Energy-based modeling to study population growth rate and production for the midge Chironomus riparius in ecotoxicological risk assessment.

Effects of toxicants are commonly assessed at individual level, whereas the aim of ecotoxicology is to protect ecosystems. We recently built energy-based models to describe and predict growth, emergence and reproduction of the midge Chironomus riparius [Péry (2002) Environ. Toxicol. Chem. 21, 2507-13]. Here we use these models to derive effects at the population level from effects at individual levels. The first endpoint we consider is population growth rate, which is a commonly studied endpoint at the population level. This parameter is informative relative to the risk of disappearance of the population. We also examined the production of organisms per generation or per time unity at population equilibrium, the study of which is allowed by our models and accounts for energy transfers. Such a study is crucial to predict effects on ecosystems, for species of the Chironomidae are keystone species, being the main food source of many other species, including birds and fish. We show in our study that the disappearance of the population can only occur in cases of very severe toxicity (99% decrease of reproduction, more than 97% mortality during a 10 days survival test or a difference of mean emergence times between males and females of more than 10 days). Concerning production of organisms, we show that reproduction decrease has little effect on it, that mortality of young larvae has an impact that cannot be neglected and that mortality of old larvae and delay of emergence has a strong effect. Our study suggests that bioassays should focus on an EC50 for reproduction, a LC30 for young instars, and a No Effect Concentration for old instars (growth and mortality) to prevent effects at the population level.