You are here:
Reducing uncertainty in species sensitivity distributions with interspecies toxicity estimation models
Citation:
Awkerman, J., Sandy Raimondo, CrystalR Jackson, AND M. Barron. Reducing uncertainty in species sensitivity distributions with interspecies toxicity estimation models. Presented at SETAC North America 33rd Annual meeting, November 11 - 15, 2012.
Impact/Purpose:
Recommendations on use of Interspecies Correlation Estimation models in species sensitivity distributions
Description:
Determining harmful levels of contaminants for a wide range of species is limited by available toxicological data. Ecological risk assessments typically depend on empirical data from only a few species that represent a narrow range of the life history spectrum. Extrapolation models such as Interspecies Correlation Estimation (ICE) models estimate unknown toxicity values to an untested species from a well-tested surrogate, increasing the utility of available data. . Developing species sensitivity distributions (SSD) with ICE-estimated toxicity values further extends data applicability by demonstrating cumulative proportion of toxicity for many species to a certain chemical. The hazard concentration HC5 represents LC50 lower than that of 95% of the species represented in the SSD. Estimation methods introduce uncertainty, such that error propagation should be addressed in extrapolation models used for risk assessment. Here we determine the effect on HC5 estimation of toxicity estimation, ICE model robustness, and surrogate species commonly used in risk assessment. We examine the potential tradeoff of using extrapolation models to increase sample size within an SSD and the subsequent propagation of uncertainty in HC5 estimates. In general, multi-surrogate ICE SSD provide recommended taxonomic proximity between surrogate and predicted species for toxicity estimation while increasing the species diversity represented within an SSD.