Chapter 5
5.3. Results Terrestrial environment
Table 5.12.
Geometric means
and 95% confidence
interval of effective rates
of accumulation of heavy
metals in reindeer tissues
(ľg/g ww per year).
tionship are given in Table 5.12, and these can be con-
(b) Heavy metals
sidered as the effective rates of accumulation of metals
HMs occur in detectable concentrations in all samples,
in reindeer tissues. The lowest values normally occur in
except for Hg in muscle tissue. Calculated
the Taymir Peninsula, however, the values do not differ
male/female concentration ratios for Hg, Pb, and Cd
by more than a factor of 3 and the differences are there-
do not differ significantly from unity and are neither
fore of no great significance.
site nor tissue specific. Mean concentrations for HMs
were, therefore, calculated using data for both sexes.
Concentrations of Cd and Hg in muscle tissue sampled
The distribution of the three HMs between tissues is
in 2001 are very close to those found in 1994/1995 in
similar for both sexes, and approximates that found in
the Russian North (AMAP, 1998), while levels of Pb in
reindeer. Relative levels of contaminants in muscle,
muscle are an order of magnitude lower than those
liver and kidney are in the ratio of 1:11:5 for Pb and
reported earlier. In comparison with 1995 values,
1:26:160 for Cd (based on geometric means of ratios
liver/muscle concentration ratios calculated in 2001
for pooled samples).
are significantly higher for all HMs, and are similar to
those measured in other regions of the Arctic (AMAP,
Levels and trends
1998). Levels of all HMs are similar to, or slightly lower
than those determined in the Canadian Arctic in 1998-
(a) Organochlorines
2001 (CACAR, 2003). However, as concentrations of
The levels of OCs that were generally above detection
HM in Canadian reindeer are reported on a dry weight
limits in tissues of hare (HCB, p,p'-DDT, p,p'-DDE, PCB-
basis, direct comparison is not possible; on a dry wt
118 and PCB-153) did not follow any geographical
basis, absolute values are typically up to an order of
trend. Geometric means of OC concentrations range
magnitude greater.
from 0.06 ng/g ww (p,p'-DDT, p,p'-DDE) to 0.12 ng/g
ww (HCB). Concentrations of PCB-138, PCB-180, as
5.3.3. PTS in the Arctic hare
well as - and - HCH occur at similar levels at several
Tissues of Arctic hare (Lepidus timidus) were sampled
sites. In a few samples, some of the other OCs (PCBs,
at all sites, except for coastal Chukotka. The number
Mirex, and cyclodienes) were found in concentrations
of single samples of each tissue, collected at a given
close to the detection limit. Concentrations of all
site and used in the preparation of pooled samples,
detectable OCs in hare tissues are 2-4 times lower than
ranged from 4 to 10 (see Table 5.1). Equal numbers
those in reindeer, and are far below the limit values for
of male and female animals, all younger than these substances established in Russia.
3 years, were sampled at each site. The muscle, liver
and kidney of hare were analyzed for all PTS listed in
Section 1.2.4.
PTS concentration relationships with hare sex
and tissue type
(a) Organochlorines
Only a few OCs (HCB, p,p'-DDT, p,p'-DDE, PCB-118,
PCB-138 and PCB-153) were detectable at all sites and
in the majority of samples. No significant concentra-
Table 5.13. Concentrations (expressed as TEQ) of PCDD/Fs in hares
tion relationship to either sex or tissue type was identi-
in the Russian Arctic in 2001.
fied for these OCs in hare.
* ratio of PCDD/F concentration in pg WHO TEQ/g to that in pg/g
99

5.3. Results Terrestrial environment
Chapter 5
(b) PCDD/Fs
ˇ
Grazers (geese that graze mainly on aquatic and ter-
Concentrations of 2,3,7,8-substituted PCDD/Fs were
restrial vegetation): bean goose (Anser fabalis),
analyzed in pooled samples of hare tissues collected at
white-fronted goose (Anser albifrons), goldeneye
each site. Results are presented in Table 5.13 and
duck (Bucephala clangula) and ptarmigan (Lagopus
Figure 5.12. PCDDFs in hare tissues follow the same
sp.);
spatioal distribution pattern as for reindeer. All con-
ˇ
Omnivores (surface-feeding ducks with a varied
centrations are at levels below the (maximum) permis-
diet consisting mainly of aquatic vegetation): pintail
sible level for meat.
(Anas acuta), wigeon (Anas penelope), and teal (Anas
crecca);
(c) PAH
ˇ
Molluscivores (diving ducks feeding mainly on
Hare tissues were analyzed for the same PAH set as were
invertebrates): eider (Somateria mollissima), and
reindeer tissues. In contrast to OCs, most PAH concen-
long-tailed duck or oldsquaw (Clangula hyemalis);
tration levels in hare are either comparable with those
ˇ
Piscivores (diving ducks feeding mainly on fish):
in reindeer or are higher. Only phenanthrene concen-
scaup (Aythya marila), merganser (Mergus sp.), scoter
trations in hare muscle are, for most sites, found to be
(Melanitta sp).
several time lower than those in reindeer.
For most sites, equal numbers of male and female birds
(d) Brominated flame-retardants
of each groupwere harvested. Exceptions were: omni-
Samples of hare tissue were analysed for 2,2',4,4'-tetra-
vores (1 male pintail) and piscivores (1 female scoter)
bromodiphenyl, 2,2',4,4',5-pentabromodiphenyl, 2,2',
in eastern Taymir; grazers (1 female goose) at inland
4,4'-tetrabromodiphenyl ether and 2,2',4,4',5-pentabro-
Chukotka; and molluscivores (1 male eider) at coastal
modiphenyl ether. In all samples, concentrations of these
Chukotka.
substances were below the detection limit of 0.2 ng/g ww.
PTS concentration relationships with bird sex
(e) Heavy metals
No significant concentration dependence on sex was
HM concentrations measured in hare tissues are given
identified for any of the detectable OCs, for Hg and Cd
in Table 5.14. Concentrations are usually several times
in all bird groups, and for Pb in waterfowl. Pb concen-
lower than those found in reindeer, with the exception
trations in the muscle tissue of male browsers were con-
of Cd in kidney tissue, for which levels in hare and rein-
sistently about twice as high as those measured in
deer are comparable. Spatial distribution patterns
females at all six sites; the male/female ratios ranging
observed for all three HMs are similar to those for rein-
from 1.7 to 2.6, with a geometric mean of 2.1 for the six
deer. The lowest concentrations occur in Chukotka
values. Since the male/female ratios for Pb in browsers
and eastern Taymir. Concentrations of HMs in hare tis-
are not particularly high, the ratio is independent of
sues measured in this study are similar to those report-
site, and the same sample pattern for sexes was fol-
ed for hares in Finland in 1995 (AMAP, 1998).
lowed at all sites (50% male and 50% female), all geo-
Differences in levels between the data from Finland in
metric means, including those for Pb concentrations
1995 and Russia in 2001 are relatively small (within a
in browsers, were calculated using data for both sexes
factor of two), with the exception of Pb in muscle,
which is 4.4 times higher in the Russian North.
Levels and trends
5.3.4. PTS in birds
(a) Organochlorines
Waterfowl and terrestrial game birds harvested by
Concentrations of OCs in birds are shown in Tables
indigenous peoples in the Russian Arctic for the proj-
5.15a and 5.15b and Figures 5.14 and 5.15. The lowest
ect, were analyzed for all contaminants listed in Section
contamination levels are found in the muscle tissue of
1.2.4. Samples tissues of the following groups of birds
browsers. The only OC that was repeatedly detected in
were collected:
these birds was p,p'-DDE. Levels of PCB15 in browsers
Table 5.14.
Concentrations (mean and
standard deviation; ľg/g
ww, n=2) of HMs in tissues
of the Arctic hare (< 3 years
of age) in the Russian Arctic
in 2001.
a The range (in brackets)
is given where the standard
deviation is larger than 50%
of the mean.
100

Chapter 5
5.3. Results Terrestrial environment
Table 5.15a. Concentrations (geometric mean and range; ng/g ww) of OCs in the muscle of birds in the Russian Arctic in 2001.
a In at least one sample, more than half of the concentrations were below the detection limit. Concentrations below the detection limit were set to zero or to the detection limit
when determining lower and upper limits of concentration ranges.
b The range is given only for oldsquaw (n=2). Concentrations in eider were below the detection limit.
Figure 5.14. Geometric means and ranges of OC concentrations in molluscivores.
Figure 5.15. Geometric means and ranges of OC concentrations in omnivores.
PCB=PCB15, HCH=HCH, CHLOR=CHLOR, and DDT=DDT.
PCB=PCB15, HCH=HCH, CHLOR=CHLOR, and DDT=DDT.
Table 5.15b. Concentrations (geometric mean and range; ng/g ww) of OCs in the muscle of birds in the Russian Arctic in 2001.
a In at least one sample, more than half of the concentrations were below the detection limit. Concentrations below the detection limit were set to zero or to the detection limit
when determining lower and upper limits of concentration ranges.
b The range is given only for oldsquaw (n = 2). Concentrations in eider were below the detection limit.
c The geometric mean and range is given only for pintail (n = 2). Concentrations in mallard are below the detection limit.
101

5.3. Results Terrestrial environment
Chapter 5
were about 1 ng/g ww, about 2-3 times higher than
DDT at all sites. Other OCs occurred at concentra-
tions below or close to the detection limit in all sam-
ples. As for OCs in reindeer, there was no evident geo-
graphical trend for OCs in terrestrial birds.
OC levels in waterfowl are up to an order of magnitude
greater that those in browsers. Clear maximum concen-
trations of all OCs in molluscivores are found in eastern
Taymir, near Khatanga. In other bird groups, OC levels at
this site are comparable to those found at other sites.
Maximum concentrations range from 0.08 ng/g ww for
Mirex, and 1-3 ng/g ww for HCB, CHLOR, and HCH,
to about 10 ng/g ww for DDT and PCB15. Similar pat-
terns of OC concentrations are seen in other waterfowl at
all sites. Concentrations of p,p'-DDE found at all sites, are
significantly higher than those found in reindeer muscle,
while concentrations of other OCs are comparable with
those in reindeer. The lowest concentrations occur, as a
rule, at sites in Chukotka. Contamination levels in most
cases, decrease in the following order: molluscivores >
omnivores > piscivores > grazers. All OC concentrations
in birds are far below the (maximum) permissible levels
for bird meat established in Russia.
Table 5.16. Concentrations (expressed as TEQ) of PCDD/Fs in bird muscles
(b) PCDD/Fs
in the Russian Arctic in 2001.
Concentrations of 2,3,7,8-substituted PCDD/Fs were
* ratio of PCDD/F concentration in pg WHO TEQ/g to that in pg/g
analyzed in pooled samples of bird muscle tissue col-
lected from each site. Results are presented in Table
2001 are given in Table 5.17. Concentrations of 2-
5.16 and Figure 5.12. PCDD/F concentrations in birds
methylnaphthalene and fluorene in waterfowl are sev-
follow the same geographic distribution pattern as they
eral times higher, whilst concentrations of phenan-
do in reindeer, although the spatial differences are less
threne and pyrene are several times lower than those
pronounced in birds. All concentrations occur at levels
found in browsers. Naphthalene and fluoranthene
which are far below the maximum permissible levels
occur in the two bird groups in comparable concentra-
for these substances in meat.
tions. Those PAHs not included in Table 5.17, were
found in concentrations close to their detection limits,
(c) PAH
and then only in few samples of waterfowl. In contrast,
Bird tissues were analyzed for the same PAH set as rein-
chrysene, benzo[k]fluoranthene, benzo[a]pyrene,
deer. Geometric means and ranges of PAH concentra-
benzo[ghi]perylene, and biphenyl were detectable in
tions found in bird muscles in the Russian Arctic in
most of the browser muscle samples in concentrations
Table 5.17. Concentrations (geometric mean and range; ng/g ww) of PAHs a in muscle of birds in the Russian Arctic in 2001.
a NAP = Naphthalene, NAP2M = 2 Methylnaphthalene, FLE = Fluorene, PA= Phenanthrene, FLU = Fluoranthene, PYR = Pyrene
102

Chapter 5
5.3. Results Terrestrial environment
Table 5.18. Concentrations (geometric mean and range; ľg/g ww) of HMs in muscle of birds in the Russian Arctic in 2001.
a Data for one sample which was contaminated by lead shot was discarded. Pb concentrations in contaminated samples range from 0.5 to 11 ľg/g wet weight.
from 0.5 to 5 ng/g ww. Levels of all PAHs in browsers,
with the exception of fluorene, were about twice as
high as those found in reindeer. No noticeable geo-
graphic trend was observed for any of the PAHs. For
these substances, the variability between samples was
always comparable with the variability between sites.
(d) Brominated flame-retardants
Samples of bird tissues were analyzed for 2,2',4,4'-tetra-
bromodiphenyl, 2,2',4,4',5-pentabromodiphenyl, 2,2',
4,4'-tetrabromodiphenyl ether, and 2,2',4,4',5-penta-
bromodiphenyl ether. In all samples, these substances
were found at concentrations below the detection limit
of 0.2 ng/g ww.
Figure 5.16. Geometric means and ranges of Hg concentrations in muscle
(e) Heavy metals
of waterfowl in the Russian Arctic in 2001. The red line indicate the maximum
In most of the waterfowl samples, levels of Hg exceed-
permissible concentrations of Hg allowed by food safety standards. The number
ed the MPC for this metal, and only in browsers were
of values used for calculating means varied between 1 and 4.
Hg concentrations below the detection limit at all sites
(see Figure 5.16). Levels of Pb and Cd in terrestrial
used in the preparation of pooled samples ranged
birds and waterfowl are comparable (see Table 5.18).
from 3 to 9 (see Table 5.1). Most of the soils sampled
No pronounced geographic trend was observed for any
were of peat litter.
of the HMs in any bird group. Concentration differ-
ences occurring between any two sites, for a given HM
(a) Organochlorines
and bird group, were not statistically significant,
It is well known that POPs that are present in abiotic
despite inter-sample variability being quite low at
media (soil, water and air) can be taken up by living
almost all sites. The only notable exception to this, was
organisms, and subsequently transferred within food
the concentration of Cd measured in omnivores and
chains. In most cases, intake via the diet is the major
piscivores in eastern Taymir, which was found to be sig-
pathway for human POP exposure. In a steady state sys-
nificantly lower than at other sites. Concentrations of
tem, POPs are distributed throughout the environ-
Pb and Cd in birds were normally below maximum per-
ment according to the fugacity capacities of the various
missible levels for these metals (0.5 and 0.05 mg/kg,
environmental compartments. For POPs, fugacity
respectively), and only in few samples were concentra-
capacities are proportional to the lipid concentration
tions found to be higher (up to twice the MPC level).
in a given biological compartment (Sharpe and
Mackay, 2000; McLachlan, 1996). In this section, the
5.3.5. PTS transfer in the terrestrial food chain
partitioning of OCs between abiotic media and the tis-
For the estimation of soil-to-lichen and water-to-fish
sues of terrestrial organisms in the Russian Arctic is
transfer coefficients, pooled samples of soil were col-
considered, mainly using data on p,p'-DDT and p,p'-
lected at all 6 sites. The number of pooled samples
DDE levels. These two OCs, are convenient reference
ranged from 1 to 5, and the number of single samples
compounds, being detectable in most biotic and abiot-
103

5.3. Results Terrestrial environment
Chapter 5
ic samples collected. In addition, the ratio of p,p'-DDE
KOA is the octanol-air partition coefficient
to p,p'-DDT levels in soil is widely used to estimate the
(the ratio of volume concentrations when
age of the contamination and can, therefore, serve as
at equilibrium).
an indicator of the relative rate of p,p'-DDT metabolic
transformation in organisms. Most other OCs follow
Taking into account of the relatively low variability in
similar patterns of uptake and transport in food chains
lipid content in plant tissues, Equation 5.1 indicates
that have been studied.
that POP concentration ratios between two species
can be considered as being independent of site. If the
The soil-lichen-reindeer food chain is one of the most
kinetic limitations of uptake and depuration are
important in the Arctic. An example of levels of p,p'-
ignored, it could also be expected that the same ratio
DDT and p,p'-DDE in this chain are given for Khatanga
value would apply for all POPs. A comparison of the
in eastern Taymir in Figure 5.17. Geometric means of
concentrations of all OCs (excluding PCB) in lichens
soil-to-lichen transfer factors (concentration ratios) for
and mosses supports these assumptions (see Figure
both p,p'-DDT and p,p'-DDE are equal to 2.9 ranging
5.18). From Figure 5.18, it is evident that OC con-
from 0.7 (for p,p'-DDT on the Kola Peninsula) to 15
centrations in lichens show a clear relationship to
(for p,p'-DDE in western Taymir). The soil-to-lichen
those in mosses. The best correlation is seen for DDT
transfer factors for other OCs show a similar degree of
metabolites, probably because these contaminants
variability and similar geometric means (1.1 for PCB-
are present at higher concentrations and there is a
28 and PCB-153, 1.9 for HCB, and 2.5 for HCH iso-
lower detection error. The lichen/moss concentra-
mers). High levels of within site variability in the ratios
tion ratio for POPs, obtained using linear regression
are most probably explained by a relatively high vari-
analysis, is equal to 0.97. In other words, POPs con-
ability of p,p'-DDT and p,p'-DDE concentrations in soils.
centrations in lichen can be used as a direct estimate
Variability in OC concentrations at a particular site is
of the POP concentrations in mosses in the study
significantly less in lichens and mosses than soils.
area, and vice versa.
Lichens and mosses uptake pollutants primarily from
the air, which, in the absence of local sources of pollu-
tants, has relatively uniform contamination levels.
Even though lichens do not take up OCs directly from
the soil, soil-to-lichen transfer coefficients can still be cal-
culated, is based on the fact that, in the air of remote
areas, OC levels are proportional to those in soil, espe-
cially surface soil. Proportionality of OC levels between
soils and lichens can, therefore, also be expected.
Uptake from the air is the main route by which POPs
contaminate, not only mosses and lichens, but also other
plants; as POPs are highly lipophilic compounds and,
once adsorbed on the root surfaces, they tend not to be
translocated to the aboveground parts of plants
(McLachlan, 1996). Concentrations of OCs in air were
Figure 5.17.
not measured in the current study, therefore, only rela-
Absolute and relative levels of p,p' DDE and p,p' DDT for the soil
lichen reindeer food chain in the Khatanga area. Geometric means and ranges of DDE
tive air-to-plant transfer factors (based on interspecies
and DDT levels in soil and lichen are provided on a dry weight basis, while levels in
concentration ratios) could be calculated. Using such
reindeer muscle are on a wet weight basis; 95% confidence limits are shown for ratios.
ratios, once the OC concentration has been measured in
one particular plant species at a given site, the concen-
tration in any other species can be estimated using the
corresponding species/ species concentration ratio.
This approach relies on the similarity of the uptake
mechanism for POPs in different plant species. The
predominant pathway for uptake of POPs by plants is
dry gaseous deposition from the atmosphere (Paterson
et al., 1994; McLachlan, 1996). The POP concentration
in plants (CP) can be related to that in air (CA) by the
following equation:
CP = L ˇ KOA ˇ CA
(5.1)
where:
Figure 5.18. The relationship between OC concentrations in mosses
L is the lipid fraction in the plant tissue
and lichen (dry weight) in the northern Russia. 1 DDT and its metabolites,
(volume/volume);
2 HCH isomers, 3 Heptachlor, 4 Mirex, 5 HCB.
104

Chapter 5
5.3. Results Terrestrial environment
Berries and mushrooms are both components of the
feed and cow tissues takes place. Data obtained in the
human diet in the Russian North. Assessment of their
current study indicated that OC distribution between
contribution to human OC exposure in the Arctic
soil, lichen, and reindeer tissues was also close to a
requires data on contamination levels. As the available
steady state. The observed lack of dependence of OC
data set for berry and mushroom contamination in the
concentrations in reindeer tissues on animal age, and
Arctic is much more limited than that for lichen and
also the similarity of values calculated for OC concen-
mosses, the calculated lichen/berry and lichen/mush-
tration and lipid content ratios for reindeer/hare and
room concentration ratios are useful. Sufficiently reli-
reindeer/birds also support this conclusion (see
able data for calculation of the berry/lichen concentra-
Figure 5.19). All differences found between rein-
tion ratio were obtained only for p,p'-DDT and HCB
deer/hare and reindeer/bird concentration ratios for
(although data for the Kola Peninsula site where an
lipids and any of the OCs were small and within a fac-
anomalously high p,p'-DDT level was measured were
tor of two, and OC ratio ranges agreed closely with
excluded). The berry/lichen concentration ratio for
those for lipids.
p,p'-DDT was found to be practically the same as that for
Figure 5.19.
HCB. The geometric mean of the nine ratios calculated,
Concentration ratios
using data on both p,p'-DDT and HCB, is equal to 0.27.
(geometric mean and 95%
confidence limits) for OCs
Lichen-to-reindeer transfer factors (TF
and lipid content
LR) were also of
in reindeer/hare
the same value (0.3, based on wet wt. and dry wt. con-
and reindeer/birds,
centrations, respectively) for both p,p'-DDT and p,p'-
for all sample sites.
DDE. They ranged from 0.1 (for p,p'-DDE in western
Taymir) to 1.8 (for p,p'-DDE in Chukotka). In the
Canadian Arctic, in 1993, this factor was found to range
from one to values in the tens (based on lipid wt. and
dry wt. concentrations, respectively) for different OCs
The DDE/DDT ratio in soils immediately following
(CACAR, 1997). The geometric mean of lipid content
application of DDT pesticide is normally about 0.1 or
measured in reindeer muscle in the current study was
less (Harner et al., 1999). As a result of the microbio-
about 5%. Using this value to convert the transfer fac-
logical transformation of p,p'-DDT into p,p'-DDE, and
tors based on wet wt. concentrations to their lipid wt.
other metabolites, this ratio increases with time. In
equivalents yields values that are close to those reported
temperate zones, the DDE/DDT ratio in soils 30 years
for Canada. TFLRs for other OCs that were found at con-
after the last pesticide application ranges from 0.7 to 2
centrations above detection limits in the current study
(Harner et al., 1999). Ratios in the Russian Arctic are,
are of similar values to those determined for p,p'-DDT
as a rule, at the lower end of this range (see Table
and p,p'-DDE (0.2 for PCB-28, 0.8 for PCB-153, 0.5 for -
5.19a). This may indicate that fresh use of DDT is still
and -HCH, and 0.3 for HCB, based on wet wt. and dry
contributing to contamination of the Russian Arctic.
wt. concentrations, respectively).
However, lower ratio values can be also explained by
the slower rate of metabolic processes which occur in
All transfer factors obtained for p,p'-DDT and p,p'-DDE
Arctic soils.
in the soil-lichen-reindeer chain agree reasonably well
with the values expected on the basis of corresponding
Metabolic transformation of p,p'-DDT also takes place
concentration ratios for lipids. This result is reasonable
in higher organisms (WHO, 1982) and the DDE/DDT
since POP concentrations in the soil surface and in
ratio in their tissues can serve as an indicator of the rel-
plants will generally be at close to equilibrium with
ative rate of p,p'-DDT transformation in different
POP concentrations in the air, and will reflect the lipid
species. DDE/DDT ratios in terrestrial food chains in
contents of the soil and plants (McLachlan, 1996).
the Russian Arctic are provided in Figure 5.17 and
Relatively large deviations from equilibrium are
Tables 5.19a and 5.19b. No statistically significant dif-
observed only for concentrations of POPs which have a
ferences were observed between the ratio values, either
very high molecular weight, but even then, these devi-
in soils, lichens, or reindeer tissues, or, between the 6
ations are similar in both plants and soils. POP absorp-
sites (see Table 5.19a and 5.19b). This indicates that
tion and depuration by mammals is significantly slower
the p,p'-DDT transformation rate in reindeer tissues is
than that by plants. For example, the depuration half-
comparable with that in soils and lichens. The ratio
life for PCDD in the human body can be as long as sev-
values for terrestrial birds are, as a rule, somewhat
eral years (Masuda, 2001). However, given relatively sta-
higher than in reindeer (see Table 5.19a), probably
ble levels of air contamination, POP concentrations in
due to a faster rate of metabolic p,p'-DDT transforma-
mammal tissues and in vegetation used for food
tion in birds.
should, in general, be comparable, after correction for
lipid content (McLachlan, 1996). For example, cow
(b) Heavy metals
milk/fodder fugacity quotients measured in Germany
Some HMs, such as copper (Cu) and zinc (Zn) are
were, with a few exceptions, close to unity for HCB,
essential elements for both plants and animals and as
PCBs, and PCDD/Fs (McLachlan, 1996). This indi-
such, their levels in tissues are under homeostatic con-
cates that a steady state partitioning of OCs between
trol (Yagodin et al., 1989; Speidel and Agnew, 1982).
105

5.3. Results Terrestrial environment
Chapter 5
Table 5.19a.
DDE/DDT ratios (geometric
means and 95% confidence
interval) in terrestrial food
chains.
n. d. not detected
Non-essential elements, such as Hg and Cd, do not
As for OCs, the soil-lichen-reindeer food chain is one
appear to be well-regulated by living organisms. Thus,
of the most important pathways for human exposure
tissue concentrations of Hg and Cd proportional to
to HMs in the Arctic. Lichen is able to assimilate min-
environmental (or food) contamination levels can be
eral substances from any material to which it adheres.
expected. The use of transfer factors is based on this
However, the similar pattern of HM concentration
assumption. However, deviations from direct propor-
ratios in lichens, mosses and soils (Hg: Cd: Pb in the
tionality do occur, and quite often are more pro-
ratio of 2:3:95) indicates that mosses, as well as
nounced at higher levels of exposure, especially for Pb
lichen, take up most of their HM burden from the
(WHO 1989a, 1989b, 1991, 1992, 1995). A possible
air, apparently from windblown soil and dust.
explanation for this is that HMs in high concentrations
Geometric mean values of lichen/moss concentra-
are toxic for all organisms and their transfer through
tion ratios ranged from 0.5 to 0.6 for all three HMs.
cell membranes may be limited when tissue contami-
This ratio can be considered indicative of the greater
nation exceeds some critical level. In addition, Pb con-
ability of mosses to intercept particles. The geomet-
tent in tissue is probably under at least some degree of
ric mean for the lichen/moss concentration ratio,
homeostatic regulation, since it belongs to a group of
calculated using the pooled set of data for all three
so-called `conditionally essential elements' (Yagodin et
metals was equal to 0.56.
al., 1989; Speidel and Agnew, 1982). As a result, the HM
transfer coefficient for a particular link in a food chain
An example of HM concentration patterns for the soil-
depends not only upon environmental conditions, but
lichen-reindeer food chain is given in Figure 5.20.
also upon the HM concentration in abiotic media (or
food). It follows that use of HM transfer coefficient val-
ues obtained at low exposure levels, for tissue concen-
tration assessment at high exposure levels, can lead to
an overestimation of the concentration of that HM.
Another important condition for the applicability of the
transfer factor approach, is the absence of kinetic limi-
tations. The biological half-life of HMs in mammals is
difficult to estimate (WHO 1989a, 1989b, 1991, 1992,
1995). The biological half-life of Hg and Pb in blood and
the soft tissues of mammals normally ranges from sever-
al weeks to several months. However, significantly slower
Hg and Pb elimination rates have also been reported.
For example, the half-life of Hg in brain tissue and of Pb
in bone ranges from years to decades, and for a mammal
to eliminate 50% of absorbed Cd can take as long time
as 30 years. Based on these elimination rates, HM food-
Figure 5.20. HM distribution patterns in the soil lichen reindeer food chain
on the Kola Peninsula in 2001. Concentrations of HMs in soil and lichen and their
to-mammal transfer factors can be expected to show a
ranges are given on the dry weight basis, while those for reindeer muscle are provided
significant degree of dependence on mammal age.
on the wet weight basis.
Table 5.19b.
DDE/DDT ratios (geometric
means and 95% confidence
interval) in terrestrial food
chains.
n. d. not detected
106

Chapter 5
5.4. Freshwater environment
Soil-to-lichen transfer coefficients for Hg and Pb are
estimated by applying equation 5.3 to experimental
similar (the geometric mean equal to 0.6, ranging from
data. Because of the small number of age groups and
0.3 to 1.3 based on dry wt. concentrations). The geo-
narrow age intervals recorded, the accuracy of such
metric mean for Cd was about twice as high (1.3), but
estimates using data obtained in this study is low.
the difference is of low statistical significance. Soil-to-
However, it is clear that the elimination half-life for all
lichen transfer factors for HMs are several times lower
three HMs is at least several years, and could be in the
than those for OCs. This is consistent with the hypoth-
order of 10 years.
esis that soil dust interception is the main pathway of
HM uptake by lichens, whilst organic chemicals, in
Using a typical rate of lichen consumption by reindeer
addition to this mechanism, are also absorbed through
(i.e. 40 g dw/kg live weight per day; White et al., 1999)
dry gaseous deposition. Uptake of HMs from soil by
and HM concentrations from Table 5.7, the total annu-
mushrooms is more strongly affected by the chemical
al uptake of HMs from lichen by reindeer can be cal-
state of the metal. Geometric mean values of soil-to-
culated. Based on geometric means, this yields values
mushroom transfer factors (based on dry weight con-
of 0.51, 29, and 1.0 mg/kg live weight per year for
centrations) ranged from 0.12 and 0.25 for Pb and Hg,
uptake of Hg, Pb and Cd, respectively. Comparison of
respectively, to 1.5 for Cd. The geometric mean of the
these values with effective deposition rates from Table
soil-to-berry transfer factor for Pb (0.006) is two orders
5.12 indicate that less than 0.1% of Pb from consumed
of magnitude lower than that calculated for lichens.
lichens is transferred into the muscle, while the effec-
tiveness of Hg and Cd transfer to muscle is up to an
Significant differences between Pb and Hg and Cd also
order of magnitude greater, with values of 0.4% and
occur in the transfer of HMs from lichen to reindeer
1.0%, respectively.
tissues. Geometric means of lichen-to-reindeer transfer
factors for Hg and Cd are similar (0.4 and 0.5, respec-
5.4. Freshwater environment
tively) and an order of magnitude higher than that cal-
culated for Pb (0.03). All transfer factors as a function
5.4.1. PTS in fish
of herd, vary within an order of magnitude. This vari-
Fish were obtained from Lake Lovozero (Kola
ability can be partially explained by differences in the
Peninsula), the Pechora River, the Yenisey River (west-
mean age of animals sampled. The age dependence of
ern Taymir), the Khatanga River (eastern Taymir),
a pollutant concentration in an animal tissues can be
and the Kanchalan River (inland Chukotka). Fish age
described by the following simple model:
ranged from 5 to 14 years. The number of individual
samples of tissue collected at a given location for use
in the preparation of pooled samples ranged from 1
(5.2)
to 13 (see Table 5.1). The following fish species were
sampled:
Where:
ˇ
pike (Esox lucius)
C is the pollutant concentration in the animal,
ˇ
burbot (Lota lota)
ng/g ww;
ˇ
perch (Perca fluviatilis)
ri is the pollutant accumulation rate,
ˇ
ide (Leuciscus idus)
ng/g ww per year;
ˇ
whitefish (Coregonus lavaretus)
ke is the pollutant elimination rate constant,
ˇ
Arctic cisco (Coregonus autumnalis)
per year;
ˇ
broad whitefish (Cerogonus nasus)
t is the animal age, years.
ˇ
Arctic char (Salvelinus alpinus)
ˇ
inconnu (Stenodus leucichthys nelma)
Assuming that the intake rate is constant, this can be
ˇ
grayling (Thymallus thymallus)
expressed as:
Fish muscle and liver tissues were analyzed for all PTS
listed in Section 1.2.4. Results of analysis were divided
(5.3)
into groups according to sex (male or female), age
(either two or three classes), and tissue type (muscle or
When ket is small (i.e., elimination is slow), equation
liver). Age differences within groups ranged from 1 to
5.4 can be simplified and the concentration depend-
2 years. The difference between the mean ages of fish
ence on age becomes directly proportional:
in the oldest and youngest groups was always less than
a factor of two.
(5.4)
PTS concentration relationships to fish sex,
As was shown in section 5.3.2, HM levels in reindeer are
age and tissue type
directly proportional to age. This means that the elim-
ination rate is quite slow (ket is small) during at least the
(a) Organochlorines
first few years of life, and effective rates of HM accu-
Male/female concentration ratios of OCs that could
mulation in reindeer tissue (see Table 5.12) provide an
be reliably quantified (p,p'-DDT, p,p'-DDE, PCB-138,
estimate of ri. Values for the elimination rate (ke) can be
PCB-153, and HCB) were calculated using data from
107

5.4. Freshwater environment
Chapter 5
fish of the same age groups. No statistically significant
difference was found between the geometric mean of
the ratios and unity, for any OCs or any species.
Calculated age ratios (for middle/young and
old/young age groups) are, with a few exceptions,
slightly higher than unity and range from 0.8 to 2.8.
However, in all cases, the standard deviation was com-
parable to, or larger than the mean ratio value. Taking
into account the relatively small number of values
Hginliver,mg/kg
included in the average, this implies that the statistical
significance of any observed age dependency is very
low, and that data for all age groups can be combined
in the calculation of geometric mean OC concentra-
tions. The geometric means of OC liver/muscle con-
centration ratios are close to unity for salmon species
Figure 5.21. Hg concentration in fish liver as a function of the fish age
and range from 2 to 5 for pike, perch and ide. The
in northern Russia in 2001.
highest absolute and relative concentrations of all OCs
from all sites were found in burbot liver samples. In the
those in other species. The liver/muscle concentration
liver of both male and female burbot, fished from
ratio for burbot varied from site to site within two
Yenisey River, OC concentrations were as high as orders of magnitude. The geometric mean for detect-
580 ng/g for PCB15, 470 ng/g for DDT, and ed OCs in burbot ranges from 50 to 160. These obser-
39 ng/g for CHLOR. Levels of OCs in the liver of
vations are explained by the fact that the lipid content
other fish species were much lower. In contrast, OC
of burbot liver is significantly higher than that of the
concentrations in burbot muscle were very close to
other species studied, whilst lipid levels in the muscle
Table 5.20.
Geometric means and 95%
confidence interval of effec
tive rates of accumulation
of heavy metals in fish
tissues (ľg/g ww per year).
108