Ecotoxicological Mechanisms and Models in an Impact Analysis Tool for Oil Spills

Abstract

In an international collaborative effort, an impact analysis tool is being developed to predict the effect of accidental oil spills on recruitment and production of Atlantic cod (Gadus morhua) in the Barents Sea. The tool consisted of three coupled ecological models that describe (1) plankton biomass dynamics, (2) cod larvae growth, and (3) fish stock dynamics. The discussions from a series of workshops are presented in which variables and parameters of the first two ecological models were listed that may be affected by oil-related compounds. In addition, ecotoxicological algorithms are suggested that may be used to quantify such effects and what the challenges and opportunities are for algorithm parameterization. Based on model exercises described in the literature, survival and individual growth of cod larvae, survival and reproduction of zooplankton, and phytoplankton population growth are denoted as variables and parameters from the ecological models that might be affected in case of an oil spill. Because toxicity databases mostly (67%) contain data for freshwater species in temperate environments, parameterization of the ecotoxicological algorithms describing effects on these endpoints in the subarctic marine environment is not straightforward. Therefore, it is proposed that metadata analyses be used to estimate the sensitivity of subarctic marine species from available databases. To perform such analyses and reduce associated uncertainty and variability, mechanistic models of varying complexity, possibly aided by new experimental data, are proposed. Lastly, examples are given of how seasonality in ecosystems may influence chemical effects, in particular in the subarctic environment. Food availability and length of day were identified as important characteristics as these determine nutritional status and phototoxicity, respectively.

Acknowledgments

Frederik de Laender is a postdoctoral research fellow from the Research Foundation–Flanders (FWO). The authors thank Statoil for financially supporting the workshops during which the concepts and ideas presented in this paper were formulated. The authors also thank the scientific partners that are involved in other aspects of the ongoing project, for constructive feedback and interesting discussions during these workshops. In particular, we wish to acknowledge Akvaplan-niva for leading the development of this impact analysis tool. The two reviewers are thanked for their constructive remarks and suggestions.