- Email: [email protected]
Distribution of 90Sr and 137Cs in Arctic soil profiles polluted by heavy metals
Journal of Environmental Radioactivity 81 (2005) 295e306 www.elsevier.com/locate/jenvrad
Distribution of 90Sr and 137Cs in Arctic soil profiles polluted by heavy metals M. Puhakainena,*, T. Heikkinena, E. Steinnesb, H. Thørringb, I. Outolac a
STUK-Radiation and Nuclear Safety Authority, P.O. Box 14, FIN-00881 Helsinki, Finland b Department of Chemistry, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway c Laboratory of Radiochemistry, Department of Chemistry, University of Helsinki, P.O. Box 55, FIN-00014 Helsinki, Finland Accepted 5 January 2005
Abstract Effects of industrial pollution on the behaviour of radionuclides in spruce forest ecosystems were studied along a gradient from of a copperenickel smelter in Monchegorsk, NW Russia. A reference site was situated in Lapland, Finland, 152 km west of Monchegorsk. Most of the total 137Cs activity in soil was in mineral (E and B) horizons, except at the reference site where the major part was still in the organic surface layer. Most of the total 90Sr activity still remaining in the soil profile was found in the surface layer, but the relative amount decreased with increasing level of industrial pollution. Pollutants from the smelter clearly affected the chemical speciation of radionuclides. Smaller amounts of exchangeable radionuclides were present in the organic surface layer at the most polluted sites. The decline of 137Cs with decreasing distance from the smelter correlated strongly with a similar depletion in exchangeable K and Mg. Total concentrations of 137Cs and 90Sr showed high correlations with exchangeable cations, particularly in the E and upper B horizon. A sudden change in
* Corresponding author. Tel.: C358 9 759 881; fax: C358 9 759 88433. E-mail address: marketta.puhakainen@stuk.fi (M. Puhakainen). 0265-931X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvrad.2005.01.006
296
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
behaviour of 137Cs in the lower B horizon may be associated with changes in clay mineralogy along the soil profile caused by weathering. Ó 2005 Elsevier Ltd. All rights reserved. Keywords:
90
Sr;
137
Cs; Soil; Pollution; Copper; Nickel; Exchangeable fractions; Nutrient cations
1. Introduction Industrial pollution from copperenickel smelters constitutes a major environmental impact on the spruce forest ecosystems in Kola Peninsula, northwest Russia. During the period 1987e1991 the atmospheric Ni and Cu emissions in Kola Peninsula were 3700 and 2600 tons per year respectively, most of which originated from the Monchegorsk smelter. In 1996 the corresponding amounts were 1300 and 700 tons per year. Studies around the Harjavalta smelter in south-western Finland (Derome and Nieminen, 1998; Derome and Lindroos, 1998) and the Nikel smelter in the Kola Peninsula (Reimann et al., 1998; Steinnes et al., 2000) show that the emissions from these smelters can have severe effects on the surrounding soils and the plants growing on these soils. In the surroundings of the Monchegorsk smelter the forest is severely affected by the smelter emissions, and in the nearest few kilometres to the south and west it is totally destroyed. Effects of heavy metals from these smelters are also likely to affect the fate of other airborne substances in these ecosystems. Effects of such smelter emissions on the behaviour of man-made radionuclides in the soil/plant system were investigated for the first time in a project carried out along a gradient from the Monchegorsk smelter (Suomela et al., 1999). Previous papers (Bunzl et al., 2001; Puhakainen et al., 2001; Riekkinen and Jaakkola, 2001) described some classical radioecological aspects related to this project. The aim of this presentation is to look more closely at the distribution of total and readily exchangeable 90Sr and 137Cs in different soil horizons in relation to parameters describing chemical processes in the same horizons.
2. Material and methods Soil samples were collected in 1997 from four spruce forest sites on podzolic soils located at different distances south of the Monchegorsk smelter and from a reference site in Lapland, Finland near the Russian border. The distance of the sampling sites from the smelter were 7 km (A), 16 km (B), 21 km (C), 28 km (D), and 152 km (REF). The exact locations (N. latitude; E. longitude) of the sites were: A (67 51#; 32 48#), B (67 46#; 32 48#), C (67 44#; 32 51#), D (67 40#; 32 47); REF (67 26#; 29 27#). The total inventories of Cu in the upper 20 cm of the soil increased from 0.93 at the reference site to 18.3 g mÿ2 at the most polluted site; the corresponding
297
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
levels for Ni were 3.5 to 33.5 g mÿ2 (Bunzl et al., 1999). A more detailed description of the sites is given elsewhere (Rahola et al., 1999; Thørring et al., 1999). At each site samples from different soil horizons (Of, Oh, E, B) were collected at five different plots. The total sampled soil depth was about 20 cm (range 12e30 cm) and the mineral horizons were divided vertically into two layers each (E1, E2 and B1, B2, respectively). The E horizon was divided into two layers so that the first 2 cm was collected separately, whereas the second layer was taken according to the natural thickness down to the B horizon (about 2e5 cm). The same procedure was also carried out with the B horizon, where the first layer, on the contrary was, of 5 cm thick (Rahola et al., 1999). In the vicinity of the polluted sites the thickness of the litter layer increased while that of the humus layer (OfCOh) decreased. The total thickness of the humus layer increased along the gradient from 1e3 cm at site A, 2e3 cm at B, 4e6 cm at C, 3e5 cm at site D and 4e11 cm at site REF. This is associated with the inhibited conversion of litter into humic substances by soil microorganisms at the most polluted sites (Bunzl et al., 2001). The contrasts in soil horizon development between the two extreme sites are illustrated in Fig. 1. Soil samples were oven dried at 40 C. Samples from the organic layer were milled and sieved to a particle size of !3 mm. Samples from the mineral soil were sieved to !2 mm particle size. The activity concentration of 137Cs was determined by gamma spectrometry using high purity Ge-detectors. The concentration of 40K was measured in National Research Centre for Environmental and Health (GSF) in Germany. Before 90Sr measurements were made, all samples were dry-ashed at 450 C. The ash from organic layer samples was dissolved in hot 3 M HCl and a carbonate fusion was performed for the insoluble part. Mineral layer samples were extracted twice with 18% HCl and after a radiochemical separation (Bryant et al., 1959) the activity concentration of 90Sr was determined using a Guardian (Wallac Oy) liquid scintillation spectrometer. The readily exchangeable fraction of 137Cs in the Of, Oh and E1 layers and that of 90Sr in the Of layer were determined after extraction with 1 M ammonium acetate, pH 7. The methods used for determination of chemical parameters (nitric acid-soluble Cu and Ni, parameters related to soil acidification, and ammonium nitrate exchangeable fractions of pollutant metals and major nutrient metals) and the results from these analyses are presented elsewhere (Thørring et al., 1999).
Litter
Of
Oh
E1
E2
B1
B2
REF A 5
0
-5
-10
-15
-20
-25
-30
Average plot depth cm Fig. 1. Mean thickness of horizons at the most polluted site A and the reference site REF (thickness of litter at the REF site is estimated).
298
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
3. Results and discussion 3.1. Relation to previous work at these study sites Before presenting the data from the present investigation some relevant results from previous publications relating to the study sites will be briefly reviewed: In Monchegorsk and at the reference site in Finland only 10% of 137Cs is Chernobylderived, and the rest presumably originates from global fallout (Bunzl et al., 2001). The contribution of Chernobyl-derived 90Sr to the total 90Sr inventory is negligible in this study area. The total sum of 137Cs and 90Sr activity in the upper 20e25 cm of the soil profile was in the range 1e2 kBq mÿ2 and 0.3e0.9 kBq mÿ2, respectively (Bunzl et al., 1999). At most plots at the reference site, 100% of the deposited 90Sr was still present above a depth of 25 cm. This amount decreased regularly with decreasing distance from the smelters. Thus at site A where the soil was most strongly polluted, only about 40% of the deposited 90Sr was still be found above a depth of 25 cm. A fraction of 90 Sr had transported into deeper soil layers (Bunzl et al., 2001). A few samples were also taken with a corer below the normal sampling layer. 90Sr was detected at site A below a depth of 40 cm. In the case of radiocaesium deposited on the soil surface almost all the whole amount could still be found above a depth of 25 cm at all sites. In the present paper the distributions of 137Cs and 90Sr between different soil horizons and their relation to some key soil chemistry parameters are discussed. 3.2. Distributions of total pollution gradient
137
Cs and
90
Sr between soil horizons along the
The amounts of 137Cs and 90Sr in the litter and their distribution between the different soil horizons along the pollution gradient are presented in Fig. 2. For both
Bq m-2
800 600
A
B
C
D
137Cs
REF
400 200 0
Litter
Of
Oh
E1
E2
B1
500
Bq m-2
400
A
B
C
D
B2 90Sr
REF
300 200 100 0
Litter
Of 137
Oh 90
ÿ2
E1
E2
B1
B2
Fig. 2. Distributions of Cs and Sr (Bq m ) in the litter and in the different soil horizons at different distances from the Monchegorsk smelter. The results are presented as the mean of values from five plots. The uncertainty bars represent the standard deviation of the mean.
299
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
nuclides there is a clear tendency of greater movement to deeper horizons with decreasing distance to the smelter. As for the total balance of the retained amount in the soil profile the fraction leached from the humus layer to the underlying mineral horizons is greater for 137Cs than for 90Sr. Within the humus layer, however, there is more 137Cs in the Of than in the Oh horizon whereas the opposite is the case for 90Sr. In the case of 137Cs the pool in the litter horizon increases strongly towards the smelter. This is associated with the diminished conversion of litter into humic material. High concentrations of heavy metals are toxic for soil micro-organisms, and heavy metal accumulated in the soil affect nutrient cycling throughout the ecosystem by inhibiting litter decomposition (Fritze et al., 1989). At increasing load of chemical pollutants as the smelter is approached, more radiocaesium accumulates in the litter layer and correspondingly less in the Of and Oh horizons because the reduced conversion of litter into humic material. This effect, however, is prominent for 137Cs and not for 90Sr, possibly because the thick litter layer near Monchegorsk due to more intense surface soil erosion also contains more mineral matter that would have strong binding sites for 137Cs. 3.3. Distributions of exchangeable 137Cs and correlations with chemical parameters
90
Sr in surface horizons and
Pollutants from the smelter have clearly affected the chemical speciation of radionuclides. In a previous paper (Puhakainen et al., 2001), the chemical fractionation of 137Cs in the humus (OfCOh) and uppermost mineral (E1) layers and of 90Sr in the uppermost humus layer (Of) was studied by sequential extraction according to the scheme of Tessier et al. (1979). In the present work the readily exchangeable fractions of 137Cs (1 M NH4OAc, pH 7) were studied in more detail by extracting samples from the Of, Oh, E1, and some E2 horizons, and results are presented in Fig. 3. The corresponding coefficient of variation (CV) in 2e3 parallel
% extracted
70
70 Exchangeable 90Sr
60
60
50
50
40
40
30
30
20
20
10
10
0
Exchangeable 137Cs
A
B
C
D
REF
0
Of Layer
Of
Oh
E1
E2
Layer
Fig. 3. Percentage distributions of readily exchangeable 137Cs and 90Sr (1 M NH4OAc) in different soil horizons. (Readily exchangeable fraction of 90Sr was determined only in the Of layer).
300
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
extraction of 137Cs varied in the O horizons within the range of 0e26% (9e44% for site A), whereas values up to 73% were observed in the E horizon samples where the 137 Cs activity was extremely low in many cases. Analysis of variance shows that the readily exchangeable fraction of 137Cs at REF site is significantly different from those at sites A-D. A significance difference is also evident between sites A and D, but no significant difference between the five sites occurs in the readily exchangeable fraction from the E1 layer (Puhakainen et al., 2001). Lower amounts of exchangeable radionuclides are present in the humus layer at the most polluted sites. The fractions of 137Cs occurring in the readily exchangeable form in the Of and Oh horizons were 57% and 46% at the reference site, and decreased as a function of pollution load being 17% and 24%, respectively at the most polluted site. There was a clear positive correlation in the Of horizon between the percentage of 137Cs in the readily exchangeable fraction and the distance from the smelter. There was also a moderate positive correlation in the Oh horizon, whereas in the E1 layer no such correlation was evident. For 90Sr, the exchangeable fraction was lower at the most polluted site (35%) compared to those at the other sites (43e46%). Correlations between the different parameters using all chemical and radionuclide data available, exchangeable as well as total concentrations, were calculated using the Spearman correlation test. When the amount of 137Cs in a given soil horizon was compared with those of heavy metals and nutrients, higher correlations were generally obtained when the exchangeable 137Cs was used instead of the total 137Cs concentration. Particularly high positive correlations (p!0:002) were observed with exchangeable K and Mg in the Of and Oh horizons, as illustrated in Fig. 4. On the other hand exchangeable 137Cs was strongly negatively correlated with total concentrations of Ni and Cu in the Of (p!0:001) and Oh (p!0:005) horizons. 3.4. Connections between total contents of radionuclides and chemical parameters Correlations between total amounts of radionuclides and corresponding data for some chemical parameters are shown for 137Cs (Bq/m2) in Table 1 and for 90Sr (Bq/m2) in Table 2. Generally high correlations with exchangeable major nutrient cations (Mg, K, Ca) are observed in most horizons for both radionuclides. A closer look at these associations with relation to the pollution gradient is possible in Figs. 5e7, where the amounts of total 137Cs and 90Sr are compared with those of exchangeable K and Ca in the combined O, E, and B horizons, respectively. In the O horizon (Fig. 5) the gradients of 137Cs and 90Sr are almost identical and very similar to that of exchangeable K, with strong depletion near the smelter. Exchangeable Ca in the O horizon is much less affected by the smelter emissions, which is also reflected by lack of significant correlations with 137Cs in the Of and Oh horizons (Table 1) and with 90 Sr in the Oh horizon (Table 2). In the E horizon (Fig. 6) the 137Cs concentration with distance from the smelter is still very similar to that of exchangeable K, with a maximum at site D (28 km distance) whereas that of 90Sr appears more similar to exchangeable Ca with a maximum at site C (21 km distance). In the combined B horizon the associations appear much weaker, although the correlation analysis
301
exhangeable Cs Bq m-2
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
1000 Mg
Of-layer
K 100
10 100
1000
10000
exhangeable Cs Bq m-2
1000
Oh-layer
Mg K
100
10 100
1000
10000
mg m-2 Fig. 4. Amount of readily exchangeable 137Cs (1M NH4OAc) as a function of exchangeable Mg and K in the Of and Oh horizons.
(Table 2) shows highly significant correlations between 90Sr and exchangeable Ca both in B1 and B2. Considering the associations with exchangeable cations there is one important difference between 137Cs and 90Sr: Whereas 90Sr shows a clear positive correlation all the way down the profile not only with K, Ca, and Mg but even with Mn, Fe, and Ni, the similar correlations with 137Cs only extend to the B1 horizon and totally
Table 1 Spearman correlation coefficients between
Sr (HNO3) K (exch.) Mg (exch.) Ca (exch.) Mn (exch.) Fe (exch.) Al (exch.) Ni (exch.) Cu (exch.)
137
Cs (Bq/m2) and selected chemical parameters (amounts/m2)
Of
Oh
E1
E2
B1
B2
e
0.71*** 0.75*** 0.63** e e 0.64** 0.66*** e e
e 0.70*** 0.56** 0.69*** e e e e e
e 0.82*** 0.70*** 0.79*** 0.85*** 0.47* e 0.62** e
0.46* 0.59** 0.62*** 0.67*** 0.65*** 0.62** 0.41* 0.77*** 0.62***
e e e e e e e e e
0.76*** 0.46* e 0.42* e e ÿ0.48* e
Only significant correlations are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
302
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
Table 2 Spearman correlation coefficients between
Sr (HNO3) K (exch.) Mg (exch.) Ca (exch.) Mn (exch.) Fe (exch.) Al (exch.) Ni (exch.) Cu (exch.)
90
Sr (Bq/m2) and selected chemical parameters (amounts/m2)
Of
Oh
E1
E2
B1
B2
0.77*** 0.76*** 0.75*** 0.68*** e e e e e
0.57* 0.49* 0.56* e e e 0.48* e e
0.43* 0.76*** 0.75*** 0.58** e e e e e
e 0.87*** 0.68*** 0.78*** 0.71*** 0.49* e 0.52* e
e 0.65*** 0.48* 0.71*** 0.57** 0.59** 0.48* 0.65*** 0.42*
0.42* 0.68*** 0.51** 0.69*** 0.50* 0.42* 0.52* 0.45* e
Only significant correlations are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
disappear in the B2 horizon. In an earlier paper it was suggested that the reason for the decline of the exchangeable 137Cs when approaching the smelter in the O layer is similar to that for the nutrient cations, i.e. leaching to lower soil layers due to cation exchangeable replacement by heavy metals from the smelter (Puhakainen et al., 2001). The high degree of association with the exchangeable pool of other cations suggests that significant fractions of 90Sr and 137Cs are present in exchangeable form even in mineral horizons. This is at variance with the general radioecological literature, where almost invariably 137Cs is stated to be very strongly bound to clay minerals in the soil. However, the fixation of radiocaesium varies considerably between different clay minerals (Maes et al., 1999), and extensive mineral weathering may cause great changes in soil clay mineralogy over time even under natural conditions (Egli et al., 2003). Thus it is conceivable that in the present soils the clay mineralogy in the E and upper B horizon has been sufficiently modified to cause less
O-horizon 100000
K ex (mg/m2)
Ca ex (mg/m2)
Cs-137 (Bq/m2)
Sr 90 (Bq/m2)
10000
1000
100 A
B
C
D
REF
sampling sites Fig. 5. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the O horizon (OfCOh) at various distances from the smelter.
303
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
E-horizon 10000
K ex (mg/m2)
Cs-137 (Bq/m2)
Ca ex (mg/m2)
Sr 90 (Bq/m2)
1000
100
10 A
B
D
C
REF
sampling sites Fig. 6. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the E horizon (E1CE2) at various distances from the smelter.
strong fixation of 137Cs, whereas the lower B horizon which has been less affected by weathering processes has retained a greater fixation potential for 137Cs. Correlations between total 137Cs, 90Sr, and 40K in the different horizons are shown in Table 3. The two fallout radionuclides are positively correlated in all horizons except B2, most strongly in E2 and B1. Assuming that 0.0118% of all natural K is 40 K, the variation in 40K results are similar to the variation to the total K in the sample. 40K and thereby total K correlated in the Oh and E1 horizon with 90Sr and
B-horizon 100000
K ex (mg/m2)
Cs-137 (Bq/m2)
Ca ex (mg/m2)
Sr 90 (Bq/m2)
10000
1000
100
10 A
B
C
D
REF
sampling sites Fig. 7. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the B horizon (B1CB2) at various distances from the smelter.
304
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
Table 3 Correlations between
137
Cs,
90
Sr, and
40
K in different soil horizons 137
90
0.60** e
e
0.54* 0.92***
0.49*
0.64** e
0.50*
0.82*** e
e
0.63*** e
e
e e
e
Cs
Of
Cs Sr 40 K 90
Oh
137
Cs Sr 40 K 90
E1
E2
137
Cs 90 Sr 40 K 137
Cs Sr 40 K 90
B1
137
Cs Sr 40 K 90
B2
Sr
137
137
Cs 90 Sr 40 K
Only significant correlation coefficients are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
only in the Oh horizon with 137Cs. Correlations between 40K and exchangeable K (not shown) are weak or non-existent in all horizons except Oh. This observation casts some doubt on the utility of soil/plant transfer factors calculated on the basis of measured 40K activity concentrations in the soil. As indicated in Fig. 2 the pollutants from the smelter, presumably Cu and Ni in particular, have strongly depleted the 90Sr content in the upper soil horizons. A similar depletion is evident for the exchangeable pool of the major nutrients K, Mg, and Ca (Thørring et al., 1999), a trend that has also been observed in other studies around Cu-Ni smelters (Fritze et al., 1989; Derome and Lindroos, 1998). The extent of this depletion at different distances from the smelter may be emphasized by calculated O-horizon/B-horizon ratios as illustrated in Fig. 8. In the case of 90Sr this ratio decreases from a value of 7.2 at the reference site to 0.7 at site A. Very similar depletion values are observed for the exchangeable cations. 4. Conclusions The increase in deposition of pollutants when approaching the smelter clearly affected the distribution of radionuclides in the soil profile. 90Sr was strongly depleted in the humus layer, whereas 137Cs was less depleted and to some extent accumulated in the litter layer. The chemical speciation of the radionuclides was affected as well. There was less 137Cs in readily exchangeable form in the humus layer (Of and Oh horizons) close to the smelter. For 90Sr the effect was smaller, but the
305
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
O-horizon/B-horizon ratio Sr-90
10
K
Mg
Ca
8 6 4 2 0 A
B
C
D
REF
Sampling site Fig. 8. Relative depletion of total 90Sr (Bq/m2) and exchangeable K, Mg, and Ca (mg/m2) with decreasing distance from the smelter illustrated as O/B concentration ratios.
exchangeable fraction was lower at the most polluted site than at the other sites. Strong correlations were evident between exchangeable 137Cs and the exchangeable pools of K and Mg with distance from the smelter. Total concentrations of both radionuclides in the soil horizons were also strongly correlated to exchangeable cations, particularly K and Mg (O, E, and B1) and Ca (E and B1). For 90Sr these correlations were also evident in the B2 horizon, but for 137 Cs not. Altogether this indicates that a significant fraction of 90Sr and even of 137 Cs must be present in exchangeable form in these horizons. The different behaviour of 137Cs in the B2 horizon may be associated with differences in clay mineralogy between horizons exposed too different degrees of weathering. The ratio of 90Sr concentrations in the Of relative to the B2 horizon decreased by a factor of 10 from the reference site to the site nearest the smelter. Similar surface depletions were evident for the exchangeable pools of K, Mg, and Ca. Acknowledgements This work was carried out as part of the project ‘‘Effects of industrial pollution on the distribution dynamics of radionuclides in boreal understorey ecosystems’’ in the EU-framework ‘‘Nuclear fission safety’’, Contract No. F14P-CT96-0039. References Bryant, F.J., Morgan, A., Spicer, G.S., 1959. The determination of radiostrontium in biological materials. United Kingdom Atomic Energy Authority Report AERE-R 3030, Harwell, UK. Bunzl, K., Albers, B.P., Shimmack, W., Rissanen, K., Suomela, M., Puhakainen, M., Rahola, T., Steinnes, E., 1999. Soil to plant uptake of fallout 137Cs by plants from boreal areas polluted by industrial emission from smelters. Science of the Total Environment 234, 213e221. Bunzl, K., Puhakainen, M., Riekkinen, I., Karhu, P., Schimmack, W., Heikkinen, T., Jaakkola, T., Pavlov, V., Rahola, T., Rissanen, K., Suomela, M., Tillander, M., A¨yra¨s, M., 2001. Fallout 137Cs, 90Sr
306
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
and 239C240Pu in soils polluted by heavy metals: Vertical distribution, residence half-times, and external gamma-dose rates. Journal of Radioanalytical and Nuclear Chemistry 247, 15e24. Derome, J., Lindroos, A-J., 1998. Effects of heavy metal contamination on macronutrient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter, SW Finland. Environmental Pollution 99, 225e232. Derome, J., Nieminen, T., 1998. Metal and macronutrient fluxes in heavy-metal pollution Scots pine ecosystems in SW Finland. Environmental Pollution 103, 219e228. Egli, M., Mirabella, A., Fitze, P., 2003. Formation rates of smectites derived from two Holocene chronosequences in the Swiss Alps. Geoderma 117, 81e98. Fritze, H., Niini, S., Mikkola, A., Ma¨kinen, A., 1989. Soil microbial effects of a Cu-Ni smelter in southwestern Finland. Biology and Fertility of Soils 8, 87e94. Maes, E., Vielvoye, L., Stone, W., Delvaux, B., 1999. Fixation of radiocaesium traces in a weathering sequence micaevermiculiteehydroxy interlayered vermiculite. European Journal of Soil Science 50, 107e115. Puhakainen, M., Riekkinen, I., Heikkinen, T., Jaakkola, T., Steinnes, E., Rissanen, K., Suomela, M., Thørring, H., 2001. Effect of chemical pollution on forms of 137Cs, 90Sr and 239,240Pu in Arctic soil studied by sequential extraction. Journal of Environmental Radioactivity 52, 17e29. Rahola, T., Albers, B., Bergman, R., Bunzl, K., Jaakkola, T., Nikonov, V., Pavlov, V., Rissanen, K., Schimmack, W., Steinnes, E., Suomela, M., Tillander, M., A¨yra¨s, M., 1999. General characterisation of study area and definition of experimental protocols. Report STUK-A166, Radiation and Nuclear Safety Authority (STUK), Helsinki. Reimann, C., A¨yra¨s, M., Chekushin, V., Bogatyrev, I., Boyd, R., de Caritat, P., Dutter, R., Finne, T.E., Halleraker, J.H., Jaeger, O.E., Kashulina, G., Lehto, O., Niskavaara, H., Pavlov, V., Ra¨isa¨nen, M.L., Strand, T., Volden, T., 1998. Environmental Geochemical Atlas of the Central Barents Region. Riekkinen, I., Jaakkola, T., 2001. Effect of industrial pollution on soil-to-plant transfer of plutonium in a Boreal forest. The Science of the Total Environment 278, 161e170. Suomela, M., Bergman, R., Bunzl, K., Jaakkola, T., Rahola, T., Steinnes, E., 1999. Effect of Industrial Pollution on the Distribution Dynamics of Radionuclides in Boreal Understorey Ecosystems (EPORA). Report STUK-A168, Radiation and Nuclear Safety Authority (STUK), Helsinki. Steinnes, E., Lukina, N., Nikonov, V., Aamlid, D., Røyset, O., 2000. Gradient study of 34 elements in the vicinity of a copperenickel smelter in the Kola Peninsula. Environmental Monitoring and Assessment 60, 71e88. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry 51, 844e851. Thørring, H., Steinnes, E., Nikonov, V., Rahola, T., Rissanen, K., 1999. A summary of chemical data from the EPORA project. Report STUK-A167, Radiation and Nuclear Safety Authority (STUK), Helsinki.
Distribution of 90Sr and 137Cs in Arctic soil profiles polluted by heavy metals M. Puhakainena,*, T. Heikkinena, E. Steinnesb, H. Thørringb, I. Outolac a
STUK-Radiation and Nuclear Safety Authority, P.O. Box 14, FIN-00881 Helsinki, Finland b Department of Chemistry, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway c Laboratory of Radiochemistry, Department of Chemistry, University of Helsinki, P.O. Box 55, FIN-00014 Helsinki, Finland Accepted 5 January 2005
Abstract Effects of industrial pollution on the behaviour of radionuclides in spruce forest ecosystems were studied along a gradient from of a copperenickel smelter in Monchegorsk, NW Russia. A reference site was situated in Lapland, Finland, 152 km west of Monchegorsk. Most of the total 137Cs activity in soil was in mineral (E and B) horizons, except at the reference site where the major part was still in the organic surface layer. Most of the total 90Sr activity still remaining in the soil profile was found in the surface layer, but the relative amount decreased with increasing level of industrial pollution. Pollutants from the smelter clearly affected the chemical speciation of radionuclides. Smaller amounts of exchangeable radionuclides were present in the organic surface layer at the most polluted sites. The decline of 137Cs with decreasing distance from the smelter correlated strongly with a similar depletion in exchangeable K and Mg. Total concentrations of 137Cs and 90Sr showed high correlations with exchangeable cations, particularly in the E and upper B horizon. A sudden change in
* Corresponding author. Tel.: C358 9 759 881; fax: C358 9 759 88433. E-mail address: marketta.puhakainen@stuk.fi (M. Puhakainen). 0265-931X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvrad.2005.01.006
296
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
behaviour of 137Cs in the lower B horizon may be associated with changes in clay mineralogy along the soil profile caused by weathering. Ó 2005 Elsevier Ltd. All rights reserved. Keywords:
90
Sr;
137
Cs; Soil; Pollution; Copper; Nickel; Exchangeable fractions; Nutrient cations
1. Introduction Industrial pollution from copperenickel smelters constitutes a major environmental impact on the spruce forest ecosystems in Kola Peninsula, northwest Russia. During the period 1987e1991 the atmospheric Ni and Cu emissions in Kola Peninsula were 3700 and 2600 tons per year respectively, most of which originated from the Monchegorsk smelter. In 1996 the corresponding amounts were 1300 and 700 tons per year. Studies around the Harjavalta smelter in south-western Finland (Derome and Nieminen, 1998; Derome and Lindroos, 1998) and the Nikel smelter in the Kola Peninsula (Reimann et al., 1998; Steinnes et al., 2000) show that the emissions from these smelters can have severe effects on the surrounding soils and the plants growing on these soils. In the surroundings of the Monchegorsk smelter the forest is severely affected by the smelter emissions, and in the nearest few kilometres to the south and west it is totally destroyed. Effects of heavy metals from these smelters are also likely to affect the fate of other airborne substances in these ecosystems. Effects of such smelter emissions on the behaviour of man-made radionuclides in the soil/plant system were investigated for the first time in a project carried out along a gradient from the Monchegorsk smelter (Suomela et al., 1999). Previous papers (Bunzl et al., 2001; Puhakainen et al., 2001; Riekkinen and Jaakkola, 2001) described some classical radioecological aspects related to this project. The aim of this presentation is to look more closely at the distribution of total and readily exchangeable 90Sr and 137Cs in different soil horizons in relation to parameters describing chemical processes in the same horizons.
2. Material and methods Soil samples were collected in 1997 from four spruce forest sites on podzolic soils located at different distances south of the Monchegorsk smelter and from a reference site in Lapland, Finland near the Russian border. The distance of the sampling sites from the smelter were 7 km (A), 16 km (B), 21 km (C), 28 km (D), and 152 km (REF). The exact locations (N. latitude; E. longitude) of the sites were: A (67 51#; 32 48#), B (67 46#; 32 48#), C (67 44#; 32 51#), D (67 40#; 32 47); REF (67 26#; 29 27#). The total inventories of Cu in the upper 20 cm of the soil increased from 0.93 at the reference site to 18.3 g mÿ2 at the most polluted site; the corresponding
297
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
levels for Ni were 3.5 to 33.5 g mÿ2 (Bunzl et al., 1999). A more detailed description of the sites is given elsewhere (Rahola et al., 1999; Thørring et al., 1999). At each site samples from different soil horizons (Of, Oh, E, B) were collected at five different plots. The total sampled soil depth was about 20 cm (range 12e30 cm) and the mineral horizons were divided vertically into two layers each (E1, E2 and B1, B2, respectively). The E horizon was divided into two layers so that the first 2 cm was collected separately, whereas the second layer was taken according to the natural thickness down to the B horizon (about 2e5 cm). The same procedure was also carried out with the B horizon, where the first layer, on the contrary was, of 5 cm thick (Rahola et al., 1999). In the vicinity of the polluted sites the thickness of the litter layer increased while that of the humus layer (OfCOh) decreased. The total thickness of the humus layer increased along the gradient from 1e3 cm at site A, 2e3 cm at B, 4e6 cm at C, 3e5 cm at site D and 4e11 cm at site REF. This is associated with the inhibited conversion of litter into humic substances by soil microorganisms at the most polluted sites (Bunzl et al., 2001). The contrasts in soil horizon development between the two extreme sites are illustrated in Fig. 1. Soil samples were oven dried at 40 C. Samples from the organic layer were milled and sieved to a particle size of !3 mm. Samples from the mineral soil were sieved to !2 mm particle size. The activity concentration of 137Cs was determined by gamma spectrometry using high purity Ge-detectors. The concentration of 40K was measured in National Research Centre for Environmental and Health (GSF) in Germany. Before 90Sr measurements were made, all samples were dry-ashed at 450 C. The ash from organic layer samples was dissolved in hot 3 M HCl and a carbonate fusion was performed for the insoluble part. Mineral layer samples were extracted twice with 18% HCl and after a radiochemical separation (Bryant et al., 1959) the activity concentration of 90Sr was determined using a Guardian (Wallac Oy) liquid scintillation spectrometer. The readily exchangeable fraction of 137Cs in the Of, Oh and E1 layers and that of 90Sr in the Of layer were determined after extraction with 1 M ammonium acetate, pH 7. The methods used for determination of chemical parameters (nitric acid-soluble Cu and Ni, parameters related to soil acidification, and ammonium nitrate exchangeable fractions of pollutant metals and major nutrient metals) and the results from these analyses are presented elsewhere (Thørring et al., 1999).
Litter
Of
Oh
E1
E2
B1
B2
REF A 5
0
-5
-10
-15
-20
-25
-30
Average plot depth cm Fig. 1. Mean thickness of horizons at the most polluted site A and the reference site REF (thickness of litter at the REF site is estimated).
298
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
3. Results and discussion 3.1. Relation to previous work at these study sites Before presenting the data from the present investigation some relevant results from previous publications relating to the study sites will be briefly reviewed: In Monchegorsk and at the reference site in Finland only 10% of 137Cs is Chernobylderived, and the rest presumably originates from global fallout (Bunzl et al., 2001). The contribution of Chernobyl-derived 90Sr to the total 90Sr inventory is negligible in this study area. The total sum of 137Cs and 90Sr activity in the upper 20e25 cm of the soil profile was in the range 1e2 kBq mÿ2 and 0.3e0.9 kBq mÿ2, respectively (Bunzl et al., 1999). At most plots at the reference site, 100% of the deposited 90Sr was still present above a depth of 25 cm. This amount decreased regularly with decreasing distance from the smelters. Thus at site A where the soil was most strongly polluted, only about 40% of the deposited 90Sr was still be found above a depth of 25 cm. A fraction of 90 Sr had transported into deeper soil layers (Bunzl et al., 2001). A few samples were also taken with a corer below the normal sampling layer. 90Sr was detected at site A below a depth of 40 cm. In the case of radiocaesium deposited on the soil surface almost all the whole amount could still be found above a depth of 25 cm at all sites. In the present paper the distributions of 137Cs and 90Sr between different soil horizons and their relation to some key soil chemistry parameters are discussed. 3.2. Distributions of total pollution gradient
137
Cs and
90
Sr between soil horizons along the
The amounts of 137Cs and 90Sr in the litter and their distribution between the different soil horizons along the pollution gradient are presented in Fig. 2. For both
Bq m-2
800 600
A
B
C
D
137Cs
REF
400 200 0
Litter
Of
Oh
E1
E2
B1
500
Bq m-2
400
A
B
C
D
B2 90Sr
REF
300 200 100 0
Litter
Of 137
Oh 90
ÿ2
E1
E2
B1
B2
Fig. 2. Distributions of Cs and Sr (Bq m ) in the litter and in the different soil horizons at different distances from the Monchegorsk smelter. The results are presented as the mean of values from five plots. The uncertainty bars represent the standard deviation of the mean.
299
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
nuclides there is a clear tendency of greater movement to deeper horizons with decreasing distance to the smelter. As for the total balance of the retained amount in the soil profile the fraction leached from the humus layer to the underlying mineral horizons is greater for 137Cs than for 90Sr. Within the humus layer, however, there is more 137Cs in the Of than in the Oh horizon whereas the opposite is the case for 90Sr. In the case of 137Cs the pool in the litter horizon increases strongly towards the smelter. This is associated with the diminished conversion of litter into humic material. High concentrations of heavy metals are toxic for soil micro-organisms, and heavy metal accumulated in the soil affect nutrient cycling throughout the ecosystem by inhibiting litter decomposition (Fritze et al., 1989). At increasing load of chemical pollutants as the smelter is approached, more radiocaesium accumulates in the litter layer and correspondingly less in the Of and Oh horizons because the reduced conversion of litter into humic material. This effect, however, is prominent for 137Cs and not for 90Sr, possibly because the thick litter layer near Monchegorsk due to more intense surface soil erosion also contains more mineral matter that would have strong binding sites for 137Cs. 3.3. Distributions of exchangeable 137Cs and correlations with chemical parameters
90
Sr in surface horizons and
Pollutants from the smelter have clearly affected the chemical speciation of radionuclides. In a previous paper (Puhakainen et al., 2001), the chemical fractionation of 137Cs in the humus (OfCOh) and uppermost mineral (E1) layers and of 90Sr in the uppermost humus layer (Of) was studied by sequential extraction according to the scheme of Tessier et al. (1979). In the present work the readily exchangeable fractions of 137Cs (1 M NH4OAc, pH 7) were studied in more detail by extracting samples from the Of, Oh, E1, and some E2 horizons, and results are presented in Fig. 3. The corresponding coefficient of variation (CV) in 2e3 parallel
% extracted
70
70 Exchangeable 90Sr
60
60
50
50
40
40
30
30
20
20
10
10
0
Exchangeable 137Cs
A
B
C
D
REF
0
Of Layer
Of
Oh
E1
E2
Layer
Fig. 3. Percentage distributions of readily exchangeable 137Cs and 90Sr (1 M NH4OAc) in different soil horizons. (Readily exchangeable fraction of 90Sr was determined only in the Of layer).
300
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
extraction of 137Cs varied in the O horizons within the range of 0e26% (9e44% for site A), whereas values up to 73% were observed in the E horizon samples where the 137 Cs activity was extremely low in many cases. Analysis of variance shows that the readily exchangeable fraction of 137Cs at REF site is significantly different from those at sites A-D. A significance difference is also evident between sites A and D, but no significant difference between the five sites occurs in the readily exchangeable fraction from the E1 layer (Puhakainen et al., 2001). Lower amounts of exchangeable radionuclides are present in the humus layer at the most polluted sites. The fractions of 137Cs occurring in the readily exchangeable form in the Of and Oh horizons were 57% and 46% at the reference site, and decreased as a function of pollution load being 17% and 24%, respectively at the most polluted site. There was a clear positive correlation in the Of horizon between the percentage of 137Cs in the readily exchangeable fraction and the distance from the smelter. There was also a moderate positive correlation in the Oh horizon, whereas in the E1 layer no such correlation was evident. For 90Sr, the exchangeable fraction was lower at the most polluted site (35%) compared to those at the other sites (43e46%). Correlations between the different parameters using all chemical and radionuclide data available, exchangeable as well as total concentrations, were calculated using the Spearman correlation test. When the amount of 137Cs in a given soil horizon was compared with those of heavy metals and nutrients, higher correlations were generally obtained when the exchangeable 137Cs was used instead of the total 137Cs concentration. Particularly high positive correlations (p!0:002) were observed with exchangeable K and Mg in the Of and Oh horizons, as illustrated in Fig. 4. On the other hand exchangeable 137Cs was strongly negatively correlated with total concentrations of Ni and Cu in the Of (p!0:001) and Oh (p!0:005) horizons. 3.4. Connections between total contents of radionuclides and chemical parameters Correlations between total amounts of radionuclides and corresponding data for some chemical parameters are shown for 137Cs (Bq/m2) in Table 1 and for 90Sr (Bq/m2) in Table 2. Generally high correlations with exchangeable major nutrient cations (Mg, K, Ca) are observed in most horizons for both radionuclides. A closer look at these associations with relation to the pollution gradient is possible in Figs. 5e7, where the amounts of total 137Cs and 90Sr are compared with those of exchangeable K and Ca in the combined O, E, and B horizons, respectively. In the O horizon (Fig. 5) the gradients of 137Cs and 90Sr are almost identical and very similar to that of exchangeable K, with strong depletion near the smelter. Exchangeable Ca in the O horizon is much less affected by the smelter emissions, which is also reflected by lack of significant correlations with 137Cs in the Of and Oh horizons (Table 1) and with 90 Sr in the Oh horizon (Table 2). In the E horizon (Fig. 6) the 137Cs concentration with distance from the smelter is still very similar to that of exchangeable K, with a maximum at site D (28 km distance) whereas that of 90Sr appears more similar to exchangeable Ca with a maximum at site C (21 km distance). In the combined B horizon the associations appear much weaker, although the correlation analysis
301
exhangeable Cs Bq m-2
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
1000 Mg
Of-layer
K 100
10 100
1000
10000
exhangeable Cs Bq m-2
1000
Oh-layer
Mg K
100
10 100
1000
10000
mg m-2 Fig. 4. Amount of readily exchangeable 137Cs (1M NH4OAc) as a function of exchangeable Mg and K in the Of and Oh horizons.
(Table 2) shows highly significant correlations between 90Sr and exchangeable Ca both in B1 and B2. Considering the associations with exchangeable cations there is one important difference between 137Cs and 90Sr: Whereas 90Sr shows a clear positive correlation all the way down the profile not only with K, Ca, and Mg but even with Mn, Fe, and Ni, the similar correlations with 137Cs only extend to the B1 horizon and totally
Table 1 Spearman correlation coefficients between
Sr (HNO3) K (exch.) Mg (exch.) Ca (exch.) Mn (exch.) Fe (exch.) Al (exch.) Ni (exch.) Cu (exch.)
137
Cs (Bq/m2) and selected chemical parameters (amounts/m2)
Of
Oh
E1
E2
B1
B2
e
0.71*** 0.75*** 0.63** e e 0.64** 0.66*** e e
e 0.70*** 0.56** 0.69*** e e e e e
e 0.82*** 0.70*** 0.79*** 0.85*** 0.47* e 0.62** e
0.46* 0.59** 0.62*** 0.67*** 0.65*** 0.62** 0.41* 0.77*** 0.62***
e e e e e e e e e
0.76*** 0.46* e 0.42* e e ÿ0.48* e
Only significant correlations are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
302
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
Table 2 Spearman correlation coefficients between
Sr (HNO3) K (exch.) Mg (exch.) Ca (exch.) Mn (exch.) Fe (exch.) Al (exch.) Ni (exch.) Cu (exch.)
90
Sr (Bq/m2) and selected chemical parameters (amounts/m2)
Of
Oh
E1
E2
B1
B2
0.77*** 0.76*** 0.75*** 0.68*** e e e e e
0.57* 0.49* 0.56* e e e 0.48* e e
0.43* 0.76*** 0.75*** 0.58** e e e e e
e 0.87*** 0.68*** 0.78*** 0.71*** 0.49* e 0.52* e
e 0.65*** 0.48* 0.71*** 0.57** 0.59** 0.48* 0.65*** 0.42*
0.42* 0.68*** 0.51** 0.69*** 0.50* 0.42* 0.52* 0.45* e
Only significant correlations are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
disappear in the B2 horizon. In an earlier paper it was suggested that the reason for the decline of the exchangeable 137Cs when approaching the smelter in the O layer is similar to that for the nutrient cations, i.e. leaching to lower soil layers due to cation exchangeable replacement by heavy metals from the smelter (Puhakainen et al., 2001). The high degree of association with the exchangeable pool of other cations suggests that significant fractions of 90Sr and 137Cs are present in exchangeable form even in mineral horizons. This is at variance with the general radioecological literature, where almost invariably 137Cs is stated to be very strongly bound to clay minerals in the soil. However, the fixation of radiocaesium varies considerably between different clay minerals (Maes et al., 1999), and extensive mineral weathering may cause great changes in soil clay mineralogy over time even under natural conditions (Egli et al., 2003). Thus it is conceivable that in the present soils the clay mineralogy in the E and upper B horizon has been sufficiently modified to cause less
O-horizon 100000
K ex (mg/m2)
Ca ex (mg/m2)
Cs-137 (Bq/m2)
Sr 90 (Bq/m2)
10000
1000
100 A
B
C
D
REF
sampling sites Fig. 5. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the O horizon (OfCOh) at various distances from the smelter.
303
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
E-horizon 10000
K ex (mg/m2)
Cs-137 (Bq/m2)
Ca ex (mg/m2)
Sr 90 (Bq/m2)
1000
100
10 A
B
D
C
REF
sampling sites Fig. 6. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the E horizon (E1CE2) at various distances from the smelter.
strong fixation of 137Cs, whereas the lower B horizon which has been less affected by weathering processes has retained a greater fixation potential for 137Cs. Correlations between total 137Cs, 90Sr, and 40K in the different horizons are shown in Table 3. The two fallout radionuclides are positively correlated in all horizons except B2, most strongly in E2 and B1. Assuming that 0.0118% of all natural K is 40 K, the variation in 40K results are similar to the variation to the total K in the sample. 40K and thereby total K correlated in the Oh and E1 horizon with 90Sr and
B-horizon 100000
K ex (mg/m2)
Cs-137 (Bq/m2)
Ca ex (mg/m2)
Sr 90 (Bq/m2)
10000
1000
100
10 A
B
C
D
REF
sampling sites Fig. 7. Mean values of 137Cs and 90Sr (Bq/m2) and exchangeable K and Ca (mg/m2) in the B horizon (B1CB2) at various distances from the smelter.
304
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
Table 3 Correlations between
137
Cs,
90
Sr, and
40
K in different soil horizons 137
90
0.60** e
e
0.54* 0.92***
0.49*
0.64** e
0.50*
0.82*** e
e
0.63*** e
e
e e
e
Cs
Of
Cs Sr 40 K 90
Oh
137
Cs Sr 40 K 90
E1
E2
137
Cs 90 Sr 40 K 137
Cs Sr 40 K 90
B1
137
Cs Sr 40 K 90
B2
Sr
137
137
Cs 90 Sr 40 K
Only significant correlation coefficients are shown. Significance levels: *p!0:05, **p!0:01, ***p!0:001.
only in the Oh horizon with 137Cs. Correlations between 40K and exchangeable K (not shown) are weak or non-existent in all horizons except Oh. This observation casts some doubt on the utility of soil/plant transfer factors calculated on the basis of measured 40K activity concentrations in the soil. As indicated in Fig. 2 the pollutants from the smelter, presumably Cu and Ni in particular, have strongly depleted the 90Sr content in the upper soil horizons. A similar depletion is evident for the exchangeable pool of the major nutrients K, Mg, and Ca (Thørring et al., 1999), a trend that has also been observed in other studies around Cu-Ni smelters (Fritze et al., 1989; Derome and Lindroos, 1998). The extent of this depletion at different distances from the smelter may be emphasized by calculated O-horizon/B-horizon ratios as illustrated in Fig. 8. In the case of 90Sr this ratio decreases from a value of 7.2 at the reference site to 0.7 at site A. Very similar depletion values are observed for the exchangeable cations. 4. Conclusions The increase in deposition of pollutants when approaching the smelter clearly affected the distribution of radionuclides in the soil profile. 90Sr was strongly depleted in the humus layer, whereas 137Cs was less depleted and to some extent accumulated in the litter layer. The chemical speciation of the radionuclides was affected as well. There was less 137Cs in readily exchangeable form in the humus layer (Of and Oh horizons) close to the smelter. For 90Sr the effect was smaller, but the
305
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
O-horizon/B-horizon ratio Sr-90
10
K
Mg
Ca
8 6 4 2 0 A
B
C
D
REF
Sampling site Fig. 8. Relative depletion of total 90Sr (Bq/m2) and exchangeable K, Mg, and Ca (mg/m2) with decreasing distance from the smelter illustrated as O/B concentration ratios.
exchangeable fraction was lower at the most polluted site than at the other sites. Strong correlations were evident between exchangeable 137Cs and the exchangeable pools of K and Mg with distance from the smelter. Total concentrations of both radionuclides in the soil horizons were also strongly correlated to exchangeable cations, particularly K and Mg (O, E, and B1) and Ca (E and B1). For 90Sr these correlations were also evident in the B2 horizon, but for 137 Cs not. Altogether this indicates that a significant fraction of 90Sr and even of 137 Cs must be present in exchangeable form in these horizons. The different behaviour of 137Cs in the B2 horizon may be associated with differences in clay mineralogy between horizons exposed too different degrees of weathering. The ratio of 90Sr concentrations in the Of relative to the B2 horizon decreased by a factor of 10 from the reference site to the site nearest the smelter. Similar surface depletions were evident for the exchangeable pools of K, Mg, and Ca. Acknowledgements This work was carried out as part of the project ‘‘Effects of industrial pollution on the distribution dynamics of radionuclides in boreal understorey ecosystems’’ in the EU-framework ‘‘Nuclear fission safety’’, Contract No. F14P-CT96-0039. References Bryant, F.J., Morgan, A., Spicer, G.S., 1959. The determination of radiostrontium in biological materials. United Kingdom Atomic Energy Authority Report AERE-R 3030, Harwell, UK. Bunzl, K., Albers, B.P., Shimmack, W., Rissanen, K., Suomela, M., Puhakainen, M., Rahola, T., Steinnes, E., 1999. Soil to plant uptake of fallout 137Cs by plants from boreal areas polluted by industrial emission from smelters. Science of the Total Environment 234, 213e221. Bunzl, K., Puhakainen, M., Riekkinen, I., Karhu, P., Schimmack, W., Heikkinen, T., Jaakkola, T., Pavlov, V., Rahola, T., Rissanen, K., Suomela, M., Tillander, M., A¨yra¨s, M., 2001. Fallout 137Cs, 90Sr
306
M. Puhakainen et al. / J. Environ. Radioactivity 81 (2005) 295e306
and 239C240Pu in soils polluted by heavy metals: Vertical distribution, residence half-times, and external gamma-dose rates. Journal of Radioanalytical and Nuclear Chemistry 247, 15e24. Derome, J., Lindroos, A-J., 1998. Effects of heavy metal contamination on macronutrient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter, SW Finland. Environmental Pollution 99, 225e232. Derome, J., Nieminen, T., 1998. Metal and macronutrient fluxes in heavy-metal pollution Scots pine ecosystems in SW Finland. Environmental Pollution 103, 219e228. Egli, M., Mirabella, A., Fitze, P., 2003. Formation rates of smectites derived from two Holocene chronosequences in the Swiss Alps. Geoderma 117, 81e98. Fritze, H., Niini, S., Mikkola, A., Ma¨kinen, A., 1989. Soil microbial effects of a Cu-Ni smelter in southwestern Finland. Biology and Fertility of Soils 8, 87e94. Maes, E., Vielvoye, L., Stone, W., Delvaux, B., 1999. Fixation of radiocaesium traces in a weathering sequence micaevermiculiteehydroxy interlayered vermiculite. European Journal of Soil Science 50, 107e115. Puhakainen, M., Riekkinen, I., Heikkinen, T., Jaakkola, T., Steinnes, E., Rissanen, K., Suomela, M., Thørring, H., 2001. Effect of chemical pollution on forms of 137Cs, 90Sr and 239,240Pu in Arctic soil studied by sequential extraction. Journal of Environmental Radioactivity 52, 17e29. Rahola, T., Albers, B., Bergman, R., Bunzl, K., Jaakkola, T., Nikonov, V., Pavlov, V., Rissanen, K., Schimmack, W., Steinnes, E., Suomela, M., Tillander, M., A¨yra¨s, M., 1999. General characterisation of study area and definition of experimental protocols. Report STUK-A166, Radiation and Nuclear Safety Authority (STUK), Helsinki. Reimann, C., A¨yra¨s, M., Chekushin, V., Bogatyrev, I., Boyd, R., de Caritat, P., Dutter, R., Finne, T.E., Halleraker, J.H., Jaeger, O.E., Kashulina, G., Lehto, O., Niskavaara, H., Pavlov, V., Ra¨isa¨nen, M.L., Strand, T., Volden, T., 1998. Environmental Geochemical Atlas of the Central Barents Region. Riekkinen, I., Jaakkola, T., 2001. Effect of industrial pollution on soil-to-plant transfer of plutonium in a Boreal forest. The Science of the Total Environment 278, 161e170. Suomela, M., Bergman, R., Bunzl, K., Jaakkola, T., Rahola, T., Steinnes, E., 1999. Effect of Industrial Pollution on the Distribution Dynamics of Radionuclides in Boreal Understorey Ecosystems (EPORA). Report STUK-A168, Radiation and Nuclear Safety Authority (STUK), Helsinki. Steinnes, E., Lukina, N., Nikonov, V., Aamlid, D., Røyset, O., 2000. Gradient study of 34 elements in the vicinity of a copperenickel smelter in the Kola Peninsula. Environmental Monitoring and Assessment 60, 71e88. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry 51, 844e851. Thørring, H., Steinnes, E., Nikonov, V., Rahola, T., Rissanen, K., 1999. A summary of chemical data from the EPORA project. Report STUK-A167, Radiation and Nuclear Safety Authority (STUK), Helsinki.