Comparison of species sensitivity distributions constructed with predicted acute toxicity data from interspecies correlation estimation models and measured acute data for Benzo[a]pyrene

Benzo[a]pyrene (BaP) is a priority Polycyclic Aromatic Hydrocarbon (PAH), which is toxic to aquatic organisms and has been widely detected in the environment. However, ecological risk assessment for BaP is hard to perform because of the absence of water quality criteria (WQC) and lack of toxicity data for this chemical. To fill in the data gaps, a interspecies correlation estimation (ICE) model was developed by USEPA to predict toxicity values for multiple species from the toxicity estimate for one species. In order to validate the applicability of the ICE model for BaP, measured-based-species sensitivity distributions (SSDs) generated using eight Chinese native aquatic species were compared with ICE-based-SSDs generated using the data predicted from three surrogate species (Lepomis macrochirus, Cyprinus carpio and Daphnia magna). The results showed that there were no significant differences between the two SSD curves and the two hazardous concentrations for the 5% of species (HC5) derived from measured acute toxicity data and ICE-based predicted data. The ICE model was verified as a valid approach for generating SSDs with limited toxicity data. (C) 2015 Elsevier Ltd. All rights reserved.