Comparison of aquatic system models using outdoor mesocosm data for ecological risk assessment, part I: methodology

Mesocosm studies are conducted in the context of higher-tier ecological risk assessments (ERAs) to integrate environmental conditions and study species interactions within waterbodies in agricultural landscapes. Aquatic system models (ASMs) could provide tools to extrapolate the dynamics and effects of chemicals observed in mesocosm studies to a wider range of environmental conditions and exposure scenarios. In this paper, we present the methodology of a ring study with four ASMs (Aquatox, CASM, StoLaM+, Streambugs) applied to data from mesocosm studies, while the results of the ring study are presented in a companion paper published alongside this manuscript. The ring study aimed to test the feasibility and capability of ASMs to represent mesocosm data and to evaluate if such models can be used as an extension of mesocosm experiments in ERAs. The ring study methodology allowed for model comparison and identified models' strengths and limitations in representing the ecosystem dynamics in control and treated mesocosm studies. Groups of species were defined to map the diversity of the mesocosm ecosystem in the taxa represented by the models, and a consensus trophic web was agreed upon to ensure harmonization among the models. Control and effect calibration criteria were developed to evaluate and compare model performances for each species group, explicitly considering the variability in mesocosm data. Our proposed methodologies and the challenges in using mesocosm data to inform modeling approaches are discussed. Finally, we derived recommendations for future research to facilitate the development of models representing mesocosms to support ERAs.