1. Ecotoxicological models used to assess the effects of radionuclides on f
reshwater ecosystems and the potential transfer to humans depend on accurat
e laboratory estimates of the elimination rate of radionuclides in fish.
2. We determined experimentally the elimination rate (excretion) of caesium
(Cs-134) of Atlantic salmon Salmo salar (two populations), Arctic charr Sa
lvelinus alpinus (two populations), whitefish Coregonus lavaretus and bream
Abramis brama by single-dose caesium labelling and measures of subsequent
samples of groups of fish maintained at different temperatures. Elimination
rates between populations of the same species and among species were compa
red.
3. Populations of the same species from similar thermal environments had si
milar elimination rates (Atlantic salmon) whereas those from different ther
mal environments differed significantly (Arctic charr).
4. Caesium elimination rates varied among fish species. Bream had slower el
imination rates than the other species at all temperatures, whereas Atlanti
c salmon had quicker elimination rates than the others at 16 degrees C. The
remaining differences among species were not consistent across temperature
s, indicating differences in thermal performance.
5. Approximately 45% of the estimated elimination rates differed significan
tly from those predicted by a general caesium elimination model. We conclud
e that a single model for caesium elimination, valid for a wide range of sp
ecies and populations, may be difficult to attain.