Can Ecological Models accurately predict mixture toxicity at the population level or is clarivoyance still needed?

In the natural environment, chemical substances are rarely present as single compounds, hence it is no surprise that at a regulatory level, mixture toxicity is currently one of the major themes. At ARCHE Consulting, our experts are fully engaged in this topic. In collaboration with Ghent University, population models were developed that accurately predict the combined effects of multiple chemicals on water flea populations. This novel approach offers multiple advantages over the statistical techniques that are currently in use.

Contrary to standard laboratory tests, the ecological reality of exposure in the field is much more complex. For example, the presence of multiple chemicals and interactions with various organisms increase the uncertainty of current risk assessment methods. Population models bridge the gap between the lab and the reality in the field. Our research applied population models to long-term experimental data on the combined effects of four chemicals (pyrene, dicofol, alfa-hexachlorocyclohexane, and endosulfan) on water flea (Daphnia magna) populations.

Using only standard toxicity data of the single compounds, the model could predict how different combinations of these chemicals affect Daphnia populations, both after constant exposure and a short-term pulse. Comparison with traditional statistical analysis of the population data showed that the population model performs at least as good as the statistical approach. Considering that the modelling approach only requires chronic toxicity data on the individual chemicals, this novel technique offers clear advantages over the statistical approach. The use of population modelling can be pivotal to gear up toward a novel and realistic approach for the ecological risk assessment of mixture exposure scenarios.