Long-term trends of radiocesium activity concentrations in vegetation in Irish semi-natural ecosystems.

137Cs activity concentrations were determined in vegetation from four Irish blanket bog ecosystems during the summer months of 1989-1997. The dominant and abundant vegetation species were sampled at each site and the data were used to estimate long-term trends of 137Cs activity concentrations. A general decline in the 137Cs activity concentrations in vegetation was observed from all sites sampled. The fastest and most consistent long-term rates of 137Cs decline were observed in the bryophyte and lichen vegetation group with 137Cs effective half-life (T ef) values of 2.2 to 10.7 y. The ericoid group with the exception of one case also showed a significant long-term decline in 137Cs activity concentrations with T ef values of 3.5 to 12.4 y as did the rush species Juncus squarrosus (T ef range 9.3 to 12.8 y). The sedge and grass groups showed the slowest and the least consistent rates of 137Cs decline, with T ef values of 2.9 to 59.8 y. Preliminary evaluation of 137Cs activity concentrations in sheep from some of the studied sites suggests that the decline found in vegetation is reflected in sheep.

An evaluation of ratio systems in radioecological studies.

Replicate samples of soil, Juncus squarrosus and Calluna vulgaris were taken within a grid system on an organic peatland soil site in Ireland. A similar sampling survey was carried out on an organic rich forest soil site in Sweden, where Vaccinium myrtillus and Vaccinium vitis-idaea with corresponding soil samples were taken within a Picea abies stand. The data were used to investigate the relationship between soil and plant 137Cs content and to examine the validity of using ratios to describe this relationship. Findings from both countries were in agreement. There were no significant changes in plant 137Cs concentration associated with increasing soil content. When data from both countries were merged and treated as a single data set, a significant overall positive correlation (95% confidence level) between soil and plant 137Cs levels was observed. Concentration ratios and transfer factors were calculated for 137Cs and 40K uptake from soils to plants. Both ratio types exhibited a clearly defined decrease associated with increasing soil concentrations for both radionuclides. Findings demonstrate serious problems with the use of ratios for the evaluation of radionuclide transfer.

Radionuclide uptake by red deer (Cervus elaphus) on mountain grazing.

Forty-two red deer (Cervus elaphus) were shot during the 1992 annual cull in Glenveagh National Park, Ireland. Samples of rumen, kidney and faeces were removed from each animal. Kidney samples were used to estimate flesh radiocaesium (137Cs) concentrations and the 95% confidence interval for the mean was 203 +/- 12 Bq kg-1. The maximum recorded 137Cs concentration in kidney was 367 Bq kg-1 (fresh weight). The altitude of the cull, the age and sex of each animal were recorded. Neither age nor sex correlated with concentrations of 137Cs in rumen, kidneys or faeces. Despite the limited altitudinal range of the study and the free ranging behaviour of deer, there was a highly significant positive correlation between rumen, kidney and faecal 137Cs concentrations and the altitude of the cull. 40K concentrations in rumen, kidney and faeces did not correlate with the altitude of cull, age or sex of slaughtered animals. Significant 137Cs concentration differences were identified in the sequence: rumen < faeces = faeces. 137Cs concentrations in rumen, kidney and faeces for individual animals were all significantly correlated. Statistical testing showed that the concentration sequence for 40K was: rumen < kidney = faeces; a sequence which differs from that of 137Cs. A comparison of ratios test for rumen:faecal ratios demonstrated that significantly more 137Cs was excreted in faeces than was the case for 40K. The concentration of 137Cs excreted in faeces relative to concentrations in forage (rumen), is approximately twice that for 40K. Linear regression of faecal 137Cs concentrations (y) on kidney concentrations (x) was carried out, the regression equation is y = -86.90 + 0.97x. This equation (R2 = 0.73, F1,40 = 107) may be used to predict 137Cs concentrations in flesh by measurement of faecal concentrations. This is a useful preliminary assessment method, particularly with herds of wild animals that prove difficult to capture for in vivo monitoring.

The variability in fallout content of soils and plants and the design of optimum field sampling strategies.

Sampling was carried out at an upland peat site in Ireland to assess the variability in 137Cs deposition to soils and activity concentration in individual vegetation species. A 3,600 m2 quadrat was laid out, and a soil core was extracted within each of sixteen 20 x 20 cm soil monoliths. The coefficient of variation values for 137Cs deposition and activity in individual soil horizons were in all cases greater than 30%, and the relative variability between corresponding horizons tended to increase with depth. Samples of Calluna vulgaris and Juncus squarrosus were taken at each soil sampling location and coefficients of variation for 137Cs activity in these species were 12 and 20%, respectively. The data was used to examine the relationship between the number of samples analyzed in a survey, the variability within those samples, and the way those two factors affect our ability to detect between-site differences in 137Cs contamination levels. A methodology is developed for detection of between-site differences in 137Cs contamination under different experimental criteria. The data show that, of the sample types examined, between-site differences could be detected most efficiently with C. vulgaris. Because of the variability associated with the 137Cs content of each sample type, between-site differences could be detected with fewer plant samples than soil samples. Plant-soil concentration ratios, transfer factors, and plant-plant ratios were calculated from the data, and the variability associated with each parameter was assessed. The variation among values for plant-plant ratios was less than among concentration ratios, and the greatest spread of values was associated with plant-soil transfer factors.

Persistence and prediction of radiocaesium levels in animals grazing semi-natural environments.

In vivo monitoring of radiocaesium levels in a flock of mountain sheep grazing a heather-dominant peatland ecosystem was undertaken over a 3-year period from 1989 to 1991. Highest values were observed during summer and autumn and measurable radiocaesium activities were present throughout the year. Comparably high mean and maximum values recorded in November 1989 are thought to be a result of fungi consumption. Analysis of variance indicated that an increase of 4% occurred in radiocaesium levels in muscle between 1989 and 1990 followed by an overall decrease of 35% between 1990 and 1991. In order to test an existing model to predict radiocaesium levels in muscle from measured activities in faeces, an independent data set was produced. The radiocaesium content of faecal samples collected from the grazing range of a sheep flock was compared with the results of in vivo measurements on the same animals. The correlation over time between the predicted and observed flock mean and maximum values was found to be significant, thereby confirming the value of faeces sampling as an inexpensive and rapid means of assessing the distribution of radiocaesium activities in sheep flocks grazing semi-natural ecosystems.