Integrative Modeling of Contaminant Accumulation in Highly Altered Aquatic EcosystemsPublic Deposited
The world's aquatic food webs are currently in crisis due to the cumulative, interacting effects of anthropogenic pressures including non-indigenous species invasions, chemical contamination, overfishing, habitat disruption and global climate change. Efforts to restore or protect the integrity and sustainability of these systems are limited by a lack of knowledge about how ecosystem structure and function are altered by multiple stressors. In this work three models are presented that progressively integrate data and mechanistic descriptions of food web structure (predator-prey interactions), chemical partitioning and species-specific physiology in order to better describe chemical accumulation in highly altered aquatic ecosystems. The first model presents a steady-state picture that reveals the importance of diet-related ecological details for the quantitative prediction of chemical accumulation in top predator species. The second model describes non-steady-state accumulation, tracing the trajectory of contaminant accumulation in an organism through its life cycle to highlight the life history parameters that most influence contaminant transfer. Finally, the third model uses data on widely distributed species assemblages to predict how climate change may affect contaminant transfer in aquatic food webs. Thus, the three models effectively describe aquatic bioaccumulation in the past, present and future: the first model uses a steady-state description to gain insight on how bioaccumulation has been affected by previous species invasions and legacy (in-place) contaminants; the second model describes how contaminant concentrations in a species born in such an environment may evolve in response to changes in its physiology (growth, maturation) or in its food web (species invasion); the third model explores how these species-specific patterns of contaminant accumulation might change as a result of increased temperatures related to climate change. The models and their results reveal the importance of species-level seasonal and life cycle resolution, and demonstrate that scenario-based forecasting models are useful tools for describing structure-function relationships in complex, dynamic ecosystems where uncertainty can be substantial.