Radioactive Contamination of Wild Boars: Measurements and Modelling
Ulrich Fielitz *
Forstweg 11, D-29313 Hambühren, Germany
* Corresponding author.
E-mail address: firstname.lastname@example.org
Buenos Aires 10, Argentinia
The radioactive contamination of wild boar caused by 137Cs has been measured in the Bavarian Forest, which is a German region particularly affected by the Chernobyl accident. The radioactive contamination of wild boar (Sus scrofa) is relatively high compared with other game and shows a slow increase for many years. Additional experimental data are available by analyses of the composition of the wild boar’s stomach content, of the food contamination and of the soil contamination. In order to explain the high contamination of wild boars and the unusual temporal course of the activity concentration a radio-ecological model has been developed. The model contains a soil model with several soil layers and considers the soil-to-plant transfer, food intake and the transfer to the wild boar’s meat. The model quantifies the importance of deer truffles (Elaphomyces granulatus), which is one of the food components and predominantly responsible for the contamination of wild boar. Model assumptions with respect to the food intake can explain the large variation of the contamination of wild boar. Furthermore, the change of the food spectrum in so-called mast years is taken into account. Different scenarios were developed which lead to prognosis of the future contamination of wild boar. It can be shown that the future contamination is strongly related to the processes of radio-caesium, which take place in the soil layers.
Keywords: 137Cs; Wild boar; Gamma-spectrometric measurements; Radioecological model; Chernobyl accident
The Bavarian Forest is a mountainous region in Germany near the frontier to the Czech Republic. The forests near the villages Bodenmais and Bayerisch Eisenstein as well as the Northwestern part of the National Park Bavarian Forest were affected particularly by the Chernobyl accident in 1986. For this, in the area numerous examinations of the radioactive contamination of the soil, plants and wildlife have been carried out regularly since 1987. The examination area has an extension of approximately 400 km2. A mean value of 53 000 Bq 137Cs per m2 can be estimated for the activity, which has been deposited there few days after the Chernobyl accident. However, still in May 2000 in some parts of the area activities of about 73 000 Bq m-2 were measured. The 137Cs activity concentrations of plants, beers, mushrooms and game belong to the highest, which have been measured in Germany. The activity concentrations of wild boars, but also of roe deer and red deer, have been measured in several research projects. The measurements show, as do similar examinations in other European areas (Hecht, 1997; Hecht, 2000; Vilic et al. 2005), that wild boars are particularly contaminated.
In order to explain the high radioactive contamination numerically and to be able to forecast future contaminations a radioecological was needed. This model should be able to take into account the experimental data of food habits and experimental data, which were acquired in the past. Although there exist radioecological models for animals no model is available which puts special emphasis on the aspects, which have to be considered for wild boars living in wilderness. The radioecological model, which has been developed, bases partly on the ECOSYS model (Müller and Pröhl, 1993). The radioecological model describes processes of radio-caesium in the soil, the root uptake of radio-caesium by plants, which serve as food for wild boar and the transfer of radio-caesium to the animal. The strong variation of the measured activity concentrations can be modelled by using Monte-Carlo methods, which allow simulating the variation of model parameters.
The activity measurements of the soil, plant and animal samples have been carried out at the Institute of Forest Botany of the University Göttingen with two germanium detectors and a germanium borehole detector (Hersteller, Typenbezeichnungen). The activities of 137Cs were usually measured until a statistical counting rate error of 5 % was achieved at a confidence level of 95 %.
2.1 Soil measurements
Soil samples were taken in the examination area with a drilling tube, in such a way, that the vertical profile remained conserved. The total depth of the soil profiles was 20 cm to 30 cm. Each series consisted of 10 drilling cores, which were divided into layers of 2 cm. The layers of a series with the same depth were mixed, dried at 105 oC and sieved (mesh size 2 mm). The fine soil particles were weighted and pulverized; afterwards the activity has been measured in measuring cups. In order to meet the above mentioned counting rate error, for some soil samples, which were taken from a depth of more than 15 cm, measuring times of more than 200 000 were necessary due to low activities with less than 5 Bq 137Cs kg-1. In the activity range of 0-2 Bq the counting rate error was 5-20%.
2.2 Measurement of plants
At each time of sampling 20 to 40 single leaves of each plant species have been collected in the examination area. The leaves of a species were mixed, stored in polyethylene bags and cool-stored during the mostly 4 days lasting outdoor activities. Later, the fresh weight and the dry weight of the sampled leaves were measured in the laboratory. If necessary, soil particles at the roots were rinsed with water before. The drying of the samples took place at 105 C in drying cupboards. For the gamma-spectrometric measurements the samples were pulverized in a mill and filled into measuring cups.
2.3 Analysis of the stomach’s content and measurements of the meat of wild boar
The stomachs of wild boars were taken out in one piece, afterwards deep-frozen until the analysis. Later in the laboratory the stomachs were opened and emptied, remaining food rests were peeled carefully from the inner stomach sides. The content of the stomach was then weighted. In order to determine the composition of the stomachs a pre-defined scheme of spot checks was used, which separates coarse components, mash of plants and soil components (Fielitz, 2005). In many cases parts of leaves, grasses, ferns and mushrooms could be found, making easier the determination of the stomach content.
Although the stomach liquids digest the tissue of leaves, epidermises were often kept preserved. Differences in size and structure of the cell types, hairs and wax layers are characteristic for different plants types and thus allow the identification. A determination key (Zettel, 1974), furthermore samples conserved in the laboratory and the experience in handling food components of stomachs acquired in many years were useful to identify the plant fragments.
Wild boars do not spurn animals as part of the food. Food components from animals were identified through the characteristic structure of the animal tissue and through characteristics of kemp hairs (Grannenhaare). For the identification a determination key is available (Day, 1966). When bigger parts of animal tissue were found (e.g. hare), the scheme of spot checks has not been used. In these cases the fresh weight was determined directly and put in relation to the total weight of the total stomach content.
Furthermore, the contribution of soil in the stomachs was determined. The analysis methods have been different for mineral soil and the organic humus soil (Fielitz, 2005). In The latter case a method has been applied to distinguish between plants and organic material in the soil.
For the measurement of the activity of wild boars, only muscle meat from the forearm (Musculus flexor digitorum superficiales/profundus) was used. The muscle meat was taken from the bone; adherent sinews and fat were eliminated. Then the meat was homogenized and examined gamma-spectrometric in measuring cups. Finally, the activity data were calculated in relation to the fresh weight.
2.4 Radioecological model
2.4.1 Soil model and transfer of 137Cs to plants
In order to describe the vertical distribution of 137Cs in the soil a model was developed which divides the soil into layers of 2 cm thickness, which lie one above the other. The nuclides migrate from a layer into the next below one. If nuclides are fixated in the soil, they cannot participate in the migration process, until they are desorbed again. For each layer a differential equation can be formulated, which describes the processes (migration, fixation and desorption) of 137Cs in the soil. For the most upper layer, litter fall and dripping water from the crowns of trees are taken into account. Both result in an additional entry of nuclides into the soil. The system of differential equations for the fluxes of 137Cs in the soil layers is the given by:
for i = 1 (1a, 1b)
for i = 2, ..., NT (2a, 2b)
i: Index of soil layer
lm(i): Time constant of migration (a-1) from soil layer (i) to soil layer (i + 1)
lf(i), ld(i): Time constants (a-1) of fixation and desorption in soil layer (i)
lb: Time constant of litter fall and water from the crowns of trees (a-1).
The numerical solution of the system of differential equations (Törnig, 1979) with initial values n(i)(t=0) und f(i)(t=0), which take into account the contamination from nuclear weapon tests until the early 1960s, leads to time functions of the activity concentration in the soil layers:
Radionuclides available for root uptake (3)
Fixated radionuclides (4)
Activity concentration (Bq kg-1) of soil layer (i)
Adep: Deposited activity (Bq m-2)
r(i): Density (kg m-3) of soil layer (i)
d = 2 cm: Thickness of soil layer
t = 30 a: Half-life of radioactive decay 137Cs
Plants are mainly contaminated by root uptake of radionuclides. The activity concentration of plants can assumed to be proportional to the concentration in the soil, where the plant transfer factor is the proportional factor. Due to the multi-layer soil model it has to be summed up over the layers in the root zone. For most of the plants the same number of soil layers has been considered. However, for deer truffles, which fructify subterranean, another set of soil layers has been chosen.
2.4.2 Food intake and processing
The resorption of radionuclides through the skin and the inhalation of radionuclides are reasons for the contamination of animals. However, apart from the early phase after a nuclear release, the by far most important pathway of contamination is the ingestion of radionuclides via the gastrointestinal tract of the animal. In the following it is thus assumed that the radionuclides reach the organism of the animal by intake of contaminated food. The activity intake of wild boars is given by
A(t): Activity intake of wild boar (Bq d-1)
NF: Number of food components
Ck(t): Activity concentration (Bq kg-1) of food component k
Ik(t): Rate of intake (kg d-1) of food component k
The activity is determined essentially by the activity of the ingested food, but also by the processing of 137Cs in the organism of the animal. The processing of 137Cs, e.g. the transition of radionuclides into the meat and the excretion of radionuclides, is described in the model by the animal transfer factor and the biological half-life, which are both nuclide-dependent model parameters. The activity concentration of wild boars is calculated by integration of the temporal varying food activity (Müller and Pröhl, 1993):
Cm(T): Activity concentration (Bq kg-1) of wild boars at time T
Tm: Animal transfer factor (d kg-1)
tbiol: Biological half-life (d)
2.4.3 Variability of model parameters
The results of the model predictions should be able to consider the variability of model parameters. For this purpose probability density functions were assigned to selected model parameters. With Monte-Carlo methods NMC=100 values are generated, so that finally NMC parameter sets are available. Latin Hypercube Sampling was applied in order to optimise the distribution of the parameter values (McKay et al., 1979). The radioecological model is then carried out with each parameter set, so that for each pint in time NMC different activity concentrations were calculated. These activity concentrations can finally be used to calculate percentile values (medians, 5% and 95 % percentiles), which reflect the influence of the variability of the model parameters. In this work the influence of variable food intake rates on the model output was examined. The variability was described mostly by triangular probability density functions.
3. Results and discussion
The parameters of the multi-layer soil model have been adapted in such a way, that the modelled activities in the soil layers agree well with the measured values. The best agreement has been achieved with a migration rate of 0.59 a-1 in the upper (organic) layers (0–6 cm) and 0.50 a-1 in the deeper (mineral) layers (>6 cm), a fixation rate of 0.59 a-1 and a desorption rate of 0.68 a-1. From measured data (Fielitz, 1994), the rate caused by litter fall and water dripping from the crowns of trees has been estimated to 0.016 a-1. Furthermore, densities of the soil layers as measured in Bodenmais have been used (fig. 1).
Fig. 1: Mean 237Cs distribution of 10 soil profiles in 2004. The error bars represent measured minimum and maximum values.
The analysis of the stomach’s content yields a large variety of food. For the application in the radioecological model, the food components have been subsumed to food groups, and mean values of the corresponding activity concentrations have been calculated. These activities were then used to adapt the plant transfer factors (table 1). From the stomach analyses the mean values of the intake rate contribution of each food group can be estimated (table 2a). For so-called mast years the food spectrum has to be changed in accordance with the stomach analysis (table 2b).
Measured 137Cs activities, soil ranges and plant transfer factors, which were used in the radioecological model.
Food intake of wild boars as portions of a total amount of 1.8 kg d-1.
Food intake of wild boars from October 2003 to April 2004 (mast period) as portions of a total amount of 1.8 kg d-1.
The activity concentrations of wild boars have remained on a high level since the Chernobyl accident. The 137Cs contamination of wild boar increased from 1987 to 2004 statistically non-significant with a half-life of +78 years. The mean contamination in 2004 was approximately 6710 Bq kg-1 (91 wild boars). In 1988 the mean value was 4810 Bq kg-1 (34 wild boars) in the fresh substance. The contamination of roe deer and red deer however clearly decreased. While in ruminants, as are roe deer and red deer, bacteria in the gastro-oesophageal vestibule process the food, wild boars are dependent on food decomposition by enzymes in saliva, stomach and the small intestine (Kirchgiesser, 1999). Due to the high solubility in water, caesium is resorbed in the stomach and transferred by the bloodstream to the tissue of the wild boar, in particular to the muscle meat. The indigestion system is thus a reason for the higher animal transfer rate. The second reason is the high body mass of wild boars, which surpasses those of roe deer and red deer noticeably. It is known that, the higher the body mass is, the higher is the transfer rate (Nalezinski, 1995). The third reason for the high contamination of wild boars is the complex food spectrum. Wild boars are omnivores and they like in particular to dig up deer truffles, which is mushroom specie widespread in the Bavarian Forest. Deer truffles have to be considered as the most important source of contamination, since they contribute predominantly (with 82%) to the 137Cs input of wild boars.
Fig. 2: Measured values and model predictions of the 137Cs activity concentration of wild boars in the examination area Bavarian Forest, 1987-2004.
The activity concentrations of wild boars have been calculated with the radioecological model by using an animal transfer factor of 2.0 d kg-1 and a biological half-life of 20 d (fig. 2). The modelled data show the median as well as the 5% and 95% percentiles. The model predicts a decrease of the 137Cs contamination in the first years after the Chernobyl accident, because of the decreasing activities of the food spectrum, which is caused by fixation of radiocaesium in the soil. The decrease of the 137Cs contamination in this time period is also indicated by the measured values. Since 1990 the contamination of wild boars increases slightly, because the radionuclides migrate increasingly into deeper soil layers and become thus available for the uptake through the mycels of deer truffles. At the same time the activity of the aboveground fructifying food components was decreasing. With regard to the intake of radio-caesium in the whole food spectrum, the high-contaminated deer truffles however have more than compensated this decrease. Since deer truffles contribute predominantly to the contamination of wild boars, the vertical distribution of the activity in the soil is decisive for the question, how much radionuclides become available for the mycel of deer truffles. In order to answer this question, two scenarios with different model parameters have been calculated. In the first scenario the soil parameters (rates of migration, fixation and desorption) are the same as mentioned above. In the second scenario the migration rate was set to 0.1 a-1 for all soil layers, i.e. set to a value approximately 5 times smaller than in the first scenario.
Fig. 3: Measured values and model predictions for two scenarios (a) and (b) of the 137Cs activity concentration of wild boars in the examination area Bavarian Forest, 1987-2030.
The first scenario is a best-case scenario (fig. 3a). The radio-caesium is transported into deeper soil layers and afterwards leaving the zone of the deer truffle mycels. In the second scenario, the worst-case scenario, a much smaller migration rate is assumed. As consequence the radionuclides are available more time for the uptake by the mycels. This leads to stagnating 137Cs contamination of deer truffles and thus also of wild boars (fig. 3b). The median of the measured values has been 4175 Bq kg-1 in 2004. This value agrees well with the median model prediction of both scenarios, 3 940 Bq kg-1 and 4070 Bq kg-1, respectively. The first scenario predicts a slow decrease of the contamination of wild boars in the coming decades. However, in 2030 the contamination will still be 1720 Bq kg-1. According to the worst-case scenario the contamination of wild boars will increase slightly to 4189 Bq kg-1 in 2010, and then decrease slowly. The model predicts in this case a median of 3351 Bq kg-1 in 2030. In both scenarios even the 5 % percentiles of the activity will be above the limit value of 600 Becquerel.
In mast periods wild boars are fed mainly with beechnuts. This results in a drastic decrease of the 137Cs contamination of wild boars, as show the measured values from October 2003 until April 2004. Due to the assumed change of the food spectrum, the decrease is reproduced well by the modelled data. The data furthermore show, that only in mast periods the meat of wild boar can be expected to have values below the limit value of 600 Bq kg-1.
The measured activities of the meat depend strongly on the food components, which wild boars intake at a specific point in time. For example, in 4 samples of meat, activities between 81 Bq kg-1 und 39539 Bq kg-1 have been measured. The variation of the food intake, represented in the model by probability density functions, explains largely the differences of the measured values. It should be noted however, that other factors, such as the spatial variation of the deposited activities, would overlap and enlarge the modelled variation. Summarizing the performance of the radioecological model, it can be stated, that by using the measured 137Cs data of the food components the modelled medians are able to describe the temporal course of the radioactive contamination of wild boars.
Finally, it should be noted, that winning of representative data about the inventory and dynamics of radiocaesium is still a problem. Sources of uncertainty are not only gamma-spectrometric measurements or analyses, but to a large extent the sampling of soil and also of plants and animal meat. For this, efforts to develop standards in the sampling and the processing of samples have to be supported (Barbizzi, 2004; Belli, 2001; IUPAC, 2003).
4. Summary and conclusions
Gamma-spectrometric measurements of soil, plants and meat of wild boars, furthermore the analyses of the stomach content of wild boars have supported the development of a radioecological model. The model starts with the mathematical description of physical processes in soil layers; the model finally leads to the calculation of the activity concentrations of wild boars. The predominant influence of deer truffles in the subterranean on the future course of the activity concentration of wild boars could be shown. The results obtained demonstrate, that the physical transport of 137Cs through the soil leads to sources of environmental radioactivity below the ground surface, which may affect plants and animals. Future work should thus study the behaviour of radionuclides in the root zone, in particular for deep-root plants.
Wild boars show a high reproduction rate, also under changing environmental conditions. This led to a high wild boar population in the past years. As a consequence there is a growing number of wild boars, whose meat is unsuitable as food for humans. The model results show, that the risk to eat high-contaminated meat will remain for many years, even decades. It should thus be taken into consideration to measure the radioactivity of the meat of each wild boar, which is shot with the intention to serve as food for humans.
Barbizzi, S., Zorzi, P., Belli, M., Sansone, U., Stellato, L., Barbina, M., Delusa, A., Menegon, S., Sandro, A., Coletti, V., 2004. Characterisation of a reference site for quantifying uncertainties related to soil sampling. Environmental Pollution. 127(1), 131-135.
Belli, M., Zori, P., 2001. Assessment of uncertainty associated with soil sampling in agricultural, semi-natural, urban and contaminated einvironments. In: IUR Newsletter No. 37. International Union of Radioecology.
Day, M.G., 1966. Identification of hair and feather remains in the gut and feaces of stoats and weasels. J. Zool.. 148, 201-217.
Fielitz, U., 1994. Radioaktivität in Wildtieren. Schriftenreihe Reaktorsicherheit und Strahlenschutz. BMU-1994-408.
Fielitz, U., 2005. Untersuchungen zum Verhalten von Radiocäsium in Wildschweinen und anderen Biomedien des Waldes. Abschlussbericht an das Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit.
Hecht, H., 1997. Radiocäsiumverteilung und –trends bei Wildbret in Bayern. Abschluss-bericht zum Forschungsvorhaben im Auftrag des Bayerischen Staatsministeriums für Landesentwicklung und Umweltfragen. Bundesanstalt für Fleischforschung, Kulmbach.
Hecht, H., 2000. Radiocäsiumgehalte des Schwarzwildbretes speziell im Gebiet des Bayerischen Waldes. Abschlußbericht zum Forschungsvorhaben. Bundesanstalt für Fleischforschung, Kulmbach.
IUPAC; 2003: Terminology in soil sampling, Draft, International Union of Pure Applied Chemistry.
Kirchgiesser, 1999. ???
McKay, M.D., Conover, W.J., Beckman, R.J., 1979. A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code. Technometrics 21, 239-245.
Müller, H., Pröhl, G., 1993. ECOSYS-87: A Dynamic Model for Assessing Radiological Consequences of Nuclear Accidents. Health Physics 64(3), 232-252.
Nalezinski S., Rühm, W., Wirth, W., 1995: Development of a general equation to determine the transfer factor feed-to-meat for radiocesium of the basis of the body mass of domestic animals. Health Physics 70(5), 717-721.
Törnig, W., 1979. Numerische Mathematik für Ingenieure und Physiker. Band 2: Eigenwertprobleme und numerische Methoden der Analysis. Berlin: Springer Verlag.
Vilic, M., Barisic, D., Kraljevic, P., Lulic, S., 2005. 137Cs concentration in meat of wild boars (Sus scofa) in Croatia a decade and half after the Chernobyl accident. Journal of Environmental Radioactivity 81, 55-62.
Zettel, J.; 1974. Mikroskopische Epidermiskennzeichen von Pflanzen als Bestimmungs-hilfen. Mikrokosmos 63(4), 106-111, ( 5) 136-140, (6) 177-181, (7) 201-206.