In contrast to single-generation experiments, natural populations are chronically exposed to pollutants over many generations and numerous factors influence the long-term response to chemical exposure in the field. C. riparius is a convenient species to assess contaminated sediments toxicity in multi-generations test. Moreover, data on uranium effects on freshwater benthic macroinvertebrates are scarce. Hence, the aim of this study was to assess the effects of uranium exposure on seven successive generations of C. riparius exposed to environmentally relevant concentrations. To this purpose, an individual-based model of C. riparius was developed. It permitted to explore mechanisms of population adaptation and to test hypotheses on the selective forces inducing adaptation. At the four tested concentrations, four experimental populations, coming from the same laboratory culture, were reared simultaneously over seven generations. Several life-history traits were recorded at the end of each generation. The model simulated the control population dynamics and the exposed populations ones over the seven generations. Our data showed significant generational trend of larval mortality, a decreased of the emergence time and a decreased of the adult size over the seven generations. In our model, we proposed to relate the survival rate to both emergence time and uranium concentration. The simulations described accurately the generational trend of all measured life-history traits. Hence, uranium exposure in controlled conditions is likely to select individual with a short emergence time. To conclude, modelling the population dynamics of C. riparus over many generations allowed to propose rational explanations for the experimental data in links with likely mechanisms of adaptation.
