Modelling population-level consequences of chronic external gamma irradiation in aquatic invertebrates under laboratory conditions

We modelled population-level consequences of chronic external gamma irradiation in aquatic invertebrates under laboratory conditions. We used Leslie matrices to combine life-history characteristics (duration of life stages, survival and fecundity rates) and dose rate-response curves for hatching, survival and reproduction fitted on effect data from the FREDERICA database. Changes in net reproductive rate R₀ (offspring per individual) and asymptotic population growth rate λ (dimensionless) were calculated over a range of dose rates in two marine polychaetes (Neanthes arenaceodentata and Ophryotrocha diadema) and a freshwater gastropod (Physa heterostropha). Sensitivities in R₀ and λ to changes in life-history traits were analysed in each species. Results showed that fecundity has the strongest influence on R₀. A delay in age at first reproduction is most critical for λ independent of the species. Fast growing species were proportionally more sensitive to changes in individual endpoints than slow growing species. Reduction of 10% in population λ were predicted at dose rates of 6918, 5012 and 74,131 μGy·h⁻¹ in N. arenaceodentata, O. diadema and P. heterostropha respectively, resulting from a combination of strong effects on several individual endpoints in each species. These observations made 10%-reduction in λ a poor criterion for population protection. The lowest significant changes in R₀ and λ were respectively predicted at a same dose rate of 1412 μGy h⁻¹ in N. arenaceodentata, at 760 and 716 μGy h⁻¹ in O. diadema and at 12,767 and 13,759 μGy h⁻¹ in P. heterostropha. These values resulted from a combination of slight but significant changes in several measured endpoints and were lower than effective dose rates calculated for the individual level in O. diadema and P. heterostropha. The relevance of the experimental dataset (external irradiation rather than contamination, exposure over one generation only, effects on survival and reproduction only) for predicting population responses was discussed.