19
Chapter 3
Radioactive Contamination
and Vulnerability of Arctic Ecosystems
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
centration in environmental compartment, if dose to
3.1. Introduction
man is not applicable) following a unit radionuclide in-
Monitoring the levels and trends of man-made radionu-
put to the environment. The estimation of integrated trans-
clides in Arctic environments is a central part of the
fer coefficients (Bq/m3 per kBq/m2) in the first AMAP as-
AMAP programme. The first AMAP assessment pre-
sessment is an example of such analysis.
sented several radionuclide time series for the Arctic. Al-
Five site-specific studies are included here, covering
though, some of these have been extended in the present
Arctic areas that were inadequately dealt with in the first
assessment, the main emphasis has been on the provi-
assessment or for which significantly improved data are
sion of new information. Milk is a key foodstuff which
now available.
was not specifically addressed in the monitoring and
trend section of the first assessment. To compensate for
3.2. Atmosphere
this, a number of time trends in 137Cs and 90Sr activity
concentrations in milk are described below.
Measurements of activity concentrations in the atmos-
Although well known, the presence of 99Tc and 129I
phere provide one of the best means of detecting recent
in the Arctic from fallout and earlier Sellafield dis-
releases of radionuclides. Measurements in the surface
charges was not addressed in the first AMAP assessment
atmosphere have not detected any significant new re-
because they were considered of less radiological signifi-
leases of artificial radioactivity over the last six years; in
cance than other radionuclides. However, increased rates
the Russian Arctic, for example, low mean annual concen-
of discharge from European reprocessing plants have
trations in the surface atmospheric layer, of < 0.1 10ญ5
raised awareness of these radionuclides, especially tech-
Bq/m3 of 137Cs and < 0.02 10 ญ5 Bq/m3 of 90Sr, have been
netium, and brought the issue into political focus. Also,
measured. Recently reported data for 239/240Pu deposi-
new information on remobilization of sedimented ra-
tion in Arctic Finland after the Chernobyl accident are
dionuclides from earlier discharges has made 239,240Pu
also low at < 0.25 Bq/m2 (Paatero et al., 2002).
activity concentrations more relevant for current moni-
The first AMAP assessment contained few air meas-
toring activities.
urement data for Arctic areas in the early-1960s; the pe-
Available data have improved in terms of their quan-
riod of most intensive atmospheric nuclear weapons
tity, the range of variables monitored, and the length of
tests. Atmospheric 137Cs activity concentrations in Fin-
time series. This has enabled analysis of vulnerability for
land are now available for a 40-year period (Figure 3ท1).
Arctic pathways of radiation with respect to three crite-
This dataset is based on measurements obtained at six
ria: spatial variation in transfer rates, spatial variation in
sites and for different time periods. Because there is little
ecological half-lives, and variation in contamination be-
spatial variation in precipitation across Finland, the data
tween species. Radioecological vulnerability, which is
for the two southerly sites (Nurmijไrvi and Seutula)
also referred to as radioecological sensitivity (Howard,
from 1971 until the Chernobyl accident in 1986 are typ-
2000), is considered in terms of the dose to man (or con-
ical of Arctic conditions. The post-1995 data indicate
137Cs concentration in atmosphere, ตBq/m3
10 000
Ivalo
Kevo
Seutula
1000
Nurmijไrvi
Rovaniemi
Karhutunturi
100
10
1
Figure 3ท1. Atmospheric activity con-
centrations of 137Cs at several sites in
0.1
Finland.
1960
1970
1980
1990
2000

---

20
AMAP Assessment 2002: Radioactivity in the Arctic
lower levels in winter, due to snow cover, and higher lev-
99 Tc releases from Sellafield, TBq/yr
els when snow is not present, due to resuspension. The
99 Tc concentration in Fucus, Bq/kg dw
newly available data indicate higher values during the
early-1960s than previously reported by AMAP for this
200 400
period. Measurements at Ivalo during the peak period of
global fallout are highly variable (Aaltonen et al., 2002a).
Some new measurements from summer 1986 have
Sellafield
also been made available since the first assessment (Aal-
150 300
tonen et al., 2002b). After the Chernobyl accident, at-
Fucus
mospheric 137Cs activity concentrations in Arctic Fin-
land were comparable to the peak values recorded in the
100 200
1960s, but the spatial variability was much greater.
137Cs concentration in precipitation, Bq/m2/yr
50 100
10 000
Rj๚pnahๆd, Iceland
1000
0
0 1990
1992
1994
1996
1998
2000
2002
Figure 3ท3. Temporal variation in 99Tc activity concentrations in
Fucus at Hilles๘y (northern Norway) and releases from the Sella-
100
field reprocessing plant (Kolstad and Lind, 2002).
range 1 105 to 2.6 105, as equilibrium between 99Tc
10
in the water and seaweed appears to occur slowly (Kol-
stad and Lind, 2002).
In 1999, 99Tc activity concentrations observed along
1
the Norwegian coast were associated with the peak re-
1960
1965
1970
1975
1980
1985
leases from Sellafield in 1995, whereas those observed in
Greenlandic and Icelandic waters were mainly due to re-
Figure 3ท2. Activity concentrations of 137Cs in precipitation at
leases from Sellafield and La Hague throughout the
Rj๚pnahๆ (near Reykjavํk). The points are quarterly measure-
ments and the line a four-period moving average.
1970s and 1980s. Global fallout levels are at least an
order of magnitude lower than discharges from Sella-
For Iceland, 137Cs activity concentrations in precipi-
field (Dahlgaard, 1994; Dahlgaard et al., 1995).
tation at Rj๚pnahๆ (near Reykjavํk) during the global
Karcher et al. (2003) simulated the 99Tc distribution
fallout period are shown in Figure 3ท2. Although the rel-
arising from Sellafield discharges between 1990 and
evant time period is not shown on this graph, there was
1999. This was done using a three-dimensional coupled
no detectable Chernobyl fallout in Iceland.
ice-ocean model forced with daily variable atmospheric
data (Karcher et al., 2002). The 99Tc was assumed to
3.3. Marine environment
have been released from January 1990 onwards, with
actual Sellafield discharge data input to the model at the
3.3.1. Technetium-99
North Channel in the northern Irish Sea. Figure 3ท4 com-
Discharges from Sellafield in the 1970s and global fall-
pares the model results for June 1999 with measured
out were mainly responsible for the initial occurrence of
99Tc activity concentrations in the Northeast Atlantic in
99Tc in Arctic seas. The recent increases in discharges
1999. Predicted activity concentrations in the southern
from Sellafield are now the major source. Activity con-
Barents Sea are in reasonable agreement with measured
centrations of around 20 to 25 Bq/kg dw were meas-
values. Values measured around Greenland, Iceland, and
ured in the seaweed Fucus vesicolosus between 1994
the Faroe Islands are mainly influenced by pre-1990 dis-
and 1996 along the northern Norwegian coast (Yiou et
charges and are not represented by the model.
al., 2002). Using 1 105 as a concentration factor for
the uptake, this corresponds to a seawater activity con-
3.3.2. Iodine-129
centration of approximately 0.02 Bq/m3.
A time series of 99Tc activity concentrations in sea-
Levels of the naturally occurring long-lived iodine iso-
water at Hilles๘y, a coastal location in northern Nor-
tope 129I have been elevated substantially over the past de-
way, showed an increase from 0.46 Bq/m3 in summer
cades, initially due to atmospheric nuclear weapons tests
1997 to a maximum of 2.0 Bq/m3 in early 2001. Follow-
and subsequently due to emissions from nuclear fuel re-
ing that peak, the levels steadily decreased to < 1 Bq/m3.
processing (see Figure 2ท3). Atmospheric weapons tests
For Fucus vesiculosus, activity concentrations increased
resulted in an increase in the ratio between radioactive 129I
from around 50 Bq/kg dw in July 1997 to > 400 Bq/kg
and its stable counterpart, 127I, in the surficial compart-
dw in January 2001. Concentrations then appeared to
ments (i.e., surface soils, surface seawater, and freshwa-
level off during 2001 and decrease throughout 2002, al-
ters) of the Northern Hemisphere from 10ญ12 to 10ญ10.
though more observations will be needed to confirm this
The iodine isotope ratio (and other radionuclide ra-
trend (Figure 3ท3). A comparison of 99Tc activity con-
tios such as 129I :137Cs and 129I :99Tc) can be used to
centrations in Fucus and seawater from this site indi-
trace the movement of water masses from the Norwe-
cates a concentration factor from water to Fucus in the
gian Coastal Current into the Arctic basin and the At-

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
21
70ฐ0'0"W
80ฐ0'0"N
50ฐ0'0"W
20ฐ0'0"W
20ฐ0'0"E
50ฐ0'0"E
80ฐ0'0"N
70ฐ0'0"E
80ฐ00'N
80ฐ00'N
70ฐ00'W
70
70ฐ ฐ0
00'E'0"E
60ฐ0'0"W
60ฐ0'0"E
70
70ฐ ฐ0'
00'N 0"N
50ฐ0'0"W
50ฐ00'W
50
50ฐ ฐ0
00'E'0"E
70ฐ0'0"N
70ฐ00'N
40ฐ0'0"W
99 Tc concentration in
Hilles๘y
40ฐ0'0"E
upper water column,
Bq/m2
Modelled
0-0.001
ing
0.001-0.025
0.025-0.05
30ฐ0'0"W
60
60ฐ ฐ0'
00'N 0"N
01
0.05-0.1
60ฐ0'0"N
60ฐ00'N
0.1-0.25
000000 0.025
0.25-0.5
000000 0.05
30
30ฐ ฐ0
00'E'0"E
0.5-1
000000 0.1
1-2.5
ing
000000 0.25
2.5-5
000000 0.5
5-10
10-25
000000 1
01
25-50
000001 2.5
1
000001 5
Measured values
000001 10
50
50ฐ ฐ0'
00'N 0"N
0.01
50ฐ0'0"N
50ฐ00'N
0
000001 25
0.1
0000
in 01
g
50
20ฐ00'W
10
20ฐ0'0"W
10ฐ00'W
00ฐ00'
10ฐ00'E
20ฐ00'E
10ฐ0'0"W
0ฐ0'0"E
10ฐ0'0"E
20ฐ0'0"E
Figure 3ท4. Simulated distribution of 99Tc concentrations in the upper 20 m of the water column in June 1999 (Karcher et al., 2003). Con-
centration is represented by the color coding. The red circles show measured values in 1999.
129 I concentration in seawater, Bq/m3
lantic Deep Water. Along the northern Norwegian and
10 ญ1
northwest Russian coasts, 129I levels, and hence the iso-
tope ratio, have been orders of magnitude higher than
the global fallout level since the late-1970s. Measure-
10 ญ2
From European reprocessing sources
ments taken one month after the Kursk submarine acci-
dent in August 2000 did not indicate any leakage of 129I
10 ญ3
from the Kursk, but confirmed the high levels and in-
Halocline
Arctic Ocean
creasing trend of 129I activity concentrations in Atlantic
Norwegian coast
Northwest Russian coast
water entering the Arctic Ocean (Figure 3ท5). The activ-
10 ญ4
Beaufort Sea
ity concentration of 129I from European reprocessing
Surface
sources in water entering the Arctic Ocean may be pre-
10 ญ5
dicted using dilution factors of 1 to 2 10ญ14 yr/m3 and
transit times of 2 to 3 yr (Cap de la Hague) and 4 to 5 yr
Natural and fallout
(Sellafield) (Figure 3ท5). This is in agreement with earlier
10 ญ6
studies on other radionuclides (e.g., AMAP, 1998; Dahl-
gaard, 1994). In the Beaufort Sea, levels in the halocline
10 ญ7
water masses have increased as a result of 129I trans-
ported from the Atlantic Ocean through the Arctic
1920
1930
1940
1950
1960
1970
1980
1990
2000
Ocean; whereas in surface waters it is mainly fallout-
Figure 3ท5. Activity concentrations of 129I in the Arctic Ocean near
derived 129I of Pacific origin that has been detected. The
the entry of Atlantic Water (Norway and northwestern Russia) and
transport time for the Atlantic halocline waters across
the Beaufort Sea (Cooper et al., 1998, 1999; Matishov et al., 2002;
Raisbeck and Yiou, 1999; Smith et al., 1998, 1999; Yiou et al.,
the Arctic Basin is in the order of one to two decades
2002).
(AMAP, 1998; Smith et al., 1998, 1999).

---

22
AMAP Assessment 2002: Radioactivity in the Arctic
Concentration in seawater, Bq /m3
3.3.3. Cesium-137 and 90Sr
30
Changes with time in 137Cs and 90Sr activity concen-
East Greenland Current
trations in surface seawater in the East Greenland Cur-
25
90Sr
137
rent (where the highest levels generally occur around
Cs
Greenland) are shown in Figure 3ท6. Whereas 90Sr activ-
20
ity concentrations in seawater decrease with an effective
15
ecological half-life (Teff ; Box 3ท1) of approximately nine
years, those of 137Cs level off and even increase during
10
the late-1980s and early-1990s due to fallout from the
Chernobyl accident and inputs from European repro-
5
cessing plants.
Recent measurements of 137Cs activity concentra-
0
tions in Fucus spp. in the Barents, Pechora, and White
1960
1970
1980
1990
2000
Seas are shown in Figure 3ท7. The highest values were
Figure 3ท6. Time trends for 137Cs and 90Sr in the East Greenland Cur-
detected in the White Sea.
rent (between 10.7ฐE and 22.35ฐE, and 70ฐN and 81.9ฐN).
Recent data (1993 to 1998) on the spatial distribu-
tion of 137Cs in sediments from Arctic areas indicate
considerable variability (Figure 3ท8). Activity concentra-
137Cs concentration in Fucus, Bq/kg dw
tions in areas with no known local sources of anthro-
pogenic radionuclides are < 20 Bq/kg dw. Higher levels
occur in the White Sea and off the Norwegian coast
(up to 60 Bq/kg dw) in areas strongly affected by Cher-
0.6 Svalbard
nobyl fallout, in the outer parts of the Yenisey estuary
(up to 80 Bq/kg dw), near the Atomflot base in the Kola
fjord (up to 200 Bq/kg dw) and in the dumping areas
along the western coast of Novaya Zemlya (up to 105
Bq/kg dw).
Consistently low 137Cs activity concentrations have
been reported in a range of marine fish species (Table
3ท1). Except for dab (Limanda limanda) and capelin
Pechora Sea 1.6
(Mallotus villosus), no significant differences were found
between the various species. Dab (with low levels) were
1.2
Table 3ท1. Activity concentrations for 137Cs in marine fish species
Kola Peninsula
(Bq/kg ww) for 1995 to 2000 (AMAP Data Centre).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
n
Mean ฑ SD
3.3
White Sea
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Haddock (Melanogrammus aeglefinus)
65
0.25 ฑ0.11
Cod (Gadus spp.)
394
0.22 ฑ0.08
Shorthorn sculpin (Myoxocephalus scorpius)
10
0.31 ฑ0.16
Flounder (Platichthys flesus)
6
0.33 ฑ0.06
Capelin (Mallotus villosus)
3
0.16 ฑ0.08
Figure 3ท7. Activity concentrations of 137Cs in Fucus (Rissanen et
Dab (Limanda limanda)
247
0.09 ฑ0.03
al., 1995).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Box 3ท1. Effective ecological half-lives
The effective ecological half-life (Teff) describes the time required for the activity concentration of a radionuclide in an envi-
ronmental compartment (often a food product) to be reduced to one half of its original activity concentration. It therefore
incorporates physical decay. The ecological half-life (Tec) does not take physical decay into account, and thus can be adapted
for different isotopes of the same element. For example, the Teff of 134Cs and 137Cs will differ because of the differences in
physical half-lives, while the Tec would be identical. The relationship between Teff and Tec for a radionuclide with a physical
half-life (Tphys) will be:
Teff = (Tphys ท 1 / Tec) / (Tphys + 1 / Tec)
Where data on long time trends are available, it is possible to model changes with time in activity concentrations. As a given
radionuclide often exists in the environment in different physical/chemical forms, with different mobilities, a multiple expo-
nential function is needed to describe the changes. In practice, a double exponential is often used, of the form:
ln2 ln2
A(t) = A0 ท (a ท exp(ญ ญญญญ ท t) + (1 ญ a) ท exp(ญ ญญญญ ท t))
Eqn. 3.1
T1
T2
where A(t) is the activity concentration at time t, A0 is the initial activity, T1 and T2 are the effective ecological half-lives
(Teff1 and Teff2) or the ecological half-lives (Tec1 and Tec2); and a is a parameter partitioning the decay between the two half-
lives. In this case, the shorter half-life will dominate the decay for the first period, and the longer half-life will govern the
process on a longer time scale.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
23
137Cs concentration
in sediment,
Bg / kg dw
< 1
1- 5
5 - 20
20 -100
100-100 000
Figure 3ท8. Distribution of 137Cs in surface sediments from 1993 to 1998.
137Cs concentration in cod and haddock,
239, 240Pu concentration in seawater
Bq/ kg ww
2 mBq /m3
0.12
66 mBq /m3
0.37
Barents
Sea
Norwegian Sea
Figure 3ท9. Average 137Cs activity concentrations in cod and had-
dock from 1995 to 2000.
sampled near Iceland, where 137Cs levels in seawater are
generally low. Figure 3ท9 shows the spatial variation in
North
mean 137Cs activity concentrations in cod (Gadus spp.)
Sea
and haddock (Melanogrammus aeglefinus) from 1995 to
2000.
Figure 3ท10. Distribution of 239,240Pu in surface seawater of the
northern seas in 1995 (Gr๘ttheim, 2000).
3.3.4. Plutonium isotopes
3.10
The 239,240Pu activity concentration in surface waters of
been transported with the currents into the Norwegian
northern seas ranged from 2 to 66 mBq/m3 in 1995 (Fig-
Sea and the Barents Sea and, to a lesser extent, even the
ure 3ท10). The highest 239,240Pu activity concentrations
Greenland Sea. The ratios indicate that Sellafield could
occurred in surface seawater off the north and northeast
have been a major contributor to the Pu concentrations
coasts of Scotland. This is consistent with findings on re-
observed in Scottish and Norwegian waters (Gr๘ttheim,
mobilized plutonium (Pu) from Irish Sea sediments as a
2000; Herrmann et al., 1998; Kershaw et al., 1999),
newly identified source of Pu, with a typical 238Pu :
whereas the primary source in the Arctic Ocean and the
239,240Pu isotope ratio of around 0.2 (see Section 2.2.3).
Greenland Sea remains global fallout. However, in two
The 238Pu : 239,240Pu isotope ratios in waters, particu-
surveys of Pu in seawater in various parts of the Arctic
larly in the Norwegian Sea and Barents Sea, are elevated
Ocean, Vintr๓ et al. (2002) explained the observed con-
above the expected fallout ratio of ~ 0.04 and are, in
centrations by advection of global fallout Pu from mid-
some cases, close to Irish Sea ratios. The isotope ratios
North-Atlantic latitudes to the Arctic Ocean and failed
therefore suggest that Sellafield-derived Pu may have
to see evidence of Sellafield Pu.

---

24
AMAP Assessment 2002: Radioactivity in the Arctic
239,240Pu concentration in sediment,
Bq/ kg dw
0.01 - 0.1
0.1- 0.5
0.5 - 2
2 -10
10 - 28.4
Figure 3ท11. Activity concen-
trations of 239,240Pu in surface
sediments 1996 to 2000.
Figure 3ท11 shows Pu activity concentrations in sur-
et al., 1997). During sea ice formation, dissolved con-
face sediments. Global fallout levels in marine sediments
taminants are rejected together with salt. Levels of dis-
depend on many factors such as sediment characteris-
solved contaminants are therefore lower in sea ice than
tics, depth, and proximity to river outflows. As is the
in seawater (Weeks, 1994). Consequently, the main fo-
case for seawater, isotope ratios may indicate different
cus in ice transport studies has been on the transport and
origins, but the actual global fallout concentrations in
fate of radionuclides associated with sediment particles
surface sediments vary considerably. Cooper et al. (2000)
incorporated into the ice, derived both from particles
found Pu isotope ratios in Arctic Ocean sediments to be
suspended in seawater and from bottom sediments.
due to global fallout in most cases, and failed to show
A radiological assessment of sea ice transport has been
any indications of a Sellafield contribution, in accord-
considered by Iosjpe and Borghuis (2000) and Iosjpe
ance with the results for seawater (Vintr๓ et al., 2002).
(2002). Their approach is based on box modelling (Iosjpe
Rissanen et al. (2000) analyzed 92 fish samples from
et al., 1997, 2002) that incorporates the various trans-
commercial operations in the Barents Sea and found
fers of radioactivity: from the liquid phase to ice; from
239,240Pu in one sample only: a ray, Raja radiata, con-
suspended sediment and bottom sediment to ice; through
taining 7.9 and 4.9 mBq/kg ww in flesh and bones, re-
sea ice transport between sea areas; and into seawater
spectively. Plutonium activity concentrations were below
during ice melt.
detection limits of 1 to 3 mBq/kg ww in all other samples.
The potential significance of sea ice transport in the
dispersion of radionuclides within the marine environ-
ment was illustrated by modelling the transport of
3.3.5. Radionuclide behavior in marine systems
radionuclides from the Kara Sea to the Fram Strait
3.3.5.1. Partitioning and uptake
through the Arctic Ocean following the simulated re-
Water movement and sedimentation are of major impor-
lease of 1 TBq of specific radionuclides into Ob Bay in
tance in determining transport pathways and the fate of
the Kara Sea. Dispersion through the Kara Sea to the
radioactive material released to, or transported within,
Fram Strait is particularly interesting because the model
the marine environment. One of the important geochem-
indicated that sea ice transport of contaminants in this
ical factors affecting the transport of radionuclides is
region may represent a more rapid transport pathway
particle-water exchange. In water, radionuclides are par-
than water. Furthermore, modelling showed the effect to
titioned between the dissolved and particulate phases.
be directly proportional to the partition coefficient (Kd)
Partitioning depends on the chemical form of the radio-
of each radionuclide. Thus, sea ice transport seems more
nuclide, the physical ญ chemical properties of the chem-
relevant for high-Kd elements such as 241Am and 60Co
ical analogues (usually elemental analogues), and the char-
and less important for 137Cs and 90Sr which are princi-
acteristics of the environment.
pally associated with the dissolved phase (Iosjpe and
Biological uptake and associated dose assessments
Strand, 2002).
for human exposure generally depend on the levels of
dissolved radionuclides in water, although filter feeders
3.3.6. Vulnerability in marine pathways
(mussels and oyster) accumulate radionuclides from the
particulate phase. For biota that are closely associated
Marine ecosystems are relatively less vulnerable to at-
with marine sediments, such as benthic infauna and epi-
mospheric inputs of radiocesium than freshwater and
fauna, sediment concentrations can exert the primary in-
terrestrial environments. This is due to the capacity of
fluence on the extent of uptake.
most marine ecosystems to rapidly dilute an input of ra-
dioactive contaminants through processes such as advec-
tion and mixing, coupled with the large volumes of
3.3.5.2. Transport of radionuclides in sea ice
water generally involved, and the high ionic strength of
Sea ice transport is a unique pathway in polar areas
the saline waters. Thus, short-term consequences are
(Pfirman et al., 1995, 1997; Strand et al., 1996) that is
likely to be more important in marine ecosystems as di-
partially independent of water mass movement (Pfirman
lution is a long-term process. Vulnerable marine ecosys-

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
25
tems include those into which liquid discharges are re-
from snow melt can be highly heterogeneous and radio-
leased ญ especially if their exchange with the World
nuclide contents in soil can be higher in areas where
Ocean is slow or restricted, as is the case for some fjords
water from melting snow accumulates. The effect varies
and inlets. In an Arctic context, this could include areas
with the size of the catchments and lakes. Ice overlying
that receive radionuclides transported from nuclear fa-
lakes can, to some extent, protect freshwater biota from
cilities by marine currents or rivers such as the Ob and
radioactive contamination. However, when the ice melts
Yenisey. In addition to man-made radionuclides, marine
there is a sudden influx of contamination, in addition to
areas may receive natural radioactivity from non-nu-
that from melting snow. Thus, there can be a delayed
clear industries.
pulse of enhanced activity concentrations in spring.
Radioecological vulnerability in marine ecosystems is
Currently, atmospheric and surface terrestrial activ-
affected by a number of factors, such as water exchange
ity concentrations of 129I in the Northern Hemisphere
rates; residence times for radionuclides in the water col-
exceed those due to global fallout by up to six orders of
umn; sediment and sedimentation properties, including
magnitude near point sources (Sellafield, Cap de la Hague,
bioturbation and resuspension; freshwater inflow; salin-
Chernobyl), three to four orders of magnitude in most of
ity; oxygenation of the water column and sediments;
Western Europe, and at least an order of magnitude out-
and ice conditions.
side Europe (e.g., Moran et al., 1999). In the Arctic, lit-
Marine zones with high biological productivity are
tle is known about terrestrial 129I levels, but some indi-
considered radioecologically vulnerable, when consider-
cations based on analyses of freshwaters and precipita-
ing collective doses. From an economic point of view,
tion suggest an increase of about two orders of magni-
highly productive areas such as the Barents Sea and
tude above the nuclear weapons test (NWT) contami-
areas used for aquaculture of mollusks, fish, or crus-
nated background levels (Moran et al., 1999), and four
taceans are potentially important.
orders of magnitude above the natural background level
in most of the Arctic. For freshwaters, where current
129I :127I ratios are typically up to an order of magnitude
3.4. Freshwater environment
lower than in terrestrial food chains within the same re-
Estimates of the initial activity concentration in water
gion, very few data are available for rivers and estuaries
bodies following radionuclide deposition can be made
without evident contamination (Beasley et al., 1997;
by assuming dilution of activity `deposited' to the river
Fehn and Snyder, 2000; Meili et al., 2002). Levels are far
or lake surface. Therefore, deep rivers and lakes would
higher in some Russian rivers into which discharges
be expected, initially, to be less vulnerable than shallow
from point sources, such as the Mayak reprocessing
water bodies. However, deposition times can be long, as
plant, occur, reaching about three and four orders of
was the case for global fallout, compared to river water
magnitude above the NWT background level near the
transit times. Catchment runoff can also make a signifi-
mouths of the Yenisey and Ob rivers, respectively (Coch-
cant long-term contribution to water activity concentra-
ran et al., 2000; Cooper et al., 1998).
tions. Activity concentrations in runoff water decline
significantly with time after deposition. At a given point
3.4.1. Rivers
in time, the activity concentration in river or lake water
per unit of deposition to the catchment (the runoff coef-
The first AMAP assessment reported temporal changes
ficient) is a measure of radioecological vulnerability of
in 137Cs and 90Sr activity concentrations in Finnish and
the catchment. Organic, boggy catchments, such as those
Russian rivers. Figure 3ท12 shows the extended time
prevailing in some Arctic areas, have much higher 137Cs
series for 90Sr activity concentrations in the water of dif-
runoff coefficients than catchments with a high propor-
ferent Russian rivers. Current levels are low in all areas
tion of mineral soils (Kudelsky et al., 1996).
of the Russian Arctic.
Loss of the initial input of radioactivity from lake
90Sr concentration in river water, Bq/m3
and reservoir water may be estimated using the water
200
residence time of the lake and simple models for the re-
moval of radioactivity to sediments. Long-term activity
concentrations in lakes with relatively short water resi-
Ob
dence times are primarily controlled by inputs of radio-
150
activity from the surrounding catchment. Long-term ac-
tivity concentrations in closed lakes, which have relatively
long water residence times, are controlled by the transfer
of radioactivity to and from bottom sediments. Lakes
100
with high vulnerability for radiocesium include shallow
lakes with long water residence times, especially those
Sev.
Pechora
having large catchments with a high percentage of organic
Dvina
boggy soils, as these allow higher runoff than catchments
50
with other soil types. In areas such as Finnish Lapland,
vulnerability to radiocesium is high due to the presence
Lena
of boggy soils (30 to 60% of catchments) combined with
Yenisey
a hilly topography that may increase surface runoff.
0
When contaminated snow melts, radionuclides may
1960
1965
1970
1975
1980
1985
1990
1995
2000
be transported with runoff waters and may then con-
Figure 3ท12. Changes in 90Sr activity concentrations in Russian rivers
taminate soils and freshwater systems. Contamination
since the mid-1960s.

---

26
AMAP Assessment 2002: Radioactivity in the Arctic
Concentration in river water, Bq / m3
Data for 90Sr and 137Cs in two Finnish rivers (Saxen,
60
2003) show that post-Chernobyl Teff values for 137Cs
are about a factor of two shorter than the bomb-fallout
137Cs
Kemijoki
half-lives (Table 3ท3). As different periods of time were
Tornionjoki
50
used to calculate the half-lives in Tables 3ท2 and 3ท3, the
90Sr
first components of the Teff for the Finnish and Russian
Kemijoki
Tornionjoki
rivers are not comparable, but the second components
are not significantly different.
40
3.4.2. Fish
30
Long-term data for Russia which enable a comparison
of 137Cs activity concentrations in freshwater fish from
two Arctic regions indicate that concentrations in fish
20
from the Kola Peninsula are statistically significantly
137Cs concentration in freshwater fish, Bq/kg fw
180
10
Kola Peninsula
Nenets AO
160
0
1960
1965
1970
1975
1980
1985
1990
1995
2000
140
Figure 3ท13. Annual mean levels of 137Cs and 90Sr in two Finnish
rivers since 1960.
120
Updated information for two Finnish Arctic rivers
(Figure 3ท13) shows further declines over the last five
years, following the Chernobyl accident (1986). In the
100
global weapons testing period, 90Sr activity concentra-
tions in rivers were consistently two- to three-fold higher
than for 137Cs. Chernobyl deposition reversed this situa-
80
tion such that activity concentrations of 137Cs were
higher than 90Sr for some years following deposition.
60
However, owing to the shorter effective half-life (Teff;
Box 3ท1) of 137Cs compared with 90Sr, activity concen-
trations of 90Sr have been slightly higher than for 137Cs
40
since 1994.
Table 3ท2 compares Teff values for 90Sr in various
Russian rivers.
20
Table 3ท2. Teff values for 90Sr in Russian rivers.
0
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1965
1970
1975
1980
1985
1990
1995
2000
Teff1 (yr)
Teff2 (yr)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Figure 3ท14. Activity concentrations (ฑSD) for 137Cs in freshwater
Severnaya Dvina
12.9 ฑ1.7 *
fish from the Kola Peninsula and Nenets AO.
Ob
0.7 ฑ0.3
13.2 ฑ2.4
Lena
0.2 ฑ0.15
14.3 ฑ2.2
higher than in fish from lakes and rivers in the Nenets
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Autonomous Okrug (NAO), both after the period of glo-
Half-lives calculated using non-linear least squares
(www.r-project.org).
bal fallout and after the deposition of Chernobyl fallout
* It was not possible to calculate a double exponential for Sever-
(Figure 3ท14). A contributory factor was the higher Cher-
naya Dvina.
nobyl fallout on the Kola Peninsula than in the NAO.
Table 3ท3. Teff values (yr) for 90Sr and 137Cs in Finnish rivers from weapons test and Cher-
nobyl fallout (based on Saxen, 2003).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Weapons tests
Chernobyl
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1965-1985
1988-2000
1965-2000
------------------------------------------------------------------------------------------------------------------------
90Sr
Tornionjoki
9.1
11.6
9.5
Kemijoki
10.5
12.8
11.6
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1965-1985
1986-1988
1989-2001
------------------------------------------------------------------------------------------------------------------------
137Cs
Tornionjoki
9.3
0.6
4.2
Kemijoki
10.4
1.0
5.1
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
27
137Cs concentration in freshwater fish, Bq/kg ww
137Cs concentration in freshwater fish, Bq/kg ww
400
400
Inarijไrvi
ฤkไsjไrvi
Burbot
Roach
Grayling
300
300
Trout
Perch
Vendance
Pike
200
Whitefish
200
100
100
0
0
1960
1970
1980
1990
2000
1960
1970
1980
1990
2000
400
400
Jerisjไrvi
Apukkajไrvi
300
300
200
200
100
100
0
0
1960
1970
1980
1990
2000
1960
1970
1980
1990
2000
Figure 3ท15. Changes in 137Cs activity concentrations with time in various fish species in four lakes in Arctic Finland.
Also, fish from the Kola Peninsula are mainly caught in
to fish also depend on feeding habit, with 137Cs activity
lakes (particularly, Lake Lovozero), whereas they are
concentrations in the more radioecologically-vulnerable
mainly from rivers in the NAO, and fish caught in lakes
predatory fish generally a factor of two or more higher
are generally more contaminated than fish from rivers.
than for non-predatory fish. The Finnish Arctic data are
diverse and thus difficult to analyze, however, it is clear
that some species generally contain lower radionuclide
3.4.2.1. Species differences
activity concentrations than others. For each lake sam-
Activity concentrations are available for 137Cs in a large
pled, 137Cs activity concentrations are higher in pike (a
number of different fish species for four Finnish lakes:
predatory species) and perch (partially predatory) than
Inarijไrvi, a large lake in northeast Finland (Pasvik area,
in other species, and whitefish contamination is consis-
Barents Sea catchment); Apukkajไrvi, a small highly eu-
tently low.
trophic lake in the vicinity of Rovaniemi (Kemijoki
catchment area); and ฤkไsjไrvi and Jerisjไrvi, which are
3.4.2.2. Migratory fish
two medium-sized lakes in the Tornionjoki system 200
km northwest of Rovaniemi (Figure 3ท15). In Inarijไrvi,
Salmon (Salmo salar) from Arctic seas migrate up the
the only lake sampled in Arctic Finland during the
Tana River, which flows from northern Finland, through
global fallout period, 137Cs activity concentrations of up
Norway to the Barents Sea, to spawn. They are an im-
to 356 Bq/kg ww were measured in pike in 1964 (Ko-
portant traditional food resource for the indigenous peo-
lehmainen et al., 1966). In 1982, activity concentrations
ples living near the river and the present total catch
of 25 Bq/kg ww occurred in whitefish from Inarijไrvi
varies from 90 to 180 t/yr. Annual samples show the ac-
(STUK, 1983). After the Chernobyl accident, when more
tivity concentrations of 137Cs in flesh to have decreased
data became available, there was a single high measure-
from 1 Bq/kg ww in 1988 to 0.3 Bq/kg ww in 2000,
ment of 305 Bq/kg ww in perch, but otherwise 137Cs ac-
with Teff values of 6 yr over the study period. In com-
tivity concentrations were much lower than in the 1960s
parison, Baltic salmon from the more heavily Chernobyl-
with maximum values of ~100 Bq/kg ww.
contaminated Bothnian Bay had activity concentrations
Radiocesium activity concentrations in fish are in-
100 times higher (Rissanen and Ikไheimonen, 2000).
versely related to the potassium (K) concentration of the
surrounding water (e.g., Blaylock, 1982; Kolehmainen
3.5. Terrestrial environment
et al., 1967). Similarly, an inverse relationship has been
found between 90Sr activity concentrations in fish and
3.5.1. Soil and humus
water calcium (Ca) concentrations (e.g., Blaylock, 1982).
Recent data demonstrate that 137Cs deposition from 1995
High K or Ca concentrations in water are often a result
to 2002 at the Zapolyarie and North monitoring sta-
of the runoff of agricultural fertilizers, but this is not
tions, covering the Asian and European parts of the Rus-
particularly relevant to Arctic ecosystems. Transfer rates
sian Arctic respectively, has not exceeded 1.3 Bq/m2/yr.

---

28
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration in humus,
Bq/kg dw
5000-11500 (6)
3000-5000 (9)
1000-3000 (52)
300-1000 (76)
200-300 (66)
150-200 (70)
100-150 (116)
50-100 (132)
0-50 (30)
Figure 3ท16. Activity concentrations of 137Cs in the
upper 3 cm humus layer (Paatero et al., 2002).
Numbers in brackets indicate the number of sam-
ples within each concentration range.
3.16
134Cs concentration in humus,
Bq/kg dw
36-78 (6)
18-36 (17)
12-18 (15)
8-12 (25)
5-8 (10)
3-5 (25)
2-3 (21)
0-2 (53)
n.d. (385)
Figure 3ท17. Activity concentrations of 134Cs in the
upper 3 cm humus layer (Paatero et al., 2002).
Number in brackets indicate the number of samples
within each concentration range.
A comprehensive survey of 137Cs and 134Cs in the
5 cm of soils near the coast in northwest Russia and
upper 3 cm humus layer in Finland and northwest Rus-
Svalbard between 1993 and 1996. The variation was due
sia in 2000 (Figures 3ท16 and 3ท17) found 134Cs (from
to the type and density of the surface vegetation (Rissa-
the first plume) from the Chernobyl accident could still
nen et al., 2001). The 238Pu : 239,240Pu ratio at these sites
be detected in southern Finland and near St Petersburg.
and in Franz Josef Land suggests the primary source is
The presence of 134Cs suggests that the elevated 137Cs
global fallout.
levels in this area are also due to the Chernobyl accident.
The time of deposition is critical for Arctic ecosys-
Paatero et al. (2002) found a similar pattern to that of
tems, particularly for short-term deposition, such as
the humus survey for Chernobyl-derived Pu fallout in
might occur after an accident. In subarctic areas, snow
Finland.
cover is present for about seven to eight months of the
Activity concentrations of 9 to 32 Bq/m2 were found
year, the period during which the land is snow-covered
for 239,240Pu in surface vegetation and in the upper 3 to
increasing with increasing latitude and factors such as

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
29
altitude, and distance from coasts. When deposition oc-
systems will enhance or reduce radionuclide mobility,
curs onto snow, the radioactivity is available for uptake
nor is the effect of permafrost known.
by vegetation and biota only after the snow has melted.
Surface contamination of plants, lichens, and mosses oc-
3.5.2. Mushrooms
curs as snow melts, by a process similar to interception
of wet deposition. The extent of foliar uptake depends
After the Chernobyl accident, the potential importance
on the rapidity of snow melt, the topography of the
of mushroom consumption as a source of radiocesium
landscape and the morphology of the vegetation. Subse-
intake became apparent, especially in the mid- to long-
quent lateral transport of radionuclides from melting
term after the accident. The first AMAP assessment found
snow depends on the extent of interception and catch-
the importance of mushroom consumption to vary con-
ment characteristics at both the large and small scale. If
siderably between countries and population groups. Al-
deposition occurs in the few months when vegetation is
though a potentially important source, data on 137Cs con-
exposed to the atmosphere, then the transfer of radioac-
tamination of mushrooms at Arctic sites were not gener-
tivity to herbivores is more rapid. Furthermore, for
ally available. Such data are now available for Finland,
short-lived radionuclides, especially 131I, deposition only
Russia, and Norway.
leads to contamination of foodstuffs if it occurs just
prior to, or during, the growing season.
3.5.2.1. Finland
The uptake of radioactivity by plants from soil oc-
curs via the soil solution. The processes controlling ra-
Figure 3ท18 shows the results of an extensive survey in
dionuclide transfer between soil components and the soil
Arctic Finland of fruiting bodies from a wide range of
solution are critical for bioavailability. For example,
mushroom species (Rissanen et al., 2002) The samples
sorption of many radionuclides on non-specific cationic
were obtained from 1983 onwards from four forest
exchangeable sites is weaker than on more specific sites
types at a site 70 km southeast of Rovaniemi. In 1993,
such as clay minerals. In addition, soil solution composi-
the average 137Cs deposition to the soil at the site was
tion is important because of the competition between ra-
0.8 to 0.9 kBq/m2 declining to 0.7 to 0.8 kBq/m2 in
dionuclides and their stable analogues, e.g., strontium
1999. Owing to the low level of Chernobyl fallout in
and Ca, and cesium and K. Therefore, soils with low po-
Finnish Lapland, about half the total 137Cs present was
tassium and clay mineral content will be more radioeco-
due to global fallout.
logically vulnerable to radiocesium than soils with high
There was no significant difference in the 137Cs activ-
potassium and clay mineral content. Strong sorption en-
ity concentration of the mushroom species between the
hances retention of radionuclides in upper soil layers
four types of forest stand. Analysis of the 134Cs content
where most roots absorb nutrients. It is not clear whether
showed that significant pre-Chernobyl 137Cs was present
the presence of a thin, organic layer in many Arctic eco-
in many species, again with about half the total 137Cs
137Cs concentration in mushrooms, Bq/kg dw
137Cs concentration in mushrooms, Bq/kg dw
10000
10000
Cortinarius armillatus
Mixed forest
Birch
Leccinum scabrum
Rozites caperata
Russula decolorans
Suillus variegatus
Lactarius rufus
Leccinum versipelle
Russula vinosa
Lactarius torminosus
5000
Leccinum vulpinum
5000
Russula xerampelina
Lactarius trivialis
Lactarius vietus
0
10000
10000
Pine
Spruce
5000
5000
0
0
1982
1985
1990
1995
2000
1982
1985
1990
1995
2000
Figure 3ท18. Activity concentrations of 137Cs in a range of mushroom species in Finnish forest stands.

---

30
AMAP Assessment 2002: Radioactivity in the Arctic
Box 3ท2. Aggregated transfer coefficients
The transfer of radionuclides is quantified using the aggregated transfer coefficient (Tag) defined as the activity concentration
in an environmental compartment (often a food product) (in Bq/kg) divided by the corresponding radionuclide deposition in
soil (in Bq/m2); with units of m2/kg.
Activity concentration
Tag = ญญญญญญญญญญญญญญญญญญญญญญ
Deposition
Tag values were most commonly used in the former Soviet Union to quantify transfer to food products. In other countries,
they are most often used for semi-natural products. They are therefore the most commonly used transfer quotient for Arctic
ecosystems. High Tag values, such as those derived for highly organic soils for radiocesium, indicate radioecologically vul-
nerable areas. Tag values are time dependent, and can be combined with ecological half-lives to quantify changes with time.
They are not appropriate for use when considering surface depositions onto plants during fallout.
The Tag may be combined with the effective ecological half-life as follows:
ln2 ln2
C(t) = A ท Tag ท exp (ญ
0
r ท t) ท (a1 ท exp(ญ ญญญญ ท t) + (1 ญ a1) ท exp(ญ ญญญญ ท t))
Eqn. 3.2
Teff1
Teff2
where C(t) is the activity concentration of a given radionuclide in a food product at time t (Bq/kg); A is the surface deposi-
tion of radionuclide p (Bq/m2); Tag is the initial value of the aggregated transfer coefficient (m2/kg);
0
r is the radioactive
decay constant for radionuclide p (1/days); t is the time after deposition (days); and the other parameters are the same as in
Equation 3.1 (Box 3ท1).
The values of parameters Tag , a
0
1, Teff 1, and Teff 2 can be estimated on the basis of long-term measurements of 137Cs and
90Sr activity concentrations in different food products during the period of global fallout (about 40 years) and after the
Chernobyl accident.
Using these techniques, it is possible to calculate integrated transfer coefficients (ITC) for 137Cs and 90Sr in different
regions (AMAP, 1998). The calculation, assuming a single deposition event uses the formula:
C(t)dt
ln2
ln2
ITC = ญญญญญญ = Tag ท exp (ญ
0
r ท t)ท (a1 ท exp(ญ ญญญญ ท t) + (1 ญ a1) ท exp(ญ ญญญญ ท t))dt Eqn. 3.3
A
T
0
0
eff 1
Teff2
or, after integration:
Teff1
Teff2
ITC = Tag ท (a
0
1 ท ญญญญญ + (1 ญ a1) ท ญญญญญ ) , (Bq/yr)/kg per kBq/m2
Eqn. 3.4
ln2
ln2
In the case of 137Cs, some areas received an additional input from the Chernobyl accident. As this input was a pulse input
and possibly in a different chemical form, separate half-lives must be calculated for the Chernobyl input after removing the
contribution from global fallout by extrapolating the pre-Chernobyl data using the model.
from Chernobyl and half from global fallout. The high-
3.5.2.2. Russia
est pre-Chernobyl value was 2390 Bq/kg dw in Lactar-
ius trivialis in 1983. Although the variable nature of the
Activity concentrations of 137Cs in mushroom fruiting
data makes derivation of half-lives difficult, it is clear
bodies from northwest Russia in 1989 to 1999 were low
that 137Cs uptake persists for many years in a wide vari-
compared to those recorded in temperate areas of Eu-
ety of mushroom species.
rope. Highest activity concentrations were recorded in
After the Chernobyl accident, the most highly con-
Lactarius flexuosus and Xerocomus spp. Over the pe-
taminated species in all four forests was the non-edible
riod 1987 to 2000, 137Cs Tag values (Box 3ท2) from soil
Cortinarius armillatus, with a maximum recorded value
to mushroom for two regions in northwest Russia were
of 9030 Bq/kg dw in 1993. Of the edible species, the
relatively constant (RTCP, 1999, 2000; Shutov et al.,
most contaminated were Rozites caperata, Lactarius tri-
1999). Therefore, in Table 3ท4, data from the whole
vialis, and Suillus variegatus, which is consistent with
sampling period have been collated to quantify transfer
data for mushrooms in temperate areas.
for the Kola Peninsula and the Mezen and NAO regions.
Table 3ท4. Activity concentrations for 137Cs in mushroom species (Bq/kg, mean ฑ SD, air dw) in northwest Russia for 1992 to 2000
and associated Tag values (m2/kg ฑ SD) for soil to mushroom (Borghuis et al., 2002).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Kola Peninsula
Mezen and Nenets AO regions
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
Activity
Activity
n
concentration
Tag
n
concentration
Tag
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Leccinium aurantiacum
126
27.6 ฑ1.7
0.013 ฑ0.0007
27
21.8 ฑ3.9
0.0072 ฑ0.0012
Leccinium scabrum
63
55.2 ฑ5.3
0.026 ฑ0.0025
28
27.8 ฑ5.4
0.0082 ฑ0.0017
Suillus bovines
ญ
ญ
ญ
2
35.0 ฑ7.1
0.0104 ฑ0.0031
Russula spp.
70
63.9 ฑ4.5
0.029 ฑ0.002
19
54.4 ฑ11.3
0.016 ฑ0.0032
Paxillus spp.
ญ
ญ
ญ
5
45.3 ฑ20.3
0.016 ฑ0.0073
Xerocomus spp.
36
117 ฑ 9
0.053 ฑ0.0044
4
77.4 ฑ20.9
0.028 ฑ0.0075
Lactarius rufus
6
133 ฑ21
0.067 ฑ0.01
5
59.4 ฑ8.5
0.018 ฑ0.0025
Lactarius flexuosus
3
235 ฑ69
0.144 ฑ0.042
4
98.3 ฑ36.8
0.035 ฑ0.013
Lactarius necator
14
50.3 ฑ9.5
0.026 ฑ0.0048
2
40.3 ฑ3.7
0.014 ฑ0.0013
Boletus edulis
1
8.9
0.0047
5
22.4 ฑ5.3
0.0075 ฑ0.0017
Suillus luteus
11
80 ฑ14
0.037 ฑ0.0066
1
40.0
0.0130000.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
31
137Cs concentration in mushrooms, Bq/kg dw
Overall, the data indicate lower activity concentra-
1000
tions in Arctic mushroom species than in those from
nearby temperate areas of NW Europe due to the low
Kola Peninsula
Mezen and Nenets AO
levels of global and Chernobyl fallout in most of the
areas sampled. Nevertheless, the 137Cs activity concen-
trations are higher in mushrooms than in many other
Arctic foodstuffs. There is, however, evidence of large
Tag values. Although activity concentrations in the most
contaminated species vary between areas, the data are
100
not directly comparable as different weight bases have
been used.
3.5.3. Berries
Activity concentrations of 137Cs in the berries of cloud-
berry (Rubus chamaemorus), bilberry (Vaccinium myr-
tillus), and cowberry (Vaccinium vitis-idaea) were meas-
ured at four sites in Lapland. In 1980 to 1981, before
10
the Chernobyl accident, 137Cs activity concentrations at
Kittilไ were 25 to 45 Bq/kg ww in cloudberry, 11 to 19
Bq/kg ww in bilberry, and 6 to 22 Bq/kg ww in cowberry
spp.
spp.
spp.
(Rissanen et al., 1987). Sufficient data were available at
one of the four sites, Salla-Kuusamo, to indicate a slow
Suillus luteus
Paxillus Russula
change in concentration since the Chernobyl accident
Boletus edulis
Suillus bovines
Lactarius rufus
Xerocomus
(Figure 3ท20) and higher 137Cs activity concentrations in
Lactarius necator
Leccinium scabrum
Lactarius flexuosus
cloudberry berries than in those of bilberry or cowberry.
Leccinium aurantiacum
137
Figure 3ท19. Activity concentrations of 137Cs in various mushroom
Cs concentration in berries, Bq/kg ww
species from the Kola Peninsula and the Mezen and Nenets AO re-
50
gions (Borghuis et al., 2002).
40
The average 137Cs Tag values from soil to mushroom on
the Kola Peninsula were significantly higher than for the
NAO region (P = 0.00024, Table 3ท4 and Figure 3ท19).
30
Cloudberry
3.5.2.3. Norway
20
Activity concentrations of 137Cs in mushroom species
from Troms and Finnmark collected in 1998 to 1999 are
Bilberry
given in Table 3ท5. Activity concentrations vary signifi-
10
cantly between species. The activity concentration in
species such as Rozites caperata, which is known to ac-
Cowberry
0
1985
1990
1995
2000
Table 3ท5. Activity concentrations for 137Cs in mushroom species
(Bq/kg dw ฑ SD) in Arctic Norway for 1998 to 1999 and corre-
Figure 3ท20. Changes over time in 137Cs activity concentrations in
sponding Tag values (m2/kg ฑ SD) for soil to mushroom (Salbu, pers.
three species of berries at Salla-Kuusamo in Lapland, Finland (Ris-
comm., 2001).
sanen et al., 1987).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Activity
The range in activity concentrations in these species
n
concentration
Tag
at the four sites in Lapland since the Chernobyl accident
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Boletus edulis
1
2097
2.17
is shown in Table 3ท6. Despite the difficulties in compar-
Leccinum spp.
2
354 ฑ175
0.37 ฑ0.14
ing data with some time dependency after the Chernobyl
Rhozites caperata
4
3929 ฑ3557
4.38 ฑ4.47
accident, the sampling frequencies for the four species
Russula spp.
5
595 ฑ417
0.48 ฑ0.43
are sufficiently similar to conclude that there are differ-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
ences between the species and that cloudberry is the
cumulate cesium, was a factor of ten to fifteen higher
most contaminated.
than in Leccinum spp. The activity concentration of
137Cs in fungi with a high cesium transfer was about 500
Table 3ท6. Range in 137Cs activity concentrations (Bq/kg ww) for
berry species in Lapland for 1986 to 2001 (Rissanen, pers. comm.,
times higher than in plant species from the same site,
2002).
whereas the factor was approximately 50 for fungi with
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
a moderate transfer.
Rovaniemi
Inari
Salla-Kuusamo
Kittilไ
The Tag values also varied between the different spe-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
cies. Species with a known high uptake of 137Cs had a
Cloudberry
30-63
16-31
21-49
30-38
ten-fold higher T
Bilberry
6-15
2-9
5-13
10-16
ag value for 137Cs than those with a
Cowberry
4-25
2-7
4-8
2-12
lower uptake.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

---

32
AMAP Assessment 2002: Radioactivity in the Arctic
Table 3ท7. Average 137Cs activity concentrations for various berry species (Bq/kg ww; mean ฑ SE) in
northwest Russia for 1998 to 1999 and associated Tag values (m2/kg) for soil to berries (Borghuis et al.,
2002).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Kola Peninsula
Mezen and Nenets AO regions
------------------------------------------------------------------------------------------------------------------------------------
Activity
Activity
n
concentration
Tag
n
concentration
Tag
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Cloudberry 28
31.6ฑ1.4
0.014ฑ0.0007
50
26.9ฑ2.3
0.0091ฑ0.0008
Bilberry
64
11.1ฑ1.0
0.0048ฑ0.0004
24
14.2ฑ1.3
0.0045ฑ0.0004
Cowberry 192
7.2ฑ0.4
0.0032ฑ0.0002
40
3.7ฑ0.4
0.0013ฑ0.0002
Cranberry 5
17.9ฑ4.7
0.0091ฑ0.0024
38
11.6ฑ1.1
0.0037ฑ0.0004
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Activity concentrations and Tag values for 137Cs in
are on a dry and not wet weight basis). Although, there
various berry species collected in 1998 to 1999 in north-
was a tendency for higher 137Cs activity concentrations
west Russia are shown in Table 3ท7. Activity concentra-
in cloudberry than bilberry the difference was not
tions and transfer of 137Cs to cloudberry are higher than
statistically significant. Some samples also contained
for the other species, including bilberry which is often
134Cs from the Chernobyl accident (with maximum val-
the most contaminated berry in temperate areas.
ues of 33 Bq/kg dw in cloudberry and 45 Bq/kg dw in
Recent data on 137Cs activity concentrations in the
bilberry).
berries of bilberry and cloudberry from Arctic Norway
The Finnish, Russian, and Norwegian data are com-
in 1998 to 1999 are shown in Table 3ท8 (note the data
pared in Figure 3ท21. Overall, the data indicate that
cloudberry, which is a species typical of Arctic ecosys-
Table 3ท8. Activity concentrations for 137Cs (Bq/kg dw (Bq/kg ww*))
tems, has the highest 137Cs activity concentrations and is
in bilberry and cloudberry in Arctic Norway for 1998 to 1999
relatively vulnerable to radiocesium deposition com-
(Salbu, pers. comm., 2001).
pared to the other berry species. Cloudberry (and cran-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
n
Mean
Minimum
Maximum
berry; Vaccinium oxycoccus) grow on wet, highly or-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
ganic bogs, conditions which would be expected to lead
Cloudberry
11
74.8 (11.2)
32.0 (4.8)
175.0 (26.5)
to high radiocesium plant uptake from soil. Bilberry and
Bilberry
15
63.1 (9.5)
nd
167.0 (25.1)
cowberry grow on dry land, mainly in forests, in which
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
the depth of the organic layer varies and sand is often
nd: not detectable;
* fresh weight values estimated assuming 85% water content in
present under the upper organic horizons.
berries.
The limited Finnish data suggest that the reduction
with time in the 137Cs content of Arctic berries is slow.
137Cs concentration in berries, Bq/kg ww
35
3.5.4. Milk
Finland
Norway
Sufficient data are now available on 137Cs activity con-
Mezen and Nenets AO
Kola Peninsula
centrations in milk to report on changes over time dur-
30
ing the period of global fallout and after the Chernobyl
accident for several locations in Finland, Sweden (annual
averages), the Faroe Islands, Iceland, Russia, and Nor-
way. The sample sites are shown in Figure 3ท22. For 137Cs,
25
all time series show a peak in activity concentrations in
the early-1960s with nearly 100 Bq/L detected in the
Faroe Islands and Iceland. After the Chernobyl accident,
20
some fallout was detected in milk in parts of Sweden,
Arctic Finland, northern Norway, northwest Russia, and
the Faroe Islands, with peak values of up to 20 Bq/L.
In most time series with an adequate sampling fre-
15
quency, a strong seasonal signal can be seen with higher
137Cs activity concentrations in the summer, when cows
are put out to pasture or fed fresh grass. In some cases,
10
the completeness of directly comparable time series has
been affected by dairies closing down and consequent
changes in the collection areas for those remaining.
Long-term time series on 90Sr activity concentrations
5
in milk are available for northern Finland, Sweden, Nor-
way, and the Faroe Islands. Peak 90Sr activity concentra-
tions in the late-1960s were around 2 Bq/L in Fenno-
scandia. Little input from the Chernobyl accident was
0
Cloudberry
Cranberry
Bilberry
Cowberry
detected, although the weighted mean value in Sweden
for 90Sr in milk increased marginally from 0.10 Bq/L just
Figure 3ท21. Activity concentrations of 137Cs in berries from 1998
and 1999. Russian data: mean ฑ SD. Finnish and Norwegian data:
before the Chernobyl accident to 0.13 Bq/L just after. In
mean and range. < : Under detection limit.
all time series with adequate sample numbers, a strong

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
33
Naryan-Mar
Vads๘
Mezen
Grafarnes
Bl๘ndu๓s
Kautokeino
Mๅlselv
Borgarnes
Akureyri
Arkhangelsk
Kursu
Reykjavik
Egilsstadir
Apukka/Vikajไrvi
Selfoss ICELAND
Bod๘
Vittangi
Rovaniemi
Onega
Hornafj๘rdur
Tไrnaby
Kargopol
FINLAND
RUSSIA
Klaksvํk
SWEDEN
T๓rshavn
Tv๘royri
NORWAY
Figure 3ท22. Sites at which
milk samples were collected.
seasonal signal in 90Sr activity concentrations is seen
milk were recorded in Lapland even though the deposi-
with higher values in the summer.
tion of 90Sr and 137Cs was no greater than in the rest of
The 137Cs and 90Sr activity concentration data are
Finland. This was mainly due to the high proportion of
summarized in Table 3ท9. Despite more 90Sr being de-
peat soils and nutrient deficiency of the pastures in Lap-
posited during the period of global fallout, there were
land. Cs-137 deposition after the Chernobyl accident in
consistently higher 137Cs activity concentrations in milk
1986 in Lapland was less than 1 kBq/m2, and 90Sr depo-
due to a roughly 15-fold higher transfer of 137Cs to milk
sition was so low that there was no detectable increase
compared to 90Sr.
in the 90Sr concentration in milk.'
Although total 137Cs deposition from nuclear weap-
3.5.4.1. Finland
ons tests was similar in parts of Finland Lapland to that
for Chernobyl fallout, 137Cs activity concentrations in
Milk has been sampled at several sites in Finnish Lap-
milk during the 1960s were considerably higher than after
land since the 1960s. Sampling was undertaken from 1963
the Chernobyl accident. The pre-Chernobyl fallout, char-
to 1987 in Kursu, an area with boggy soils. Samples of
acterized by an annual maximum in summer, resulted
milk powder were collected from a dairy in Rovaniemi
in significant direct contamination of growing crops,
from 1966 to 1975 and dairy milk from 1986 onwards.
whereas the Chernobyl deposition occurred before the
Samples from individual farms were collected from
start of the growing season in Lapland. The direct con-
Apukka from 1975 to 1977 and 1986 to 1991 and from
tamination of growing crops resulted in higher contami-
Vikajไrvi from 1991 onwards. Kostiainen and Rissanen
nation levels compared to the short-term deposition fol-
(2003) summarized the findings as follows: `The highest
lowing the Chernobyl accident when, before the start of
90Sr and 137Cs concentrations in the 1960s in Finnish
the growing season, the food chain is hay to milk. The
Table 3ท9. Activity concentrations for 137Cs and 90Sr in milk (Bq/L; mean (range)) (AMAP Data
Centre).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1964
1986
1998
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
137Cs
Faroe Islands
51 (22-97)
5.8 (0.9-19)
0.8 (0.2-1.9)
Arctic Finland
31 (27-36)
2.4 (0.5-5.2)
0.55 (0.5-0.6)
Iceland
27 (7-83)
ญ
1.4 (0.9-2.4)
Arctic Norway
16 (7-37)
6.2 (1-20)
ญ
NW Arctic Russia
ญ
6.4 (4.0-10)
1.3 (0.04-0.3)
Arctic Sweden
22 (14-30)
6.0 (2.0-13)
2.5 (1.5-4)
----------------------------------------------------------------------------------------------------------------------------
90Sr
Faroe Islands
7.2 (3.3-12)*
ญ
0.04 (0.04-0.05)*
Arctic Finland
1.1 (0.9-1.4)
ญ
0.05 (0.04-0.05)
Iceland
ญ
ญ
ญ
Arctic Norway
1.7 (1.3-2.2)
ญ
ญ
NW Arctic Russia
ญ
ญ
0.17 (0.05-0.54)
Arctic Sweden
1.4 (1.0-1.8)**
ญ
0.06 (0.03-0.09)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
** the Faroese 90Sr data were calculated from values in Bq/kg Ca, using an average of 1.2 g Ca
in 1 kg milk;
** data from 1965, for Sweden, on average, 90Sr levels declined by 18% from 1964 to 1965.

---

34
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration in milk, Bq/L
Table 3ท10. Teff values (yr) for 137Cs and 90Sr in milk at the Kursu
40
and Rovaniemi dairies after the period of global fallout and the
Kursu dairy
Chernobyl accident (Kostiainen and Rissanen, 2003).
Rovaniemi dairy
Apukka MTT
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Vikajไrvi
137Cs
90Sr
30
-------------------------------------------------------------------------------------
Kursu
Rovaniemi
Kursu
Rovaniemi
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1963 to 1966
2.0
2.2
20
1966 to 1975
4.8
3.2
5.1
6.1
1975 to 1985
5.3
10
1987 to 1989
2.2
14
10
1989 to 1993
4.1
6.9
1993 to 2001
7.6
16
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
0 1965 1970 1975 1980 1985 1990 1995 2000
about the same during the 1990s (at 7 to 8 yr), as were
those for 90Sr during the 1970s. There were larger fluc-
Figure 3ท23. Time series for 137Cs activity concentrations in milk
tuations in the monthly activity concentrations of 137Cs
from sites in northern Finland (Kostiainen and Rissanen, 2003).
in farm milk than in dairy milk.
90Sr concentration in milk, Bq/L
Kostiainen and Rissanen (2003) conclude that the
2.5
whole of Lapland is vulnerable to radioactive contami-
Kursu dairy
Rovaniemi dairy
nation. The transfer of 137Cs into milk from peat soils
2.0
Apukka MTT
was more than twice that for the clay soils of southern
Vikajไrvi
Finland, and the Teff values for milk in Lapland are
1.5
twice those for intensively-cultivated clay pastures.
1.0
3.5.4.2. Sweden
0.5
Annual average 137Cs activity concentrations at two
sites in Arctic Sweden are shown in Figure 3ท25. These
0 1965 1970 1975 1980 1985 1990 1995 2000
indicate consistently higher values at Tไrnaby than at
Vittangi. A peak due to the Chernobyl accident can be
Figure 3ท24. Time series for 90Sr activity concentrations in milk
from sites in northern Finland (Kostiainen and Rissanen, 2003).
seen in the values for Tไrnaby milk.
A regular seasonal variation was evident for 137Cs,
time trends in 137Cs and 90Sr activity concentrations in
and to a lesser degree for 90Sr, during the 1950s and
milk are shown in Figures 3ท23 and 3ท24, respectively.
1960s. Generally, peak levels occurred during the third
The increase in the activity concentrations of 137Cs in
quarter of the year coincident with the months of high-
milk due to the Chernobyl fallout was clearly visible in
est precipitation. Seasonal variations are especially pro-
July 1986, and the peak lasted until summer 1987. In
nounced in milk from the dairy in Tไrnaby where cows
early-1987, activity concentrations in Kursu dairy milk
graze in summer on natural pastures and in forests. A
were twice as high (5 to 7 Bq/L) as those from the Ro-
similar, but not as regular, seasonal variation was appar-
vaniemi dairy (3.5 Bq/L), which collects milk from most
ent after the Chernobyl accident.
of Lapland. The difference was due to the high fre-
Teff values for 137Cs in Swedish milk, estimated for
quency of peat soils and to the higher fallout in the area
the period from the peak of the atmospheric testing fall-
of the Kursu dairy compared to the average deposition
out until the Chernobyl accident, exhibit a fast and a slow
in Lapland. The peak activity concentrations in the Apuk-
component. The Teff1 values were 1.4 and 1.8 yr, and
ka farm milk were about the same as in Kursu dairy
the Teff 2 values 9.1 and 6.2 yr for Tไrnaby and Vitti-
milk, and decreased after summer 1988 to below the
angi, respectively. The Teff 2 of 9.1 yr for Tไrnaby is the
level of the Rovaniemi dairy milk. The activity concen-
longest Teff 2 for milk found in this study, but is not sig-
trations of 137Cs in Vikajไrvi farm milk were similar to
nificantly longer than Teff 2 values found in northern
those in Rovaniemi dairy milk.
Norway and on the Faroe Islands. Insufficient data were
In contrast, 90Sr in milk in Lapland is mainly from
available to calculate post-Chernobyl Teff values.
the fallout from nuclear weapons tests and the activity
137Cs concentration in milk, Bq/L
concentrations in milk from different areas of northern
25
Finland were similar.
Tไrnaby
After the period of nuclear weapons tests, fallout
20
continued to be deposited at lower, but not negligible
levels, for several years. The T
15
eff values for 137Cs and
90Sr in milk after the peak values in 1963 were similar, at
2 yr in 1963 to 1966 and 5 yr in 1966 to 1975 (Table
10
3ท10). During 1975 to 1985, the effective half-life of
5
137Cs was still 5 yr, but for 90Sr had increased to 10 yr.
Vittangi
The effective half-lives of 137Cs in milk after the peak
0
concentrations resulting from the Chernobyl accident
1960
1965
1970
1975
1980
1985
1990
1995
were similar to those in the 1960s to 1970s. The effec-
Figure 3ท25. Annual average 137Cs activity concentrations in milk
tive half-lives for 137Cs in dairy milk and farm milk were
for two sites in northern Sweden.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
35
90Sr concentration in milk, Bq/L
137Cs concentration in milk, Bq/L
2.0
100
T๓rshavn
Tไrnaby
Klaksvํk
1.6
Tv๘royri
80
1.2
60
0.8
Vittangi
40
0.4
20
0
1960
1965
1970
1975
1980
1985
1990
1995
Figure 3ท26. Annual average 90Sr activity concentrations in milk for
0 1965 1970 1975 1980 1985 1990 1995 2000
two sites in northern Sweden.
Figure 3ท27. Weekly 137Cs activity concentrations in milk at three
High precipitation combined with the high transfer
locations in the Faroe Islands.
of 137Cs via the grass cow food chain in natural pas-
tures and forest environments was the main cause of the
90Sr concentration in milk, Bq/kg Ca
high activity concentrations and relatively slow decrease
10000
rates in milk at Tไrnaby.
T๓rshavn
Klaksvํk
Annual average 90Sr activity concentrations at the
8000
two sites in Arctic Sweden are shown in Figure 3ท26.
Again, consistently higher values occurred at Tไrnaby
than Vittangi. For 90Sr, the short component (T
6000
eff1) after
1963 was 2.1 yr, while the longer component (Teff2) be-
fore and after the Chernobyl accident was 9.1 and 9.2
4000
yr, respectively.
2000
3.5.4.3. Faroe Islands
0
Milk has been sampled weekly in T๓rshavn, Klaksvํk,
1965
1970
1975
1980
1985
1990
1995
2000
and Tv๘royri. The Klaksvํk and Tv๘royri samples are
Figure 3ท28. Weekly 90Sr activity concentrations in milk (per kg Ca)
from locally produced milk, while the T๓rshavn samples
3.28
from two sites in the Faroe Islands. The average calcium content of
are from a dairy which collects milk from most of the
milk in the Faroe Islands is 1.2 g/L.
country. The 137Cs activity concentrations are shown in
Figure 3ท27.
components in the Teff for 90Sr were found to be 1.1 to
The Teff for 137Cs in milk from the Faroe Islands has
1.4 and 5.2 to 5.5 years, respectively.
been calculated to 6.5 to 8.8 yr for the long (second)
component in global fallout, whereas the short (first)
3.5.4.4. Iceland
component was between 1.0 and 1.8 yr. For Chernobyl
fallout, the Teff was from 1.3 to 1.8 yr.
The 137Cs activity concentrations in Icelandic milk in the
Activity concentrations of 90Sr in milk have been
mid-1960s are shown in Figure 3ท29. The average Teff
measured as Bq/kg Ca (Figure 3ท28). With an average Ca
for 137Cs in milk from these dairies during 1964 to 1967
concentration of 1.2 g/L, the peak values of ~ 10000
was 3.0 yr, similar to that in Kursu in northern Finland.
Bq/kg Ca correspond to ~12 Bq/L. The short and long
Soon after this period, the rate of reduction of Cs con-
137Cs concentration in milk, Bq/L
100
Selfoss
Reykjavik
Borgarnes
Grafarnes
Akureyri
Egilsstadir
50
Hornafj๖rdur
20
10
Figure 3ท29. Activity concentrations for
137Cs in Icelandic milk (adapted from
Pแlsson, 1996). Note logarithmic scale.
1964
1965
1966
1967

---

36
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration in milk, Bq/L
100
50
Selfoss, Iceland
Borgarnes, Iceland
Akureyri, Iceland
T๓rshavn, Faroe Islands
10
5
1
0.5
Figure 3ท30. A comparison of 137Cs activ-
0.1
ity concentrations in Icelandic and Faroe
Island milk.
1960
1970
1980
1990
2000
centration in milk decreased. The Icelandic data exhibit a
Icelandic data sets during the 1990s. The explanation may
high spatial variability between sample areas, with rela-
lie in the lower cesium-binding properties of the young
tively high maxima compared to other Arctic areas. In par-
volcanic Icelandic soils compared to those of the Faroe
ticular, the values at Grafarnes are relatively high for many
Islands (Sigurgeirsson et al., 2002). It is thus concluded,
years although there are occasional dips. A comparison
as was also the case for Lapland, that Iceland is vulnera-
with the 137Cs deposition map in Section 3.7.2.2 (Figure
ble to radioactive contamination due to high transfer of
3ท52) shows good agreement between the pattern of depo-
137Cs from soil to milk and very high Teff values.
sition and 137Cs activity concentrations in milk, with the
highest values in the southwest and the lowest in the north.
3.5.4.5. Norway
Activity concentrations of 137Cs in Icelandic (Selfoss,
Borgarnes, and Akureyri) and Faroe Island (T๓rshavn)
Figures 3ท31 and 3ท32 show 137Cs and 90Sr activity con-
milk are compared in Figure 3ท30. This shows that activ-
centrations in milk from four dairies in Arctic Norway,
ity concentrations in milk in the 1960s were slightly lower
since 1960. Because precipitation in Arctic Norway is
in Iceland than the Faroe Islands. However, in the 1990s,
low compared to precipitation on the southwest coasts,
the levels in Iceland were higher in the three study areas
the deposition of global fallout in Arctic Norway is also
than in T๓rshavn, although the relative difference in
relatively low compared to other parts of the country.
concentration between the milk from the three Icelandic
In general, the spatial variation in activity concentra-
regions remained the same (Pแlsson et al., 2002b). This
tions broadly follows that for precipitation and there-
is despite the Faroe Islands receiving some fallout from
fore global fallout. The highest values were measured at
the Chernobyl accident (see Figure 3ท30), in contrast to
Bod๘, a coastal area. However, there is an anomaly in
Iceland where it was hardly detectable. The comparison
that levels at Kautokeino are higher than would be ex-
indicates that the reduction in 137Cs activity concentra-
pected from the low rate of precipitation in this area.
tions in milk has been slower in Iceland than the Faroe
A possible explanation is that cows in Kautokeino were
Islands. This is also reflected in the lower slope of the
fed lichen as fodder (Eikelmann, pers. comm., 2002).
137Cs concentration in milk, Bq/L
50
Bod๘
Mๅlselv
Vads๘
Kautokeino
20
10
5
2
1
1960
1965
1970
1975
1980
1985
1990
Figure 3ท31. Activity concentrations of 137Cs in milk at four locations in northern Norway.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
37
90Sr concentration in milk, Bq/L
3.5.4.7. Trends
20
Bod๘
The data in sections 3.5.4.1 to 3.5.4.6 were collected
Mๅlselv
using various techniques over different periods of time,
Vads๘
Kautokeino
which makes direct comparisons difficult. Nevertheless,
a rough comparison between countries is attempted in
Table 3ท11 using double exponential Teff values and a
consistent method for the various sites in the different
10
countries. All the half-lives were calculated using the sta-
tistical package `R', and its nls- (nonlinear least squares)
library. The Teff values are estimated for global fallout
data from 1964 to 1985, and for Chernobyl fallout from
1986 onward. The global fallout values are expected to
be greater than for a single pulse input as global fallout
continued to be deposited after the peak values in the
0
1960
1965
1970
1975
mid-1960s.
A comparison of 137Cs activity concentrations in
Figure 3ท32. Activity concentrations of 90Sr in milk at four locations
milk from a range of locations in northern Scandinavia
in northern Norway.
and the Faroe Islands in 1964, the year for which most
The impact of the Chernobyl accident is variable, with
data were available, shows that even though some sam-
fallout most noticeable at Bod๘.
ples came from dairies taking milk from a wide geo-
graphical area, a relationship is still evident between
milk contamination and global fallout, which is linked
3.5.4.6. Russia
to precipitation rates (Figure 3ท33). Thus, for instance,
Sporadic data on 137Cs activity concentrations in Rus-
within Norway and Iceland the west coast, which re-
sian milk from 1986 onwards support the findings in
ceives the highest rate of precipitation, produced the
sections 3.5.4.1. to 3.5.4.5. Due to the limited amount
most contaminated milk. Some of the lowest values were
of data available for milk in the Russian Arctic, a de-
recorded in the areas of Norway closest to the Novaya
tailed assessment is not possible.
Zemlya test site.
Table 3ท11. Teff values (yr ฑ SD) for 137Cs and 90Sr activity concentrations in milk from various Arctic areas.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Faroe Islands
Finland
Norway
Sweden
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Klaks-
T๓rs-
Tv๘r-
Rova-
Mๅls-
Kauto-
Tไrna-
Vitti-
vํk
havn
oyri
Apukka Kursu
niemi
Bod๘
Vads๘
elv
keino
by
angi
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
137Cs
Global fallout
Teff1
1.5ฑ0.1 1.0ฑ0.1 1.8ฑ0.2
1.0ฑ0.1
1.9ฑ0.6 1.6ฑ0.4 1.5ฑ0.3 1.1ฑ0.3
1.4ฑ0.4 1.8ฑ0.6
Teff2
7.1ฑ0.5 6.5ฑ0.4 8.8ฑ0.7
4.5ฑ0.7
4.5ฑ1.2 5.1ฑ1.3 6.1ฑ1.2 6.0ฑ2.0
9.1ฑ1.1 6.2ฑ1.0
Chernobyl fallout
Teff
1.3ฑ0.1 1.8ฑ0.1 1.8ฑ0.1
0.7ฑ0.1
3.4ฑ0.1
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
90Sr
Teff1
1.0ฑ0.1 1.4ฑ0.1
1.3ฑ0.1
1.8ฑ0.6
n.a.
n.a.
1.5ฑ0.4
3.0ฑ1.0 1.4ฑ0.3
Teff2
5.2ฑ0.4 5.5ฑ0.5
8.4ฑ0.3
4.0ฑ1.0
n.a.
n.a.
4.6ฑ1.3
9.0ฑ2.0 8.5ฑ1.0
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
n.a.: no statistically valid half-life could be calculated.
137Cs concentration
in milk,
Bq/L
50-60
40-50
30-40
20-30
10-20
Figure 3ท33. Activity concentrations of 137Cs in milk
0-10
from different locations in Nordic areas in 1964.

---

38
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration
in milk, Bq/L
16-20
12-16
8-12
4-8
Figure 3ท34. Maximum 137Cs activity
<4
concentrations in 1986 to 1987 in
milk from different Nordic areas.
The highest 137Cs activity concentrations in milk, after
137Cs concentration in reindeer meat, Bq/kg ww
the Chernobyl accident, are shown in Figure 3ท34. In all
4000
Arctic areas, levels were higher in the global fallout pe-
riod than after the Chernobyl accident.
Kola Peninsula
There were no statistically significant differences in
3000
the average 137Cs Tag values from soil to milk (and beef)
produced in two Russian Arctic regions (Table 3ท12)
after global fallout.
2000
Table 3ท12. 137Cs Tag values (10ญ3 m2/kg) for beef and milk in north-
west Russia (Borghuis et al., 2002).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
1000
n
Mean ฑ SD
t-test
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Nenets AO
Beef, 1978-1985
0
Kola
7
0.28 ฑ0.08
1960
1970
1980
1990
2000
P = 0.10
Nenets AO
34
0.24 ฑ0.12
Figure 3ท35. Changes with time in 137Cs activity concentration in
Milk, 1974-1978
reindeer meat collected in the Kola Peninsula and NAO regions
Kola
21
0.14 ฑ0.07
since 1961 (Borghuis et al., 2002).
P = 0.68
Nenets AO
7
0.12 ฑ0.05
3ท1 and 3ท2), for lichen and reindeer meat, in northern
Milk, 1978-1985
Russia and northern Fennoscandia. These values are
Kola
12
0.082 ฑ0.030
P = 0.081
based on the assumption that the interception rate was
Nenets AO
39
0.062 ฑ0.035
the same for global fallout and Chernobyl fallout. Over
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
the 40-year observation period, activity concentrations
in lichen and reindeer meat are well described in all re-
3.5.5. Lichen and reindeer
gions by a double exponential fit.
The long-term trend in 137Cs activity concentrations in
The Tag values in Tables 3ท13 and 3ท14 were esti-
0
reindeer meat from the Kola Peninsula and the NAO re-
mated on the basis of annual data. However, significant
gion in Russia is shown in Figure 3ท35. Over a 40-year
seasonal fluctuations are possible. Thus, for example, if
period, the 137Cs activity concentrations in meat from
the deposition in winter occurs onto snow, lichen is con-
the Kola Peninsula were consistently statistically signifi-
taminated during the period of snowmelt and the inter-
cantly higher than in the NAO region. One reason for
ception fraction decreases through runoff as part of the
the difference is the greater amount of global fallout on
activity is lost in snowmelt. In contrast, for a single dry
the Kola Peninsula than in the NAO region, but there
deposition event, the interception fraction can be much
may also be other contributory factors connected with
higher than the annual mean value. Annual Tag values
0
pasture conditions.
must therefore be used with care as they are likely to un-
Tables 3ท13 and 3ท14 present parameters describing
derestimate single pulse deposition and do not take sea-
the transfer and temporal variability in 137Cs activity
sonal variation or (wet versus dry) deposition conditions
concentrations, such as Tag , A
into account.
0
0, T1, and T2 (see Boxes

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
39
Table 3ท13. Initial values of the aggregated transfer coefficient (m2/yr), ecological half-lives (Tec , yr), and ef-
fective ecological half-lives (Teff , yr) for 137Cs and 90Sr in lichen (dw). a is a parameter partitioning the decay
between the two half-lives (see Box 3ท1 and Box 3ท2)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Tag0
a1
Tec1
Tec2
Teff1
Teff2
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
137Cs
Kola Peninsula
1.4
0.80
2.0
20
1.9
12
Northern Sweden
1.4
0.52
3.4
14
3.0
10
90Sr
Kola Peninsula
0.7
0.72
0.70
20
0.7
12
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Table 3ท14. Initial values of the aggregated transfer coefficient (m2/yr), ecological half-lives (yr), and effective
ecological half-lives (yr) for 137Cs activity concentrations in reindeer meat (ww).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Tag0
a1
Tec1
Tec2
Teff1
Teff2
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Kola Peninsula
1.7
0.82
2.0
18.0
1.9
11
Nenets AO
1.2
0.81
1.8
15.6
1.5
10
Kautokeino (Norway)
1.8
0.83
1.2
13.0
1.1
9.0
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Activity concentrations of137Cs in reindeer meat also
If contamination occurs during deep snow cover,
vary over the year due to changes in food selection. In
then reindeer will ingest contaminated snow. The ex-
summer, reindeer eat herbaceous vegetation. In autumn
tent of intake is determined by the amount of snow
they can eat large quantities of mushrooms. In winter,
ingested, but also by further snowfalls that may be
they mainly eat ground and arboreal lichens, which have
less contaminated. The extent of contamination re-
a higher radiocesium content than herbaceous vegeta-
ceived from lichen depends on whether the contami-
tion which they dig out from under the snow. The high
nated snow overlying the lichen melts allowing the
interception of radionuclides by lichen, particularly ra-
lichen to intercept the radioactivity. There can be a sig-
diocesium, is one of the key factors contributing to the
nificant delay in reindeer attaining high levels of radio-
most vulnerable Arctic food pathway, lichen reindeer
cesium from lichen owing to the protection of lichen
man (Figures 3ท36 and 3ท37; ลhman and Nyl้n, 1998).
by snow.
137Cs concentration in reindeer meat, Bq/kg ww
20000
Jiingevaerie, Sweden
Summer values
15000
Winter values
Simulated
10000
5000
Figure 3ท36. Activity concentra-
tions of 137Cs in fresh reindeer
meat from the Jiingevaerie herd-
ing district in Sweden (ลhman
0
and Nyl้n, 1998).
1
2
3
4
5
6
7
8
9
10
11
12
Years after the Chernobyl accident
Tag value for reindeer, m2/kg
0.8
May-July
August-September
0.7
October-December
January-April
0.6
0.5
0.4
0.3
0.2
Figure 3ท37. Seasonal variation
0.1
in Tag values for reindeer follow-
ing the Chernobyl accident (ลh-
0
man and Nyl้n, 1998).
0
1000
2000
3000
4000
Days after the Chernobyl accident

---

40
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration in reindeer meat, Bq/kg ww
1000
Highland areas
Highland areas
Coast
Small
Small
confined
confined
area,
area,
100
summer
winter
10
Figure 3ท38. 137Cs activity concentra-
tions in meat from free-ranging rein-
deer from Iceland for different hunting
seasons. Datasets for 1965 to 1991 are
1
from highland areas and the coast, and
the 1992 data from a more homoge-
neous restricted area.
1965
1966
1967
1990
1991
1991
1992
1992
137Cs wholebody content, Bq
60000
Kautokeino, Norway
50000
Males
Females
40000
30000
20000
10000
Figure 3ท39. Wholebody measurements
of 137Cs in reindeer herders from Kau-
0
tokeino, Norway.
1965
1970
1975
1980
1985
1990
1995
2000
Significant changes from year to year due to changes
Table 3ท15. Wholebody measurements (Bq) for reindeer herders
in food selection and variability in the 137Cs activity
(STUK; NRPA; Borghuis et al., 2002).
concentrations in plants are also clearly visible in a study
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Males
Females
Average
of wild free-ranging reindeer in Iceland. They are rela-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
tively few in number (around 3000) and roam within a
Russia
highland region of different types of vegetation. A few
Kola
1999
3250 ฑ250
samples were obtained each year by inspectors monitor-
---------------------------------------------------------------------------------------
ing the hunting together with information about the lo-
Finland
cation in which the animals had been shot and where
Inari
1995
3300
1997
3000
they had grazed (Figure 3ท38).
---------------------------------------------------------------------------------------
If the herd had grazed a large area with different
Norway
types of vegetation, this was reflected in greater variabil-
Kautokeino 1996
2600 ฑ1400
1400 ฑ600
ity in the activity concentrations in the meat. If the herd
1999
2200 ฑ800
1100 ฑ400
had grazed a relatively small or uniform area, then activ-
2002
1414
872
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
ity concentrations showed little variability. The grazing
areas selected by the herd varied from year to year, influ-
enced for example by climate and the state of the vegeta-
tion. The lowest values (<1 Bq/kg ww) occurred in sam-
3.7. Site-specific data
ples from 1992 from a herd that had grazed a relatively
homogeneous area confined by glacial rivers and close
3.7.1. Faroe Islands
to the glacier Vatnaj๖kull.
The Faroe Islands were not addressed in the first AMAP
radioactivity assessment. They comprise 18 islands be-
tween 6ฐ15'W and 7ฐ41'W and 61ฐ20'N and 62ฐ24'N
3.6. Humans
with a total land surface area of 1399 km2 (Figure 3ท40).
Trends in wholebody measurements of 137Cs presented
The land is mountainous, with the highest peak 882 m
in the first AMAP assessment have been extended by
above sea level. There were 46 180 inhabitants on 31
new data on wholebody measurements for northern
December 2000. There is no woodland on the Faroe Is-
Norway, Finland, and the Kola Peninsula and NAO re-
lands. Land cover is dominated by rough, semi-natural
gions of northwest Russia (see Table 3ท15). An example
pasture, and is grazed throughout the year by around
of the temporal trend since 1965 is shown in Figure 3ท39.
70 000 sheep and some cattle.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
41
Lamb meat
Figures 3ท43 and 3ท44 show annual average 137Cs and
90Sr activity concentrations in lamb meat, based on a
few samples collected mainly in October. Trend analyses
Funnings-
fj๘rdur
Klaksvํk
Nordoyri
were not possible because the samples were collected
Hvalvํk
from different places and there is significant spatial vari-
Skแli
B๘ur
137Cs flux via precipitation, Bq/m2/ yr
10 000
T๓rshavn
Velbastadur
Klaksvํk
1000
Sandur
100
62ฐN
T๓rshavn
10
Atlantic Ocean
1
Hvalba
0
Tv๘royri
1960
1970
1980
1990
2000
20 km
Figure 3ท41. Annual average 137Cs levels in precipitation from T๓rs-
havn and Klaksvํk, Faroe Islands.
Sumba
7ฐW
90Sr flux via precipitation, Bq/m2/ yr
Figure 3ท40. The Faroe Islands.
10 000
Klaksvํk
3.7.1.1. Climate
1000
The climate is milder than might be expected at a lati-
100
tude of 62ฐN due to the influence of the North Atlantic
T๓rshavn
Current (the `Gulf Stream'). Measurements at synoptic
10
weather stations within 100 m of sea level, indicate an
average annual air temperature of 6 to 7ฐC, with aver-
1
age winter and summer air temperatures of 3 to 4ฐC and
0
9 to 10ฐC, respectively (Cappelen and Laursen, 1998;
1960
1970
1980
1990
2000
Lysgaard, 1969). Only minor differences in air tempera-
Figure 3ท42. Activity concentrations of 90Sr in precipitation from
ture occur between the synoptic stations. There is signif-
T๓rshavn and Klaksvํk, Faroe Islands.
icant spatial variation in precipitation rates, however,
due to the combined effects of meteorology and topog-
137Cs concentration in lamb meat, Bq/ kg ww
raphy. These synoptic data are not totally representative
1000
of the Faroese climate however, as the zone within 100
Faroe Islands
m of sea level only covers 10% of land area. The cli-
matic conditions change gradually from cool temperate
100
oceanic conditions at the coast to Arctic conditions in
the mountains (Mortensen, 2002). The annual average
temperature at a new weather station on the mountain
10
Sornfelli, 722 m above sea level, was 1.71ฐC in 2000
(Mortensen, 2002).
0
1960
1970
1980
1990
2000
3.7.1.2. Cesium-137 and 90Sr
Figure 3ท43. Annual average 137Cs activity concentrations in lamb
in precipitation and foodstuffs
meat, Faroe Islands.
Measurements of environmental radioactivity have been
90
carried out on samples from the Faroe Islands since 1962,
Sr concentration in lamb meat, Bq / kg ww
100
with an emphasis on terrestrial and marine foodstuffs.
Faroe Islands
10
Precipitation
Monthly precipitation samples have been obtained for
1
radioactivity analyses since the 1960s. Annual average
137Cs deposition rates in T๓rshavn in the central part of
0.1
the country and in Klaksvํk in the north, shown in Fig-
0.01
ure 3ท41, were highest in the early-1960s and showed a
pronounced peak following the Chernobyl accident in
0
1986. Pre-1986 137Cs data are based on 90Sr measure-
1960
1970
1980
1990
2000
ments (using a 137Cs : 90Sr fallout ratio of 1.6); after 1986
Figure 3ท44. Annual average 90Sr activity concentrations in lamb
they are actual measurements (Figures 3ท41 and 3ท42).
meat, Faroe Islands.

---

42
AMAP Assessment 2002: Radioactivity in the Arctic
ation in contamination across the country (Joensen,
3.7.1.3. Transfer of 137Cs within the lamb food chain
1999). During the 1990s however, the samples were col-
in semi-natural pastures
lected consistently from the same places. Figure 3ท43 in-
dicates increased 137Cs activity concentrations after the
Spatial variability in 137Cs transfer to lamb has been
Chernobyl accident. In contrast, 90Sr activity concentra-
evaluated by comparing its characteristics at nine uncul-
tions in lamb meat are lower and are not affected by
tivated pastures during 1990 to 2000. Their locations
Chernobyl fallout (Figure 3ท44).
are shown in Figure 3ท40 (locations marked by black
symbols). Teff values and transfer factors for various
Drinking water
components of the lamb food chain were estimated for
Figure 3ท45 shows annual average activity concentrations
the nine pastures. The soil at each site was previously
for 90Sr in tap water from T๓rshavn and Klaksvํk since
characterized by Hove et al. (1994) and Joensen (1999).
1962. Faroese drinking water is obtained from surface
water. Sampling frequency has varied from monthly in
Soil
the early-1960s to an annual summer value in the 1990s.
There are large temporal and spatial variations within
No 90Sr from Chernobyl was observed in drinking water.
and between pastures in terms of 137Cs deposition in the
upper 10 cm soil layer (Figure 3ท46). Between 50 and
90Sr in drinking water, Bq/m3
80% of the deposition in this layer occurs in the upper 5
100
cm. Deposition ranges from 2000 to 8000 Bq/m2. There
T๓rshavn
was no consistent pattern of change with time for the
nine pastures; with clear declines in some pastures, but
not in others.
10
Soil pH in the nine pastures was between 4.4 and 5.3,
Klaksvํk
and loss on ignition was 50 to 70% (Joensen, 1999). Thus
these soils are acidic with a high organic matter content:
conditions that favour a high uptake of radiocesium.
1
Grass
Activity concentrations of 137Cs in mixed grass decreased
in most pastures during the 1990s, with the highest lev-
els in Hvalvํk and the lowest in Hvalba. Although Fig-
0
ure 3ท46 shows little clear difference in deposition be-
1960
'65
'70
'75
'80
'85
'90
'95
2000
tween the nine sites, change over time in the grass varied
Figure 3ท45. Annual average activity concentrations for 90Sr in drink-
widely (Figure 3ท47). In some pastures early declines
ing water from T๓rshavn and Klaksvํk.
were evident, which were presumably due to the declin-
3 45
ing Chernobyl input, while others showed no overall de-
Effective ecological half-lives
cline, and others an approximate 20-fold decline.
Table 3ท16 presents Teff values for 137Cs and 90Sr in milk,
Soil-to-grass Tag values for the nine pastures are
lamb meat, precipitation, and drinking water. These are
shown in Figure 3ท48. The highest values occurred in
estimated by regressing the logarithm of the measured
Hvalvํk and the lowest in Hvalba and Sandur. A multi-
activities against time. The 137Cs Teff values in foodstuffs
ple linear regression analysis between Tag values and pH,
range from 4.9 to 8.7 yr, while those for 90Sr range from
loss on ignition, and potassium, showed loss on ignition
3.7 to 4.5 yr. Figures 3ท41 to 3ท45 and Table 3ท16 indi-
to be the most significant factor (Joensen, 1999), and
cate a tendency for increasing Teff values with time. De-
that the regression coefficient is negative for pH and
spite the small geographical extent of the Faroe Islands,
potassium, and positive for loss on ignition. There were
spatial variation in the Teff values is apparent.
two orders of magnitude between the lowest and highest
Table 3ท16. Estimated Teff values (yr) based Figures 3.41 to 3.45. r2 (shown in brackets) from linear regres-
sion of the natural logarithm of the measured activity concentrations against time.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Milk
Lamb meat
Precipitation
Drinking water
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
137Cs*
Klaksvํk
6.2 (0.846)
T๓rshavn
4.9 (0.959)
Tv๘royri
5.8 (0.971)
`Faroes'
5.5 (0.555)
------------------------------------------------------------------------------------------------------------------------------------------
137Cs**
Klaksvํk
5.3 (0.980)
3.3 (0.940)
T๓rshavn
6.4 (0.977)
4.3 (0.932)
Tv๘royri
7.1 (0.987)
`Faroes'
8.7 (0.772)
------------------------------------------------------------------------------------------------------------------------------------------
90Sr***
Klaksvํk
4.4 (0.980)
2.8 (0.941)
5.1 (0.958)
T๓rshavn
4.4 (0.985)
3.2 (0.953)
6.9 (0.966)
Tv๘royri
4.5 (0.986)
`Faroes'
3.7 (0.935)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
*only pre-Chernobyl data; **all data except for 1986 to 1992 (to avoid the Chernobyl peak); ***all data.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
43
137Cs concentration in soil, Bq/m2
10 000
8000
6000
4000
2000
0 1990
2000
B๘ur
Velbastadur
Hvalvํk
Skแli
Funnings-
Nordoyri
Sandur
Hvalba
Sumba
fj๘rdur
Figure 3ท46. Annual average (ฑSE) 137Cs deposition to surface soil (upper 10 cm) between 1990 and 2000 (1999 for Hvalba and Sumba) in
the Faroe Islands (for locations see Figure 3.40).
137Cs concentration in mixed grass, Bq/kg dw
400
350
300
250
200
150
100
50
0 1990 B๘ur 2000 Velbastadur
Hvalvํk
Skแli
Funnings-
Nordoyri
Sandur
Hvalba
Sumba
fj๘rdur
Figure 3ท47. Annual average (ฑSE) 137Cs activity concentrations in mixed grass between 1990 and 2000 (1999 for Hvalba and Sumba) in the
Faroe Islands (for locations see Figure 3.40).
137Cs soil-to-grass Tag value, m2/ kg dw
3 47
0.12
0.10
0.08
0.06
0.04
0.02
0 1990
B๘ur
2000
Velbastadur
Hvalvํk
Skแli
Funnings-
Nordoyri
Sandur
Hvalba
Sumba
fj๘rdur
Figure 3ท48. Annual average and ranges of soil-to-grass Tag values for 137Cs between 1990 and 2000 in the Faroe Islands (for locations see
Figure 3.40).

---

44
AMAP Assessment 2002: Radioactivity in the Arctic
137Cs concentration in lamb meat, Bq/ kg ww
90
80
70
60
50
40
30
20
10
0 1990
2000
B๘ur
Velbastadur
Hvalvํk
Skแli
Nordoyri
Sandur
Hvalba
Sumba
Figure 3ท49. Annual average (ฑSE) 137Cs activity concentrations in lamb meat between 1990 and 2000 (1999 for Hvalba and Sumba) in the
Faroe Islands (for locations see Figure 3.40).
137Cs soil-to-meat Tag value, m2/ kg ww
0.025
0.020
0.015
0.010
0.005
0 1990
B๘ur
2000
Velbastadur
Hvalvํk
Skแli
Nordoyri
Sandur
Hvalba
Sumba
Figure 3ท50. Tag values for 137Cs transfer to lamb meat at nine different sites in the Faroe Islands (for locations see Figure 3.40). Yearly aver-
ages and ranges 1990-2000.
Tag values over the ten-year period. Some individual pas-
The Tag values at the nine sites are shown in Figure
tures showed a similar degree of variation. There was no
3ท50. Again, the highest values occurred at Hvalvํk,
clear time dependency in Tag values at most sites.
Skแli, and Noroyri. There is significant variation in Tag
values within and between sites.
Lamb meat
Figure 3ท49 shows 137Cs activity concentrations in lamb
Effective ecological half-lives
meat for 1990 to 1999. Large standard errors reflect
Effective ecological half-lives were derived for 137Cs ac-
large individual variation between animals. The highest
tivity concentrations in grass and meat (Table 3ท17).
values occurred at Hvalvํk, Skแli, and Noroyri.
They could only be estimated for some pastures.
Table 3ท17. Teff values (yr) based on measurements for 1990 to 2000. r2 (shown in brackets) from linear regression of the natu-
ral logarithm of the measured 137Cs activity concentrations against time. No estimates given when r2 < 0.3.
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
B๘ur
Velbastaur
Hvalvํk
Skแli
Noroyri
Sandur
Hvalba
Sumba
Grass
-
5.3
ญ
ญ
5.3
3.1
ญ
3.6
(0.027)
(0.306)
(0.235)
(0.167)
(0.93)
(0.379)
(0.005)
(0.667)
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
Meat
5.1
ญ
ญ
ญ
ญ
6.9
8.0
ญ
(0.668)
(0.033)
(0.199)
(0.031)
(0.060)
(0.392)
(0.873)
(0.069)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
45
Conclusion
The uneven surface of the rangeland areas and the
Even though the Faroe Islands cover a small geographi-
sparse vegetation can make it difficult to obtain repre-
cal area, there was considerable spatial and temporal
sentative deposition estimates by sampling the soils.
variation in the transfer of 137Cs from soil to both grass
The first AMAP assessment (AMAP, 1998) identified
and lamb meat. Owing to this high variability it is inap-
Iceland as one of the Arctic areas receiving the most fall-
propriate to use a single Tag value for either grass pas-
out from atmospheric nuclear weapons tests, owing to re-
ture or lamb meat. In other countries, there is generally
latively high precipitation rates compared with much of
greater variation in the key soil characteristics influ-
the rest of the Arctic and subarctic (Wright et al., 1999).
encing radiocesium uptake than was measured at these
The 137Cs in the Icelandic terrestrial ecosystem origi-
sites. Even higher variability could thus be expected in
nates almost entirely from nuclear weapons tests carried
other countries and country-wide generalizations about
out in the atmosphere until the early-1960s. Fallout was
transfer are open to considerable error.
greatest in the mid-1960s. Additional fallout from the
Chernobyl accident was relatively small (Pแlsson, 1996).
This section provides data on radionuclide contamina-
3.7.2. Iceland
tion in Iceland and uses recently acquired data to test the
3.7.2.1. Site description
methodology and conclusions of the first AMAP assess-
Iceland is the second largest island in Europe, located in
ment regarding global fallout.
the North Atlantic just south of the Arctic Circle. The total
Measurements of fallout from nuclear weapons tests
surface area is 103 000 km2, of which 23 805 km2 (23%)
in soil, vegetation, and agricultural products started in
are vegetated, 11 922 km2 glaciers, 2757 km2 lakes, and
Iceland over 40 years ago (Pแlsson, 1996). Considerable
the remaining 64 538 km2 (63%) barren. The coastline,
variability was present in the results, even between adja-
including fjords and inlets, is about 4970 km long.
cent sites, probably due to the mountainous terrain, var-
Iceland is the most sparsely populated country in Eu-
iable and strong winds, and highly variable levels of pre-
rope with an average of 2.8 inhabitants per km2. On 31
cipitation. This variability is particularly noticeable for
December 2000 the number of inhabitants was 283 361.
soils. Early measurements of nuclear fallout were re-
The Icelandic diet is western European in most re-
stricted to cultivated lowland areas. The importance of
spects. Nevertheless it retains some characteristics of a
uncultivated rangelands in Icelandic agriculture (e.g., for
subarctic region, making it unique among European na-
sheep farming) makes their inclusion desirable for cur-
tions. Fish, meat, and milk are traditionally the main
rent and future estimates of radionuclide transfer into
foods produced in Iceland. Icelanders consume more fish
agricultural products.
than any other nation in Europe (73 g/d/cap) and, in gen-
Since summer 2000, spatial variation in 137Cs depo-
eral, food of animal origin constitutes a large proportion
sition in Iceland has been studied systematically. The ob-
of the Icelandic diet. During the 1990s, the consumption
jectives of the study are to measure the spatial variation
of lamb meat decreased, beef consumption increased
of radiocesium inventories in Icelandic soils and to com-
slightly, and the consumption of pork and poultry in-
pare the results with predicted 137Cs soil levels (Sigur-
creased significantly (50 to 100%). This is reflected in
geirsson et al., 2002).
agricultural production figures, since most of the prod-
In summer 2000, soil samples were collected to a
ucts are consumed domestically.
depth of 25 cm at 14 sites. The sites were located close
In 2000, there were 466 000 sheep and 72 000 cattle
to meteorological measurement stations so that repre-
on Iceland. During summer the sheep graze freely on
sentative precipitation data were available. Deposition
rangelands in the interior, the same applies for a propor-
at each site was estimated by assuming the 137Cs activity
tion of the 74 000 horses and around 4000 wild reindeer
concentration in precipitation was the same at all sites
(Rangifer tarandus, the original herd imported from
during any given period. Thus, deposition at each site is
Scandinavia between 1771 and 1787) which inhabit the
estimated by measuring the activity concentration of
northeast of the country. Although reindeer constitute a
137Cs in precipitation at one reference site and then esti-
minor part of the Icelandic diet, hunting is an increas-
mating deposition at the other sites by summing the
ingly popular sport, which also provides an important
product of precipitation (in m) at the site and 137Cs in
source of income for local communities.
precipitation (in Bq/m3) at the reference site for the pe-
Thus, sheep, horses, and reindeer would be affected
riod of interest (Pแlsson et al., 2002a,b).
by contamination of the highland areas of Iceland,
A reference station close to Reykjavํk (Rj๚pnahๆ,
whereas lowland contamination would affect cattle, and
location 7 on Figures 3ท51 and 3ท52) was used for pre-
pig and chicken farming.
dicting fallout in Iceland, and quarterly measurements of
Volcanic eruptions are frequent in Iceland, producing
fallout radioactivity in precipitation were undertaken reg-
lava fields and volcanic ash deposits of various extent.
ularly by the U.K. Atomic Energy Authority from 1959
The unstable barren areas of the highlands and the
to 1982 (Pแlsson, 1996). Precipitation data for the refer-
floodplains of glacial rivers act as sources of aeolian ma-
ence site were supplied by the Icelandic Meteorological
terial. The parent materials of Icelandic soils are largely
Office (Verแttan 1959-1983). The reference station data
of volcanic origin. Icelandic soils are mostly andosols,
show that 82.9% of the decay-corrected deposition of
which are characterized by low cohesion and a high ca-
137Cs occurred during the first eight years, i.e., 1960 to
pacity to absorb water (>100% on a dry weight basis).
1967. The emphasis of the study was thus placed on me-
This high water-holding capacity intensifies freezing ef-
teorological stations that were operational during this
fects, resulting in solifluction, landslides, needle ice
eight-year period; estimates of deposition for these years
formation, and the formation of hummocks (Arnalds,
were based on 1960 to 1967 precipitation data. The re-
1999).
sults were subsequently scaled up to cover the entire study

---

46
AMAP Assessment 2002: Radioactivity in the Arctic
137
137
Cs deposition, Bq/m2
Cs deposition for Iceland, based on the AMAP meth-
5000
odology is given in Figure 3ท52.
Iceland
Overall, the AMAP methodology has been successful
and has the advantage that it is possible to predict the
4000
137Cs deposition at any location in any year since 1960.
;;;
Allthough fallout did occur prior to 1960, this was at
3000
lower levels than during the 1960s.
Predicted
2000
4
;;
;;
;;
;;;;
Measured
1
1000
2
;;
;;
;;
;;;;
;;
;;
;;;;
3
5
6
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Sampling site
11
;;
;;
;;
;;;;
;;
;;
;;;;
Figure 3ท51. Comparison of predicted 137Cs deposition based on
12
3.51
7
precipitation data and measured values at fourteen sites (see Figure
8
3ท52 for locations) close to meteorological stations in Iceland (Pแls-
9
10
13
son et al., 2002a).
14
period, 1960 to 1982, assuming in all cases that 82.9%
of the deposition had occurred during the first eight
137Cs, estimated cumulative deposition, Bq/m2
years used for calculations. Fallout data prior to 1960
were not included in this study and thus the measured
500
1000
2500
5000
10 000
25 000
values should be slightly higher than those predicted.
Figure 3ท52. Preliminary map of estimated cumulative deposition of
The measured 137Cs content per unit area of soil var-
137Cs from atmospheric nuclear weapons tests, decay-corrected to
ied from 900 to 4700 Bq/m2, with deposition greater in
1995 (AMAP Data Centre). The map is based on a preliminary esti-
the south of Iceland which receives more precipitation.
mate of the average annual precipitation in Iceland, using a model
There are various ways of estimating the correlation
developed by Crochet (2002) and precipitation data from 1960 to
between predicted and measured 137Cs deposition. The
1990. The conversion to deposition was achieved using a method
equivalent to the AMAP method (Pแlsson et al., 2002a). Numbers
method used in the comprehensive AMAP study (Wright
indicate locations of sites represented in Figure 3ท51.
et al., 1999) was to force the regression line through the
origin and calculate correlation coefficients on that
3.7.3. Amchitka Island
basis. This gives a higher value for the correlation coeffi-
cient than for an unbound regression line, but can be
In November 1971, the project Cannikin was conducted
justified in that the assumption being tested is that depo-
by the U.S. Atomic Energy Agency, now the Department
sition is directly proportional to precipitation. This ap-
of Energy, at the Amchitka Island underground nuclear
proach was used in the present study. Figure 3ท51 com-
test area; this was its largest underground nuclear test,
pares measured and predicted deposition at the 14 sites.
with a yield of about 5 Mt. Preceding Project Cannikin
The correlation between predicted and measured val-
were Projects Long Shot and Milrow; tests of approxi-
ues for Iceland was much stronger than that reported by
mately 80 kt and 1 Mt. These three tests represented an
Wright et al. (1999). The AMAP study was based on 50
estimated 15 to 16% of the total effective yields of all
samples obtained from Greenland, Norway, and Russia
the U.S. underground nuclear tests. In total effective
between 1961 and 1985. A line through the origin was
yield, this site is the second largest and the only island
fitted to the data using least squares regression and gave
underground nuclear test area in the United States. The
an r2 value of 0.51 based on a coarse precipitation data
location is shown in Figure 3ท53.
set and disparate sources of measured 137Cs deposition
using different sampling methods. In Iceland, a compar-
N
ison of predicted and measured values gave a correspon-
ding r2 value of 0.96. This same value was obtained
Bering Sea
using both the AMAP method and when prediction was
A
10 km
m
based on average annual precipitation for 1960 to 1967.
c h
Cannikin Site
i
Some of the improved correlation relative to the AMAP
t k a
Long Shot Site
study is probably due to the proximity of meteorological
Pacific Ocean
stations, where precipitation has been measured in a
consistent manner. Also, soil sampling was conducted by
Amchitka
the same team, with a consistent, rigorous methodology
over a short period of time. However, the strength of the
Milrow Site
correlation is surprising considering that dry deposition
is not accounted for, although the high precipitation rate
in much of Iceland means dry deposition is unlikely to
contribute much to the total 137Cs deposition. In addi-
tion, lateral transport by erosion would be expected in
Figure 3ท53. Amchitka Island, Alas-
some Icelandic areas (Arnalds, 1999). A map of predicted
ka, underground nuclear test areas.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
47
ties in analyses and environmental processes must be
3.7.3.1. Sampling
fully assessed before making conclusions. Further work
Since the late-1970s there has been no marine sampling
is needed to determine whether there are any other
around Amchitka to assess the status of the anthro-
sources of Pu to the Bering Sea and North Pacific regions
pogenic radionuclides present or to determine trends.
besides global fallout.
Groundwater contaminated by the three underground
Results of the 1997 and 1998 sampling based on the
nuclear tests is transported toward discharge points on
measured 240Pu : 239Pu ratios and 3H levels do not pro-
the ocean floor. Conceptual groundwater transport mod-
vide any evidence for leakage of 241Am or other ra-
els have shown, based on a range of geohydrological as-
dionuclides from the underground test shot cavities into
sumptions, that discharge of radionuclides could have
the terrestrial or freshwater environments on Amchitka
started as early as 1975; ten years after the first test.
Island (Dasher et al., 2002). In addition, the hydrogeo-
In 1996, Greenpeace reported that leakage of 241Am
logical regime as understood for Amchitka does not pro-
and 239+240Pu had been detected from these under-
vide the physical means to transport transuranics from
ground test sites to the terrestrial and freshwater envi-
the test cavities to the reported surface locations.
ronment (Miller and Buske, 1996). The marine environ-
Clearly, these results do not mean that leakage from
ment was not specifically addressed in the Greenpeace
the Amchitka underground nuclear tests is not occur-
report. In response, a federal, state, tribal, and non-gov-
ring or will not occur into the North Pacific Ocean or
ernmental team conducted a terrestrial and freshwater
the Bering Sea. Hydrogeological modelling predicts
radiological sampling program in 1997. Additional radio-
leakage can begin initially for 3H from the test sites into
logical sampling was conducted in 1998. An assessment
the marine water over periods of 20 to 3000 years (Claas-
of the reported leakage to the freshwater environment
sen, 1978; Dudley et al., 1977). These periods bridge the
was evaluated by assessing tritium (3H) values in surface
various hydrogeological parameter assumptions that
waters and 240Pu : 239Pu ratios in various sample media
can be made. No sampling has been conducted in the
(Dasher et al., 2002). Tritium values ranged from 0.41
marine environment surrounding Amchitka since the
Bq/L ฑ0.11 (2 SD) to 0.74 Bq/L ฑ0.126 (2SD) at the sur-
late-1970s and, thus, it remains an important area to be
face water sites sampled, including the reported leakage
addressed.
sites. Only at the Long Shot test site, where leakage of
radioactive gases to the near surface occurred in 1965,
3.7.3.2. Geological forces
were higher 3H levels of 5.8 Bq/L ฑ0.19 (2SD) still ob-
served in 1997; in mud pit # 3 (Faller and Farmer, 1998).
Since the underground nuclear tests, dramatic changes
The mean 240Pu : 239Pu ratio for all Amchitka samples
have taken place in the field of geosciences and in the
was 0.199 ฑ0.014 (1 SD), with values ranging from
understanding of the geological forces acting on the
0.182 ฑ0.0007 (1 SD) to 0.24 ฑ0.02 (1 SD).
Aleutian Islands, including Amchitka (Eichelberger et
For the macroalga Fucus distichus the mean
al., 2002). In the 1960s, the site was considered geo-
240Pu : 239Pu atom ratio of 0.217ฑ0.016 (1 SD) is
logically stable, showing little evidence of vertical tec-
slightly outside the 95% confidence interval (ฑ2 SD) of
tonic motion or massive slope failures. Large ongoing
the reported global ratio of 0.176 ฑ 0.014 (1 SD) (Krey
horizontal displacements were not considered a pos-
et al., 1976). The mean 240Pu : 239Pu atom ratio of 0.216
sibility.
ฑ0.023 (1 SD) for the littoral zone marine sediment
A paradigm shift occurred in the years following the
samples was consistent with the higher ratio seen for F.
nuclear tests with the acceptance of the theory of plate
distichus. Deviations from the global fallout mean
tectonics. Amchitka Island is now understood to be a
240Pu : 239Pu atom ratio observed in marine algae, sedi-
fragment of an island arc crest at the intersection of the
ment, and pooled Amchitka samples may suggest an-
subduction of the Pacific Plate beneath the North Amer-
other source of Pu to the marine environment. In an in-
ican Plate (Figure 3ท54). Recent field measurements (Ei-
vestigation of Bering Sea sediments Hameedi et al.
chelberger et al., 2002) indicate that Amchitka is under-
(1999) reached similar conclusions. However, uncertain-
going westward movement of about 1 cm/yr. This sug-
Russia
Alaska
N o r t h A m e r i c a n P l a t e
Amchitka
Figure 3ท54. The Aleutain Vocanic Arc.
Orange circles indicate active volcanoes.
Arrows show the motion of the Pacific
P a c i f i c P l a t e
Plate relative to the North American
Plate (Eichelberger et al., 2002).

---

48
AMAP Assessment 2002: Radioactivity in the Arctic
gests potential major faults in Amchitka Pass and also a
strike-slip boundary north of Amchitka that is moving
at a rate approximately two-thirds that of the San An-
Barents Sea
dreas fault. The stresses induced would tend to open
a
fractures perpendicular to the island leading into the
y
l
marine environment.
m
e
The acceptance that Amchitka is part of a crustal
Z
block rotating clockwise within the fore-arc of an
Sukhoy Nos
Kara Sea
Zone C
obliquely converging subduction zone raises concerns
over possible enhancement of `fast pathways' for the re-
Matochkin Shar
Zone B
a
lease of radionuclides from the underground nuclear test
y
a
sites to the marine environment. Figure 3ท55 provides a
v
schematic illustration of the Cannikin underground nu-
o
clear test site over the adjacent Teal Creek Fault. A lim-
N
ited survey of the coastline near the Cannikin test area
Chernaya Bay
Zone A
100 km
revealed a low density of joints, but did identify appar-
ent fluid transport along these structures in the geologi-
Figure 3ท56. The central test site of the Russian Federation at No-
cal past. A better understanding of the geological forces
vaya Zemlya.
affecting Amchitka is needed to determine their poten-
tial impact on leakage and in designing appropriate
3.7.4. Novaya Zemlya
long-term monitoring programs.
Since the first AMAP assessment, two further sources of
information on Novaya Zemlya have been published: a
3.7.3.3. Summary
report by Ivanov et al. (1997) and a major review of the
The knowledge that environmental pathways, poten-
Novaya Zemlya test site by Logachev (2000). The fol-
tially enhanced by geological forces, exist on Amchitka
lowing information is based on these sources.
Island to transport radionuclides into the nearshore ma-
Novaya Zemlya was one of two major nuclear test
rine environment is a cause for concern. The nearshore
areas for the former Soviet Union. Altogether, 130 nu-
environment is commercially and environmentally im-
clear tests had been carried out at the north test site by
portant. An independently conducted radiological as-
25 October 1990 (the date the effective moratorium was
sessment is required to protect the indigenous peoples of
announced). Nuclear tests were conducted in three areas
the region (the Aleut), U.S. and Russian citizens, and
(Figure 3ท56):
people living off the vast commercial fisheries of the re-
Zone A. A series of nuclear tests were conducted from
gion (State of Alaska and Aleutian/Pribilof Island Asso-
1955 to 1962. The first underwater nuclear test occur-
ciation, 2001). The recent radiological assessment of the
red on 21 September 1955, the date of commencement
French nuclear test sites in the South Pacific provides an
for the site. The explosion comprised the experimental
example of what is needed (IAEA, 1998b).
blasting of a T-5 torpedo with a warhead of about 3.5 kt
at a depth of about 12 m. Another underwater test was
SE
NW
Teal Creek Fault
conducted later. Six underground nuclear tests were con-
Cannikin Lake
ducted in vertical blast holes between 1972 and 1975,
and there was a near-surface nuclear explosion on 7 Sep-
tember 1957.
Zone B. Thirty-three underground nuclear tests were
Collapse
conducted in tunnels within mountains between 1964
chimney
and 1990.
Zone C. A series of elevated and atmospheric nuclear
Limit of
elastic
tests, were undertaken commencing 24 September 1957
deformation
and ending 25 December 1962. The largest nuclear test
explosion ever, a 58 Mt atmospheric blast, occurred in
this zone on 30 October 1961.
Cavity
3.7.4.1. Soil contamination
The data used by Ivanov et al. (1997) and Logachev
Limit of
(2000) are mainly from surveys by the E.K. Fedorov In-
fracturing
stitute of Applied Geophysics (in 1976 to 1978 and 1990)
and the V.G. Khlopin Radium Institute (in 1992).
The 137Cs contamination density outside the test areas
Limit of
500 m
melt injection
ranged from 1.7 to 5.6 kBq/m2 (45 to 150 mCi/km2) in
the 1976 to 1978 survey, with a mean of 3.4 kBq/m2 (91
mCi/km2). The contamination density in 1990 ranged
Figure 3ท55. Schematic cross-section showing the effects of the Can-
nikin underground test, based on work by R. Laczniak and col-
from 1.5 to 6.7 kBq/m2 (40 to 180 mCi/km2) with a mean
leagues at the Nevada Test Site (Eichelberger et al., 2002).
of 3.3 kBq/m2 (90 mCi/km2).
3 55

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
49
54ฐ40'E
55ฐ20'E
56ฐ00'E
0.03
dose rate, ตSv/ h
0.05
0.1
0.04
0.03
August 1964
0.04
0.05
0.03
71ฐ00'N
0.04
0.1
0.1
0.03
0.05
0.2
0.05
0.04
0.3
0.03
0.05
0.4
0.05
70ฐ50'N
0.1
0.1
0.1
0.4
0.2
0.3
0.1
0.04
2 1
0.05
0.03
0.03
2
0.04
1
Figure 3ท57. Radiation situation in
0.5
0.05
0.03
0.1
0.05
the area around Chernaya Inlet in Au-
0.03
gust 1964 (Ivanov et al., 1997; Lo-
0.04
gachev, 2000). Contour lines show
0.04
10 km
0.1
0.3
70ฐ40'N
dose rates 1 m above ground.
54ฐ40'E
55ฐ20'E
56ฐ00'E
0.03
dose rate, ตSv/ h
September 1977
0.03
71ฐ00'N
0.03
19
0.03
15
0.03
0.03
70ฐ50'N
0.04
18
14
13
11
17
12
0.1 0.04
0.1
0.03
Figure 3ท58. The radiation situation
0.2
0.3
9
6
8
10
in the area of the Chernaya Inlet in
5
September 1977 (Ivanov et al., 1997).
23
16
24 23
0.1
Contour lines show dose rates 1 m
1
0.2 22
4
above ground. Red dots and associ-
0.03
25
21
ated numbers indicate sampling sites
10 km
20
70ฐ40'N
referred to in Table 3ท19.
The surveys placed particular emphasis on studying
The epicenter of one of the explosions conducted in
radioactive traces due to deposition of radioactive prod-
1957 was about 100 m from the shore. The trace of the
ucts after the nuclear explosions. The zones of increased
radioactive fallout from this explosion, as followed in
radioactive contamination detected during the survey in-
the first three days after the explosion to a distance of
clude:
1500 km from the epicenter, covered part of the South
Island and the Yamal, Gydansky, and Taymir Peninsu-
ท the area around Chernaya Bay (Zone A);
las. In 1964, another aerial survey was carried out over
ท Sukhoy Nos Peninsula (Zone C);
the extent of the trace covering 70 km along the axis and
ท the area around Bashmachnaya Inlet (Zone B); and
the results are shown in Figure 3ท57. Analyses revealed
ท the tidal area of the Matochkin Shar Strait (Zone B).
the presence of the following radionuclides: 137Cs,
The present assessment contains improved data for
144Ce, 125Sb, 106Ru, 90Sr, 60Co, 152Eu, 154Eu, and 155Eu.
the first two sites; data for the others are available in the
The epicenter of this near-surface explosion is the most
first AMAP assessment (AMAP, 1998).
contaminated zone on the archipelago. The measure-
ments in this zone were repeated in 1976 to 1977. Re-
sults of the 1977 aerial survey and the location of soil
3.7.4.2. Area around Chernaya Bay
sampling points are shown in Figure 3ท58. The level of
Of the 90 atmospheric nuclear weapons tests conducted
gamma radiation near the 1957 explosion funnel was as
on the archipelago between 1955 and 1962 at least four
high as 5 Sv/h (the transverse size of the zone is several
were in contact with the underlying surface, and all oc-
tens of meters). The distribution of the radionuclides
curred in the area of Chernaya Bay.
was investigated using depth samples collected from two

---

50
AMAP Assessment 2002: Radioactivity in the Arctic
Table 3ท18. Radionuclide profiles for soils (Bq/kg (10ญ9 Ci/kg)) col-
In addition to the trace resulting from the near-sur-
lected in 1977 from the epicenter of the near-surface nuclear explo-
face explosion of 1957, traces of radioactive contamina-
sion conducted in 1957 in the area of Chernaya Inlet, Novaya
tion from an above-water explosion in 1961 and an un-
Zemlya (Ivanov et al., 1997; Logachev, 2000).
derwater explosion in 1955 can be seen in Chernaya
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Bay. Data on the density of contamination by radionu-
Depth, cm
137Cs
60Co
152Eu
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
clides in these traces obtained in 1977 are shown in
Site 1
Table 3ท19.
0-1
67000 (1800)
52000 (1400)
Trace
1-2
70000 (1900)
44000 (1200)
Trace
2-3
44000 (1200)
33000 (900)
17000 (460)
3.7.4.3. Sukhoy Nos Peninsula
3-4
19000 (520)
22000 (590)
18000 (480)
4-5
12000 (320)
22000 (590)
22000 (590)
Most of the atmospheric nuclear weapons tests were
10-15
2300 (61)
27000 (720)
54000 (1450)
conducted above the Sukhoy Nos Peninsula to the north
20-25
ญ
8700 (235)
17000 (460)
of the Matochkin Shar strait. Several areas of contami-
30-35
70 (1.9)
13000 (360)
2500 (67)
nation resulted from these tests. A gamma survey and
40-45
330 (8.9)
190 (5.1)
440 (12)
soil sampling campaign carried out in 1977 showed four
50-55
230 (6.1)
ญ
ญ
contaminated areas (see Figure 3ท59):
60-65
160 (4.2)
ญ
ญ
Site 2
ท the western trace is an area (0.5 km2) 3 km east of the
0-2
41000 (1100)
32000 (870)
4400 (120)
Fedorov Mountain (Site 1);
2-4
46000 (1250)
26000 (710)
6700 (180)
ท the central trace (0.3 km2) is in the center of the penin-
4-6
50000 (1350)
26000 (715)
7000 (190)
sula (Site 2);
6-8
27000 (725)
16000 (425)
8000 (215)
8-10
23000 (610)
12000 (325)
7800 (210)
ท the northern trace (0.3 km2) is 10 km from the Tsi-
10-15
16000 (430)
13000 (360)
6800 (185)
volki Cape (Site 3); and
20-25
13000 (360)
4100 (110)
8500 (230)
ท the eastern trace (0.4 km2) is 12 km northeast of the
30-35
190 (5)
ญ
1900 (50)
Klochkovsky Peninsula (Site 4).
40-45
110 (3)
120 (3.3)
250 (6.8)
50-55
ญ
ญ
ญ
Table 3ท20 shows the man-made radionuclides in soil
60-65
46000 (3.8)
ญ
ญ
from the eastern trace. These data suggest that contami-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
nation after the elevated air explosions, during which
pits; located 10 m southwest of the funnel crest (site 1,
mineral (soil) particles were generally not entrained in
Figure 3ท58) and 100 m north of the funnel (site 2, Fig-
the plume, was primarily due to radionuclides formed as
ure 3ท58). Data on the local distribution of -emitters
a result of interaction between the neutron flux of the
for these samples are shown in Table 3ท18.
penetrating radiation arising from the nuclear explosion
The extent of the radioactive trace is decreasing with
and the soil.
time. The area of the trace exceeding 0.1 ตSv/h decreased
from 133 km2 in 1964 to 10 km2 in 1977 (see Figures
3.7.5. Thule
3ท58 and 3ท59). Moreover, the trace has become hetero-
geneous in character, due to radioactive decay of short-
In January 1968, a B-52 aircraft carrying four nuclear
lived -emitters.
weapons crashed onto the sea ice ~11 km from Thule
54ฐ00'E
54ฐ20'E
54ฐ40'E
55ฐ00'E
Air Base in northwest Greenland (see Section 2.4). As a
consequence, the benthic marine environment of Bylot
dose rate, ตSv/ h
Sound (180 to 230 m deep) became contaminated by
September 1977
239,240Pu. The site was revisited in August 1997, 29
years after the accident. Sections 3.7.5.1. to 3.7.5.4. are
extracted from Dahlgaard et al. (2001).
3
0.03
73ฐ50'N
0.03
3.7.5.1. Plutonium in water and seaweed
Activity concentrations of 239,240Pu in Fucus distichus
(a brown alga) around Thule and 750 km to the south
2
0.04
near Uummannaq ranged from 0.15 to 1.14 Bq/kg dw.
4
The source of most of this Pu is global fallout ญ except
0.04
0.03
possibly the highest value seen in a single sample near
1
the accident site. With the exception of a near-bottom
0.04
water sample taken at the point of impact containing
0.03
73ฐ40'N
30 mBq/m3, no clear effect of the accident was seen in
any of the water samples. Of this elevated level, 42%
was particulate indicating that resuspended sediments
containing accident-related Pu are an important
10 km
source. The general level of 239,240Pu within Bylot
Sound was 5 to 10 mBq/m3 in unfiltered surface water,
which is regarded as global fallout background. These
Figure 3ท59. The radiation situation on the Sukhoy Nos Peninsula
data indicate that Pu from contaminated sediments
in September 1977. The radiation dose rate was in excess of natural
background (Logachev, 2000). Red dots and associated numbers in-
is not transported into surface waters in significant
dicate contaminated sites referred to in the text.
quantities.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
51
Table 3ท19. Radionuclide profiles for soils (kBq/m2 (mCi/km2)) collected in 1977 from within the traces of nuclear explosions in the
Chernaya Bay area (Ivanov et al., 1997).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
No. of site in
Sample
in Fig 3ท58
No.
Depth, cm
137Cs
90Sr
60Co
152Eu
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Epicenter zone
1
0-5
4800 (1.3 105)
ญ
2800 (7.5 104)
780 (2.1 104)
and near-surface
2
0-5
4100 (1.1 105)
ญ
2400 (6.4 104)
280 (7600)
explosion trace
3
0-3
26 (690)
ญ
160 (4200)
230 (6300)
1957
3-6
20 (540)
ญ
160 (4200)
190 (5200)
6-10
7.8 (210)
ญ
130 (3500)
200 (5400)
4
0-5
14 (380)
ญ
ญ
ญ
5
0-5
85 (2300)
ญ
59 (1600)
ญ
6
1
0-5
160 (4400)
100 (2700)
190 (5200)
ญ
2
0-2
21 (560)
ญ
24 (650)
5.4 (140)
2-4
5.2 (140)
ญ
4.8 (130)
0
4-6
ญ
ญ
ญ
ญ
3
0-2
37 (1000)
ญ
78 (2100)
8.1 (220)
2-4
ญ
ญ
ญ
ญ
7
0-5
74 (2000)
ญ
89 (2400)
ญ
8
0-5
11 (310)
4.8 (130)
10 (280)
ญ
9
0-5
7.8 (210)
ญ
5.9 (160)
ญ
10
0-5
30 (800)
ญ
24 (660)
ญ
11
0-5
2.6 (69)
ญ
ญ
ญ
12
0-2
11 (310)
ญ
10 (270)
ญ
2-4
3.5 (94)
ญ
3.3 (88)
0
4-6
ญ
ญ
ญ
0
13
0-5
5.7 (155)
0.93 (25)
2.7 (72)
ญ
14
0-5
9.3 (250)
ญ
5.2 (140)
ญ
15
0-5
3.9 (106)
3.7 (100)
ญ
ญ
16
0-5
2.6 (69)
ญ
ญ
ญ
Trace of above-
17
0-5
48 (1300)
70 (1900)
ญ
ญ
water explosion
18
0-5
5.2 (140)
ญ
ญ
ญ
1961
19
0-5
2.3 (61)
ญ
ญ
ญ
Spot resulting from 20
1
0-5
ญ
ญ
120 (3300)
220 (5900)
near-surface 10-15
ญ
ญ
56 (1500)
110 (2900)
explosion
20-25
ญ
ญ
12 (330)
23 (630)
30-35
ญ
ญ
ญ
ญ
2
0-1
0.96 (26)
ญ
4.4 (120)
7.4 (200)
1-2
ญ
ญ
13 (350)
23 (610)
2-3
ญ
ญ
28 (770)
46 (1250)
3-5
ญ
ญ
63 (1700)
96 (2600)
21
0-5
2.6 (69)
2.9 (79)
ญ
4.4 (120)
Trace of the
22*
0-3
300 (8000)
ญ
48 (1300)
ญ
the underwater
3-6
160 (4400)
ญ
34 (930)
ญ
explosion 1955
6-10
44 (1200)
ญ
10 (270)
ญ
23
0-2
410 (11000)
ญ
56 (1500)
ญ
2-4
160 (4300)
ญ
29 (780)
ญ
4-6
26 (700)
ญ
7 (190)
ญ
6-8
41 (1100)
ญ
9.8 (265)
ญ
8-10
59 (1600)
ญ
ญ
ญ
24
0-5
44 (1200)
3.7 (100)
8.5 (230)
ญ
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
* no vegetation on the sampling location.
Table 3ท20. Radionuclides in soil (kBq/m2 (mCi/km2)) from the eastern part of the exper-
imental zone (Site 4) on the Sukhoy Nos Peninsula in early-1993 (Ivanov et al., 1997; Lo-
gachev, 2000).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Sample No.
Depth, cm
137Cs
60Co
152Eu
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
#1
0-1
1.4 (37)
Trace
2.7 (72)
1-2
1.4 (37)
Trace
3.7 (100)
2-3
0.85 (23)
Trace
4.4 (120)
3-4
Trace
Trace
5.6 (150)
4-5
Trace
Trace
5.9 (160)
#2
0-5
2.6 (70)
19 (520)
36 (980)
10-15
Trace
Trace
23 (610)
20-25
Trace
Trace
Trace
30-35
Trace
Trace
Trace
#3
0-5
2.2 (60)
15 (400)
14 (380)
#4
0-5
3.5 (95)
Trace
20 (550)
#5
0-5
4.6 (125)
12 (330)
31 (830)
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ

---

52
AMAP Assessment 2002: Radioactivity in the Arctic
239, 240Pu concentration in sediment, Bq/kg dw
0.1
1
10 100 1000
0.1
1
10 100 1000
0.1
1
10 100 1000
0.1
1
10 100 1000
0.1
1
10 100 1000
0.1
1
10 100 1000
Depth, 0
cm
10
20
20
25
s4
Y
P
23
30
0
10
20
22
C
V2
X2
H'
Hn๘
30
0
10
Figure 3ท60. Sediment 239,240Pu activity
20
concentration profiles at sites in Bylot
Sound (shown in Figure 3ท61) and back-
D
Ny-3
Thule-1412
Schades ุer
30
ground sites in 1997 (Eriksson, 2002).
3.7.5.2. Sediments
to particles within the gut and adheres to the surface struc-
Figure 3ท60 shows Pu depth profiles for contaminated
ture of the animals. One single bivalve sample had a much
sediment cores from Bylot Sound, plus background cores
higher level, which was probably due to a `hot particle'.
taken outside Bylot Sound (Ny-3, Thule-1412, and Scha-
des ุer; the latter about 750 km southeast of Thule). In
3.7.5.4. Isotope ratios
all cases, the Pu appears well mixed throughout the
upper 3 to 5 cm layer (note the logarithmic concentra-
A number of the sediment samples were analyzed for
tion axis in Figure 3ท60). Despite logarithmic axes, the
240Pu : 239Pu atom ratios by High Resolution Inductively
large variation in Pu concentrations is clear. This is
Coupled Plasma/Mass Spectrometry. These had ratios in
caused by `hot particles'. In a recent thesis Eriksson
the range 0.027 to 0.057. The calculated uncertainties
(2002) stated that these `hot particles' hold more Pu
for most of the samples were 2 to 10%. The samples
than previously anticipated (see Section 2.4).
with highest activity ญ which have been identified as
Plutonium concentrations in surface (0 to 3 cm) sed-
containing `hot particles' ญ show significant variation in
iments are shown in Figure 3ท61. The highest concentra-
the 240Pu : 239Pu atom ratios, i.e., there is a variation in
tions are centered on the accident site, with a fairly even
Pu isotope ratios in the Thule debris significantly above
distribution in the remaining deep part of Bylot Sound,
measurement error. This supports the conclusion by
and almost fallout background concentrations outside
Mitchell et al. (1997) that the Thule Pu originates from
Bylot Sound. The water depths at the accident site ญ close
at least two sources of different quality. Plutonium con-
to location V2 ญ are 180 to 230 m. The two assumed
centrations in the samples for this study were dominated
background sites outside Bylot Sound, Ny-3 and Thule-
by the Thule weapons accident. Therefore, the higher
1412, have depths of 500 and 640 m. A surface (0 to 3
cm) 239,240Pu activity concentration of 0.12 Bq/kg dw
Table 3ท21. Plutonium (239,240Pu) concentration ratios (based on
values in Bq/kg dw) for benthic biota and surface (0 to 3 cm) sedi-
occurred 750 km further south near Schades ุer. Sur-
ments (Dahlgaard et al., 2001).
face concentrations outside Bylot Sound (at Thule-1412
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
and Ny-3) are an order of magnitude higher. It is unclear
Mean
SD
n
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
whether this reflects accident Pu or a natural perturba-
Mollusks
Bivalves
0.025
0.024
13
tion caused by differences in sedimentological parame-
Macoma calcarea 37*
1
ters. At other Arctic marine locations similar levels of Pu
Snails
0.0033
0.0018
9
have been attributed to global fallout (see Section 3.3.4).
Squid, Rossia sp.
0.00036
1
Echinoderms
Starfish
0.0094
0.0139
9
3.7.5.3. Benthic biota
Brittle stars
0.013
0.016
4
Feather stars
0.0070
0.0060
4
Plutonium concentrations in biota have been compared
Sea urchins
0.12
0.16
4
with concentrations in surface (0 to 3 cm) sediments (Fig-
Sea cucumber
0.0080
0.0083
4
ure 3ท61) to give `concentration ratios' (Table 3ท21). Al-
Crustaceans
Shrimp
0.0048
0.0088
4
though the biota live within or on the sediments the con-
Various
0.038
0.039
4
centration ratios indicate that the bioavailability of the
Annelids
Pectinaria sp.
0.068
0.052
4
weapons Pu is low. Most of the concentration ratios fall
Various
0.023
0.033
10
within the range 0.01 to 0.1, i.e., Pu concentrations in
Tube
0.28
0.29
6
benthic biota are around one to two orders of magnitude
Fish
Liparis sp.
0.00035
1
lower than in surface sediments. Furthermore, a signifi-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
cant proportion is probably not metabolized but is bound
* outlier, probably caused by `hot particle'.

---

Chapter 3 ท Radioactive Contamination and Vulnerability of Arctic Ecosystems
53
73ฐW
72ฐW
71ฐW
70ฐW
69ฐW
2
A'
1.6
Cary Islands
1.1
76ฐ40'N
B
1.9
C
23
Ny-3
0.9
D
2.6
1.4
Saunders ุ
d
n
u
o
S
Thule
t
lo
y
B
Air Base
s12
14
Wolstenholme ุ
s14
5.1
76ฐ20'N
239, 240 Pu concentration
(bold numbers)
in surface sediment,
2.3
Thule 1412
2.3
Bq/kg dw
G
64
2 km
Hn๘
68
H
30
I
41
H'
76
J
12
23
45
X2
127
Figure 3ท61. Activity concentrations of 239,240Pu in
V2
642
surface sediments (0 to 3 cm layer) near Thule in 1997.
Y
617
239, 240
20
150
Pu concentration
Location names are shown in italics, concentrations in
(bold numbers)
25
520
in surface sediment,
bold. The point of impact was on the sea ice (180 m
22
26 s4 43
Bq/kg dw
water depth) at the location marked V2 (Eriksson,
P
43
2002).
240Pu : 239Pu atom ratio observed in global fallout, ap-
activity ratios for sediments (Table 3ท22) with those for
proximately 0.18, will not affect these results. Any in-
benthic biota (Table 3ท23), it is evident that some biota ap-
fluence of the higher 240Pu : 239Pu atom ratios in Sell-
pear to have a higher uptake of americium (Am) than Pu.
afield discharges, up to around 0.25, is even more un-
This appears to be the case for mollusks ญ bivalves as well
likely as the Sellafield-derived Pu concentration in the
as snails ญ and for some echinoderms, namely brittle stars
Thule area is accepted to be less than global fallout.
(Ophiuroidea) and starfish (Asteroidea), but not sea ur-
Average isotope ratios for 240Pu : 239Pu atom ratios,
chins (Echinoidea) or sea cucumber (Holothurioidea). This
and 238Pu : 239,240Pu and 241Am : 239,240Pu activity ratios
greater affinity for Am than Pu is not new. The Interna-
for sediment samples containing > 20 Bq 239,240Pu/kg, i.e.,
tional Atomic Energy Authority reported higher concen-
at least an order of magnitude above the fallout back-
tration ratios for Am than Pu in mollusks (IAEA, 1985).
ground, are given in Table 3ท22. The reference date is the
sampling date, 1997. By comparing the 241Am : 239,240Pu
3.8. Summary
Table 3ท22. Isotope ratios in Thule sediment samples with > 20 Bq
239,240
Since the first AMAP assessment, monitoring of various
Pu/kg in August 1997 (Dahlgaard et al., 2001).
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
man-made radionuclides in the Arctic environment has
Mean
SD, %
n
continued to a variable degree. In general, concentra-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
tions of radionuclides derived from global fallout, from
240Pu : 239Pu
atom ratio
0.045
15
30
238
the Chernobyl accident, and from earlier discharges
Pu : 239,240Pu
activity ratio
0.014
53
223
241Am : 239,240Pu
activity ratio
0.13
61
114
from European reprocessing plants are slowly decreas-
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
ing, as expected. This is especially evident for 90Sr, where
global fallout is still the dominant source. It is also the
Table 3ท23. 241Am : 239,240Pu activity ratios in sediment-dwelling
case for 137Cs, although the contribution from the Cher-
benthic biota, Thule 1997 (Dahlgaard et al., 2001).
nobyl accident and reprocessing discharges during the
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
Mean
SD, %
n
1970s and 1980s has added significantly to the fallout
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
level and still constitutes a source to the Arctic marine
Benthos
All
0.39
76
84
environment owing to remobilization and relocation
Mollusks
All
0.63
62
24
processes. More unexpectedly, evidence is mounting that
the Pu being remobilized from Irish Sea sediments is
Polychaetes
All
0.28
77
16
now responsible for a major proportion of the Pu con-
Crustaceans
All
0.22
36
5
tamination in the Norwegian and Barents Seas.
Echinoderms
Brittle stars
0.55
24
4
Data for the Faroe Islands and Iceland were not ade-
Starfish
0.41
34
11
quately dealt with in the first AMAP assessment.They
Sea urchins
0.17
8
5
have therefore been addressed in significantly greater de-
Sea cucumber
0.13
67
2
ญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญญ
tail in the present assessment.

---