, milk, mushrooms, berries and freshwater fish.
Radioactive doses to populations are derived via two major
Using the values for radiocaesium activity concentration
exposure routes, external and internal; the latter including
in dietary components provided in section 8.4, the relative
both inhalation and ingestion pathways. For collective ex-
importance of different groups of dietary components to
ternal exposure, distance from source and population den-
137Cs intake in various Arctic countries has been plotted for
sity are the major factors which determine vulnerability to
the most recent five-year data series available (1990-94),
a radionuclide release. In the Arctic, specific environmental
and is shown for both the average (Figure 8·69) and spe-
and morphological factors, such as extent of forest and
cially selected populations (Figure 8·70). These figures have
urban areas or presence of snow cover, could also influence
necessarily been derived from data (which for some food-
vulnerability to external dose; but in general, assessments
stuffs/countries is rather limited) supplied by each country
of potential external dose can be made from considering
on the basis of measurements from the Arctic area.
potential sources. Internal doses are, however, readily af-
For the average populations (with the notable exception
fected by environmental influences, including biological,
of Canada, where intake is dominated by the consumption
physical, climatic and socio-economic factors. For this rea-
of caribou meat), a range of different products contribute
son, this study will concentrate on internal doses, in partic-
to the total 137Cs intake. The lowest estimated 137Cs intake
ular the ingestion pathway, aiming to identify areas or pro-
by an average population is for Greenland, where a large
cesses which would be vulnerable to deposition from a pos-
proportion of the diet is derived from marine ecosystems,
sible future radioactive release.
the products of which (fish, seal, whale) generally contain
Arctic ecosystems vary in their land cover, land usage
low 137Cs activity concentrations. The highest estimate of
and soil type. Furthermore, the Arctic environment is ex-
137Cs intake by the average population is for Canada. Also,
ploited by humans in different ways, according to climate,
intake in Fennoscandia and western Russia is enhanced by
resource availability and socio-economic factors. Variations
recent additional contamination from Chernobyl.
598
AMAP Assessment Report
137Cs Bq/y
as the Greenland selected populations is hypothetical and
5 000
other selected groups are not representative of all indigenous
people in that country (e.g., Canada). Where hypothetical or
critical groups have been used to calculate intakes by `the in-
4 000
digenous population' as in Canada and Greenland, calcu-
lated intakes will not be applicable to the indigenous popu-
lation as a whole. For instance, estimates of 137Cs intake by
3 000
the Canadian selected groups are based on consumption of
foodstuffs by an inland Inuit group who subsist almost en-
tirely on caribou and eat no marine fish. This is not repre-
2 000
sentative of all Canadian Inuit, many of whom live closer to
the sea and consume larger quantities of seafood, or of the
Indian or Metis indigenous peoples. Indeed, the Canadian
1 000
group could be considered representative of a critical popu-
lation for the whole Arctic, namely people who consume ex-
tremely large quantities of reindeer meat. It is likely that such
0
groups could be found in other areas of the Arctic, such as
Canada
Greenland
Norway
Sweden
Finland
Russia
in Fennoscandia or Russia. Under such circumstances, the
Freshwater fish
Mushrooms/Berries
Other
intake by hypothetical individuals consuming reindeer meat
at similar rates to the Canadian group during the 1990-94
Sheep and goat milk
Reindeer/Game
period have been calculated (using the 137Cs activity concen-
Figure 8·69. The relative contribution to the 137Cs content of the diet of
trations in reindeer meat for each country for 1980-84 in the
y
the `average' population of various Arctic areas for the period 1990 to
Annex tables) and are shown in Table 8·47.
1994.
8£87d01
Table 8·47. Annual intake of 137Cs by individuals consuming large quantities
of reindeer/caribou meat.
137Cs Bq/y
Estimated annual 137Cs intake (kBq) from consumption
60 000
of reindeer/caribou meat at a rate of 1 kg/d
Country/
1980-1984
1990-1994
50 000
region
(pre-Chernobyl)
(post-Chernobyl)
Arctic Canada
164
55
Arctic Sweden
88
290
40 000
Arctic Norway
157
160
Arctic Finland
164
144
Arctic Russia (West)
182
108
Arctic Russia (East)
95
104
30 000
Table 8·47 demonstrates the importance of food intake
from such semi-natural products. A comparison of pre-
20 000
Chernobyl estimates of intake given in Table 8·47 with the
137Cs deposition in each country (Table 8·3) shows some dis-
10 000
crepancies. These probably arise from variation in location
and representativeness of the 137Cs data for reindeer. In par-
ticular, few data are available for 137Cs in reindeer meat in
0
Canada, relative to its area.
Canada Greenland Norway
Sweden
Finland
East
West
Russia
Russia
8.7.2. Spatial distribution of Arctic communities
Freshwater fish
Mushrooms/Berries
Other
Other
Sheep and goat milk
Reindeer/Game
The likelihood and extent to which different Arctic popula-
tions are affected by radionuclide deposition will be influ-
Figure 8·70. The relative contribution to the 137Cs content of the diet of
enced by the size, ethnic and spatial compositions of com-
the `selected' population of various Arctic areas for the period 1990 to
munities. The selected populations in section 8.4 (compris-
1994.
ing mostly indigenous people) have higher dietary intakes of
In Figures 8·69 and 8·70 the effects of regional dietary
radiocaesium than average populations. Therefore, the larger
preferences are clear. For example, goat cheese is an impor-
the proportion of indigenous people in an area, the higher
tant source of dietary radiocaesium in Norway only; mush-
will be the net transfer of radiocaesium. However, caution
rooms and berries appear to be a more important source of
must be taken in some areas where indigenous people form
137Cs intake in Fennoscandia and Russia than elsewhere.
the bulk of the population and also, therefore, dominate the
However, information is sparse on exploitation of natural
average population.
food products in other areas.
In some of the areas being considered, assessments of typ-
In contrast to the average populations, 137Cs intake by
ical consumption and intake rates may be complicated by
the specially selected groups is consistently dominated by
the diversity of indigenous groups, with different dietary pre-
reindeer/caribou consumption throughout the Arctic area,
ferences. For instance, Arctic Russia is populated by many
although in Sweden consumption of freshwater fish also
discrete groups, from Saami in the west to Chukchi and Yu-
contributes significantly. The highest annual estimates of
piks in the extreme east. Similarly, Arctic Canada and Alas-
intake for 1990-1994 (ca. 50 kBq) were obtained for Can-
ka are populated by various different indigenous groups, in-
ada and Norway, with the smallest for Greenland (ca. 5 kBq).
cluding Inuit, Aleut, Athabascans, Indians and Metis with
However, the results must be interpreted with some caution
similar variations in culture and habits.
Chapter 8 · Radioactivity
599
137Cs Bq/y
8.7.3. Spatial differences in transfer
35 000
through pathways
The rate of transfer of radionuclides through major food-
30 000
chain pathways can vary considerably within relatively small
areas due to differences in factors such as the form of conta-
25 000
mination, soil type and nature/intensity of land use. Addi-
tionally, the relative importance of different pathways varies
among radionuclides. By considering important pathways
20 000
and the factors that influence transfer through them, the lay-
ers of information necessary for spatial modeling can be
identified and developed.
15 000
In most terrestrial ecosystems, the major transport path-
way for dietary intake will be that from soil plant ani-
10 000
mal humans, for instance via milk and meat. The extent of
contamination will be influenced by the soil type. In Arctic
ecosystems, however, the pathway lichen reindeer hu-
5 000
mans is of great importance. As lichen have no rooting sys-
tems, this pathway is not influenced by soil type. Arctic lichen
are contaminated via direct aerial contamination after which
0
the deposited radionuclides are retained by the lichen and
Individual
Yamalo-
Arctic
Taimyr
Chukotskii
related
made available for subsequent translocation to fresh growth.
Nenetskii
Sakha
assessment
Both 90Sr and 137Cs cycle readily within lichen. Hence, com-
Freshwater fish
Mushrooms/Berries
Other
paratively high levels of radioactive contamination are main-
Sheep and goat milk
Reindeer/Game
tained over long periods. Ultimately, however, 90Sr is washed
out more rapidly than 137Cs (Nevstrueva et al. 1967). During
Figure 8·71. Estimated source of dietary intake of 137Cs by various indige-
the winter, reindeer may graze almost exclusively on lichen,
nous population groups in different regions of Arctic Russia in 1993, and
especially the species Cladonia alpestris and Cladonia rangi-
the AMAP region as a whole.
ferina, and transfer can then be expected to be independent
From Figure 8·71, it is clear that, throughout Asian Arc-
of soil type. In contrast, during the summer, reindeer graze on
tic Russia, intake of 137Cs by all groups is dominated by
herbaceous vegetation, at which time their 137Cs content will
reindeer consumption, although it is also possible to identify
be derived from soil via the pathway soil plant reindeer.
regional differences in consumption of freshwater fish and
The 137Cs content of reindeer, therefore, follows a seasonal
mushrooms and berries that have an influence on total 137Cs
cycle with peak activity concentrations in the winter months.
intake. The above approach used varying values for deposi-
As reindeer are traditionally culled in winter, less spatial
tion to each indigenous autonomous area, calculated using
variation in the rate of transfer of radiocaesium might be ex-
the GIS-based method described in section 8.3. This has
pected than for other food products. Generally, in most Arc-
been combined with a mean value for aggregated transfer
tic areas, this appears to be the case. However, there are ex-
to reindeer meat of 0.3 kg/m2 across eastern Arctic Russia.
ceptions; Icelandic reindeer consume herbaceous vegetation
This mean value has been used because measurements of
and a species of lichen (Cetaria islandica) which is less effi-
reindeer in eastern Russia were too few to derive representa-
cient at trapping and retaining radiocaesium than those spe-
tive spatial or temporal values relevant to each indigenous
cies preferred by reindeer in most other regions (Palsson et
group. For additional comparison, the rates of consumption
al. 1994). Consequently, transfer rates to Icelandic reindeer
used in the dose assessment (section 8.4) to calculate the in-
are much lower than those observed elsewhere and there is
take of the selected population have also been calculated.
little seasonal variation in the level of 137Cs contamination.
The distribution of indigenous peoples and their contri-
Similar observations have been made in the Yakut region of
bution to the populations of different Arctic countries as a
Siberia. Transfer may also be influenced by such factors as
whole are shown in Table 8·48. These values are derived
grazing pressure; intensive grazing of lichen could reduce the
from chapter 5 of the AMAP assessment.
pool of radiocaesium available for ingestion by reindeer.
There is considerable variation in the proportion of in-
For other foodstuffs, differences in soil type can pro-
digenous people in each country's Arctic area, with the high-
foundly influence transfer of some radionuclides. Spatial
est proportions in Greenland and Canada. Such differences
variation in soil type will be reflected in the rate of transfer
will obviously affect the interpretation of the `average group'
to the products derived from the soils. Generally, radiocae-
estimates given previously.
sium is fixed strongly by mineral soils but can remain mo-
bile in soils with low clay mineral content, such as highly
Table 8·48. Populations of different AMAP Arctic regions.
organic and sandy soils. In contrast, radiostrontium is often
Country/region
Total
Indigenous
% Indigenous
less mobile in organic soils. Generally, improved soils usu-
ally fix 137Cs and the transfer to arable crops and animal
Alaska
481000.
73200.
15.2
products is relatively low, whereas transfer to milk and meat
Arctic Canada
93000.
47400.
51.0
Greenland
55400.
48000.
86.6
from animals grazing on unimproved pasture over poorer
Iceland
255700.
0.0
0.0
quality soils is higher. The spatial distribution of different
Arctic Finland
200700.
4000.
2.0
soil types will, therefore, be an important factor in deter-
Arctic Norway
379500.
37400.
9.9
Arctic Russia
1999700.
67200.
3.4
mining vulnerability.
Arctic Sweden
263700.
6100.
2.3
In most Arctic soils, cold wet conditions prevail due to
Faeroe Islands
43700.
0.0
0.0
restriction of drainage by permafrost and/or high precipita-
Svalbard
3200.
.00
0.
0
tion. Biological activity is reduced, resulting in development
Total
3775600
283300
7.5
of poor quality soils. Where conditions favour plant growth,
600
AMAP Assessment Report
soils with organic upper horizons may develop, varying
both on catchment size and soil type. In particular, catch-
from deep peat to shallow rankers and podzols. Where veg-
ments dominated by organic soils can act as a long-term
etation cover is more limited, poor quality unconsolidated
source of radiocaesium release into freshwater ecosystems
gravel soils may develop on parent materials such as weath-
(Smith et al. 1995b). Furthermore, although much of the
ered regolith, outwash and glacial deposits. Many of these
radioactive contaminants will sorb to sediments, partition-
poor quality soils have a low fixation capacity for radiocae-
ing between water and sediments will be affected by local
sium. In more organic soils, much of the 137Cs remains in
physical-chemical variables such as pH, and ionic concen-
the upper soil horizon and is available for uptake by plants,
tration. There is, therefore, considerable potential for spa-
whereas in coarser soils, 137Cs is more readily leached down
tial variation in transfer in the food chain water fish hu-
through the soil profile, although in some, layers of perma-
mans among different freshwater systems.
frost may act as barriers to its mobility in the soil profile.
In the most temperate Arctic regions, however, conditions
8.7.4. Changes with time
encourage sufficient mineralization and humification to pro-
duce agricultural soils. These soils are not distributed evenly
Consideration of spatial vulnerability requires an under-
through the area and are generally associated with mild
standing of how the relative importance of different transfer
coastal regions. Production of some foodstuffs is, therefore,
pathways varies with time. This may help determine whether
highly skewed toward a comparatively small and atypical
the vulnerability of an area is short- or long-term.
part of the region as a whole.
Firstly, this will be due to differences in the physical half-
Climate and soil type strongly influence land use. In areas
lives of different radionuclides. For example, immediately
of the Arctic with better quality soils, land use involves culti-
after a release, 131I may present the greatest radiological con-
vation of crops and fodders, or grazing by cattle or sheep.
cern because of its high mobility and radiotoxicity. How-
As such areas are relatively small and well characterized, it
ever, its short physical half-life renders it a short-term radio-
is theoretically straightforward to derive rates of transfer for
logical problem. In the longer term, transfer of radiocaesium
a given product and soil type. Areas with poorer quality
and 90Sr will become more significant.
soils support unimproved pasture or forest, and are exam-
Secondly, long-term vulnerability will be influenced by var-
ples of semi-natural ecosystems. The distribution of different
iation in the effective ecological half-life (T1/ 2 eff-eco) of 137Cs
semi-natural ecosystems, such as forests, tundra and moun-
in food products, potentially altering the relative importance
tain pastures, will again be important in modeling spatial
of different transfer pathways. Consequently, as time elapses
variation in transfer and radionuclide flux. Different semi-
after deposition, foodstuffs with longer T1/2 eff-eco will propor-
natural ecosystems are utilized in varying ways: for summer
tionately assume greater importance for 137Cs intake. The
grazing of sheep, goats and cattle, for herding reindeer, as
T1/2 eff-eco of 137Cs in most food products (other than rein-
sources of fungi and berries (e.g., blueberry, lingonberry and
deer) is related to the rate of depletion of the available pool,
cloudberry) and for hunting game. Some of these products
either due to fixation by the soil or to removal by runoff
are distributed in a variety of different types of semi-natural
or leaching down the soil profile. Therefore, the contribu-
ecosystem, whereas others such as moose, are largely con-
tion to the total 137Cs intake from those products with the
fined to forests. Analysis of the variation in importance of
longest T1/ 2 eff-eco, namely those such as fungi and lamb meat
transfer pathways can be refined by knowledge of the spatial
produced on pasture with organic soils, can become propor-
distribution of different ecosystems and the rate of produc-
tionately greater with time. For reindeer, grazing lichen, the
tion and exploitation of products from them. However,
T1/ 2 eff-eco can be further influenced by the depletion of 137Cs
quantifying transfer in semi-natural ecosystems is compli-
in the available pool (lichen) by grazing, as well as losses
cated by their size and diversity, particularly the intrinsic
through leaching.
variability in deposition, soil and vegetation type between
Similarly, vulnerability will be influenced by variation in
and within areas. As transfer of radiocaesium to different
the throughput of activity in environmental compartments.
vegetation species varies, factors such as diet composition,
For example, in freshwater ecosystems, throughput of 137Cs
which may itself be influenced by grazing pressure, will af-
along the pathway catchment lake river could vary con-
fect the rate of aggregated transfer to grazing animals.
siderably depending on such factors as catchment size, lake
Transfer through marine and freshwater ecosystems is
size and local physico-chemical characteristics. This will af-
dominated by the pathway water fish/mollusk/crustacean
fect the effective ecological half life of 137Cs in the lake, and
humans, although in some Arctic areas population groups
therefore in the food chain.
also eat large quantities of marine mammals (seal and whale).
Transfer in marine systems is generally low because of the
8.7.5. Transfer coefficients and relationships
magnitude of dilution and because the bulk of the activity of
8.7.5.1. UNSCEAR transfer coefficients
many radionuclides is sorbed to bottom sediments. Hence,
contamination of marine organisms is generally low, although
The transfer coefficient from deposition to diet applied by
it may be expected that benthic species would contain more
UNSCEAR was defined in section 8.2. Using data series sup-
activity than pelagic species. Spatial variation in transfer to
plied by participating countries, time-integrated transfer co-
marine products is not well defined across the Arctic Ocean
efficients to various important Arctic products have been de-
areas. Consequently this initial spatial analysis will not con-
rived. Such parameters can only be estimated when sufficient
sider variation in transfer through marine pathways in detail.
time-series data are available for a given product. Addition-
Potentially, freshwater ecosystems are important contri-
ally, use of the coefficients on a wider scale requires the as-
butors to radiocaesium intake (although less so than terres-
sumption that the raw data are representative of the wider
trial ecosystems), and will exhibit spatial variation in trans-
area. Estimation of this transfer parameter is thus useful for
fer. Deposition to freshwater catchments could vary over
broad comparisons among different areas, but of limited use
quite small distances as a result of orographic influences.
when data are scarce. Therefore, the areas and products in-
Over long periods, catchments act as a source for release of
cluded in the following analysis represent those areas for
radionuclides into streams and lakes by runoff. The extent
which adequate data were available. Analysis of the data has
of radionuclide release into water bodies will be dependent
been carried out as shown in the boxed example next page.
Chapter 8 · Radioactivity
601
Table 8·49. Integrated transfer coefficients (Bq/kg y per kBq/m2) for 137Cs
to Arctic lichen.
Example: Calculation of the integrated transfer coefficient for
137
Cs to Arctic Finnish lichen (see Table 8·49).
Time-integrated concentration,
From Figure 8·28 the mean values for 5-year intervals from
Integrated
Bq/kg y
Integrated
1960 to 1994 are obtained. These values are summed and mul-
deposition,
transfer
Area
kBq/m2
1950-59 1960-94
1995-
coefficient
tiplied by 5 (to account for the 5-year means) to obtain the time
integral 31 000 Bq / kg y for the period 1960-1994.
Arctic Finland
2.5
3100
31000
10000
18000
We get the time integral for 1951-1959 from multiplying the
Greenland
4.3
4800
25000
1700
7000
Arctic Russia
3.1
6000
18500
3000
9000
5-year mean value for 1960-1964 (700 Bq 137Cs / kg) in Figure
8·28 by 0.5
9 = 4.5, as we assume that the 137Cs concentra-
tions in lichen increased linearly from zero in 1951 to 700 Bq
137Cs / kg in 1959. The time integral becomes 3100 Bq / kg y.
Table 8·50. Integrated transfer coefficients (Bq/kg y per kBq/m2) for 90Sr to
The time integral for 1995 and onwards is calculated from the
Arctic lichen.
5-year mean value for 1990-1994 (700 Bq 137Cs / kg) assuming
Time-integrated concentration,
an effective half-live of 137Cs in Finnish Arctic lichen of 10 years
Integrated
Bq/kg y
Integrated
from 1995 and onwards, i.e., 700
10 / ln2 = 10 000 Bq / kg y.
deposition,
transfer
The three time integrals obtained above are now added:
Area
kBq/m2
1950-59 1960-94
1995-
coefficient
3100 + 31 000 + 10 000 = 44 100 Bq 137Cs / kg y, and we have the
Greenland
2.7
900
5200
300
2000
total infinite time integral of 137Cs in Arctic Finnish lichen, aris-
Arctic Russia
1.7
1100
6300
1000
5000
ing from a total (not decay-corrected) deposition of 2.5 kBq
137Cs / m2 since 1950. The so-called integrated deposition density
of 2.5 kBq 137Cs / m2 was the sum of the annual depositions of
137
Table 8·51. Integrated transfer coefficients (Bq/kg y per kBq/m2) for 137Cs
Cs in Arctic Finland as shown in Figure 8·20, which cover
to Arctic reindeer meat.
the period 1960-1994. For the years prior to 1960, the deposi-
tion of 137Cs in Arctic Finland was assumed to be proportional
Time-integrated concentration,
to the deposition measured of 90Sr in New York for these years
Integrated
Bq/kg y
Integrated
(see also the text in 8.3.2.1).
deposition,
transfer
Area
kBq/m2
1950-59 1960-94
1995-
coefficient
The integrated transfer coefficient is finally calculated by divid-
ing the infinite time integral (44 100) by the integrated deposi-
Arctic Finland
2.5
5700
34000
5500
18000
tion density (2.5) and we get 18 000 Bq / kg y per kBq / m2 as
Greenland
4.3
1500
6500
600
2000
shown in Table 8·49.
Arctic Norway
4.4
4200
31000
9500
10000
Arctic Russia
3.1
4300
21000
3300
9000
All the following calculations of time-integrated transfer
might, therefore, be expected, compared with reindeer
coefficients are derived from deposition data published by
from Finland and Norway, where animals are slaughtered
UNSCEAR. This approach to quantifying transfer has some
in winter.
limitations. A more localized problem, especially in parts of
In Russia, some reindeer are slaughtered all the year
the European Arctic, is the patchy nature of the Chernobyl
round, which could potentially lead, on average, to lower
deposit which adds uncertainty to estimates of integrated
transfer. However, the raw data for Russian reindeer were
deposition. For instance, the value of 2.5 kBq/m2 used for
collected from a range of areas, with inadequate frequency
Finland has been derived from measurements of activity de-
of sampling to establish a time series for different indigenous
posited at specific sampling sites that may not be representa-
regions. The data were, therefore, too scattered, spatially
tive of the entire area. Mobile-gamma surveys conducted in
and temporally, to explore the seasonal factors influencing
Arctic Finland after the Chernobyl accident suggest a higher
the transfer coefficient.
value would be more appropriate (Arvela et al. 1990) which
This highlights a more general problem with representa-
would give lower transfer coefficients. Whilst mobile gamma
tiveness of lichen and reindeer samples. As deposition of
measurements are not supported by soil analyses from vari-
137Cs from nuclear weapons tests was distributed fairly ho-
ous parts of Arctic Finland, they do, however, give similar
mogeneously, lichen samples collected locally at Inari in Fin-
results to the sampling of wet and dry deposition as values
nish Lapland were representative of Finnish Lapland as a
are corrected for decay.
whole, i.e., they were comparable with the reindeer samples.
Tables 8·49 and 8·50 show comparisons between time-
However, after Chernobyl, Inari received rather more (three
integrated transfer of 137Cs and 90Sr to lichen in Greenland,
times higher) fallout than most of the surrounding area.
Arctic Russia, and Arctic Finland (137Cs only). The transfer
Hence, after 1986, lichen from Inari would no longer be di-
coefficients of 90Sr are lower to lichen in Greenland than in
rectly comparable to Finnish reindeer meat samples, which
Arctic Russia, and of 137Cs, lower to lichen in Greenland
integrate deposition from a large area.
than in Arctic Russia and Arctic Finland. Furthermore, for
A further variable influencing the time-integrated transfer
Greenland and Russia, the transfer coefficient of 137Cs is al-
coefficient is the use of countermeasures. In many parts of
most 2-3 times higher than that of 90Sr, which is consistent
Norway and Sweden that received deposition from Cher-
with the known ability of lichen to intercept and retain ra-
nobyl, countermeasures were applied, including use of in-
diocaesium for longer than radiostrontium.
traruminal boli and salt licks. This could explain why 137Cs
Table 8·51 shows transfer coefficients for 137Cs to rein-
transfer to reindeer meat appears lower in Norway than in
deer. Initial comparison of the transfer coefficients for 137Cs
Finland, where such countermeasures were not introduced.
to reindeer meat and lichen reveals that transfer to reindeer
Similarly, countermeasures have been observed to reduce the
in Arctic Finland and Russia is proportional to that for
wholebody radiocaesium content of reindeer herders in Swe-
lichen but transfer to reindeer is, relatively, lower in Green-
den relative to estimates based on dietary information and
land. This is probably because reindeer are slaughtered in
measurements of integrated deposition. The greatest discrep-
Greenland during the summer, when activity concentrations
ancies were observed where Chernobyl fallout was highest,
in their meat will reflect their summer diet of green vegeta-
with corresponding countermeasure applied (R. Bergman,
tion. A lower transfer coefficient to Greenland reindeer
Sweden, pers. comm). Comparison of time-integrated trans-
602
AMAP Assessment Report
Table 8·52. Integrated transfer coefficients (Bq/m3 y per kBq/m2) for 90Sr to
much of their 137Cs intake is derived from consumption of
Arctic freshwater.
reindeer that cautious comparisons may be drawn. Com-
pared with the integrated transfer coefficient for reindeer
Time-integrated concentration,
Integrated
Bq/kg y
Integrated
meat (Table 8·51), that for the human body is significantly
deposition,
transfer
lower in Norway and Finland, but for Russia the difference
Area
kBq/m2
1950-59 1960-94
1995-
coefficient
is smaller. These differences could be due to local dietary
Arctic Finland
1.72
130
580
120
500
preferences such as differences in rate of consumption of
Greenland
2.7
500
1400
90
700
foodstuffs other than reindeer, or for some foodstuffs (e.g.,
fungi, berries), spatial variation in consumption rates of dif-
ferent species, or even in the method of preparation. Finnish
Table 8·53. Integrated transfer coefficients (Bq/kg y per kBq/m2) for 137Cs
to human body for Arctic selected groups.
Saami that consume fungi apparently select the genus Lac-
tarius specifically, and prepare them by parboiling, which
Time-integrated concentration,
reduces the 137Cs content resulting in lower intake via fungi
Integrated
Bq/kg y
Integrated
deposition,
transfer
than might otherwise have been expected. Furthermore, the
Area
kBq/m2
1950-59 1960-94
1995-
coefficient
extent of utilisation of natural foods by indigenous peoples
is, in many cases, not well known.
Arctic Finland
2.5
1400
6200
1100
3500
Arctic Norway
4.4
1100
7700
3700
3000
Figure 8·72 summarizes the relationship between 137Cs
Arctic Russia
3.1
3000
17400
3000
7500
activity concentrations in estimated Finnish deposition and
measurements in lichen, reindeer meat and the human body.
fer coefficients, therefore, requires knowledge of agricultural
practices that may result in spatial variations within the data.
8.7.5.2. Spatial and temporal variations in transfer
The integrated transfer coefficients for 90Sr to freshwater
to Arctic food products using aggregated
(Table 8·52) were surprisingly similar, even though the sam-
transfer coefficients
ples consisted of river water from Finland and Russia, but of
drinking water from Greenland. The integrated transfer co-
The analysis carried out in section 8.7.1 and 8.7.5.1 used
efficient of 137Cs to Finnish river water was estimated to be
data generalized to represent the whole Arctic area of the
900 Bq 137Cs/m3 y per kBq 137Cs/m2 which is somewhat high-
countries involved (although in section 8.7.1, eastern and
er than that observed for 90Sr in Finnish rivers.
western Russia were considered separately for the selected
As with terrestrial systems, freshwater ecosystems are
groups). However, 137Cs activity concentrations in foodstuffs
subject to uncertainties with regard to estimation of inte-
will vary spatially within, as well as among, countries, par-
grated deposition and net influx and efflux of radionuclides.
ticularly depending on such factors as local deposition rates,
Effects of catchment size and soil type, lake size and distri-
soil and vegetation type, presence of forests and dietary pref-
bution of radionuclides between sediments and solution can
erences of domestic and game animals. Rates of production
all contribute significantly to spatial variability in transfer.
or harvesting will also vary spatially, depending on ecologi-
Transfer to the human body will be influenced by the rate
cal factors such as the availability of forage for animals. In
of transfer in all the ecosystems contributing to the human
this section, such variation is considered in order to predict
diet. The integrated transfer coefficients of 137Cs for the
how both the 137Cs activity concentrations and total output
wholebody of selected groups (reindeer herders) are shown
in reindeer and milk may vary spatially both within and
in Table 8·53. The integrated transfer coefficient for the
among countries in the event of deposition from a future
Russian group is twice that for the Finnish and Norwegian
radionuclide release.
groups.
Where data are available, spatial information on such
Because human wholebody content is normally derived
variables as food production, 137Cs deposition and 137Cs ac-
from a range of dietary sources, comparison of the transfer
tivity concentrations in food products have been incorpo-
coefficient to reindeer with that to humans would not nor-
rated into a Geographical Information System (GIS). These
mally be valid. However, for the selected populations, so
layers of information have been combined to derive aggre-
gated transfer coefficients (Tag values) for the spatial unit
Lichen / Reindeer / Wholebody
Deposition
Deposition
being modeled.
137
Lichen
137
Cs Bq/kg (Lichen/Reindeer)
Cs Bq/m2
Spatial analysis has, therefore, been conducted with the
137Cs Bq (Wholebody)
Reindeer
aim of assessing the vulnerability of different regions or
50 000
Wholebody
1 000
products to 137Cs contamination. Analysis of vulnerability
can be undertaken using two approaches:
40 000
· Specific vulnerability is calculated using the specific activ-
100
ity of a radionuclide in a product at a known time from a
specified level of deposition. For large scale analysis, this
30 000
requires prior knowledge of aggregated transfer and ef-
10
fective half-lives for different soil types and products,
20 000
which are then applied generically. As aggregated transfer
is time-dependent, specific vulnerability is described as a
1
year-specific Tag. This allows simple comparison among
10 000
areas of different soil type or land use and the develop-
ment of optimal and cost-effective countermeasure strate-
gies. Alternatively, vulnerability could be demonstrated
0
0.1
1960
1965
1970
1975
1980
1985
1990
1995
through the level of deposition needed for a product to
Figure 8·72. Changes with time in 137Cs contamination in the food chain
exceed a given activity concentration. This is a similar ap-
lichen reindeer humans. The figure, showing the relationships be-
proach to that used previously in studies of soil acidifica-
tween 137Cs in deposition, lichen, reindeer meat and the human body for
tion, to assess critical loads for different soils.
northern Finland, is an example of real rather than calculated transfer.
Chapter 8 · Radioactivity
603
· Flux vulnerability takes spatial variation in productivity
release and the extent of outdoor production of agricultural
into account by estimating the total output of radionu-
products. Consequently, Tag values would be most likely to
clides from an area in each food product. Total radionu-
underestimate the flux in the first few weeks after deposition.
clide flux is calculated by applying a knowledge of differ-
Prediction of spatial differences in radionuclide flux re-
ences in food production and aggregated transfer. The flux
quires adequate data, which are, as yet, not always available.
vulnerability for an area is, therefore, a product of the
In this initial assessment, two approaches are adopted to
time-dependent aggregated transfer (specific vulnerability)
demonstrate the wider applicability of this analysis: 1) col-
and annual production of the appropriate foodstuffs, di-
lation of detailed spatial information on radiocaesium flux
vided by the area of the spatial unit being considered.
in reindeer and milk production throughout the Arctic, and
2) detailed analysis of spatial differences in radiocaesium
flux in one country, Norway, at the sub-national scale, show-
Example: To calculate the aggregated transfer coefficient (Tag)
ing how vulnerability varies spatially and temporally within
for reindeer meat.
a relatively small Arctic area.
Collate information on activity concentrations in reindeer meat
from a defined area or ecosystem. For reindeer information on
season is also important.
8.7.5.2.1. Spatial variation in total production
Collate information on ground deposition (from measurements
The rate of production of foodstuffs, or extent of exploita-
of activity in soil per unit area) in the area where the reindeer
graze. (Alternatively, transfer to reindeer can be derived from
tion via hunting or fishing, varies considerably both between
measurement of lichen samples, where lichen cover is sufficient
and within different Arctic areas of each country. Therefore,
to intercept deposited activity. For other terrestrial foodstuffs
radiocaesium flux will vary spatially, both through individual
only soil measurements are appropriate).
products and as a whole. An initial attempt has been made
The aggregated transfer coefficient (Tag) is calculated using the
at collating production and harvesting statistics for the major
following expression:
foodstuffs identified in section 8.4 for each country, and is
Activity concentration in reindeer meat (Bq/kg )
shown in Table 8·54. Much of the production information
Tag =
has been collated from national statistics gathered by GRID-
Activity of deposit (soil or lichen) per unit area (Bq/m2)
Arendal, but where information was not readily available,
with units of m2/kg.
supplementary information was provided by national experts.
From Table 8·54, it is apparent that information on pro-
duction of some foodstuffs is more readily available than
Aggregated transfer coefficients, defined, for 137Cs, as the
others. Commercial production of the agricultural and fish-
137Cs activity concentration in the food product (Bq/kg fresh
ery industries is reasonably well documented in most areas,
weight) divided by the current ground deposition (Bq/m2),
whereas, by comparison, exploitation of natural foodstuffs
were described in section 8.2. An example calculation of the
such as game, berries, fungi and freshwater fish is poorly
Tag for reindeer meat in Finland, Russia and Norway is
documented. Additionally, production of such foodstuffs as
shown in the boxed example above. In this initial spatial
berries and fungi can be extremely variable, depending on
assessment, Tags have been used for quantifying and model-
the climate during the growth season.
ing transfer to terrestrial products only. It is acknowledged
that variability in activity concentration and transfer between
Reindeer Production
individual lakes is considerable, and therefore Tag values in
Analysis of reindeer production required differentiation be-
freshwater ecosystems are subject to considerable uncertainty.
tween herding of domestic reindeer, which is the practise in
Tag values can be combined with appropriate regional es-
most of Arctic Fennoscandia, and exploitation of wild or
timates of effective ecological half life to give spatial informa-
feral reindeer herds, which is more typical of parts of Arctic
tion on short- and long-term vulnerability. The scale of the
Russia (e.g., Taimyr Peninsula), Canada and Alaska. This is
analysis presented here has been chosen according to the avail-
because the rate of cull is different, generally being higher in
ability of information at the AMAP radioactivity data center.
herded stocks, and therefore varies spatially, according to
A weakness of this approach is that predictions using Tags
the dominant mode of production.
are not applicable during the initial phase of direct foliar
For semi-domestic reindeer production, information was
contamination and rapid changes in radionuclide availability
obtained from national statistics and national experts. Infor-
in the soil following the release (Howard et al. 1996). This is
mation on the exploitation of wild reindeer, which is an im-
not a problem with predictions for the food chain lichen
portant activity in Alaska, Canada and Russia and to a les-
reindeer humans, but may be for the soil plant animal
ser extent in Greenland and Iceland, are less comprehensive
humans pathway, when the duration and importance of
and subject to greater uncertainty. For this assessment, in-
foliar interception would vary according to the season of the
formation was obtained on the worldwide distribution of
Table 8·54. Annual production and harvesting of foodstuffs in Arctic countries.
Product
Alaska
Russia
Finland
Sweden
Norway
Greenland
Iceland
Canada
Total
Milk, L
5.00 105
1.34 108
9.88 107
1.88 108
1.82 108
1.03 108
1.37 103
7.06 108
Goat Milk, L
4.82 106
1.03 107
2.68 102
1.51 107
Pork, kg
2.20 104
1.98 107
6.40 105
1.78 106
1.51 106
3.21 106
4.54 102
2.70 107
Beef, kg
1.86 104
1.98 107
3.38 106
2.10 106
8.52 106
3.06 106
1.25 104
3.69 107
Lamb, kg
9.11 102
8.00 104
2.59 105
1.80 106
2.54 104
8.80 106
1.10 107
Potato, kg
6.40 106
8.53 107
4.40 106
3.61 107
4.72 106
1.11 107
9.81 104
1.48 108
Fruit/vegetables, kg
3.15 105
2.26 106
2.57 106
Reindeer, kg
9.08 105
1.89 107
3.04 106
2.20 106
1.59 106
2.06 105
8.99 103
1.33 106
2.82 107
Moose, kg
6.45 106
2.10 106
4.07 105
8.96 106
Fungi, kg
1.05 107
1.60 106
1.21 107
Berries, kg
2.38 104
2.68 106
2.40 106
4.72 106
1.27 104
9.84 106
Freshwater fish, kg
8.00 105
1.20 106
5.80 106
4.70 104
3.15 106
1.10 107
Marine fish, kg
1.23 109
1.10 105
n.a.
3.79 108
6.50 107
1.56 109
3.23 109
604
AMAP Assessment Report
Table 8·55. Estimated total reindeer production in the AMAP area.
Wild reindeer
Semi-domestic reindeer
Country/
Estimated
Estimated Estimated
Estimated
Estimated
Estimated
region
number
hunter kill
production, kg
number
slaughter
production, kg
Alaska
597000
25000
576000
36000
14000
332000
Canada
1287000
58000
1328000
00
00
000
Finland
00
00
000
346000
132000
3040000
Greenland
18000
7000
168000
4600
1600
38000
Iceland
3000
400
9000
00
00
000
Norway
00
00
000
164000
69000
1590000
Svalbard
8800
00
000
00
00
000
Russia
887000
222000
5110000
1494000
598000
13800000
Sweden
00
00
000
229200
95000
2200000
Total
2800800
312400
7191000
2273800
909600
21000000
wild reindeer populations (Williams and Heard 1986) and
time. Ground deposition from global fallout was predicted
used to help allocate herds into appropriate spatial units in
using the GIS-based method described in section 8.3. Tempo-
the GIS. Updated information was incorporated where avail-
ral variation in Tags was calculated for each country and is
able, for Alaska (Swanson and Barker 1991) and Canada
shown in Figure 8·73. Generally, aggregated transfer has been
(Ferguson and Gauthier 1992). The rate of exploitation of
decreasing since a peak in the early 1960s corresponding to
wild reindeer stocks has been estimated largely from data
the maximum input of global fallout. This is a function of the
published by Williams and Heard (1986). In all cases, it has
declining magnitude of the pool of radiocaesium available to
been assumed that the carcass yield is broadly similar to that
the reindeer because 137Cs activity concentrations in lichens
reported for Finnish semi-domestic reindeer of 23 kg. It is
will decrease due to grazing and leaching of the radiocaesium
recognized that the extent to which the carcass is utilized
into the underlying soil. The data suggest that the highest rates
may vary spatially, but this has not been investigated.
of transfer have been recorded in Fennoscandia. However, it
Based on these data and assumptions, the estimated con-
is important to note that the quantity and quality of the data
temporary annual production of reindeer meat throughout
used to generate these Tag values varies: measurements of 137Cs
the Arctic is shown in Table 8·55.
in reindeer meat were most comprehensive from Fennoscan-
dia; in other areas coverage was less comprehensive; data
Milk production
were readily available from Russia for most years, but covered
Information on milk production was collated from various
a vast area rather thinly; for Alaska, data were only available
sources; both national statistics and data provided by na-
from the 1960s, and for some other areas data were sparse.
tional experts. Many Arctic areas have little or no milk pro-
The assessment of vulnerability has been modeled for a
duction and, in other areas, animals graze outdoors for only
hypothetical accident resulting in new uniform 137Cs deposi-
a few months of the year. Furthermore, unlike the flux
tion of 100 kBq/m2 across the entire Arctic. This level of
through reindeer meat, which is derived from an activity
contamination was typical of that experienced over large
concentration at slaughter, the flux through milk is continu-
areas of the former Soviet Union following the Chernobyl
ous and subject to seasonal variation depending on what the
accident. This has been assumed to enable comparison
animal is eating. Calculation of fluxes through milk produc-
among different areas. Obviously after an actual accident,
tion are, therefore, complicated by differences in husbandry,
deposition would vary spatially.
both spatially and temporally.
The Tag values selected to estimate flux in individual areas
The collation of estimated annual milk production in
were the pre-Chernobyl maxima, as the evidence from Fig-
Arctic countries, given in Table 8·54, shows that the major-
ure 8·73 suggested that transfer maxima were recorded dur-
ity of Arctic milk production occurs in the Nordic countries
ing the peak period of deposition of global fallout. The spa-
and Russia. Consequently, information on 137Cs contamina-
tial trends in net flux from reindeer meat for the first year af-
tion of milk is relatively sparse for the other Arctic areas.
ter deposition are shown in Figure 8·74, where it is apparent
Monitoring of milk from dairies in the Nordic countries has
that fluxes would be greatest in Fennoscandia and Russia,
provided some useful time series.
with the highest flux in Arctic Finland. Spatial variations in
flux in the Arctic regions of each country are shown more
clearly by dividing the net flux by its area to obtain the flux
8.7.5.2.2. Spatial variation in fluxes
vulnerability as shown in Table 8·56. It must, however, be em-
The raw data used to collate national production (Table
phasized that this calculation illustrates the flux vulnerabil-
8·54) incorporated into the GIS were allocated to appropri-
ity from production during the initial year following the de-
ate spatial units such as regional administrative areas, to
position event only. With adequate information on effective
give greater detail.
ecological half-lives it is possible to calculate and compare
spatial variations in fluxes of 137Cs over defined time inter-
Reindeer
vals. This has been carried out for Norway in section 8.7.8.
Reindeer production is highest in Arctic Russia and Fenno-
scandia. Taking account of the much smaller areas involved,
Milk
production is densest in Fennoscandia and western Russia.
The information on milk production has been combined
The flux of radiocaesium into the human diet via reindeer
with time series data compiled in the AMAP radioactivity
was calculated by combining production information with
data center from the monitoring of 137Cs contamination in
appropriate aggregated transfer coefficients for individual re-
milk from dairies in Sweden, Finland, Norway and Russia.
gions. These were calculated by combining measurements of
Time dependent Tag values were calculated from predictions
reindeer contamination at known times and locations (col-
of 137Cs ground deposition in the same way as for reindeer.
lated in the AMAP data center for radioactivity) with estimates
Trends for Norway, Sweden, Finland and Russia are shown
of ground deposition in the same area at the corresponding
in Figure 8·75. Tag values to milk have also decreased as a
Chapter 8 · Radioactivity
605
Calculated Tag
Bq kg-1/Bq m-2
3.0
USA
Canada
Greenland
2.5
Norway
Sweden
2.0
Finland
West Russia
East Russia
1.5
1.0
0.5
0
1960
1965
1970
1975
1980
1985
1990
1995
Figure 8·73. Temporal variation in calculated reindeer Tag val-
ues for different Arctic regions.
137Cs flux
MBq
Table 8·56. 137Cs flux and flux density through reindeer production in
0.01
1 000
Arctic regions of each country in the first year after an assumed even de-
2 500
position of 100 kBq/m2.
5 000
10 000
Total flux,
Flux density,
25 000
Arctic region
Area, km2
MBq
kBq/km2
50 000
AMAP boundary
100 000
Alaska
1410000
5.06 1010
35.9650
250 000
Arctic Canada
3990000
7.85 1010
19.7650
Figure 8·74. Spatial trends in the net flux from reindeer
500 000
Greenland
2140000
1.21 1090
0.565
meat in different Arctic areas, for the first year after an
1 000 000
Iceland
102000
5.40 1050
0.005
assumed uniform deposition of 100 kBq/m2.
Arctic Norway
167000
1.89 1011
1130.00000 AMAP boundary
No data
Arctic Sweden
164000
4.25 1011
2590.00000
Arctic Finland
98800
6.99 1011
7070.00000
Arctic Russia (west)
411000
2.02 1011
491.00000
sition event. As with reindeer, the highest pre-Chernobyl Tag
Arctic Russia (east)
6790000
1.39 1012
205.00000
has been applied to modeling the flux. However, it is impor-
Total
15272800
2.9 1 1012
00
tant to remember that for milk 137Cs transfer will vary with
fodder type and source. Nevertheless, even in winter, much
function of the decreasing pool of radiocaesium that is avail-
of the fodder hay is likely to be produced locally; concen-
able for uptake by plants and ingestion by cows. In contrast
trates, however, are likely to be imported.
to the factors influencing transfer to reindeer, the dominant
A histogram showing the relative flux of 137Cs in reindeer
mechanism in lowering the rate of transfer to milk is fixa-
and milk throughout the Arctic in the first year following de-
tion of radiocaesium in the soil.
position, is shown in Figure 8·76. Where data was inade-
The relevant Tag values have been used to estimate the
quate for calculation of accurate local Tag values, a value of
flux of 137Cs through milk in Fennoscandia and western
0.0113, based on a mean of the data from Fennoscandia, has
Russia assuming a hypothetical uniform 100 kBq/m2 depo-
been applied. It is apparent that in the event of a future re-
8 9 0
lease the flux vulnerability of the Arctic to reindeer produc-
Calculated Tag
tion is greater than that for milk production. However, there
Bq kg-1/Bq m-3
137Cs flux MBq
0.020
Finland average
1 400 000
Total 137Cs flux via milk
0.018
Western Russia average
Sweden average
1 200 000
Total 137Cs flux via reindeer meat
0.016
Norway average
0.014
1 000 000
0.012
800 000
0.010
600 000
0.008
400 000
0.006
0.004
200 000
0.002
0
0
Alaska
Canada Greenland Iceland Norway Sweden Finland Western Eastern
Russia Russia
1955
1960
1965
1970
1975
1980
1985
1990
1995
Figure 8·76. Comparison of the predicted flux of 137Cs in milk and rein-
Figure 8·75. Temporal variation in cow milk Tag values for Sweden, Nor-
deer meat in different Arctic areas, for the first year after an assumed uni-
way and Finland.
form deposition of 100 kBq/m2.
606
AMAP Assessment Report
are differences among countries. Whilst for most counties the
reindeer flux is greater, the milk flux dominates in Iceland,
whilst in Norway the milk and reindeer fluxes are similar.
Radioiodine contamination of milk
In the immediate period following a release, the radionuclide
131I is of major radiological concern due to its volatility, mo-
bility and radiotoxicity to human thyroid. The short half-life
Finnmark
of this radionuclide (8 days) means that the primary inges-
tion pathway for exposure is via fresh produce, particularly
the consumption of milk.
Generally, milk production in the Arctic is low compared
with temperate areas. Imported fodders are often used in
Troms
winter, and there is comparatively little production of sheep
and goat milk (to which transfer of radioiodine is higher
than that to cow milk). However, on the Kola Peninsula,
close to the Kola Nuclear Power Plant which is a potential
source of 131I, milk is produced on both private and collec-
tive farms. This area would, therefore, have some vulnera-
bility following an 131I release. If a release occurred during
the outdoor grazing season, individual doses to consumers
of privately produced milk might potentially be similar to
those in temperate areas. However, collective doses would
be likely to be lower, as net production is low compared to
temperate areas. Furthermore, as the animals spend rela-
tively little time grazing pasture, there is reduced probability
Nordland
of an accident happening during that period.
Generally, therefore, for human dose, Arctic ecosystems
might be expected to have lower vulnerability for 131I com-
pared to temperate regions, and the inhalation pathway may
Figure 8·77. Norwegian counties considered in the flux vulnerability case study.
be of greater relative importance than ingestion. Locally, how-
ever, incorporating the location of milk-producing areas in re-
Nordland (not all of the latter county is in the AMAP area)
lation to potential sources into spatial models could improve
(Figure 8·77). This analysis is included as an example of the
predictions of areas vulnerable to a release of 131I in the Arctic.
approach that could be used in other parts of the AMAP area.
Most major foodstuffs with relevance to 137Cs transfer have
been included, but current deficiencies are noted and discussed.
8.7.7. Sensitivity to uncertainties:
radiocaesium in fungi and berries
8.7.8.1. Production data
Previous dietary studies have demonstrated the potential im-
portance of consumption of semi-natural and natural food
Compared to much of western Europe, Arctic Norway has
products to radiocaesium intake. Fungi and berries are impor-
only a small proportion of land that is actively managed as
tant dietary components in many areas. Unfortunately, infor-
arable or pasture. Foodstuff production data (given in Table
mation on intake, and contamination, of such foodstuffs is
8·54) were entered into the GIS at the smallest spatial unit
not readily available. For fungi, there are further uncertainties.
for which they were available, which was at county level
Radiocaesium transfer to different species of fungi from
with the exception of cow and goat milk, moose and rein-
the same location can vary considerably. Currently, however,
deer. The GIS was used to aggregate the total production for
there are very few relevant data on transfer of radiocaesium
each county. Production of cow and goat milk was available
to fungi in Arctic ecosystems.
for each municipality. The numbers of moose culled in dif-
The rate of production of fungi varies greatly from year
ferent age and sex classes for each county were combined
to year, as a function of the prevailing climate. This will
with average carcass weights to calculate the total produc-
probably be reflected in consumption.
tion of moose meat. Similarly, for reindeer meat, the number
The exploitation of different natural foodstuffs varies
of animals slaughtered was combined with average carcass
spatially, according to preference and availability, and is not
weights for reindeer grazing areas. Because reindeer grazing
well documented.
areas are not based upon administrative boundaries, the pro-
In some Arctic areas, estimates of net radiocaesium flux
duction of reindeer meat was partitioned within each county
may be especially sensitive to variation in, and lack of knowl-
using the GIS, assuming an even density of meat production
edge regarding, production, transfer to, and exploitation of
per unit area for each reindeer grazing area.
semi-natural and natural foodstuffs. Unfortunately, however,
The production values in Table 8·54 for freshwater fish
information is currently too sparse to assess the extent to
only consider salmon and sea trout, and will therefore un-
which spatial variation in production and exploitation of
derestimate the total production. Furthermore, this table has
these products can influence fluxes.
notable omissions because data sources for harvesting rates
of mushrooms were not identified.
8.7.8. Flux vulnerability of Arctic Norway
8.7.8.2. Aggregated transfer coefficients
An initial attempt at a more comprehensive spatial analysis
has been carried out for the mainland Norwegian AMAP
Aggregated transfer coefficients for the counties of Arctic
area, considering the three counties Finnmark, Troms and
Norway have been derived by comparing values for relevant
Chapter 8 · Radioactivity
607
Table 8·57. Aggregated transfer coefficients for Norwegian Arctic counties.
137Cs flux GBq
Aggregated transfer coefficients, m2/kg fw
400
County
Reindeer Cow milk
Potato
Lamb
Fruit/vegetables
350
Finnmark
2.566
0.020
0.003
0.164
0.002
300
Troms
2.400
0.009
0.001
0.631
0.001
Nordland
0.971
0.014
0.005 a
0.142
0.001
250
a. Based on national Norwegian values.
200
products compiled in the AMAP data center with predicted
values using the GIS. The Tag values shown in Table 8·57 rep-
150
resent the maximum calculated values prior to the Chernobyl
accident in 1986; these have been taken to be representative
100
of transfer soon after a contamination event and would consti-
50
tute a worst-case scenario. In the AMAP assessment, the max-
imum yearly average Tag values for reindeer meat, calculated
0
at a country level, for Fennoscandia ranged between 1.1871
Reindeer
Cow milk
Lamb
Other
and 2.2981 m2/kg (Howard et al. 1996). In the absence of rel-
evant data for Arctic Norway, recommended values have
Nordland
Troms
Finnmark
been used for goat milk (0.004 m2/kg), moose (0.02 m2/kg),
Figure 8·78. Relative 137Cs output in foodstuffs from the different Arctic
beef (0.006 m2/kg) and wildfowl/small game (0.02 m2/kg)
Norwegian counties in the first year after an assumed uniform deposition
(Howard et al 1996). For freshwater fish, a value of 0.03
of 100 kBq/m2.
m2/kg was adopted based on data from Saxén et al. (1996).
The Tag value used for reindeer meat in Nordland is almost
The total 137Cs flux from each county declines in the order:
three times lower than that for Finnmark and Troms, but is
Finnmark > Nordland > Troms
based upon far fewer observations. Indeed, because sample
numbers are low and taken over many years, it is not possible
with total estimated fluxes for the first year of 3.0
1011,
to statistically test whether the transfer values used for each
2.1
1011 and 1.7
1011 Bq, respectively, from the ten prod-
county are significantly different. Tag values are highly vari-
ucts in Table 8·54, (i.e. the listed products excluding fungi,
able between seasons and years. For example, Tag values in
for which no data are available, berries and freshwater fish).
reindeer meat are higher in winter when reindeer feed exclu-
Such an approach can be used for the first year after de-
sively upon lichen that may contain high radiocaesium con-
position, although caution must be used as the initial phase
centrations. Similarly, using a single Tag value for lamb ne-
of direct ingestion of radiocaesium deposited onto grass sur-
glects the seasonal pattern of sheep grazing in Norway (sheep
faces or vegetables, if a release occurred during the growing
are stabled in the winter and fed upon stored feed) and as-
season, would not be incorporated.
sumes an exclusive consumption of locally-produced feed.
To account for the reduction in contamination levels with
However, the slaughter of reindeer and sheep normally occurs
time, the total 137Cs output over a 50-year period has been
once, during the autumn/winter, in a given year. These tempo-
calculated, using appropriate effective ecological half-lives
ral variations make it difficult to compare Tag values between
for the major three products which dominate the total 137Cs
sites in different years. To predict changes with time, Tag val-
flux in each county, namely reindeer, cow milk and lamb.
ues need to be combined with effective ecological half-lives.
The effective ecological half-lives used were based on the
most relevant Norwegian data available and were:
8.7.8.3. Total 137Cs output
Reindeer: 4 years
Cow milk: 4 years
The total 137Cs output was estimated for an assumed conta-
Lamb: 15 years
mination event of 100 kBq/m2 uniformly distributed over
the three counties. Predicted 137Cs contamination levels in
The relative importance of the three different products over
each product were calculated for each spatial unit and then
a 50-year period for each county is shown in Figure 8·79
the total Bq contained in the total output of each product
(next page). For all three counties, the relative contribution
was calculated by multiplying the predicted activity concen-
to radiocaesium output from lamb has increased due to the
tration by the weight of product.
longer effective ecological half-life. In Nordland, cow milk
The data are summarized in Figure 8·78 where it is clear
still provides the most significant contribution to radiocae-
that reindeer constitutes the most important route of 137Cs
sium output whilst in Finnmark reindeer meat still domi-
output in Finnmark and Troms, but not in Nordland. In
nates. Lamb has become the most important food product in
Nordland, the comparative importance of different food
terms of radiocaesium output for the county of Troms.
products in the first year declines in the order:
It must be emphasized that, due to limited data availa-
cow milk > reindeer > lamb > beef > goat milk > potato >
bility, the analysis does not currently include mushrooms,
moose
freshwater fish and berries. After the Chernobyl accident,
these products were shown to be potentially important
while in Troms the order of importance of different food
sources of radiocaesium intake due to their high radiocae-
products is:
sium activity concentrations (especially for mushrooms) and
reindeer > lamb > cow milk > goat milk > beef > potatoes
to high rates of consumption by certain population groups
> moose
(Skuterud et al. 1997).
Regional output does not necessarily relate directly to
and in Finnmark it is:
human consumption, because some produce will be ex-
reindeer > cow milk > lamb > freshwater fish > beef >
ported to other areas. In the same way, no estimation of
moose > potatoes.
food products imported into the three counties is consid-
608
AMAP Assessment Report
1-year period
50-year period
8.7.8.5. Conclusions
Nordland
Nordland
The comparative importance of different foodstuffs in con-
tributing to collective dose varies spatially due to differing
rates of production and transfer. In this assessment, Finn-
mark has been identified as being most vulnerable to radio-
caesium deposition due to the importance of reindeer meat,
the large proportion of indigenous people and proximity to
potential Russian sources of radioactive contamination.
Currently, there are clear limitations to this initial approach
which relies upon the homogeneity of production and trans-
fer within certain spatial units and uses conservative T
Troms
Troms
ag val-
ues, thereby constituting a worst-case scenario. Furthermore,
the method does not consider the time of year in which a re-
lease occurs. Accordingly, it would probably underestimate
radiocaesium contamination of milk and fresh vegetables in
the first month if the release occurred during the growing
season.
The integration of information on deposition, transfer
and production, within a GIS, provides an efficient and
rapid method of quantifying collective dose in the event of
Finnmark
Finnmark
a radioactive release. However, it is important to incorpo-
rate appropriate information on the spatial variation in
transfer. For comparative purposes, we have assumed an
unrealistic even deposition occurring as the result of a nu-
clear release. If information about the spatial variation of
radiocaesium deposition could be integrated with this meth-
odology, within the GIS, either using dispersion modeling
or actual measurements, it would be possible to refine cur-
Total
Total
rent predictions of the consequences of a nuclear release.
For example, the consequences of a release at the Kola nu-
clear power plant could be more accurately predicted for
Finnmark.
Further improvements to the spatial analysis of vulnera-
bility to radiocaesium could be achieved by considering the
following factors:
· Identification of areas used for different types of
agricultural production.
Reindeer
Cow milk
Lamb
· Consideration of import and export of both food
Figure 8·79. The comparative importance of different foodstuffs as sources
products and animal feedstuffs.
of 137Cs for the three Arctic Norwegian counties over the 1-year and 50-
· Identification of soil type and ecosystem dependent
year period following an assumed uniform deposition of 100 kBq/m2.
transfer values.
· Incorporation of the effect of possible counter-
ered. However, if production associated with indigenous
measures.
people living by traditional methods can be identified, the
total flux to these people could be estimated and compared
Population
with individual dose assessments. Total 137Cs intake will also
250 000
be affected by imported food but the comparative impor-
tance of these foodstuffs as a source of radiocaesium com-
pared with locally-produced food is likely to be low. To as-
200 000
sess the importance of internal cycling within discrete Arctic
areas it is important to consider the numbers and proportion
of indigenous people in each area.
150 000
8.7.8.4. Spatial distribution of the Norwegian
Arctic population
100 000
The population of the three Arctic Norwegian counties has
been divided into three categories: urban, rural-non-indige-
50 000
nous and rural-indigenous. The total number of inhabitants
of the three counties decreases with increasing latitude so
that the most northerly county, Finnmark, has the fewest
0
inhabitants and the greatest number of indigenous people
Nordland
Troms
Finnmark
(Figure 8·80).
Both the relative proportion of the total population and
Urban
Rural indigenous
Rural non-indigenous
the total number of indigenous (Saami) show increases with
Figure 8·80. Distribution of the population of the three Arctic Norwegian
latitude.
counties.
Chapter 8 · Radioactivity
609
Further data refinements would include improved consider-
wastes dumped at sea. In the case of radioactive wastes
ation of potentially significant sources such as mushrooms,
dumped at sea and releases from underground and under-
freshwater fish and berries for which data on transfer, har-
water nuclear explosions, the radionuclides remain main-
vesting and consumption are currently inadequate.
ly localized. Radionuclides released from Russian fuel re-
processing plants and in liquid radioactive wastes dumped
in the Arctic marine environment have been distributed
8.7.9. Summary
more widely. Nevertheless, the additional contamination
Arctic ecosystems and food production systems have been
of the Arctic by radionuclides from these diverse sources
considered with respect to their vulnerability to deposition
is of negligible radiological significance.
from a radioactive release. Pathways and factors influencing
· The levels of artificial radionuclides in the Arctic attained
fluxes of radionuclides in the Arctic have been discussed,
maximum values during the period 1950-1970, primarily
including the location and dietary habits of some Arctic in-
as a consequence of atmospheric nuclear weapons testing.
digenous peoples. Appropriate parameters and methods for
Following the cessation of widespread atmospheric weap-
modeling radionuclide transfer have been considered, and
ons testing in the early-1960s, other sources, such as re-
Geographical Information Systems used to collate, store and
leases from European nuclear fuel reprocessing plants, in-
analyse data. Studies have been undertaken to compare po-
creased in relative importance. A second, but lower, peak
tential fluxes through important Arctic food products in dif-
in fission product radionuclides occurred in the Arctic
ferent geographical areas.
marine environment in the early 1980s as a consequence
It is clear that the most vulnerable food product to con-
of the peak in the rates of radionuclide discharge from
tamination following a radioactive release is reindeer/cari-
Sellafield in the mid-1970s. Finally, fallout from the Cher-
bou meat, although, under some circumstances, other prod-
nobyl accident in 1986 made an additional contribution
ucts such as milk and lamb may also be important. The com-
to radionuclide contamination of the Arctic. Since then,
parative importance of different foodstuffs varies within
the levels of radionuclides have been in general, but not
Arctic regions of each country, and therefore generalizations
ubiquitous, declining.
based at a country level may be inappropriate. The potential
· The major contribution to radiation doses of Arctic resi-
contribution of foodstuffs, which are known to readily accu-
dents delivered by artificial radionuclides originates from
mulate radiocaesium, needs further consideration particu-
previous nuclear weapons explosions in the atmosphere
larly for mushrooms, freshwater fish and berries.
giving rise to global fallout. However, in some geographi-
The total number, geographical distribution and dietary
cally limited, but populated, areas of the Arctic (Fenno-
composition of indigenous peoples within different Arctic
scandia and western Russia), a substantial dose contribu-
regions are important factors affecting potential individual
tion has been made by additional fallout from the Cher-
and collective doses arising from a nuclear accident in the
nobyl reactor accident. This contribution to the dose to
Arctic. To date, little analysis has been carried out concern-
Norwegian and Swedish Arctic residents was, and contin-
ing the effect of the spatial distribution of the indigenous
ues to be, reduced through the application of justified
people, variations in diet among ethnic groups, or variations
countermeasures.
in transfer rates to major food items.
· Arctic residents, whose diets comprise a large proportion
of traditional terrestrial and freshwater foodstuffs, receive
the highest radiation exposures to both natural and artifi-
cial radionuclides in the Arctic. Doses to members of both
8.8. Conclusions and recommendations
the average population and selected indigenous population
8.8.1. Conclusions
groups in the Arctic depend on the rates of consumption
of locally-derived terrestrial and freshwater foodstuffs, in-
The overall conclusion of this assessment is that the greatest
cluding reindeer/caribou, freshwater fish, goat cheese,
threats to human health and the environment posed by hu-
berries, mushrooms and lamb. In contrast, Arctic resi-
man and industrial activities in the Arctic are associated
dents having diets largely comprising marine foodstuffs
with the potential for accidents in the civilian and military
receive comparatively low radiation exposures because of
nuclear sectors. Of most concern are the consequences of
the lower levels of contamination of marine organisms.
potential accidents in nuclear power plant reactors, during
· The vulnerability of Arctic terrestrial ecosystems results
the handling and storage of nuclear weapons, in the decom-
in a five-fold higher exposure to radioactive contamina-
missioning of nuclear submarines and in the disposal of
tion compared to that in temperate areas. Because of the
spent nuclear fuel from vessels. In the Arctic, terrestrial
unique ecology of the Arctic, the comparative importance
pathways of human exposure to radioactive contamination
of both radionuclides and exposure pathways differs
are far more important than marine pathways. The vulnera-
from those in temperate areas. For example, exposures to
bility of Arctic populations, especially indigenous peoples,
artificial radionuclides are dominated by 137Cs contained
to radiocaesium deposition is much greater than for temper-
in a wide variety of traditional Arctic (native) foods of
ate populations due to the importance of terrestrial, semi-
terrestrial and freshwater origin but most importantly
natural exposure pathways.
reindeer/caribou meat. For reindeer-herders and others
The following provides detailed conclusions arising from
consuming comparatively large quantities of caribou/rein-
the assessment:
deer meat, the dominant pathway of natural radiation ex-
· Large-scale contamination of the Arctic with artificial ra-
posure is the intake of 210Po through caribou/reindeer
dionuclides is derived from three primary sources: global
meat consumption. Furthermore, unlike the situation in
fallout from past atmospheric nuclear weapons testing;
temperate areas, where immediate exposures to radioio-
releases from European nuclear fuel reprocessing plants;
dine are of primary concern following accidents, in the
and fallout from the Chernobyl reactor accident.
Arctic the low rate of milk production reduces the signifi-
· Some localized areas of the Arctic are also contaminated
cance of this pathway.
with radionuclides from other sources such as nuclear de-
· The highest time-integrated radiation exposures to mem-
vice explosions, spent fuel storage sites, and radioactive
bers of average populations of the eight Arctic countries
610
AMAP Assessment Report
from global fallout occurred in Canada, and the lowest in
quences of accidents in contemporary activities. These
Greenland. The variations in individual dose distribution
need to be rectified to enable more authoritative and
are not primarily due to geographical heterogeneities in
comprehensive evaluations to be made of the risks posed
radionuclide fallout. Rather, they result from variations in
to human health and the environment by such accidents.
diet among Arctic residents. Indigenous peoples comprise
Major effort has been devoted to determining, with high
a relatively high proportion of the inhabitants of Arctic
degrees of confidence and precision, the consequences of
Canada, some of whom rely comparatively heavily on
radioactive waste dumping at sea. These assessments have
caribou as a source of food. In contrast, the population
clearly shown that there is little associated risk to human
of Greenland is confined to coastal areas and has a diet
health or the environment. However, the risks associated
containing a comparatively large proportion of marine
with other major activities, which have considerable po-
foodstuffs having low radionuclide contamination.
tential for widespread and serious consequences (such as
· Selected indigenous Arctic population groups can have
the operation of nuclear-powered vessels and of nuclear
individual radiation exposures up to 50 times larger than
reactors in the Arctic and the handling and carriage of
those of the members of the average populations. Individ-
nuclear weapons), have been inadequately addressed. Ide-
ual doses within these selected groups are distributed
ally, a risk assessment for all potential sources should be
among the Arctic countries in a similar manner to those
undertaken, not only those of contemporary political and
to the average populations. It cannot be ruled out that
economic concern. The priority and detail with which as-
there are small numbers of individuals within other Arctic
sessments of practices are conducted should be commen-
countries having similar dietary habits to those of the se-
surate with the probability and severity of consequences
lected Canadian community which has the highest calcu-
to humans and the environment.
lated individual doses. Accordingly, comparable, or in-
· The objective of this assessment was to obtain a balanced
deed higher, doses than those calculated for the Canadian
and objective scientific assessment of the relative risks
subgroup may exist within the Arctic.
posed by radioactivity in the Arctic. The assessment is
· Releases of radionuclides from the Thule B-52 accident,
fairly comprehensive from the perspective of sources, al-
the sunken Komsomolets nuclear submarine and radioac-
though some sources/activities are not assessed in as much
tive wastes dumped in the Arctic marine environment
detail as others. The correction of deficiencies in source/ac-
have not resulted in any significant increases in human
tivity coverage are reflected in the recommendations speci-
exposures or risks to human health of Arctic residents.
fied below. The most serious limitation in the assessment
There is minimal likelihood of significant radiological
has been related to the heterogeneity in the detail with
consequences associated with any future releases of ra-
which the individual-related assessments, the consequen-
dionuclides from dumped radioactive wastes or from the
ces of previous and potential releases of radionuclides to
sunken submarine Komsomolets. There is inadequate ra-
the environment and the estimation of radiological vul-
diological justification for remediation measures to be
nerability could be addressed. The AMAP radioactivity
taken in the case of either radioactive wastes dumped in
assessment group had to make some generalisations re-
the Kara Sea or the Komsomolets submarine.
garding prevailing conditions and human activities be-
· The greatest radiological threats to human health and the
cause of the diversity of population characteristics in the
environment in the Arctic are associated with the poten-
Arctic. It also faced difficulty with the variable reliability
tial for nuclear accidents and failures in the containment
and detail of information available on dietary habits with-
of the large inventories of radioactive materials in storage
in population groups among the Arctic countries. Finally,
such as high-level liquid and solid wastes. Issues of major
information available within, or provided by, contribut-
concern in relation to the potential for effects on the Arc-
ing countries on other relevant topics, especially the na-
tic environment and its inhabitants are:
ture, consequences and probabilities of potential releases,
Accidents at nuclear power plants sited within, or close
was of variable quality and completeness. All of these de-
to, the Arctic.
ficiencies are addressed within the recommendations for
further study.
Accidents in military operations, including the handling
and storage of nuclear weapons, decommissioning and
refueling of nuclear powered vessels and radioactive
8.8.2. General recommendations
waste storage and disposal.
· Contemporary international guidance on radiation pro-
Accidents during civilian vessel operations including
tection, nuclear safety, radioactive waste management
refueling.
and emergency preparedness should be rigorously ad-
Migration of radionuclides from major uncontained
hered to by all Arctic states to minimize the probabilities
sources in the drainage basins of the Ob and Yenisey
and consequences of accidents.
rivers.
· More authoritative and comprehensive evaluations should
Releases from contained sources situated in the
be made of the risks posed to human health and the envi-
terrestrial environment.
ronment by accidents in nuclear power installations. As-
· It should be noted that the risk of accidents in the han-
sessments of the risks of releases of radionuclides and the
dling and disposal of radioactive waste, especially spent
radiological consequences for humans and the environ-
nuclear fuel, from military vessels has probably been in-
ment should be performed for all existing nuclear instal-
creased by the accelerated rate of submarine decommis-
lations in, and near, the Arctic, including Probabilistic
sioning partly imposed by recent disarmament agree-
Safety Analyses for nuclear power reactors, preferably at
ments. These activities have imposed additional technical,
PSA Level 3.
infrastructural and financial demands on processes of
· International recommendations regarding the improvement
waste management that were already inadequate to meet
of nuclear and radiation safety in the nuclear industry,
the requirements of normal operations.
which cover reactor refueling, decommissioning, and asso-
· There are deficiencies both in the assessments of some
ciated spent fuel storage and disposal operations, should be
previous accidents and of the probabilities and conse-
extended to, and implemented in, nuclear fleet operations.
Chapter 8 · Radioactivity
611
· Additional information should be obtained regarding:
· Enhanced exchange of information under the aegis of the
the habits and diets of Arctic residents; the transfer rates
Arctic Council/Arctic Environmental Protection Strategy.
of radionuclides to terrestrial and freshwater foodstuffs;
Under AMAP, an international data center on radioactive
and spatial and temporal variations in production and
contamination and sources within the Arctic has been es-
consumption patterns of locally-produced foodstuffs.
tablished. The data center's tasks should be extended to
Such information would enable more precise estimates of
ensure that the identity and source of all relevant docu-
radiological exposures and risks to Arctic inhabitants to
ments, including policy and interpretative documents, are
be obtained and provide a basis for deciding on interven-
catalogued to make them accessible throughout the Arctic
tion measures in the event of nuclear accidents.
States on a timely basis.
· Dietary analyses and measurements of the uptake of ra-
dionuclides into radiologically significant constituents of
8.8.3. Specific recommendations
diet, carried out by all countries in a systematic manner,
The general recommendations, above, lead to the following
to enable improved dose and risk assessments for both
specific recommendations:
average residents and members of the most-exposed pop-
ulation groups in the Arctic.
· Improved information on the harvesting rates of semi-
8.8.3.1. Recommendations regarding storage of spent
natural and natural food products that are not normally
nuclear fuel and radioactive waste
included in most national statistics, particularly for mush-
· Comprehensive and detailed evaluations of the handling
rooms. In addition, the collation of information on har-
and storage of spent nuclear fuel and other high-level ra-
vesting of all natural and semi-natural products at a
dioactive waste should be continued to identify and im-
higher spatial resolution than that currently available to
plement any additional measures required to ensure that
the AMAP assessment for most Arctic countries is war-
the risks of accidental and unauthorized releases of ra-
ranted. Analysis of the precision of intake-related esti-
dionuclides to the environment are minimized.
mates of dose, and comparisons between dose estimates
· Spent nuclear fuel should be removed from decommis-
obtained from dietary calculations and wholebody count-
sioned nuclear submarines currently located in Arctic
ing, respectively, to assess the accuracy of dose estima-
marine areas and defueled reactor compartments should
tions.
be stored safely on land.
· Increased attention to:
Characterizing source terms for uncontained environ-
8.8.3.2. Recommendations regarding monitoring
mental sources and any other sources not covered by
· A harmonized system for monitoring radionuclide con-
probability safety assessments of nuclear installations.
tamination of the Arctic environment for radiological
The study and modeling of terrestrial transport path-
assessment purposes should be developed. This should
ways of radionuclides and exposure pathways to Arctic
be carried out in conjunction with an evaluation of the
residents to improve the resolution and comprehensive-
purposes and efficacy of existing trend monitoring stud-
ness of radiological assessments.
ies. Particular attention in monitoring design should be
The transport of radionuclides from land sources
paid to:
through river catchments, particularly those of major
Monitoring of radionuclides in key components of
Russian rivers.
human exposure pathways for Arctic residents (i.e.,
Estimating radiation exposures to biota and associated
reindeer, mushrooms, freshwater fish).
effects on biological populations.
Monitoring of atmospheric fallout in the Arctic.
The processes of incorporation, transport and deposi-
· Systems should be developed for:
tion of contaminants in sea ice, to enable an evaluation
Early warning monitoring of fission product releases
of the relative importance of sea ice transport of radio-
from nuclear power plants, military and civilian nuclear
nuclides compared with water and sediment transport
vessel operations.
in the Arctic marine environment.
Surveillance monitoring of areas of severe environmen-
Source term reconstruction, through the analysis of
tal contamination (such as those in the Ob and Yenisey
cores from glaciers and accumulating freshwater and
drainage basins) and large contained radioactive sources
marine sediment columns.
on land (such as radionuclide thermal generators in
Determining the contribution of nuclear testing on
navigational aids).
Novaya Zemlya to local and regional fallout.
8.8.3.3. Recommendations for further study
to correct information deficiencies
Acknowledgments
The principal deficiencies in information are of two main
Editors and lead authors
types: limitations in the availability of information; and gaps
Per Strand, Mikhail Balonov, Asker Aarkrog, Michael J.
in scientific understanding that restrict the ability to estimate
Bewers, Brenda Howard, Anneli Salo, Yuri S. Tsaturov
reliably the effects of nuclear activities. Limitations in the
availability of information relate primarily to military activi-
Additional contributors
ties and dietary habits. Gaps in scientific understanding re-
L. Amosova, R.M. Andersen, A. Belikov, R. Bergman,
late to the terrestrial and hydrological transport of radionu-
D. Dasher, R.S. Dyer, M. Filippov, K. Kouprij, I. Lisovsky,
clides in the environment, the role and importance of sea ice
S. Magnusson, H. Mehli, G. Miretsky, H. Nies, A. Nikitin,
transport, and the transfer of radionuclides among environ-
S. Pallsson, T. Paluszkiewicz, T. Rahola, P.V. Ramzaev,
mental compartments within exposure pathways. Correction
K. Rissanen, B. Salbu, M. Sickel, K-L. Sjøblom, W. Temple-
of these deficiencies requires:
ton, P. Thoresen, S. Vakulovsky, A. Walton, S. Wright.
612
AMAP Assessment Report
the Arctic Ocean during 1961-1990. In: P. Strand and A. Cooke (eds.).
Environmental Radioactivity in the Arctic. Proceedings of the second
References
International Conference on Environmental Radioactivity in the Arc-
Aaltonen, H., R. Saxén and T.K. Ikäheimonen, 1990-91. Finnish Centre
tic, Oslo, August 21-25, 1995, pp. 79-83. Norwegian Radiation Pro-
for Radiation and Nuclear Saftey (STUK), Helsinki, Finland. STUK A-
tection Authority, Østerås, Norway. 416p.
74 and STUK A-75.
Coad, S., 1994. Consumption of fish and wildlife by Canadian native peo-
Aarkrog, A., 1971. Radiological investigations of plutonium in an arctic
ples: A quantitative assessment from the published and unpublished
marine environment. Health Phys. 20: 31-47.
literature. Contract report to Health Canada, Ottawa.
Aarkrog, A., 1977. Environmental behaviour of plutonium accidentally
Dahlgaard, H. (ed.), 1994a. Nordic radioecology. The transfer of radionu-
released at Thule, Greenland. Health Phys. 32: 271-284.
clides through Nordic ecosystems to man. Elsevier, Amsterdam.
Aarkrog, A., 1978. Environmental studies on radioecological sensitivity
Dahlgaard, H., 1994b. Sources of 137Cs, 90Sr and 99Tc in the East Green-
and variability with special emphasis on the fallout nuclides Sr-90 and
land Current. J. Environ. Radioact. 25: 37-55.
Cs-137. Risø National Laboratory, Roskilde, Denmark. Risø-R-437.
DCS, 1991. Decision of the Council of State, Finnish code of statutes
Aarkrog, A., 1995. Radionuclides in the Greenland marine environment.
395/1991.
Risø National Laboratory, Roskilde, Denmark. National Assessment
Degteva, M.O., V.P. Kozheurov and M.I. Vorobyova, 1992. Dose recon-
Report, 19p.
struction for the population exposed to radiation due to discharge of
Aarkrog, A., et al., 1963-1995. Environmental radioactivity in Greenland
radioactive wastes to the Techa river. Atomic energy 72(4): 386-390.
1962-1982. Risø Report Nos. 65, 87, 109, 132, 155, 182, 203, 222,
Dobson, G.M.B., 1956. Origin and distribution of the polyatomic mole-
247, 267, 293, 307, 325, 347, 363, 388, 405, 423, 448, 471 and 489
cules in the atmosphere. Proc. R. Soc., London A. 236: 187-193.
(1963-1983). Risø National Laboratory, Roskilde, Denmark.
Ecological safety of Russia, 1996. Issue 2. Proc. of the Interagency Com-
Aarkrog, A., S. Boelskifte, H. Dahlgaard, S. Duniec, E. Holm and J.N.
mission for ecological safety (Sept. 94-Oct. 95). Yuridicheskaya Lite-
Smith, 1987. Studies of transuranics in an arctic marine environment.
ratura, Moscow.
J. Radioanal. nucl. Chem. Art. 115: 39-50.
EU/CIS JSP-6, 1996. Atlas of caesium deposition on Europe after the
Aarkrog, A., H. Dahlgaard, K. Nilsson and E. Holm, 1984. Further stud-
Chernobyl accident. EC Office of Publication, Luxembourg, EUR
ies of plutonium and americium at Thule, Greenland. Health Phys. 46:
report 16733.
29-44.
Ferguson, M.A.D. and L. Gauthier, 1992. Status and trends of Rangifer
Adushkin, V. and G. Krasilov, ND. Novaya Zemlya test site and the prob-
tarandus and Ovibos moschatus populations in Canada. Rangifer 12:
lem of the radioactive pollution of the Polar Ocean.
127-140.
AMAP, 1997. Arctic pollution issues: A state of the Arctic environment
Fillippov, M., 1995. The "Lepse" vessel - storage for used damaged nu-
report. Arctic Monitoring and Assessment Programme, Oslo. 188p.
clear fuel. In: P. Strand and A. Cooke (eds.). Environmental Radio-
Amosov, P.V., A.A. Baklanov and S.I. Mazukhina, 1995. Estimation of
activity in the Arctic. Proceedings of the second International Con-
potential risk for the population from possible accident at the Kola
ference on Environmental Radioactivity in the Arctic, Oslo, August
nuclear power plant. In: P. Strand and A. Cooke (eds.). Environmental
21-25, 1995, pp. 408-412. Norwegian Radiation Protection Authority,
Radioactivity in the Arctic. Proceedings of the second International
Østerås, Norway. 416p.
Conference on Environmental Radioactivity in the Arctic, Oslo,
Filippov, M., 1996. Handling of radioactive waste produced by the Rus-
August 21-25, 1995, pp. 365-368. Norwegian Radiation Protection
sian civilian nuclear fleet. Proceedings of IMO Workshop, London.
Authority, Østerås, Norway. 416p.
Freke, A.M., F. Morley and G.R. Tyler, 1969. A review of the liquid efflu-
Arvela, H., M. Markkanen and H. Lemmelä, 1990. Mobile survey of
ent discharges and marine monitoring programme at Dounreay: 1958-
environmental gamma-radiation and fallout levels in Finland after the
1968. United Kingdom Atomic Energy Authority, Harwell, Berks.,
Chernobyl accident. Radiat. Prot. Dosim. 32: 177-183.
Authority Health and Safety Branch Report AHSB(RP) R 88, 38p.
Beak, 1995. Review of human exposure to environmental radiation in
Gladkov, G.A., N.S. Khlopkin, V.N. Lystsov, G.A. Nejdanov, B.G.
the Canadian Arctic. Beak Consultants Ltd. Contract no. H4078-4-
Pologikh and Y.V. Sivintsev, 1994. Assessment and prognosis of the
CS86101-55. Contract report to Health Canada, Ottawa.
state of nuclear installation of submarine "Komsomolets". RRC
Bjurman, B., L.E. De Geer and I. Vintersved, 1990. The detection of
Kurchatov Institute, Moscow, Russia, Report of a Working Group
radioactive material from a venting underground nuclear explosion. J.
under the leadership of Academician N.S. Khlopkin.
Environ. Radioact. 11: 1-14.
Glagolenko, Y.V., 1995. Radioactive waste management strategy at "May-
Blindheim, 1994. Physical characteristics of the area. In: R. Sætre (ed.)
ak" PA site. Paper presented at Norwegian Academy of Science and
The sunken nuclear submarine in the Norwegian Sea a potential
Letters, Oslo, Norway, April 1995.
environmental problem? Fisken og havet, No. 7, 1994.
Gray, J., S.R. Jones and A.D. Smith, 1995. Discharges to the environment
Boedeker, B., 1991. Census counts for Alaska Natives (1990). Alaska Area
from the Sellafield Site, 1951-1992. J. Radiol. Prot. 15: 99-131.
Native Health Service, Anchorage, Alaska, 15p.
Hälinen, K. and K. Sikkilä, 1993. Poronhoitajien elintavat, työ terveys
Bradley, D.J. and U.P. Jenquin, 1995. Radiactive inventories and sources
1986. Kansaneläkelaitoksen Julkaisuja ML. 127, Helsinki. (In Finnish).
for contamination of the Kara Sea by fluvial transport. Prepared for
Harley, J.H. (ed.), 1972. HASL procedures manual. HASL-300. Health
the Office of Naval Research, Pacific Northwest Laboratory, Richland,
and Safety Laboratory, U.S. Atomic Energy Commission, Department
Washington.
of Energy. 602p.
Brewer, A.W., 1949. Evidence for a world circulation provided by mea-
HASL, 1958-1978. General reference to: Environmental Quarterly reports
surements of helium and water vapour distribution in the stratosphere.
from Health and Safety Laboratory. U.S. Atomic Energy Commission,
Quart. J. R. Met. Soc. 75: 351-363.
now Environmental Measurements Laboratory, Department of Energy,
Brynildsen, L.I., T.D. Selnæs, P. Strand and K. Hove, 1996. Countermea-
New York.
sures for radiocesium in animal products in Norway after the Cher-
Hasunen, K. and H. Möttönen, 1976. Food consumption and intake of
nobyl accident - Techniques, effectiveness and costs. Health Physics
iron and potassium by Finnish Lapps in 1971 and 1976, preliminary
70(5).
report, Dept. of Nutrition and Dept. of Radiochemistry, University of
Calmet, D. and P. Guegueniat, 1985. Les rejets d'effluents liquides radio-
Helsinki.
actifs du centre de traitement des combustibles irradies de La Hague
Hasunen, K., M. Pekkarinen, E. Koskinen, R. Seppänen and L. Bäck-
(France) et l'evolution radiologique du domain marin. In: Behaviour of
ström, 1976. The food consumption and nutrient intake in Finland in
radionuclides released into coastal waters. International Atomic Ener-
1969 and 1972. Kansaneläkelaitoksen Julkaisuja ML.9, Helsinki. (In
gy Agency, Vienna, IAEA-TECDOC-329, 183p.
Finnish).
Canadian Encyclopaedia, 1985. Hurtig Publishers Ltd., Edmonton, Al-
HMSO, 1959. The control of radioactive wastes. Her Majesty's Stationery
berta.
Office, London, Cmnd. 884.
Cawse, P.A. and A.D. Horrill, 1986. A survey of 137Cs and plutonium in
Howard, B.J., K.J. Johanson, G.S. Linsley, K. Hove, G. Prohl and J. Hory-
British soils in 1977. Her Majesty's Stationery Office, London, AERE
na, 1996. Transfer of radionuclides by terrestrial food products from
R-10155.
semi-natural ecosystems. In: 2nd report of the VAMP terrestrial work-
CCMS/CDSM/NATO, 1995. Cross-border environmental problems ema-
ing group, pp. 49-79. IAEA, Vienna, IAEA-TECDOC-857.
nating from defence-related installations and activities. Final Report
Hunt, G.J., 1995. Radiation doses to critical groups since the early 1950s
Volume 1: Radioactive contamination. Phase I: 1993-1995. NATO,
due to discharges of liquid radioactive waste from Sellafield. Proceed-
Report no. 2041995.
ings of an International Symposium on Environmental Impact of
CEC, 1990. The radiological exposure of the population of the European
Radioactive Releases, May 8-12, 1995, International Atomic Energy
Community from radioactivity in North European marine waters:
Agency, Vienna.
Project `Marina'. Commission of the European Communities. Radia-
Høibråten S. and P.E. Thoresen, 1995. The sunken submarine Komso-
tion Protection Report 47, Directorate of Nuclear Safety, Impact of
molets and its effect on the environment. In: P. Strand and A. Cooke
Industry on the Environment and Waste Management, Directorate
(eds.). Environmental Radioactivity in the Arctic. Proceedings of the
General Environment, Consumer Protection and Nuclear Safety, EUR
second International Conference on Environmental Radioactivity in
12483 EN, 571p.
the Arctic, Oslo, August 21-25, 1995, pp. 350-355. Norwegian Ra-
Chernobyl Papers VI, 1993. Doses to the Soviet population and early
diation Protection Authority, Østerås, Norway. 416p.
health effects studies. S.E. Merwin and M.I. Balonov (eds.), Research
IAEA, 1988. Assessing the impact of deep sea disposal of low-level
Enterprises, Richland, Washington.
radioactive waste on living marine resources. International Atomic
Christensen, T., R. Mustonen, H. Ehdwall, H. Hansen, E. Stranden and A.
Energy Agency, Vienna, Technical Report Series No. 288, 127p.
Sørensen, 1990. Natural radiation, nuclear wastes and chemical pollu-
IAEA, 1992. INSAG-6 Probabilistic safety assessment. International
tants. Norwegian Radiation Protection Authority, Østerås, Norway.
Atomic Energy Agency, Vienna.
Chumichev, V.B., 1995. Sr-90 discharges with main rivers of Russia into
IAEA, 1993. Risk comparisons relevant to sea disposal of low level
Chapter 8 · Radioactivity
613
radioactive waste. International Atomic Energy Agency, Vienna, IAEA-
Tchernyshev, G.A. Tsyrkov and M.P. Shumaev, 1996. USSR nuclear
TECDOC -725, 43p.
weapons tests and peaceful nuclear explosions, 1949 through 1990.
IAEA, 1994. Convention on nuclear safety. International Atomic Energy
RFNC-VNIIEF, Sarov, 63p.
Agency, Vienna, IAEA Legal Series No. 16, 13p.
Miller, K.M. and H.L. Beck, 1984. Indoor gamma and cosmic ray expo-
IAEA, 1996. Predicted radionuclide release from marine reactors dumped
sure rate perturbations using a GE spectometer and pressurized ioniza-
in the Kara Sea. Report of the Source Term Working Group of the
tion chamber. Radiation Prot. Dosim. 7(1-4): 185-189.
International Arctic Seas Assessment Project, International Atomic
Moberg, L. (ed.), 1991. The Chernobyl fallout in Sweden. Swedish Radia-
Energy Agency, Vienna, IAEA-TECDOC -XXX, (to be published in
tion Protection Institute, Stockholm, 633p.
1996).
NCRP, 1987. Exposure of the population in the United States and Canada
Ilvonen, M., J. Rossi and M. Salonoja, 1994. Radiation risk in Finland
from natural background radiation. National Council on Radiation
from Sosnovyi Bor, VTT 1597, Espoo. (In Finnish)
Protection and Measurements, Bethesda, MD. NCRP Report 94.
Ivanov, V., 1994. Sedimentological description of the Kara and Barents
NEA, 1987. Nuclear Energy Act, Finnish code of statutes 990/1997.
Seas. International Atomic Energy Agency, Vienna, IAEA-IASAP-3.
NEA, 1994. The use of quantitative safety guidelines in member countries.
Izrael, Y.A., S.M. Vakulovsky, V.A. Vetrov, V.N. Petrov, F.Y. Rovinsky and
Committee on the Safety of Nuclear Installations, Nuclear Energy
Y.D. Stukin, 1990. Chernobyl: Radioactive contamination of the envi-
Agency, Organization for Economic Co-operation and Development
ronment, Gidrometeoizdat, Leningrad, 296p. (In Russian)
(OECD), Paris, Addendum to CSNI Report No. 177, Unpublished.
Johanson, K.J. and R. Bergström, 1993. Radiocaesium transfer to man
NEA, 1995. Ten years on. Radiological and health impact. An appraisal
from moose and roedeer in Sweden. Sci. Total Environ. 157: 309-316.
by the NEA Committee on Radiation Protection and Public Health.
Jokelainen, A., 1965. Diet of the Finnish Lapps and its cesium-137 and
Nuclear Energy Agency, Organisation for Economic Co-operation and
potassium contents, Acta Agralia Fennica 103, Helsinki.
Development (OECD), Paris, 33p.
Kauranen, P. and J.K. Miettinen, 1969. Polonium and radiolead in some
NED, 1988. Nuclear Energy Decree, Finnish code of statutes 161/1988.
aqueous ecosystems in Finland. In: Radioactive foodchains in the sub-
NEFCO, 1996. Proposals for environmentally sound investment projects
arctic environment, Annual Report, 1969. University of Helsinki,
in the Russian part of the Barents Region, Volume 2: Radioactive cont-
Department of Radiochemistry, Paper No. 27.
amination. The Nordic Environment Finance Corporation, Barents
Kershaw, P.J. and A.J. Baxter, 1993a. Sellafield as a source of radioactivity
Region Environmental Programme, Phase I report, prepared by the
to the Barents Sea. Proceedings of a Conference on Radioactivity and
AMAP Expert Group. AMAP, Oslo. 41p
Environmental Security in the Oceans: New Research and Policy
Nelson, V. and V.E. Noshkin, 1973. Enewetak survey report. USAEC
Priorities in the Arctic and North Atlantic, Woods Hole Oceano-
Nevada Operations Office, Las Vegas, Nevada, NVO-140, pp. 131-
graphic Institution, Woods Hole, Massachusetts, June 7-9, 1993.
224.
Kershaw, P.J. and A.J. Baxter, 1993b. The transfer of reprocessing wastes
Nevstrueva, M.A., P.V. Ramzaev, A.A. Moiseer, M.S. Ibatullin and L.A.
from N.W. Europe to the Arctic. Deep Sea Res. 42: 1413-1448.
Teplykh, 1967. The nature of 137Cs and 90Sr transport over the lichen-
Kolehmainen, S., E. Häsänen and J.K. Miettinen, 1966. Health Physics,
reindeer-man food chain. In: B. Aberg and F.P. Hungate (eds.). Radio-
Pergamon Press, 1966; Proceedings of the first International Congress
ecological concentration processes. Pergamon Press, Oxford.
of Radiation Protection, Rome, 1966. Pergamon Press, 1968.
Nies, H. and S.P. Nielsen, 1996. Radioactivity in the Baltic Sea. Proceed-
Kolomiets, B.I., A.F. Samorodov and M.P. Filippov, 1992. Radioactive
ings from «Radionuclide in the Oceans» (RADOC 1996) Cherbourg.
waste of the nuclear ice-breakers. In: Proceedings of the Russian
NKS, 1994. Design and safety features of nuclear reactors neighbouring
Nuclear Society Conference, N. Novgorod.
the Nordic countries. E. Nonbol (ed.). The Nordic Council of Ministers,
Kolstad, A.K., 1995. Expedition to "Komsomolets" in 1993 and 1994.
Copenhagen, TemaNord 1994, 595p.
NRPA Strålevern Rapport 1995:7. Norwegian Radiation Protection
NRPA, 1989. Havari av sovjetisk ubåt, ECHO II: Beredskapstiltak, radio-
Authority, Østerås, Norway.
aktivitetsmålinger, utslipp av jod og konsekvenser. Backe, S. (ed.).
Koupri, K., 1995. Problems of radioactive and radioecological safety in
Norwegian Radiation Protection Authority, Østerås, Norway. (In
the north of Europe and Russia. In: P. Strand and A. Cooke (eds.).
Norwegian).
Environmental Radioactivity in the Arctic. Proceedings of the second
NRPA, 1996. Dumping of radioactive waste and investgation of radioac-
International Conference on Environmental Radioactivity in the Arc-
tive contamination in the Kara Sea. Results from 3 years of investiga-
tic, Oslo, August 21-25, 1995, pp. 404-405. Norwegian Radiation
tions (1992-1994) in the Kara Sea. Joint Norwegian-Russian Expert
Protection Authority, Østerås, Norway. 416p.
Group for Investigation of Radioactive Contamination in the Northern
Kozheurov, V.P. and M.O. Degteva, 1994. Dietary intake evaluation and
Areas, Norwegian Radiation Protection Authority, Østerås, Norway.
dosimetric modelling for the Techa river residents based on vivo mea-
NRPA, 1997. Sources contributing to radioactive contamination of the
surements of Strontium-90 in teeth and skeleton. Sci. Total Environ.
Techa river and areas surrounding the "Mayak" production associa-
142: 63-72.
tion, Urals, Russia. Joint Norwegian-Russian Expert Group for
Laitinen, J., S. Näyhä, K. Sikkilä and J. Hassi, 1996. Diet and cardiovas-
Investigation of Radioactive Contamination in the Northern Areas,
cular risk factors among Lapp and Finnish reindeer herders. Nutrition
Norwegian Radiation Protection Authority, Østerås, Norway.
Research vol. 16, No.7.
OPRF, 1993. Yablokov, A.V., V.K. Karasev, V.M. Rumyantsev, M.E.
Lebedev, V.A, Y.S. Volozhanin, P.V. Morozov, Y.A. Revenko, V.V. Zhidkov
Kokeev and O.J. Petrov (eds.). Facts and problems related to radioac-
and V.V. Ivanov, 1996. Deep in Siberia's mines, let naught... In: Atom
tive waste disposal in seas adjacent to the territory of the Russian
declassified. Second collection. Half a century with the bomb, pp 53-
Federation. Materials for a report by the Government Commission on
62. IPPNW, Moscow.
matters related to radioactive waste disposal at sea. Created by Decree
Leemans, R. and W.P. Cramer, 1991. The IIASA database for mean
No. 613 of the Russian Federation President of October 24, 1994.
monthly values of temperature, precipitation and cloudiness on a glob-
White Book No. 3, Small World Publisher, Moscow.
al terrestrial grid. Digital raster data on a 30 min. geographic
OTA , 1995. Nuclear wastes in the Arctic: An analysis of Arctic and other
(lat./long.) 360 x 720 grid. IIASA, Laxenburg, Austria, 9-track tape,
regional impacts from Soviet nuclear contamination. U.S. Government
10.3 MB.
Printing Office, Washington, D.C., OTA-ENV-623.
Lin, P.H., C.J. Chen, C.C. Huang and Y.M. Lin, 1986. Study of the
Palsson, S.E., K. Eglisson, S. Porisson, S.M. Magnusson, E.D. Olafsdottir
indoor cosmic radiation ionization intensity. Radiat. Prot. Dosim.
and K. Indridason, 1994. Transfer of radiocaesium from soil and
16(4): 329-332.
plants to reindeer in Iceland. J. Environ. Radioact. 24: 107-125.
Lisovsky, I., O. Petrov and A. Belikov, 1996. Radioactive contamination
Paluszkiewicz, T., L.F. Hibler, M.C. Richmond and D.J. Bradley, 1995. An
of the Arctic by the North Fleet of Russia. In: M. Balonov (ed.).
assessment of the flux of radionuclide contamination through the Ob
Radionuclides in the Russian Arctic. Part 2 of a report to the Arctic
and Yensei rivers and estuaries to the Kara Sea. In: P. Strand and A.
Monitoring and Assessment Programme, Oslo, 76p.
Cooke (eds.). Environmental Radioactivity in the Arctic. Proceedings
Lystsov, N., 1995. Environmental impact of Peaceful Underground Nu-
of the second International Conference on Environmental Radioac-
clear Explosions (PUNE) in the Russian North. Russian Research
tivity in the Arctic, Oslo, August 21-25, 1995, pp. 38-50. Norwegian
Center " Kurchatov Institute".
Radiation Protection Authority, Østerås, Norway. 416p.
MAFF, 1967. Radioactivity in surface and coastal waters of the British
Panteleyev, G.P., H.D. Livingstone, F.L. Sayles and O.N. Medkova, 1995.
Isles (1967). Ministry of Agriculture, Fisheries and Food, Fisheries
Deposition of plutonium isotopes and Cs-137 in sediment of the Ob
Radiological Laboratory, Lowestoft, Suffolk, Technical Report FRL 1,
delta from the beginning of the nuclear age. In: P. Strand and A.
45p.
Cooke (eds.). Environmental Radioactivity in the Arctic. Proceedings
Malyshev, S.V., 1995. Formation of radioactive contamination at "May-
of the second International Conference on Environmental Radioac-
ak" PA site. Paper presented at Norwegian Academy of Science and
tivity in the Arctic, Oslo, August 21-25, 1995, pp. 57-64. Norwegian
Letters, Oslo, Norway, April 1995.
Radiation Protection Authority, Østerås, Norway. 416p.
Mayak, 1993. Information given during the official Norwegian visit to
Pavlov, V.K., 1994. Oceanographical description of the Kara and Barents
"Mayak" PA, May 1993.
Seas. Working Material of the International Arctic Seas Assessment
Meese D., L. Cooper, I.L. Larsen, W. Tucker, E. Reimnitz and J. Greb-
Project, IAEA-IASAP-2, International Atomic Energy Agency, Vienna.
meier, 1995. Caesium-137 contamination in Arctic Sea ice. In: P.
Pentreath, R.J., 1988. Radionuclides in the aquatic environment. In: J.H.
Strand and A. Cooke (eds.). Environmental Radioactivity in the Arctic.
Harley, G.D. Schmidt and G. Silini (eds.). Radionuclides in the food
Proceedings of the second International Conference on Environmental
chain. Springer-Verlag, Berlin .
Radioactivity in the Arctic, Oslo, August 21-25, 1995, pp. 195-198.
Pentreath, R.J., D.S. Woodhead, P.J. Kershaw, D.F. Jeffries and M.B.
Norwegian Radiation Protection Authority, Østerås, Norway. 416p.
Lovett, 1986. The behaviour of plutonium and americium in the Irish
Mikhailov, V.N., I.A. Andryshin, V.V. Bogdan, S.A. Vashchinkin, S.A.
Sea. Rapp. P.-v. Réun. Cons. Int. Explor. Mer 186: 60-69.
Zelentsov, G.E. Zolotukhin, V.M. Karimov, V.V. Kirichenko, A.M.
Petrov, O., 1995. Radioactive wastes and spent nuclear fuel in the Navy
Matushchenko, Y.A. Silkin, V.G. Strukov, K.V. Kharitonov, A.K.
of Russia. In: P. Strand and A. Cooke (eds.). Environmental Radioac-
614
AMAP Assessment Report
tivity in the Arctic. Proceedings of the second International Conference
tions in Russia. Sci. Total Environ. 193: 237-242.
on Environmental Radioactivity in the Arctic, Oslo, August 21-25,
Smith, J.N., K.M. Ellis, A. Aarkrog, H. Dahlgaard and E. Holm, 1994.
1995, pp. 404-405. Norwegian Radiation Protection Authority, Øster-
Sediment mixing and burial of the 239,240Pu from the 1968 Thule,
ås, Norway. 416p.
Greenland nuclear weapons accident. J. Environ. Radioact. 25: 135-
Playford, K., J. Toole and L. Adsley, 1993. Radioactive fallout in air and
159.
rain; results to the end of 1991. United Kingdom Atomic Energy Au-
Smith, J.N., K.M. Ellis, K. Naes, S. Dahle and D. Matishov, 1995a.
thority, Harwell, UK, AEA-EE-0498.
Sedimentation and mixing rates of radionuclides in Barents Sea sedi-
RA, 1991. Radiation Act, 1991, Finnish code of statutes 592/1991.
ments off Novaya Zemlya. Deep Sea Res. II 42: 1471-1493.
Rahola T., T. Jaakkola, M. Suomela, M. Tillander and K. Rissanen, 1993.
Smith, J.T., J. Hilton and R.N.J. Comans, 1995b. Application of two sim-
In: P. Strand and E. Holm (eds.). Environmental Radioactivity in the
ple models to the transport of 137Cs in an upland organic catchment.
Arctic and Antarctic. Proceedings of the International Conference on
Sci. Total Environ. 168: 57-61.
Environmental Radioactivity in the Arctic and Antarctic, Kirkenes,
SSI, 1994. Activity releases from and worker doses in nuclear industry up
1993. Norwegian Radiation Protection Authority, Østerås, Norway.
to 1994. Swedish Radiation Protection Institute (SSI) reports 91-05,
Rahola, T. and K. Rissanen, 1996. Rahola, T., (Institute of Radiation
91-11, 92-15, 93-23, 94-05 and 95-13. (In Swedish).
Chemistry, University of Helsinki), STUK, Helsinki; Rissanen, K., Fin-
SSI FS, 1991. Regulation by the Swedish Radiation Protection Institute
nish Centre for Radiation and Nuclear Saftey (STUK), Rovaniemi,
(SSI), 1991. SSI FS 1977:2 revised as SSI FS 1991:5.
partly unpublished.
Steingrimsdottir L., Throgeirdottir and S. Ægisdottir, 1991. A survey of
Ramzaev, P.V., G.I. Miretsky, M.N. Troitskaja and A.A. Doudrarev, 1993.
dietary habits in Iceland 1990. Iceland Nutrition Council, Reykjavik,
Radioecological peculiarities around Novaya Zemlya (USSR) atomic
104p.
testing range (1993). Int. J. Rd. Hygiene 1(1): 1-13.
Strand, P., 1994. Radioactive fallout in Norway from the Chernobyl acci-
Rantalainen, L., 1995. Source terms of the Kola nuclear power plant and
dent. NRPA Report 1994:2. Norwegian Radiation Protection Author-
risk of severeenvironmental contamination. In: P. Strand and A. Cooke
ity, Østerås, Norway.
(eds.). Environmental Radioactivity in the Arctic. Proceedings of the
Strand, P., G. Kjølås, J.B. Reitan, T. Strand, T. Berthelsen and T.D. Selnæs,
second International Conference on Environmental Radioactivity in
1990. Doses from external and internal exposure in Norway the first
the Arctic, Oslo, August 21-25, 1995, pp. 359-364. Norwegian Radia-
year following the Chernobyl accident. National Institute of Radiation
tion Protection Authority, Østerås, Norway. 416p.
Hygiene, Østerås, Norway. SIS report 1990:2.
RCRA, 1997. Radioactive contamination in the Russian Arctic. M. Balo-
Strand, P., A.I. Nikitin, B. Lind, B. Salbu and G.C. Christensen, 1997.
nov, Y. Tsaturov, B. Howard and P. Strand (eds.). Report of Russian
Dumping of radioactive waste and investigation of radioactive contam-
experts for the AMAP. Oslo.
ination in the Kara Sea. Results from 3 years of investigations (1992-
RD, 1991. Radiation Decree. Finnish code of statutes 1512/91.
1994) in the Kara Sea. Joint Norwegian-Russian Expert Group for
Rissanen, K. Finnish Centre for Radiation and Nuclear Saftey (STUK),
investigation of radioactive contamination in the northern areas. Nor-
Rovaniemi, Unpublished.
wegian Radiation Protection Authority, Østerås, Norway.
Rissanen, K. and T. Rahola, 1993. Radiocesium in special terrestrial food-
Strand, P., A.L. Rudjord, B. Salbu, G. Christensen, L. Foyn, B. Lind, H.
chains in Finnish Lapland. In: P. Strand and E. Holm (eds.).
Bjørnstad and T. Bjerk, 1993. Survey of artificial radionuclides in the
Environmental Radioactivity in the Arctic and Antarctic. Proceedings
Kara Sea 1993. In: P. Strand and E. Holm (eds.). Environmental Ra-
of the International Conference on Environmental Radioactivity in the
dioactivity in the Arctic and Antarctic. Proceedings of the International
Arctic and Antarctic, Kirkenes, 1993. pp 375-380. Norwegian Radia-
Conference on Environmental Radioactivity in the Arctic and Antarc-
tion Protection Authority, Østerås, Norway.
tic, Kirkenes, 1993. Norwegian Radiation Protection Authority, Øster-
Rissanen, K. and T. Rahola, 1996. Radioactivity levels in foodstuffs in
ås, Norway.
Finnish Lappland. Proceedings of the 11th Meeting of the Nordic
Strand P., M. Sickel, A. Aarkrog, J.M. Bewers, Y. Tsaturov and S. Mag-
Society for Radiation Protection, Reykjavik, 26.-29 Aug.1996 (in
nusson, 1996. Radioactive contamination of the Arctic environment.
press).
Les Editions de Physique.
Risø-Report, 1970. Projected crested ice. Risø National Laboratory,
Strand, P., T.D. Selnæs, E. Bøe, O. Harbitz and A. Andersson-Sørlie,
Roskilde, Denmark. Risø-Report No. 213, 98p.
1992. Chernobyl fallout: Internal doses to the Norwegian population
Risø-Report, 1986-1993. Risø National Laboratory, Roskilde, Denmark.
and the effect of dietary advice. Health Phys. 63(4): 385-392.
Risø Report Nos 541, 550, 564, 571 and 622.
Strand, T., P. Strand and J. Baarli, 1987. Radioactivity in foodstuffs and
Romanov, G.N., 1995. Modern radioecological situation in the "Mayak"
doses to the Norwegian population from the Chernobyl Fallout. Ra-
PA site region. Paper presented at Norwegian Academy of Science and
diat. Prot. Dosim. 20: 211-230
Letters, Oslo, Norway, April 1995.
STUK, 1991. Safety assessment, based on the decision of the Council of
RPA, 1988. Radiation Protection Act, 1988. Swedish code of statutes.
State 395/91, for Loviisa NPP (STUK-B-YTO 81) and for Okiluoto
RPO, 1988. Radiation Protection Ordinance, 1988. Swedish code of sta-
(STUK-B-YTO 82). Finnish Centre for Radiation and Nuclear Safety
tutes 1988: 293.
(STUK), Helsinki.
Russian Federal Program, 1995. Handling of radioactive wastes and spent
STUK, 1992. Limitation of public exposure in the environment of and
nuclear fuel, their utilisation and disposal in 1996-2005. Approved by
limitation of radioactive releases from nuclear power plants. Finnish
the Government of Russian Federation on 23.10.95.
Centre for Radiation and Nuclear Safety (STUK), Helsinki. Guide
Salo, A., M. Puhakainen, R. Saxén and H. Aaltonen, 1966-1996a. Finnish
YVL 7.1, 9p.
Centre for Radiation and Nuclear Safety (STUK), Helsinki. SFL-44,-
STUK, 1994. Annual reports concerning operation of Finnish nuclear
A8, -A12, - A17, STL -A21,-A26, -A32,-A34, -A38, -A40, -A47, -A48,
power plants. Annual reports of environmental radioactivity and cor-
STUK-A54,- A57, -A75, -A109; Saxén, R. (unpublished)
responding data bases (up to 1994). Finnish Centre for Radiation and
Salo, A., M. Puhakainen, K. Tuomainen, R. Saxén, H. Aaltonen and U.
Nuclear Safety (STUK), Helsinki.
Koskelainen, 1966-1996b. Finnish Centre for Radiation and Nuclear
STUK-A62, 1987. Rantavaara A., T. Nygren, K. Nygren and T. Hyvönen.
Safety (STUK), Helsinki. SFL-A8,-A12, STL-A21,-A32,-A34, -A38, -
Radioactivity of game meat in Finland after the Chernobyl accident in
A40, -A47, -A48, STUK -A54, -A55, -A77, -A89.
1986. Supplement 7 to Annual Report STUK-A55. Finnish Centre for
Saxén R. and U. Koskelainen, 1996. Radioactivity of surface water and
Radiation and Nuclear Safety (STUK), Helsinki.
freshwater fish in Finland in 1991-1994. Finnish Centre for Radiation
STUK-A74, 1991. Finnish Centre for Radiation and Nuclear Safety (STUK),
and Nuclear Safety (STUK), Helsinki. STUK-A129,
Helsinki.
Sazykina, T.G. and I.I. Kryshev, 1994. Provision of site specific input data
STUK-A78, 1991. Rantavaara, A., Radioactivity of foodstuff in Finland in
for and assessment of radiological impact of waste dumping into the
1987-88. Supplement 4 to Annual Reports STUK-74 and STUK-A89.
Barents and Kara Seas. Working material of the International Arctic
Finnish Centre for Radiation and Nuclear Safety (STUK), Helsinki.
Seas Assessment Project, IAEA-IASAP-4. International Atomic Energy
STUK-94A, 1992. Saxén, R. and U. Koskelainen. Radioactivity of surface
Agency, Vienna.
water and freshwater fish in Finland on 1988-1990. Supplement to
Sinkko, K., H. Aaltonen, R. Mustonen, T.K. Taipale and J. Juutilainen,
Annual Report STUK A-89. Finnish Centre for Radiation and Nuclear
1987. Finnish Centre for Radiation and Nuclear Safety (STUK), Hel-
Safety (STUK), Helsinki.
sinki. STUK A-55 and STUK A-56.
Swanson, J.D. and M.H.W. Barker, 1991. Assessment of Alaska reindeer
Sivintsev, Y., 1994a. Study of the nuclides composition and characteristics
populations and range conditions. Rangifer 12: 33-43.
of the fuel in dumped submarine reactors and atomic icebreaker "Le-
Sweden, 1982. Decision by Government 1982-06-2782, Stockholm.
nin". Part 1 Atomic icebreaker. Working material of the Interna-
Swedish Institute of Radiation Protection and Swedish Nuclear Power
tional Arctic Seas Assessment Project, IAEA-IASAP-1. International
Inspectorate, 1995. Consequences of a severe accident in Sweden.
Atomic Energy Agency, Vienna.
Swedish Institute of Radiation Protection and Swedish Nuclear Power
Sivintsev, Y., 1994b. Study of the nuclides composition and characteristics
Inspectorate, Stockholm. (In Swedish)
of the fuel in dumped submarine reactors and atomic icebreaker "Le-
Toivonen, H., S. Klemola, J. Lahtinen, A. Leppänen, R. Pöllänen, A.
nin". Part 2 Nuclear submarines. Working material of the Interna-
Kansanaho, A.L. Savolainen, A. Sarkanen, I. Valkama and M. Jäntti,
tional Arctic Seas Assessment Project, IAEA-IASAP-5. International
1992. Finnish Centre for Radiation and Nuclear Safety (STUK),
Atomic Energy Agency, Vienna.
Helsinki. STUK A-104.
Sivintsev, Y., 1995. Description of shielding barriers at the nuclear reac-
Tracy, B.L., G.H. Kramer, J.M. Zielinski and H. Jiang, 1995.
tors dumped in the Arctic. Working material of the International
Radiocaesium body burdens in residents of Northern Canada from
Arctic Seas Assessment Project, IAEA-IASAP-8. International Atomic
1963 to 1990. Preprint of submission to Health Physics.
Energy Agency, Vienna.
Trapeznikov, A., V. Pozolotina, M. Chebotina, V.N. Chukanov, V.N.
Skuterud, L., I.G. Travnikova, M.I. Balonov, P. Strand and B.J. Howard,
Trapeznikova, N.V. Kulikov, S.P. Nielsen and A. Aarkrog, 1993.
1997. Contribution of fungi to radiocaesium intake by rural popula-
Radioactive contamination of the Techa River, the Urals. Health Phys.
Chapter 8 · Radioactivity
615
(accepted for publication in 1993).
Trapeznikov A., V. Pozolotina, V. Trapeznikova, P. Yushkov, 1995.
Distribution of artificial radionuclides in the sediments of some Urals
rivers, Ob Basin. In: P. Strand and A. Cooke (eds.). Environmental
Radioactivity in the Arctic. Proceedings of the second International
Conference on Environmental Radioactivity in the Arctic, Oslo,
August 21-25, 1995, pp. 65-67. Norwegian Radiation Protection
Authority, Østerås, Norway. 416p.
Tronstad, E., D.H. Oughton, A. Nøren, B. Salbu, L. Skipperud, A.P.
Clacher, G.N. Romanov, P. Strand and G.C. Christensen, 1995.
Mobilization of 90 Sr and other radionuclides from Mayak and
Chernobyl sediments and soils. In: P. Strand and A. Cooke (eds.).
Environmental Radioactivity in the Arctic. Proceedings of the second
International Conference on Environmental Radioactivity in the
Arctic, Oslo, August 21-25, 1995, pp. 256-260. Norwegian Radiation
Protection Authority, Østerås, Norway. 416p.
Tsaturov, Y., 1996. Answer to AMAP request, based on information from
the Federal Nuclear and Radiation Safety Authority of the Russian
Federation. (unpublished).
UNSCEAR, 1982. Ionizing radiation: Sources and biological effects.
United Nations Scientific Committee on the Effects of Atomic
Radiation, New York, 773p.
UNSCEAR, 1988. Sources, effects and risks of ionizing radiation. United
Nations Scientific Committee on the effects of Atomic Radiation. New
York.
UNSCEAR, 1993. Sources and effects of ionizing radiation. United Na-
tions Scientific Committee on the Effects of Atomic Radiation, New
York, 922p.
Vakulovsky, S., A. Nikitin and V. Chumichev, 1993. Radioactive contami-
nation of the Russian Arctic seas from results of observations during
the years 1960-1992. In: P. Strand and E. Holm (eds.). Environmental
Radioactivity in the Arctic and Antarctic. Proceedings of the Inter-
national Conference on Environmental Radioactivity in the Arctic and
Antarctic, Kirkenes, 1993. pp. 177-182. Norwegian Radiation Protec-
tion Authority, Østerås, Norway.
Walton, A., 1995. Canadian source data on radionuclides in environmen-
tal materials from the Arctic. Contract report to the Department of
Fisheries and Oceans, Canada.
Williams, T.M. and D.C. Heard, 1986. World status of wild Rangifer ta-
randus populations. Rangifer Special Issue 1: 19-28.
Wix, L.F.U., A. Fairbairn and H.J. Dunster, 1960. A review of the moni-
toring associated with the discharge of radioactive liquid effluent to
sea at Windscale from early 1953 until the end of 1958. United King-
dom Atomic Energy Authority, Harwell, Berks. Authority Health and
Safety Branch Report AHSB(RP)R 1, 62p.
Woodhead, D.S., 1982. The natural radiation environment of marine
organisms and aspects of the human food chain. J. Soc. Radiol. Prot.
2(4): 18-25.
Yefimov, E., 1994. Radionuclides composition, characteristics of shielding
barriers and analyses of weak points of the dumped reactors of subma-
Annex
rine N. 601. Working material of the International Arctic Seas Assess-
ment Project, IAEA-IASAP-6. International Atomic Energy Agency,
Vienna.
Contents
Yiou, F., G.M. Raisbeck, Z.Q. Zhou, L.R. Kilius and P.J. Kershaw, 1995.
Table 8·A1. 137Cs activity concentrations (Bq/kg) in food products
Improved estimates of oceanic discharges of 129I from Sellafield and La
consumed by the average population in Arctic Finland 616
Hague. In: P. Strand and A. Cooke (eds.). Environmental Radioactivity
Table 8·A2. 137Cs activity concentrations (Bq/kg) in food products
in the Arctic. Proceedings of the second International Conference on
consumed by Finnish Sami reindeer herders, (1550 people) 616
Environmental Radioactivity in the Arctic, Oslo, August 21-25, 1995,
Table 8·A3. 137Cs activity concentrations (Bq/kg) in some food prod-
pp. 113-116. Norwegian Radiation Protection Authority, Østerås,
ucts consumed by the Greenland average population
Norway. 416p.
and by the selected group . . . . . . . . . . . . . . . . . . . . . . . 616
Table 8·A4. 137Cs activity concentrations (Bq/kg) in food products
consumed by the average population in northern Can-
ada and by the selected group . . . . . . . . . . . . . . . . . . . 617
Table 8·A5. 137Cs activity concentrations (Bq/kg) in food products
consumed by the average population of the Russian Arctic 617
Table 8·A6. 137Cs activity concentrations (Bq/kg) in food products
consumed by reindeer herdsmen and their families in
the western part of the Russian Arctic . . . . . . . . . . . . . 617
Table 8·A7. 137Cs activity concentrations (Bq/kg) in food products
consumed by reindeer herdsmen and their families in
the eastern part of the Russian Arctic . . . . . . . . . . . . . . 617
Table 8·A8. 137Cs activity concentrations (Bq/kg) in some food prod-
ucts consumed in Arctic Norway by average population
and the selected group . . . . . . . . . . . . . . . . . . . . . . . . . 618
Table 8·A9. 137Cs activity concentrations (Bq/kg) in food products
consumed by the average population in Arctic Sweden 618
Table 8·A10. 137Cs activity concentrations (Bq/kg) in food products
consumed by Swedish reindeer herders. . . . . . . . . . . . . 618
Table 8·A11. 90SR activity concentrations (Bq/kg) in some food prod-
ucts consumed by the Greenland population . . . . . . . . 618
Table 8·A12. 90SR activity concentrations (Bq/kg) in food products
consumed by the average population of the Russian
Arctic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619
Table 8·A13. 90SR activity concentrations (Bq/kg) in food products
consumed by reindeer herdsmen and their families in
the western part of the Russian Arctic . . . . . . . . . . . . . 619
Table 8·A14. Sr-90 activity concentrations (Bq/kg) in food products
consumed by reindeer herdsmen and their families in
the eastern part of the Russian Arctic . . . . . . . . . . . . . 619
Document Outline
- Go to opening screen
- 8.1. Introduction
- 8.2. Fundamentals and definitions
- 8.2.1. Radioactivity
- þÿ
- 8.2.3. The system of radiological protection
- 8.2.4. Modeling
- 8.2.5. The AMAP assessment
- 8.3. Past and present radioactive contamination of the Arctic
- 8.3.1. Geographical distribution of radioactive contamination
- 8.3.2. Time dependence of radioactive contamination
- 8.3.3. Human wholebody measurements
- 8.3.4. Summary
- 8.4. Individual doses to man estimated from environmental measurements
- 8.4.1. Natural radiation
- 8.4.2. Radionuclide contamination
- 8.4.3. Intakes of 137 Cs through various dietary components
- 8.4.4. Summary
- 8.5. Source-related assessments of past and present releases
- 8.5.1. Nuclear explosions
- 8.5.2. Operational releases from the nuclear fuel cycle
- 8.5.3. Accidental releases
- 8.5.4. Summary
- 8.6. Source-related assessments of potential releases
- 8.6.1. Nuclear power plant reactor accidents
- 8.6.2. Potential accidental releases from nuclear vessels and nuclear storage sites
- 8.6.3. Potential releases from reprocessing plants
- 8.6.4. Radioactive wastes dumped at sea
- 8.6.5. Nuclear weapons
- 8.6.6. Radionuclide thermoelectric generators
- 8.6.7. Summary
- 8.7. Spatial analysis of vulnerability of Arctic ecosystems
- 8.7.1. Sources of radionuclide intake by humans
- 8.7.2. Spatial distribution of Arctic communities
- 8.7.3. Spatial differences in transfer through pathways
- 8.7.4. Changes with time
- 8.7.5. Transfer coefficients and relationships
- 8.7.7. Sensitivity to uncertainties: radiocaesium in fungi and berries
- 8.7.8. Flux vulnerability of Arctic Norway
- 8.7.9. Summary
- 8.8. Conclusions and recommendations
- 8.8.1. Conclusions
- 8.8.2. General recommendations
- 8.8.3. Specific recommendations
- Acknowledgments
- References
- Annex