Influence of fertilizing on the (137)Cs soil-plant transfer in a spruce forest of Southern Germany.

Fertilization with 2.5t/ha limestone: (83% CaCO(3), 8% MgO, 6% K(2)O, 3% P(2)O(5)) reduces the (137)Cs transfer from spruce forest soil into plants like fern (Dryopteris carthusiana) and blackberry (Rubus fruticosus) by a factor of 2-5 during at least 11 years as measured by the aggregated transfer factor T(ag). In 1997 and 2006 these results were confirmed by additional measurements of the (137)Cs transfer factor TF, related to the root zone (O(h) horizon), which were explained by the selective sorption of (137)Cs in the root zone by measurements of the Radiocaesium Interception Potential (RIP) in fertilized (RIP>179meq/kg) and non-fertilized soils (RIP<74meq/kg).

Seasonality of 137Cs in roe deer from Austria and Germany.

Empirical data on the (137)Cs activity concentration in meat of roe deer (Capreolus capreolus) roaming in 3 spruce forest areas and one peat bog area are presented and compared. They cover time series of nearly 20 years after a spike contamination in 1986 originating from Chernobyl. A model is presented which considers three soil compartments to describe the change of the availability of (137)Cs with time. The time-dependency of the (137)Cs activity concentration in meat of roe deer is a combination of two components: (1) an exponential decay and (2) a peak in the second half of each year during the mushroom season. The exponential decay over the years can be described by a sum of two exponential functions. The additional transfer of (137)Cs into roe deer during the mushroom season depends on precipitation. On the peat bog the (137)Cs activity concentration in roe deer is higher and more persistent than in spruce forest.

Migration and bioavailability of (137)Cs in forest soil of southern Germany.

To give a quantitative description of the radiocaesium soil-plant transfer for fern (Dryopteris carthusiana) and blackberry (Rubus fruticosus), physical and chemical properties of soils in spruce and mixed forest stands were investigated. Of special interest was the selective sorption of radiocaesium, which was determined by measuring the Radiocaesium Interception Potential (RIP). Forest soil and plants were taken at 10 locations of the Altdorfer Wald (5 sites in spruce forest and 5 sites in mixed forest). It was found that the bioavailability of radiocaesium in spruce forest was on average seven times higher than in mixed forest. It was shown that important factors determining the bioavailability of radiocaesium in forest soil were its exchangeability and the radiocaesium interception potential (RIP) of the soil. Low potassium concentration in soil solution of forest soils favors radiocaesium soil-plant transfer. Ammonium in forest soils plays an even more important role than potassium as a mobilizer of radiocaesium. The availability factor - a function of RIP, exchangeability and cationic composition of soil solution - characterized reliably the soil-plant transfer in both spruce and mixed forest. For highly organic soils in coniferous forest, radiocaesium sorption at regular exchange sites should be taken into account when its bioavailability is considered.

Ecological half-lives of 90Sr and 137Cs in terrestrial and aquatic ecosystems.

This paper describes the long-term behaviour of (90)Sr and (137)Cs in foods, feeds and a variety of environmental media. The long-term behaviour is quantified by means of the ecological half-life which integrates all processes that cause a decrease of activity in a given medium such as leaching, fixation and erosion. A large number of long-term time series of concentrations of radiocaesium and radiostrontium in these media have been identified and re-evaluated using a standardised statistical procedure to establish reference data sets of ecological half-lives. By example of undisturbed soils and marine water bodies it is shown that the ecological half-life concept is questionable if the distribution of the radionuclide of interest within the medium studied is non-uniform and if mixing and transport processes within this medium, therefore, are of considerable importance during the time period of observation.

The "AQUASCOPE" simplified model for predicting 89,90Sr, 131I, and 134,137Cs in surface waters after a large-scale radioactive fallout.

Simplified dynamic models have been developed for predicting the concentrations of radiocesium, radiostrontium, and I in surface waters and freshwater fish following a large-scale radioactive fallout. The models are intended to give averaged estimates for radionuclides in water bodies and in fish for all times after a radioactive fallout event. The models are parameterized using empirical data collected for many lakes and rivers in Belarus, Russia, Ukraine, UK, Finland, Italy, The Netherlands, and Germany. These measurements span a long time period after fallout from atmospheric nuclear weapons testing and following the Chernobyl accident. The models thus developed were tested against independent measurements from the Kiev Reservoir and Chernobyl Cooling Pond (Ukraine) and the Sozh River (Belarus) after the Chernobyl accident, from Lake Uruskul (Russia), following the Kyshtym accident in 1957, and from Haweswater Reservoir (UK), following atmospheric nuclear weapons testing. The AQUASCOPE models (implemented in EXCEL spreadsheets) and model documentation are available free of charge from the corresponding author.

Comparative study of 137Cs partitioning between solid and liquid phases in Lakes Constance, Lugano and Vorsee.

The methodology for estimating radiocaesium distribution between solid and liquid phases in lakes is applied for three prealpine lakes: Lake Constance (Germany), Lake Lugano (Switzerland) and Lake Vorsee (Germany). It is based on use of the exchangeable distribution coefficient and application of the exchangeable radiocaesium interception potential (RIPex). The methodology was tested against experimental data. Good agreement was found between estimated and measured 137Cs concentrations in Lake Constance and Lake Lugano, whereas for Lake Vorsee a discrepancy was found. Bottom sediments in Lake Vorsee are composed mainly of organic material and probably cannot be described in terms of the specific sorption characteristics attributed to illitic clay minerals.

Time-dependence of the radiocaesium contamination of roe deer: measurement and modelling.

In spruce forest and peat bog, the migration of 137Cs from soil to plants, fungi, roe deer and consumers has been surveyed. In spruce forest the 137Cs activity concentration in roe deer decreases slowly with time and has superimposed periodic maxima in autumn which are correlated with the mushroom season. The decrease with time can be described by an effective half-life of 3.5 yr caused by a fraction of the 137Cs in the soil becoming unavailable for green grazing plants with time. The additional transfer of 137Cs into roe deer meat during the mushroom season depends on precipitation in July, August and September which also determines the yield of fungi in autumn. Our model confirms the assumption that fungi also have access to a fraction of the 137Cs in the soil which is unavailable for green plants. On peat bog the 137Cs activity concentration in roe deer is higher than in spruce forest and its effective half-life is about 17 yr, due to reversible binding of 137Cs to organic matter in the peat bog.

Factors affecting the transfer of radiocaesium from soil to roe deer in forest ecosystems of southern Germany.

Since 1987, in Southern Germany, as a consequence of the Chernobyl accident, the 137Cs activity concentration in roe deer (Capreolus capreolus) has been intensively monitored. A large data set is now available with approximately 5000 samples, with information about the location of the animal, the time of slaughtering, soil characteristics and distance to intensively managed agricultural land. Both the roe deer 137Cs activity concentrations and aggregated transfer coefficients soil-roe deer (T(ag,r)) show a considerable variability with geometric mean and geometric standard deviation in meat of 270.3.5 +/- 1 Bq.kg-1 and 0.01.3.5 +/- 1 m2.kg-1, respectively. From 1987 to 1991, T(ag,r) values exhibited a decline with an ecological half-life of about 3 years. Since 1991, no further decrease of the roe deer contamination level was observed. This general trend is superimposed by an increase in T(ag,r) in autumn that has occurred in almost every year, which is probably due to consumption of mushrooms. This hypothesis is supported by a significant positive correlation between the precipitation during July and August, stimulating the fungi's growth and the height of the maximum in T(ag,r) in autumn. Animals are less contaminated when living in parts of the forest from where they have access to intensively managed agricultural land. A small, but significant correlation is observed between the T(ag,r) and soil parameters; the total potassium content of the soil and the pH of the organic soil layer. The highest transfer coefficients soil-roe deer were found on a peat bog, where most values ranged from 0.01-0.1 m2.kg-1, and neither a pronounced seasonal variation, nor a long-term decrease was observed. On sites with coniferous forests, most T(ag,r) values range from 0.005 to 0.05 m2.kg-1, on sites with a mixture of coniferous and deciduous trees, the transfer coefficients range mostly from 0.002 to 0.02 m2.kg-1. This is a further indication of the importance of the uppermost soil layers and the biological processes therein with respect to the bioavailability of caesium. On the two sites, the values at the lower boundary are more typical for animals that have access to agricultural areas, whereas the higher values were observed in autumn and for animals grazing distant from agricultural areas.