Radiocaesium soil-to-wood transfer in commercial willow short rotation coppice on contaminated farm...

www.elsevier.com/locate/jenvrad

Journal of Environmental Radioactivity 78 (2005) 267–287

Radiocaesium soil-to-wood transferin commercial willow short rotationcoppice on contaminated farm land

A. Gommersa, T. Gafvertb, E. Smoldersc,R. Merckxc, H. Vandenhovea,)

aRadiation Protection Research Department, Radioecology Section, SCK�CEN,

Boeretang 200, 2400 Mol, BelgiumbDepartment of Radiation Physics, The Jubileum Institute, University Hospital,

Lund University, SE-221 85 Lund, SwedencDepartment of Land Management, Laboratory of Soil Fertility and Soil Biology,

K.U. Leuven, K. Mercierlaan 92, 3001 Heverlee, Belgium

Received 1 January 2004; received in revised form 1 April 2004; accepted 6 May 2004

Abstract

The feasibility of willow short rotation coppice (SRC) for energy production asa revaluation tool for severely radiocaesium-contaminated land was studied. The effects ofcrop age, clone and soil type on the radiocaesium levels in the wood were assessed followingsampling in 14 existing willow SRC fields, planted on radiocaesium-contaminated land in

Sweden following Chernobyl deposition. There was only one plot where willow stands ofdifferent maturity (R6S2 and R5S4: R, root age and S, shoot age) and clone (Rapp and L78183both of age category R5S4) were sampled and no significant differences were found. The soils

differed among others in clay fraction (3–34%), radiocaesium interception potential (515–6884meq kg�1), soil solution K (0.09–0.95 mM), exchangeable K (0.58–5.77 meq kg�1) and cationexchange capacity (31–250 meq kg�1). The soil-to-wood transfer factor (TF) of radiocaesium

differed significantly between soil types. The TF recorded was generally small (0.00086–0.016kg kg�1), except for willows established on sandy soil (0.19–0.46 kg kg�1). Apart from theweak yet significant exponential correlation between the Cs-TF and the solid/liquiddistribution coefficient (R2 ¼ 0:54) or the radiocaesium interception potential, RIP

(R2 ¼ 0:66), no single significant correlations between soil characteristics and TF were found.The wood–soil solution 137Cs concentration factor (CF) was significantly related to the

) Corresponding author. Tel.: C32-14-332114; fax: C32-14-321056.

E-mail address: [email protected] (H. Vandenhove).

0265-931X/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jenvrad.2004.05.007

http://www.elsevier.com/locate/jenvrad

mailto:[email protected]

268 A. Gommers et al. / J. Environ. Radioactivity 78 (2005) 267–287

potassium concentration in the soil solution. A different relation was, however, found be-tween the sandy Trodje soils (CF=1078.8!mK

�1.83, R2 ¼ 0:99) and the other soils

(CF=35.75!mK�0.61, R2=0.61). Differences in the ageing rate of radiocaesium in the soil

(hypothesised fraction of bioavailable caesium subjected to fast ageing for Trodje soils only1% compared to other soils), exchangeable soil K (0.8–1.8 meq kg�1 for Trodje soils and 1.5–

5.8 meq kg�1 for the other soils) and the ammonium concentration in the soil solution (0.09–0.31 mM NH4

C for the Trodje soils compared to 0.003–0.11 mM NH4C for the other soils)

are put forward as potential factors explaining the higher CF and TF observed for the Trodje

soils. Though from the dataset available it was not possible to unequivocally predict the Cs-soil-to-wood-transfer, the generally low TFs observed point to the particular suitability forestablishment of SRC on radiocaesium-contaminated land.

� 2004 Elsevier Ltd. All rights reserved.

Keywords: Short rotation coppice; Willow; Radiocaesium; Soil-to-plant transfer; Alternative land use;

Chernobyl deposition

1. Introduction

The fallout of radiocaesium from the Chernobyl radioactive plume (April 26,1986) caused a large-scale and heterogeneous contamination, mainly in Belarus(70% of the deposits), Ukraine (20%) and Russia (7%) (Belli and Tikhomirov,1996). The radioactive aerosols that were released during the accident (such as radio-iodine and radiocaesium) spread through the entire Northern hemisphere. Problemswith 137Cs were, apart from the above-mentioned countries, especially met insensitive, often semi-natural or agricultural regions in Sweden (Rosen, 1986) wheredeposition levels were up to 100 kBq m�2 (McGee et al., 2000). From theenvironmental viewpoint, 137Cs is a radioactive pollutant of great concern (longphysical half-life of 30.2 years, high bio-availability, and behaviour similar topotassium) which can rapidly integrate the biological cycles and accumulate intoterrestrial ecosystems. Hence, in forests and agricultural areas, the contaminationwith 137Cs is and will be of great concern for several decades with considerable socio-economic implications for the local populations in terms of management andvalorisation of the contaminated territories.

Following the Chernobyl disaster, many studies have focused on the efficiencyand technical feasibility of physical and chemical countermeasures for contaminatedterrestrial environments (Lembrechts, 1993; Vovk et al., 1993; Nisbet, 1993; Melinet al., 1996). Various remediation options have been compared for agricultural lands(Shaw et al., 1992; Segal, 1993; Melin et al., 1996; Renaud and Maubert, 1997) andfor forest ecosystems (Davydchuk, 1997; Shaw et al., 2001) severely affected by 137Csdeposits (Belli et al., 1995; Belli and Tikhomirov, 1996). Countermeasures can alsobe based on the selection of crops which exhibit smaller radionuclide uptake (Nisbetet al., 1999; Voigt et al., 2000), on food processing or selecting for non-food cropssuch that the products from the land are radiologically acceptable (Alexakhin, 1993;

269A. Gommers et al. / J. Environ. Radioactivity 78 (2005) 267–287

Segal, 1993; GOPA, 1996). Information on the long-term effect of countermeasuresand especially the change to non-food crops is still limited. The present study alsoconcerns the cultivation of non-food crops on contaminated arable lands. It ishypothesised that the production of willow wood cultivated in short rotations (ShortRotation Coppice; SRC) to produce energy is a valuable alternative for foodproduction on agricultural land. For example, when the activity concentrations ofthe soil are such that physicochemical countermeasures are not sufficient to reducethe activity levels in crops or when they are economically, technically or socially notfeasible or acceptable.

Coppicing is a method of vegetative forest regeneration by cutting trees at thebase of the trunk at regular time intervals. Fast-growing species from the Salix genus(willows) are frequently used in coppice systems because of the ease of vegetativeregeneration and the large biomass production. Furthermore, they are tolerant toa wide range of climatic and edaphic factors (Ledin, 1996), temperature and wateravailability being the most important growth-limiting factors (Perttu, 1998). Thewood production ascertains economic revenue from the land at regular time intervals(3–5 years; Jossart et al., 1999) and may therefore be preferred to traditional forestry.Radiation doses to workers were found smaller than for agricultural systems or eventhan for other biofuel crops as oil rape seed, due to the limited time spent on the fieldfor establishment, maintenance and harvest of the crop (Vandenhove et al., 1999,2001). The perennial crop provides a protection of the land against erosion(secondary contamination is reduced) and leaching of radionuclides to thegroundwater is reduced due to the high water evaporation (700 mm y�1; Ledin,1996). Renewable energy, and in particular biofuels, has further a vital role to play inclimate stabilisation. Carbon dioxide (CO2) and other greenhouse gas emissionsmust be reduced in the European Union to 8% below the level of 1990 before theyears 2008–2012 (Kyoto protocol of 1997). Substantial experience in SRC has beenobtained in Sweden. The oil crisis in the 1970s resulted in a national researchprogramme to use biomass of rapidly growing willows for fuel (Philippot, 1996). Asurface of more than 10 000 ha of farmland in Sweden has actually been com-mercially planted with willow (Rosenqvist et al., 2000).

Willow SRC may be a suitable rehabilitation tool for radiocaesium-contaminatedland only if the Cs-activity in the wood does not exceed the exemption limits forbiofuels set in a particular country. Studies about the transfer of Cs from soil towood in a SRC system, or more generally in traditional forestry, are scarce.Moreover, most importance has been dedicated up to now to the fate ofradiocaesium deposited on foliage and soil (Bunzl et al., 1989; Ronneau et al.,1991; Melin et al., 1994) though it is suggested that in the long term, root absorptionof radiocaesium by trees is the most important contamination mechanism (McGeeet al., 2000).

Transfer of radiocaesium to willow was studied by Vandenhove et al. (1999, 2001,2004) and Gommers et al. (2000) at experimental scale and at field plots, for onerotation cycle. However, since willow short rotation coppice (SRC) is a perennialcrop and wood is only harvested after a rotation of 3–5 years, knowledge about theradiocaesium incorporation in the wood on a longer time scale is necessary.


The objective of this study is twofold: firstly, to assess the influence of crop ageand clone and soil characteristics on the radiocaesium soil-to-wood transfer incommercial willow SRC plantations planted on radiocaesium-contaminated land;secondly, to indicate if the observed soil-to-wood transfer factors offer particularsuitability for SRC establishment on contaminated land.

Therefore, existing willow cultivation systems were sampled in the Chernobyl-affected area in Sweden, selecting only those sites where SRC was established afterthe Chernobyl cloud past in order to have to deal solely with radionuclide in-corporation in the wood through root absorption.

The soil-to-plant transfer can be envisioned as a result of two processes: theequilibrium of the radionuclide between the solid and the liquid phase of the soil,expressed by the solid liquid distribution coefficient, KD [KD=(Bq l�1)/(Bq kg�1)]and the uptake of the radionuclide by the plant from the soil solution, expressed asa soil solution to plant transfer or concentration factor, CF [(Bq kg�1 plant)/(Bq l�1)]. This can be expressed by the following equation:

TF ¼ CF

KD

ð1Þ

in which TF is the transfer factor [(Bq kg�1 plant)/(Bq kg�1soil)].Radiocaesium soil-to-plant transfer depends strongly on soil properties as pH,

organic matter content, exchangeable potassium, potassium and ammoniumconcentration in the soil solution and the radiocaesium solid:liquid distribution(Maraziotis, 1992; Bilo et al., 1993; Smolders et al., 1997; Yera et al., 1999). Cor-relations between soil-to-plant transfer of radiocaesium and single soil propertieshave sometimes been found in field surveys. Soil-to-plant transfer factors (TFs) inupland herbage (Cumbria, UK) increased with increasing organic matter content(P! 0:01), and with increasing exchangeable radiocaesium or potassium contents inthe soils (P! 0:01) (Sandalls and Bennett, 1992). In contrast, Bilo et al. (1993) founda significant negative relation between the TF for cereals and exchangeablepotassium. Maraziotis (1992) found significant negative relations between the TFand pH, soil clay content or exchangeable K and a positive relation with cationexchange capacity (CEC) in a study with 33 soils and four crops (potatoes, wheat,olives and grass). In the latter study, the pH varied considerably between the soils(3.5–8.3) and adequately predicted the soil-to-plant TF. Kuhn et al. (1984) alsoobserved decreasing TFs at increasing pH values.

Nisbet and Woodman (2000), however, compiled a database for radiocaesiumsoil-to-plant transfer factors for arable crops. They found that there were no singlesoil characteristics that accounted for more than 30% of the variability in radiocae-sium TFs.

Recently it was suggested that the radiocaesium soil availability is determined bythe radiocaesium and the potassium concentration in the soil solution. Smolderset al. (1997) and Vandenhove et al. (2003) found that these two parameters predictedthe radiocaesium TF for grass rather well [respectively, log TF=b0 exp(�b1 logmK)�log KD; R

2 ¼ 0:94; and log10 TF=�1.30�1.22 log10 [KC] � log KD; R

2 ¼ 0:92,


with KD the solid/liquid radiocaesium distribution coefficient (l kg�1) and mK thepotassium concentration in the soil solution (mol l�1)]. Yera et al. (1999) estimatedthe soil-to-plant transfer from a loamy soil and a loamy-sandy soil based on theKD, the potassium and ammonium concentration in the soil solution anda plant available fraction of radiocaesium in soils. They successfully predictedrelative radiocaesium accumulation in winter barley (P! 0:001). Absalom et al.(1999, 2001) predicted the radiocaesium soil-to-plant transfer based on deposi-tion level, soil exchangeable K-content and clay content, including a time-factor.Their model explained 52% of the variation found between TFs observed and theyfound a significant fit between the transfer factor and the soil solution Kconcentration.

2. Materials and methods

Existing willow SRC cultures in Sweden were sampled during the dormant period(winter) in 1997–1998. The plots sampled represented most of the plots commerciallyplanted with willow SRC in central-east Sweden, the region most affected by theChernobyl deposition (Fig. 1). More SRC fields are planted in southern Sweden.

.... Stockholm

Gävle

Uppsala

Österfärnebo

Björklinge TierpViksta....

Trödje

Fig. 1. Location of the willow SRC fields sampled in Sweden.


However, soil contamination was too low in those regions to detect 137Cs activity inthe wood of willows. All willow SRC plantations in central-east Sweden weresampled which were established following the Cs-deposits originating from theChernobyl accident. Only those sites were sampled in order to study radiocaesiumincorporation in the biomass solely due to root uptake.

Our study of the effect of plant age and clone and of soil characteristics onradiocaesium transfer was hence limited by the availability of willow SRC systemsestablished in central-eastern Sweden after the Chernobyl accident. Plant age variedbetween 4 and 8 years. Some of the plots were cut back after the first year. Shoot agevaried from 1 to 6 years and plants were in the first or in the second rotation. Plantage will further be written as RxSy, with x the age of the root system (R) and y theage of the shoots (S). Three clones were sampled: Salix viminalis L. var. L 78183 (themost frequently planted Salix clone in Sweden), Salix viminalis L. var. L 78112 andSalix viminalis L. var. Rapp. Only at one of the plots, however, two out of the threevarieties were present (Viksta 1 with the clones L 78183 and Rapp). Information onage and varieties of the plants in the different plots is given in Table 1. Most of theplots were planted and cultivated in a similar way. Plant density was 17 500 plantsha�1, planted in double rows with spacing of 1.5 m between the double rows and0.75 m between the rows. Most plots were fertilised, mostly with an NPK fertiliser,before planting or during plant growth (amounts and time of application uncertain).

Ten plants per plot were randomly chosen and diameters of all shoots weremeasured at a height of 50 cm above ground level. One shoot of each of the 10 plants

Table 1

Sampled willow SRC fields

Site Salix clone Age

RxSya

Cut-backb

yes or no

Cutting cycle

1st or 2nd

Fertiliser Soil contaminationc

Bq kg�1

Viksta 1 Rapp R5S4 y 1st NPK 58.7G 0.6

L 78183 R6S2 y 2nd NPK 58.7G 0.6

R5S4 y 1st NPK 58.7G 0.6

Viksta 2 L 78183 R6S2 y 2nd N 71.3G 3.2

Viksta 3 L 78183 R6S2 y 2nd NPK 44.7G 1.5

Tierp 1 L 78183 R6S2 y 2nd NPK 93.7G 1.2

Tierp 2 1G Rapp R5S4 y 1st NPK 72.0G 2.6

Tierp 2 1M Rapp R5S4 y 1st NPK 127.0G 2.6

Borklinge 1 L 78183 R7S2 y 2nd NPK 116.2G 4.5

Bjorklinge 2 L 78183 R8S2 y 2nd N 94.3G 4.9

Bjorklinge 3 Rapp R6S2 y 2nd NPK 116.7G 14.0

Osterfarnabo L 78183 R7S6 y 1st NPK 41.1G 2.1

Trodje 1 L 78112 R4S4 n 1st NP 285.3G 28.1

Trodje 2 L 78112 R4S4 n 1st NPK 285.3G 28.1

Trodje 3 L 78112 R4S4 n 1st NP 285.3G 28.1

Trodje 4 L 78112 R4S4 n 1st NPK 285.3G 28.1

a RxSy: Root (R) x years old and Shoot (S) y years old.b Harvesting (cutting back) the willows at the end of the first year in order to promote shoot formation.c Soil sampling depth 25 cm.


was cut and weighed. The shoots were cut in pieces of 5–10 cm and mixedthoroughly. Three times, 1 kg of the wood was ashed and 137Cs activity in the asheswas measured by g-spectrometry (n-type HPGe-detector with 23% relative efficiencyand a resolution of 2.15 keV at 1332 keV, 48 h measurement time and 2.5%measurement uncertainty). Analysis was done in triplicate, unless activity was so lowthat ash of the three samples had to be merged for analysis. K concentration in thewood was measured (three replicates) following ashing for a minimum 24 h at 500(C, mineralisation with HCl, 36–38% and analysis by atomic absorption spec-trometry (AAS).

Ten soil samples (0–25 cm, w1 kg per sample) were randomly taken at each plot.The soil (9 kg) was thoroughly mixed over all layers and 0.5 kg was taken for furtheranalysis. Total activity in the soil was measured by HPGe (three replicates). TheCEC was measured by AgTU, as is described by Chhabra et al. (1975). Exchangeablecations and exchangeable radiocaesium were extracted by 1 N ammonium acetate(1 g soil to 18 ml NH4Ac solution) and measured by AAS and HPGe detection,respectively. Soil solution was isolated by centrifugation of soil samples (about 50 g),previously equilibrated for 24 h with distilled water at field capacity. Soil solutioncomposition was measured at field capacity. To collect the soil solution, a disposable60-ml syringe without plunger was lined with a filter and filled with soil. The syringewas transferred to a centrifuge tube and centrifuged for 30 min at 50! g. The soilsolution was filtered through a 0.45-mm membrane filter (Millipore). Cations in thesoil solution were determined by AAS. The NH4

C concentration in the soil solutioncould not be determined directly as soils were air dried and stored for several monthsbefore analysis. The NH4

C , if present in the soils, was thus possibly oxidised toNO3

�. We, therefore, proposed a semi-quantitative measure for an a posterioridetermination of the NH4

C concentration. Therefore, soils were rewetted to fieldcapacity, ryegrass was sown on the soils and the NO3

� present was removed withryegrass, ryegrass was harvested, roots were removed from the soil and the soil wasincubated (2 (C, 1 month). The soil solution was extracted by centrifugation andNH4

C in the soil solution was determined colorimetrically (indophenol bluemethod). The Radiocaesium Interception Potential (RIP) of the soils (meq kg�1)was determined as the product of KD and mK in a specific K–Ca scenario, with KD

the solid/liquid radiocaesium distribution coefficient (l kg�1) and mK the potassiumconcentration in the soil solution (mol l�1) (Wauters et al., 1996). Briefly, the soilsamples were equilibrated with a mixed solution 0.5 mM KCl/100 mM CaCl2. Theliquid phase was then labelled with trace 137CsC (50 Bq ml�1) quasi carrier free(Amersham), and the 137CsC concentration in the supernatant was measured after168 h equilibration time by g-counting, using a NaI scintillator g-counter (Auto-gamma 5000 series, Canberra; statistical counting error!5%). The texture wasanalysed by sieving and precipitation (Stokes’ law: Day, 1965). The clay fraction ofthe Ap horizon was characterized by X-ray diffraction (diffractometer using Cu Karadiation, D8 Advance Bruker diffractometer, Philips Norelco) on KC and Mg2C

saturated samples, after air-drying at 20 (C, ethylene glycol (EG) saturation andthermal treatment (550 (C). Organic carbon content (OC) was determined by loss ofignition at 450 (C for 48 h.


Transfer factors (TF, kg kg�1 or m2 kg�1) were calculated as follows:

TF ¼Cs½ �plantCs½ �soil

ð2Þ

with radiocaesium concentrations ([Cs]) in Bq kg�1 dry weight plant material and Bqkg�1 or m2 kg�1 dry weight soil. For the TF expressed on a surface basis, a 25 cm soildepth and a soil density of 1.25 kg l�1 was considered. The former expression for theTF is used in the discussion on how soil parameters influence the TF; the secondexpression in the reasoning on expected wood contamination levels with surfacecontamination.

The soil solution to plant transfer (or concentration factor, CF), is defined as:

CF ¼Cs½ �plantCs½ �solution

ð3Þ

with the [Cs]solution the radiocaesium concentration in the soil solution (Bq l�1). Thelatter parameter could not be measured directly since radiocaesium concentrations inthe soil solution were below the detection limit. The Cs-concentration in the soilsolution was estimated from the RIP and the soil solution K-concentration followingthe formula RIP=KD!mK, and with KD=[Cs]soil/[Cs]solution. The TF can then beestimated as TF=CF/KD (Eq. 1 repeated).

3. Results

3.1. Soil properties

No detailed information could be obtained on the fertilisation regimes. Except forthe Trodje soils, all soils were farmers’ fields. Lack of knowledge on detailed fer-tilisation regimes was anyway backed up by a thorough soil analysis. Also, as can bededuced from Table 2, fertilisation with NP or NPK of the Trodje soils did not sig-nificantly influence soil K-status (Table 2) or Cs-concentration in the wood (Table 4).

Selective soil characteristics are presented in Table 2. The clay content of the soilsvaried between 3 and 34%. Few sandy soils were sampled (Trodje, Tierp 1, Tierp 21M). Most soils were very clayey, differing, however, in organic carbon content andnutrient status. The RIP value was significantly correlated with the clay content(P! 0:01; R2 ¼ 0:85; data not shown). The calculated solid:liquid distribution co-efficient KD for Cs differed 111-fold between the soil types. The smallest predictedradiocaesium concentration was 6!10�5 Bq l�1 (Viksta 3) and the largest was 0.04Bq l�1 (Trodje 4). Exchangeable radiocaesium concentrations were also below thedetection limit. The K concentration in the soil solution was below 1 mM (0.09–0.95mM) for all the soils and was significantly correlated (P! 0:001) with the relativeoccupancy of the CEC by K ions

mK ¼ 0:0013ðG 0:076ÞC0:132ðG 0:029Þ!ð%KxÞ; P! 0:001; R2 ¼ 0:63 ð4Þ

with mK the K concentration in the soil solution and %Kx the relative occupancy ofthe CEC by K.

Texture (%)

!2

mm

2–20

mm

20–200

mm

200–2000

mm

4 30 22 38 9

3 32 23 43 2

1 33 25 39 3

3 6 5 86 3

3 15 13 70 2

3 4 5 30 54

4 31 26 42 1

1 22 33 38 7

3 34 30 34 2

6 14 16 48 22

2 3 2 23 71

1 3 2 23 71

2 3 2 23 71

2 3 2 23 71

H4, concentration of NH4

C in soil solution;

275

A.Gommers

etal./

J.Enviro

n.Radioactivity

78(2005)267–287

Table 2

Some soil properties of the sampled willow SRC fields in Sweden

RIP

meq kg�1CEC

meq kg�1Exch K

meq kg�1mK

mM

mNH4

mM

pH

(KCl)

OC

%

Viksta 1 6884G 499 201G 4 3.36G 0.06 0.12G 0.02 0.11 6.19G 0.02 0.44G 0.0

Viksta 2 5431G 315 217G 2 3.72G 0.09 0.20G 0.08 0.001 5.10G 0.02 0.47G 0.0

Viksta 3 5453G 241 250G 1 3.61G 0.09 0.09G 0.02 0.06 5.02G 0.03 0.53G 0.0

Tierp 1 1517G 35 81G 2 2.72G 0.06 0.38G 0.04 0.003 5.10G 0.04 0.34G 0.0

Tierp 2 1G 4055G 174 95G 1 2.39G 0.03 0.19G 0.05 0.03 5.79G 0.02 0.27G 0.0

Tierp 2 1M 908G 23 106G 2 1.47G 0.02 0.31G 0.03 0.13 5.41G 0.02 0.55G 0.2

Borklinge 1 3866G 307 144G 3 2.14G 0.02 0.34G 0.02 0.05 4.66G 0.02 0.61G 0.0

Bjorklinge 2 5233G 315 224G 2 5.77G 0.03 0.15G 0.03 0.03 6.20G 0.03 0.37G 0.0

Bjorklinge 3 4929G 65 222G 2 0.58G 0.04 0.16G 0.02 0.04 5.01G 0.02 0.57G 0.0

Osterfarnabo 3431G 633 86G 2 2.40G 0.12 0.32G 0.06 0.05 4.24G 0.04 0.44G 0.0

Trodje 1 564G 33 39G 2 1.13G 0.07 0.48G 0.06 0.09 5.20G 0.01 0.22G 0.0

Trodje 2 562G 241 45G 1 0.80G 0.01 0.29G 0.03 n.d. 6.59G 0.07 0.23G 0.0

Trodje 3 578G 8 38G 6 0.97G 0.01 0.39G 0.03 0.31 6.58G 0.01 0.08G 0.0

Trodje 4 515G 22 31G 1 1.84G 0.05 0.95G 0.12 n.d. 5.25G 0.04 0.18G 0.0

RIP, Radiocaesium Interception Potential; Exch K, exchangeable K; mK, concentration of K in soil solution; mN

n.d., not determined.


The X-ray diffraction patterns demonstrated differences in clay mineralogy ofthe soils at the different locations (Table 3, Fig. 2). Viksta and Bjorklinge soilscontained most illites as was seen from the clear D001 reflection at 1.0 nm. Illiteswere also observed in the sandy Tierp 2 1M soil, but could hardly be detected inthe Trodje soils. Kaolinite minerals (0.7 nm) were present in all soils and hydroxy-interlayered vermiculites (HIV) especially in theViksta, Bjorklinge andTierp soils. TheTrodje soil is characterised by a very small amount of clay minerals. Furthermore,micaceous minerals, such as illites are hardly detectable. This is reflected in the smallRIP of the Trodje soils. The silt fraction in the Trodje soils was also smaller than in allother soils.

3.2. Plants

From the shoots sampled, a diameter–weight relationship was established andwith the plant density, woody biomass per ha was calculated. Relations were of theform y=axb with y the biomass and x the diameter of the shoots. Correlationcoefficients (R2) varied between 0.78 and 0.99 for the different plots.

Annual biomass production varied between the fields, with highest productionsfound at Bjorklinge (17.6 t ha�1 y�1) and lowest at Tierp 2 1M (5.7 t DM ha�1 y�1).In Viksta 1, the clone Rapp produced 12.0 t ha�1 y�1, the clone L 78183 produced11.1 t ha�1 y�1. The clone Rapp produced on average 9.1G 3.2 t ha�1 y�1, the cloneL 78183 10.5G 4.1 t ha�1 y�1 (P ¼ 0:62).

The Cs concentration in the willow wood ranged from 0.1 to 128 Bq kg�1. The137Cs TFs for the willow stems ranged over about three orders of magnitude: from

Table 3

Identification key for minerals in the clay fraction of four groups of soils (Ap-horizon) sampled under

willow SRC in Sweden

D001 (nm) K sat. Mg sat. 550 (C EG sat.

Viksta 0.7 x x – x Kaolinite

1.0 x x x x Illites

Bjorklinge 0.7 x x – x Kaolinite

1.0 x x x x Illites

1.4 x x – x Chlorites

and HIV

Tierp 2 1M 0.7 x x – x Kaolinite

1.0 x x x x Illites

1.4 x x – x Chlorites

and HIV

Trodje 0.7 x x – x Kaolinite

1.0 x x x x Illites

1.4 (x) (x) (–) (x) Chlorites

and HIV

HIV, hydroxy-interlayered vermiculite.


Tierp 2 1MViksta

Trödje

0.5 1.0 1.5 2.0d001 (nm)

0.5 1.0 1.5 2.0d001 (nm)

0.5 1.0 1.5 2.0d001 (nm)

K-sat.

Mg-sat.

550 °C

EG-sat.

K-sat.

Mg-sat.

550 °C

EG-sat.

K-sat.

Mg-sat.

550 °C

EG-sat.

Fig. 2. X-ray diffractions of the clay fractions (K or Mg saturated, heated at 550 (C and ethylene glycol

saturated) of the different soil types sampled.

Table 4

Wood activity concentrations and transfer factors for willows grown as short rotation coppice on

contaminated fields in Sweden

Var. Wood 137Cs

concentration

(Bq kg�1)

GSDa TF

(kg kg�1)

TF

(m2 kg�1)

Viksta 1 L78183 (R6S2) 0.83 G 0.21 1.4!10�2 3.8!10�5

L78183 (R5S4) 0.51 G 0.37 8.6!10�3 1.9!10�5

Rapp 0.18 G 0.03 3.0!10�3 8.2!10�5

Viksta 2 L78183 0.67 G 0.05 9.4!10�3 3.1!10�5

Viksta 3 L78183 0.73 G 0.04 1.6!10�2 5.4!10�5

Tierp 1 L78183 0.95 G 0.08 1.0!10�2 3.2!10�5

Tierp 2 1G Rapp 0.40 G 0.34 5.6!10�3 1.4!10�5

Tierp 2 1M Rapp 0.56 G 0.08 4.4!10�3 1.3!10�5

Bjorklinge 1 L78183 0.27 – 2.3!10�3 9.0!10�6

Bjorklinge 2 L78183 0.57 G 0.00 6.1!10�3 3.2!10�5

Bjorklinge 3 Rapp 0.10 – 8.6!10�4 2.4!10�6

Osterfarnabo L78183 0.42 – 1.0!10�2 2.8!10�5

Trodje 1 L78112 80 – 0.286 8.9!10�4

Trodje 2 L78112 128 – 0.457 1.4!10�3

Trodje 3 L78112 94 – 0.336 1.0!10�3

Trodje 4 L78112 52 – 0.186 5.8!10�4

a Standard deviations illustrate the variation in activity measurements of wood from the same field and

are only given when triplicate measurement was carried out.


0.00086 kg kg�1 to 0.457 kg kg�1 (from 2.4!10�6 to 1.4!10�3 m2 kg�1 whenexpressed on surface basis) (Table 4). Excluding the Trodje plots, the TFs differedonly 19-fold, the same order of magnitude as the difference in KD values (21-fold:Trodje excluded). Apart from the rather weak exponential correlation between theCs-TF and the solid/liquid distribution coefficient (TF=133![KD]

�0.9996,R2 ¼ 0:54) or the RIP (TF=93![RIP]�1.61, R2 ¼ 0:66), no single significant cor-relations between soil characteristics (clay content, pH, OM, mK, exchangeable K)and TF were found. Neither were the TFs significantly correlated to biomassproduction, shoot age nor to the K concentration in the stems (ranging from 1.6 and3.1 g kg�1) (data not shown).

The K concentration in the soil solution was, however, significantly related tothe soil-solution-to-wood transfer (Fig. 3). However, the relation found differedbetween the Trodje soils and the other soils. For the Trodje soils the relationfound was:

CF ¼ 1078:8!m�1:83K ; R2 ¼ 0:99 ð5aÞ

and for the other soils:

CF ¼ 35:75!m�0:61K ; R2 ¼ 0:61 ð5bÞ

Reasons for this discrepancy may be a difference in ageing of the soils, differences inNH4

C content in the soil solution or long-term K-availability (see further).Plot Viksta 1 was the only plot where willow stands of different maturity and

clones were sampled. No significant difference was observed between the two ageclasses of the clone L 78183 (R6S2 and R5S4). The wood activity concentration inclone Rapp was smaller than the activity concentration in the clone L 78183 of thesame age (R5S4) at 10% significance level (P ¼ 0:096). Large variations in the woodCs-concentrations within a plot were, however, observed partially accounting for thelack of significant differences and correlations.

y = 1078.80x-1.83

R2 = 0.99

y = 35.75x-2.24

R2 = 0.61

0

2000

4000

6000

8000

10000

12000

14000

0 0.2 0.4 0.6 0.8 1

mK (mM)

CF

(L

kg-1

)

Other soils

Trödje

Fig. 3. Relation between the potassium concentration in the soil solution and the predicted radiocaesium

CF of willow wood.


4. Discussion

4.1. The Cs soil-to-plant transfer and its predictability

The 137Cs soil-to-wood TFs ranged over from 0.00086 kg kg�1 to 0.457 kg kg�1

(from 2.4!10�6 to 1.4!10�3 m2 kg�1 when expressed on surface basis) and may beconsidered at the lower-medium end of the Cs-TF scale compared to agriculturalcrops and trees. Nisbet and Woodman (2000) presented typical ranges of 137Cs TFs(in kg kg�1) for cereals (0.0004–0.25), tubers (0.003–0.89), leafy vegetables (0.08–1.7)and grasses (0.01–1.0). Soil-to-wood TFs observed in forests range from around2!10�4 to 2!10�3 m2 kg�1 (Kaletnyk, 1999; Goor et al., 2003).

Soil-to-wood 137Cs TFs factors recorded by Gommers (2001) in a lysimeter studyat the end of a 3-year cutting cycle with willow established were 0.0095G 0.0056 kgkg�1 on loamy soil and 0.043G 0.016 kg kg�1 on sandy soil. The value found for theloamy soils corresponds very well with the average TF observed in present study forsoils with >14% clay (0.0077G 0.0049 kg kg�1). The higher value observed in thestudy by Gommers (2001) for the sandy soil in comparison with the TFs recorded onthe Tierp 1 and Tierp 2 1M soils, may be linked with the higher RIP and soil solutionK-concentrations for the latter soils and the fact that the Cs is aged over a longerperiod (12-year-old contamination at Swedish plots and contamination 4 years oldat the end of the lysimeter experiment).

A major aim of this study was to predict the radiocaesium concentration (or TF)in the wood from soil or plant parameters. However, the 14 plots sampled were notvery distinctive in soil characteristics and could be roughly divided in soils with a lowclay content (%6%: plots in Trodje, Tierp 1 and Tierp 2 1M) and a high clay content(14–33%: remaining plots). Differences in soil properties were, therefore, possiblytoo small to establish very significant relations between soil properties and the soil-to-wood transfer factor. As mentioned, the most significant relations were foundbetween the TF and the KD and RIP, explaining, respectively, 54 and 66% of thevariation found. In the current study, the KD value varied only 110-fold (21-foldTrodje excluded) and the RIP-value only 13-fold (8-fold Trodje excluded). Smolderset al. (1997) found a 1350-fold difference in KD values for 30 soils (360-fold differencein RIP). Sanchez et al. (1999, 2002) observed a 3800-fold and 1300-fold differenceover 23 UK-soils, for KD and RIP, respectively. Delvaux et al. (2000) found a 380-fold difference in RIP over 47 soils. The corresponding ranges in TFs found in therespective studies were 1000-fold, 185 fold and 733-fold. It is, hence, striking that theRIPs (or KD) and TFs all vary in the same order of magnitude (in our study,respectively, 8-, 21- and 19-fold differences, Trodje excluded or 13-, 110- and 350-fold differences considering all soils). The variation in K concentrations in the soilsolution was also smaller than in the study of Smolders et al. (1997) (factor 10 vs.factor 167).

It should be borne in mind that the SRC cultivation system is a perennial system,and that values obtained here for the more changeable soil characteristics likepH, soil solution and exchangeable cation concentrations are momentary and maynot reflect fully the average cation status the willows were exposed to during on


rotation cycle or lifetime of the system. So, for example, the high soil solutionK-concentrations observed for the Trodje soils, despite the low exchangeable K andthe low CEC, may point to recent fertilisation.

As mentioned, the Trodje soil is characterised by a very small amount of clayminerals. Furthermore, micaceous minerals, such as illites are hardly detectable. Thisis reflected in the small RIP of the Trodje soils. In kaolinite (1:1 type clay minerals)no specific Cs-binding sites are present (Komarneni, 1978; Erten et al., 1988).Radiocaesium is thus easily extracted from these soils and ageing (radiocaesiumdiffusion to the 1.0 nm interlayers) will be very small for the Trodje soils. Tierp 2 1Mwas also characterised by very small clay contents but the presence of illites and HIVexplains the twofold higher RIP compared to the Trodje soils. RIP values for Tierp 21M values were also slightly larger than the values for the Trodje soils.

Absalom et al. (1999) proposed a model in which Cs uptake is predicted from soilsolution characteristics (Cs and K concentrations). These concentrations areestimated from the total radiocaesium concentration of the soil, the exchangeableK and the soil clay content and the time between contamination and sampling. Thismodel was used to predict the concentration in the willow wood for the fields inSweden. There are three parameters in the model that are fitted to the data. Theseparameters are plant specific and are used to describe the CF–mK relationship. Thepredicted wood activity concentrations correlate quite well with the measured values(R2 ¼ 0:74; Fig. 4). However, the good correlation was due to the high activity levelsin the wood on the plots in Trodje compared to the other plots. Without these, nosignificant correlation was found.

-2

-1

0

1

2

3

-2 -1 0 1 2 3Modelled log(Cs

plant)

(log (Bq kg-1

))

Measu

red

lo

g(C

sp

lan

t)

(lo

g (B

q kg

-1))

R2=0.74

Trödje plots

Fig. 4. Predicted vs. measured activity levels in willow wood on radiocaesium-contaminated fields in

Sweden (after Absalom et al., 1999).


Fig. 3 reveals that the CFs for the willows in Trodje were significantly higher thanthe CFs for the willows of all other plots at similar K concentrations in the soilsolution. The CFs presented were calculated based on an estimated radiocaesiumconcentration in the soil solution, based on the soil RIP and the soil solution Kconcentration. Several factors may account for the difference observed, ageing beingone of them.

The radiocaesium concentration in the soil solution was corrected for radioactivedecay and for ageing in the soil as was done by Absalom et al. (1999). The ageingfactor D (after Absalom et al., 1999) was calculated as follows:

D ¼ Pfaste�kfasttCð1� PfastÞe�kslowt ð6Þ

with kfast and kslow fixation rate constants (1.9!10�3 and 1.9!10�4 d�1,respectively) and Pfast the proportion of the bioavailable Cs at time 0 subject tofixation at the rate kfast (Z0.81). The proportional rate of decrease in the bioavail-ability of radiocaesium was assumed to be constant for all soil types, since almost allsoils contain illitic or micaceous minerals and Cs fixations process is mainlyregulated by these minerals. Lewyckyj (2000), studying apparent kinetic coefficientsquantifying the caesium sorption on very specific sites (described as edge and/orinterlayer sites), found differences between minerals (illites, vermiculites, beidillites).

A sensitivity analysis showed that small differences in the parameter Pfast resultsin large differences in the calculated soil solution radiocaesium concentration andthus in the CF. The radiocaesium concentration in the soil solution 12 years afterdeposition is underestimated with 4% when the real value of Pfast is 1% lower thanthe value of 0.81 used by Absalom et al. (1999). A value that is 10% lower un-derestimates the radiocaesium concentration in the soil solution with 30%. TheTrodje soils were characterised by a very low clay (and fine silt) content and onlyshowed traces of illites or illite/vermiculite and chlorite/vermiculite (Fig. 2). If thefraction Pfast in the Trodje soils is lower than 0.81, the predicted CF decreases andthe Trodje-curve in Fig. 4 becomes more similar to the curve of the other soils. IfPfast for the Trodje soils is 1% of the Pfast for the other soils, the curves almostcoincide. The exact ageing rate of the different soils could, however, not bequantified. The hypothesis that ageing was overestimated in the Trodje soils due tolack of specific sorption sites might thus hold but probably does not fully explain thedifferences observed with the other soils.

Also the relatively higher NH4C concentrations in the soil solution of the Trodje

soil may be at the origin of higher Cs soil-to-plant transfer. Enhanced uptake ofcaesium by plants is reported in the presence of increased amounts of NH4

C

(Jackson et al., 1965; Minotti et al., 1965; Lembrechts, 1993; Bondar and Dutov,1992; Belli et al., 1995).

As mentioned several times, the K concentration in the solution affects stronglythe uptake of radiocaesium by plants (e.g. Shaw et al., 1992; Smolders et al., 1997;Waegeneers, 2002; Vandenhove et al., 2003). The K-concentration near the rootsurface is, however, difficult to determine and does not equal the concentrationmeasured in the bulk soil (Hinsinger, 1998). Plant uptake of K generates K-depletionaround the roots. Jungk and Claasen (1986) measured K concentrations at the root


surface of maize plants as low as 2 mM, which was 70–350 times smaller than in theliquid phase of the bulk soil. Waegeneers (2002) predicted for a series of annual cropsthat at low K-availability (!0.1 mM), the K-concentrations observed at the rootsurface was only 7–32% of the concentrations observed in the bulk soil, while theywere more than 67% of the concentrations found in the bulk soil at high K-supply.In our study, the momentary K-concentration in the soil solution was higher forTrodje (range 0.29–0.95 mM; high values possibly due to recent fertilisation)compared to the other soils (range 0.09–0.38 mM). The levels of exchangeable Kwere, however, lower at the Trodje soil: 0.8–1.8 meq kg�1 compared to 1.5–5.7 meqkg�1 for the other soils. This means that the semi-long-term K-supply is much morelimited for the Trodje soil, aggravated by the low CEC of the Trodje soils. Moreover,in the long-term, in conditions of K depletion, K desorption from the solid phaseis favoured (Hinsinger, 1998). Hoagland and Martin (1933), followed by others,stressed the importance of the release of non-exchangeable K to satisfy K demands(e.g. Mengel, 1985; Markewitz and Richter, 2000). It was further shown that undersevere K depletion due to plant growth, biotites and phlogopites transform tovermiculites or kaolinites along with K mobilisation (Mortland, 1958; Hinsinger andJaillard, 1993). Potassium (and radiocaesium) mobilisation from layer silicates wasalso found for willow (Gommers, 2001), depending on the weatherability of thesilicates: muscovite! phlogopite! biotite! vermiculite. The Trodje soils do onlycontain very small amounts of illites and almost no easily weatherable chlorites andvermiculites and contain mostly kaolinites with lower K levels in their clay structure.Hence compared with the other soils, long-term K-availability may be morerestricted on the Trodje soils compared to the other soils.

Finally, in a perennial SRC-system still other parameters, beyond control, mayaffect the uptake: e.g. the microbial processes, root density and the distribution ofroots over the possible heterogeneous transport to roots (Ehlken and Kirchner, 2002).

4.2. Feasibility of SRC for biofuel use

We wanted to predict the soil to plant transfer factor from soil parameters sincetransfer of radiocaesium to willows and incorporation in the wood is of majorimportance in the evaluation of a willow short rotation coppice (SRC) cultivationsystem as possible revaluation tool for contaminated land. For the actual conditionsof the plot studied, with respect to deposition levels and TFs, the wood radiocaesiumconcentration in the wood (0.1–128 Bq kg�1) remained below the exemption limit forfuel wood as put forward in the Commonwealth of Independent States (740 Bq kg�1

dry wood; Szekely et al., 1994).In Sweden there are, to the knowledge of the authors, no limits set for Cs-levels in

fuelwood. A potential limit of 100 Bq kg�1 137Cs in fuelwood can be assumed basedon the limit of 5000 Bq kg�1 (and considering a concentration factor 50 in the ashes)proposed for application of ashes as fertiliser in order to reach an acceptableadditional dose of 0.01 mSv per annum (Hubbard, 1999). Ravila and Holm (1994)assessed that the maximum activity concentration of 137Cs be limited to the range of12–90 Bq kg�1, in order that the dose to people working on an ash deposit during


a normal year of work (2000 h) would be less than 1 mSv per year. Although thesesuggested limits are very stringent, generally the Cs-levels in the willow woodobserved are below these potential Swedish limits.

The soil-to-wood TFs for willow expressed as aggregated transfer factors(TFagg) range from 5.8!10�4–1.4!10�3 m2 kg�1 for the Trodje soils and from2.4!10�6�8.9!10�5 m2 kg�1 for all other soils. Taking an average TFagg of10�3 m2 kg�1 for the Trodje soils and of 10�5 m2 kg�1 for all other soils, soil con-tamination should be below, respectively, 740 kBq m�2 (20 Ci km�2) and 74 MBqm�2 (2 Ci km�2) in order to produce clean wood. Far less than 5% of the areacontaminated by the Chernobyl accident was contaminated at levels of 740 kBq m�2

or more. In Sweden this level of contamination did not occur.

5. Conclusions

In this study we wanted to assess the influence of crop age and clone and soilcharacteristics on the radiocaesium soil-to-wood transfer in commercial willow SRCplantations planted on radiocaesium-contaminated land and to indicate if theobserved soil-to-wood transfer factors offer particular suitability for SRC establish-ment on contaminated land. Therefore, existing willow cultivation systems estab-lished after the Chernobyl accident were sampled in the Chernobyl affected area inSweden.

Plant maturity and clone did not significantly affect the radiocaesium transferfactor.

Since the radiocaesium soil-to-SRC wood TFs are in general very small, there isusually little concern that the exemption limits for fuelwood will be exceeded unlesssoil contamination levels are extremely elevated. Only for soils with low RIP, low soilK status (and high NH4

C), as is the case for the Trodje soils, there may be someconcern that the permissible levels are exceeded when soil contamination is high.

It may, however, be important for environmental decision making in the processof defining an optimal allocation of the contaminated land, to have a tool to predictthe contamination levels in the exploitable plant compartment. In this context, theradiocaesium solid/liquid distribution coefficient, the radiocaesium interceptionpotential, the K and NH4

C concentration in the soil solution, the overall soil Kstatus and the ageing rate were determined as potential important parameters for theprediction of the soil-to-wood radiocaesium TFs. The influence of these parameterscould, however, not be fully quantified in this study.

Acknowledgements

The authors want to thank Dr. T. Verwijst and the laboratory of short rotationforestry of the Swedish University of Agricultural Sciences, Uppsala for theirassistance with the sampling of all willow SRC plantations in central-east Sweden,for their reception and the possibility they created to work in their laboratories fordrying the wood and mixing the soil samples.


References

Absalom, J.P., Young, S.D., Crout, N.M.J., Nisbet, A., Woodman, R.F.M., Smolders, E., Gillet, A.G.,

1999. Predicting soil to plant transfer of radiocaesium using soil characteristics. Environmental Science

and Technology 33, 1218–1223.

Absalom, J.P., Young, S.D., Crout, N.M.J., Sanchez, A., Wright, S.M., Smolders, E., Nisbet, A.F., Gillet,

A.G., 2001. Predicting the transfer of radiocaesium from organic soils to plants using soil

characteristics. Journal of Environmental Radioactivity 52, 31–43.

Alexakhin, R.M., 1993. Countermeasures in agricultural production as an effective means of mitigating the

radiological consequences of the Chernobyl accident. The Science of the Total Environment 137, 9–20.

Belli, M., Tikhomirov, F., 1996. The behaviour of radionuclides in natural and semi-natural environments.

EUR 16531 (final report ECP 5), ISBN 92-827-5197-X, European Commission, Brussels, p. 147.

Belli, M., Sansone, U., Ardiani, R., Feoli, E., Scimone, M., Menegon, S., Parente, G., 1995. The effect of

fertilizer applications on 137Cs uptake by different plant species and vegetation types. Journal of

Environmental Radioactivity 27, 75–89.

Bilo, M., Steffens, W., Fuhr, F., Pfeffer, K.-H., 1993. Uptake of 134/137Cs in soil by cereals as a function of

several soil parameters of three soil types in upper Swaia and North Rhine-Westphalia (FRG). Journal

of Environmental Radioactivity 19, 25–39.

Bondar, P.E., Dutov, A.I., 1992. Parameters of radiocaesium transfer into oats harvest on lime soils in

connection with the application of mineral fertilizers and chemical meliorants. Collections of Scientific

Works – Problems of Agricultural Radiology, Kiev, Ukraine, pp. 125–132.

Bunzl, K., Schimmack, W., Kreutzer, K., Schierl, R., 1989. Interception and retention of Chernobyl

derived 134Cs, 137Cs and 106Ru in a spruce stand. The Science of the Total Environment 78, 77–87.

Chhabra, R., Pleysier, J., Cremers, A., 1975. The measurement of the cation exchange capacity and

exchangeable cations in soil: a new method. In: Bailey, S.W. (Ed.), Proceedings of the International

Clay Conference on Mexico. Applied Publications, Wilmette, IL, pp. 439–449.

Davydchuk, V., 1997. Ecosystem remediation in radioactively polluted areas: the Chernobyl experience.

Ecological Engineering 8, 325–336.

Day, P.R., 1965. Particle fraction and particle size analysis. In: Black, C.A., Evans, D.D., Whit, J.L.,

Ensminger, L.E., Clarck, F.E. (Eds.), Methods of Soil Analysis – Part 1 – Physical and Mineralogical

Properties, Including Statistics of Measurements and Sampling. American Society of Agronomy,

Wisconsin, pp. 556–561.

Delvaux, B., Kruyts, N., Cremers, A., 2000. Rhizospheric mobilization of radiocaesium in soils.

Environmental Science and Technology 34, 1489–1493.

Ehlken, S., Kirchner, G., 2002. Environmental processes affecting plant root uptake of radioactive trace

elements and variability of transfer data: a review. Journal of Environmental Radioactivity 58, 97–112.

Erten, H., Aksoyoglu, S., Hatipoglu, S., Gokturk, H., 1988. Sorption of caesium and strontium on

montmorillonite and kaolinite. Radiochimica Acta 44/45, 147–151.

Gommers, A., 2001. Radiocaesium Uptake and Cycling in Willow Short Rotation Coppice. PhD.

Dissertationes de Agricultaura, 467. KU Leuven, Leuven, Belgium, pp. 133.

Gommers, A., Thiry, Y., Vandenhove, H., Vandecasteele, C.M., Smolders, S., Merckx, R., 2000.

Radiocaesium uptake by one-year-old willows planted as Short Rotation Coppice. Journal of

Environmental Quality 29, 1384–1390.

Goor, F., Davydchuk, V., Vandenhove, H., 2003. GIS-based methodology for Chernobyl contaminated

land management through biomass conversion into energy – a case study for Polessie, Ukraine.

Biomass and Bioenergy 25, 409–421.

GOPA, 1996. Belarus: Study of Alternative Biodiesel Sources in Relation with Soil Decontamination.

TACIS/REG93, pp. 165.

Hinsinger, P., 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in the

rhizosphere. Advances in Agronomy 64, 225–265.

Hinsinger, P., Jaillard, B., 1993. Root-induced release of interlayer potassium and vermiculization of

pholopite as related to potassium depletion in the rhizosphere of ryegrass. Journal of Soil Science 44,

525–534.


Hoagland, D.R., Martin, J.C., 1933. Absorption of potassium by plants in relation to replaceable,

non-repleacable and soil solution potassium. Soil Science 36, 1–33.

Hubbard, L., 1999. Occurrence of 137Cs in forest biofuels. SSI News 7 (1), 1–3.

Jackson, W.A., Craig, D., Lugo, H.M., 1965. Effects of various cations on cesium uptake from soils and

clay suspensions. Soil Science 99, 345–353.

Jossart, J.-M., Goor, F., Nerinckx, X., Ledent, J.F., 1999. Le taillis a tres courte rotation. 2e edition.

Laboratory of the Ecology on major cultures, Faculty of Agricultural Sciences, Catholic University of

Louvain la Neuve, UCL, Belgium.

Jungk, A., Claasen, N., 1986. Availability of phosphate and potassium as the result of the interaction

between root and soil in the rhizosphere. Zeitung fur Pflanzeneraehr. Bodenkunde 149, 411–427.

Kaletnyk, M.M., 1999. Forestry management in conditions of radioactive contamination: experience

and problems. In: Linkov, I., Schell, W.R. (Eds.), Contaminated Forests: Recent Developments

in Risk Identification and Future Perspectives. Kluwer Academic Publishers, The Netherlands,

pp. 281–287.

Komarneni, S., 1978. Cesium sorption and desorption behaviour of kaolinites. Soil Science Society

America Proceedings 42, 531–532.

Kuhn, W., Handl, J., Schuller, P., 1984. The influence of soil parameters on 134Cs uptake by plants from

long-term fall-out on forest clearings and grassland. Health Physics 46, 1083–1093.

Ledin, S., 1996. Willow wood properties, production and economy. Biomass and Bioenergy 11, 75–83.

Lembrechts, J., 1993. A review of literature on the effectiveness of chemical amendments in reducing the

soil-to-plant transfer of radiostrontium and radiocaesium. The Science of the Total Environment 137,

81–98.

Lewyckyj, N., 2000. Retention of Radiocaesium Traces in Clays: A Dynamic Semi-empirical Modelling

Approach. PhD Thesis. Faculty of Agricultural Sciences, Catholic University of Louvain-la-Neuve,

UCL, Belgium.

Maraziotis, A., 1992. Soil-to-plant concentration factor and dependence on soil parameters. Radiological

Protection 12, 77–84.

Markewitz, D., Richter, D.D., 2000. Long-term soil potassium availability from a Kanhapludult to an

aggrading loblolly pine ecosystem. Forest Ecology and Management 130, 109–129.

McGee, E.J., Synnott, H.J., Johanson, K.J., Fawaris, B.H., Nielsen, S.P., Horrill, A.D., Kenedy,

V.H., Barbayiannis, N., Veresoglou, D.S., Dawson, D.E., Colgan, P.A., McGarry, A.T., 2000.

Chernobyl fall-out in Swedish spruce forest ecosystems. Journal of Environmental Radioactivity

48, 59–78.

Melin, J., Wallberg, L., Suomela, J., 1994. Distribution and retention of cesium and strontium in Swedish

boreal forest ecosystems. The Science of the Total Environment 157, 93–105.

Melin, J., Moberg, L., Askbrant, S., Vandecasteele, C., Lewyckyj, N., Cremers, A., Valcke, E., Rauret, G.,

Vidal, M., Vallejo, V.R., Sandalls, J., Hinton, T., Arkhipov, N., Arkhipov, A., Loginova, L.,

Alexakhin, R., Kruglov, S., Ratnikov, A., Jigareva, T., Firsakova, S., Grebenschikova, N., Timofeev,

S., Podolyak, A., Prister, B., Loschilov, S., Ivanov, Y., Bondar, P., 1996. In: Rauret, G., Firsokova, S.

(Eds.), Transfer of Radionuclides through the Terrestrial Environment to Agricultural Products,

Including the Evaluation of Agrochemical Practices EUR 16528. European Commission, Luxem-

bourg, pp. 179.

Mengel, K., 1985. Dynamics and availability of major nutrients in soils. Advances in Soil Science 2,

477–481.

Minotti, P.L., Craig, D., Jackson, W.A., 1965. High caesium uptake in wheat seedlings cultured with

ammonia. Soil Science Society Proceedings, Notes 220–221.

Mortland, M.M., 1958. Kinitics of potassium release from biotite. Soil Science Society of America

Proceedings 22, 503–508.

Nisbet, A.F., 1993. Effect of soil-based countermeasures on solid-liquid equilibria in agricultural soils

contaminated with radiocaesium and radiostrontium. The Science of the Total Environment 137,

99–118.

Nisbet, A.F., Woodman, R.F.M., 2000. Soil-to-plant transfer factors for radiocaesium and radiostrontium

in agricultural systems. Health Physics 78, 279–288.


Nisbet, A.F., Woodman, R.F.M., Haylock, R.G.E., 1999. Recommended Soil-to-plant Transfer Factors

for Radiocaesium and Radiostrontium for Use in Arable Systems. NRPB-R304 NRPB, Didcot, UK,

pp. 50.

Perttu, K.L., 1998. Environmental justification for short rotation coppice in Sweden. Biomass and

Bioenergy 16, 291–297.

Philippot, S., 1996. Simulation models of short rotation forestry production and coppice biology. Biomass

and Bioenergy 11, 85–93.

Ravila, A., Holm, E., 1994. Radioactive elements in the forest industry. The Science of the Total

Environment 157, 339–356.

Renaud, P., Maubert, H., 1997. Agricultural countermeasures in the management of the post-accidental

situation. Journal of Environmental Radioactivity 35, 53–69.

Ronneau, C., Sombre, L., Myttenaere, C., Andre, P., Vanhouche, M., Cara, J., 1991. Radiocaesium and

potassium behaviour in forest trees. Journal of Environmental Radioactivity 14, 259–268.

Rosen, K., 1986. Transfer of radiocaesium in sensitive agricultural environments after the Chernobyl

fall-out in Sweden. II. Marginal and seminatural areas in the country of Jamtland. The Science of the

Total Environment 182, 135–138.

Rosenqvist, H., Roos, A., Ling, E., Hector, B., 2000. Willow growers in Sweden. Biomass and Bioenergy

14, 259–268.

Sanchez, A.L., Wright, S.M., Smolders, E., Naylor, C., Stevens, P.A., Kennedy, V.H., Dodd, B.A.,

Singleton, D.L., Barnett, S.L., 1999. High plant uptake of radiocaesium from organic soil due to Cs

mobility and low K content. Environmental Science and Technology 33, 2752–2757.

Sanchez, A.L., Smolders, E., Van den Brande, K., Merckx, R., Wright, S.M., Naylor, C., 2002.

Predictions of in situ solid/liquid distribution of radiocaesium in soils. Journal of Environmental

Radioactivity 63, 35–47.

Sandalls, J., Bennett, L., 1992. Radiocaesium in upland herbage in Cumbria, UK: a three-year field study.

Journal of Environmental Radioactivity 16, 147–165.

Segal, M.G., 1993. Agricultural countermeasures following deposition of radioactivity after a nuclear

accident. The Science of the Total Environment 137, 31–48.

Shaw, G., Hewamanna, R., Lillywhite, J., Bell, J.N.B., 1992. Radiocaesium uptake and translocation in

wheat with reference to the transfer factor concept and ion competition effects. Journal of

Environmental Radioactivity 16, 167–180.

Shaw, G., Robinson, C., Holm, E., Frissel, M.J., Crick, M., 2001. A cost-benefit analysis of long-term

management options for forests following contamination with Cs-137. Journal of Environmental


Smolders, E., Van den Brande, K., Merckx, R., 1997. Concentrations of 137Cs and K in soil

solution predict the availability of 137Cs in soils. Environmental Science and Technology 31,

3432–3438.

Szekely, J.G., Amiro, B.D., Rasmussen, L.R., Ford, B., 1994. Environmental Assessment of Radiological

Consequences for Forestry in Contaminated Areas of the Republic of Belarus. World Bank Report,

57 p.

Vandenhove, H., Thiry, Y., Gommers, A., Goor, F., Jossart, J.-M., Gavfert, T., Holm, E., Roed, J., 1999.

RECOVER: Relevancy of Short Rotation Coppice Vegetation for the Remediation of Contaminated

Areas. Project FI4-CT95-0021c. Final report. BLG 826, SCK�CEN, Mol, Belgium, pp. 63.

Vandenhove, H., Thiry, Y., Gommers, A., Goor, F., Jossart, J.M., Holm, E., Gavfert, T., Roed, J.,

Grebenkov, A., Timofeyev, S., 2001. Short rotation coppice for re-valuation of contaminated land.

Journal of Environmental Radioactivity 56, 157–184.

Vandenhove, H., Smolders, E., Cremers, A., 2003. Potassium bentonites reduce radiocaesium availability

to plants. European Journal of Soil Science 54, 91–102.

Vandenhove, H., Goor, F., Timofeyev, S., Grebenkov, A., Thiry, Y., 2004. Short rotation coppice as

alternative land-use of Chernobyl contaminated areas in Belarus. International Journal of

Phytoremediation 6 (2), 1–18.

Voigt, G., Eged, K., Hilton, J., Howard, B., Kis, Z., Nisbet, A., Oughton, D.H., Rafferty, B., Salt, C.A.,

Smith, J.T., Vandenhove, H., 2000. Factors to be Considered Before the Implementation of


Radioactive Countermeasures Following a Nuclear Accident. EULEP/EURADOS/IUR Concerted

Action. IUR, pp. 23.

Vovk, I.F., Blagoyev, V.V., Lyashenko, A.N., Kovalev, I.S., 1993. Technical approaches to

decontamination of terrestrial environments in the CIS (former USSR). The Science of the Total

Environment 137, 49–63.

Waegeneers, N., 2002. Soil processes Affecting the Bioavailability of Radiocaesium. PhD. Dissertationes

de Agricultaura, 523. KU Leuven, Leuven, Belgium, pp. 150.

Wauters, J., Elsen, A., Cremers, A., Konoplev, A.V., Bulgakov, A.V., Comans, R.N.J., 1996. Prediction of

the solid/liquid distribution coefficients of radiocaesium in soils and sediments. Part one: A simplified

procedure for the solid phase characterization. Applied Geochemistry 11, 589–594.

Yera, T.S., Vallego, V.R., Valcke, E., Colle, C., Forstel, H., Millan, R., Jouglet, H., 1999. 137Cs and 90Sr

root uptake prediction under close-to-real controlled conditions. Journal of Environmental


Radiocaesium soil-to-wood transfer in commercial willow short rotation coppice on contaminated farm...

Documents

Transcript of Radiocaesium soil-to-wood transfer in commercial willow short rotation coppice on contaminated farm...