Derivation of water quality criteria of phenanthrene using interspecies correlation estimation models for aquatic life in China.

Species sensitivity distribution (SSD) method has been widely used to derive water quality criteria (WQC). However, the toxicity data of some environmental pollutants are not easily accessible, especially for endangered and threatened species. Thus, it would be very desirable and economical to predict the toxicity of those species not subjected to toxicity test with the aid of a mathematical model. The interspecies correlation estimation (ICE) model (developed by the US Environmental Protection Agency (USEPA)) uses the initial toxicity estimate for one species to produce correlated toxicity values for multiple species, and it can be utilized to develop SSD and HC5 (hazardous concentration, 5th percentile). In this study, we explored the applicability of ICE to predict toxicity of phenanthrene to various species. ICE-based SSDs were generated using three surrogate species (Oncorhynchus mykiss, Lepomis macrochirus, and Daphnia magna) and compared with the metrical-based SSD. The corresponding HC5 of both models were also compared. The results showed there were no significant differences between HC5 derived from measured acute and ICE-based predicted values. The ICE model was verified as a valid approach for generating SSDs with limited toxicity data and deriving WQC for phenanthrene.