Estuarine systems are among the most impacted ecosystems due to anthropogenic contaminants; however, they present unique challenges to toxicity testing with regard to varying water quality parameters. The euryhaline amphipod species, Hyalella azteca, is widely used in toxicity testing and well suited for testing estuarine water samples. Nevertheless, the influence of relevant water quality parameters on test endpoints must be quantified in order to efficiently use this species for routine monitoring. Here, we studied the influence of five water quality parameters: electrical conductivity, pH, un-ionized ammonia, dissolved oxygen and temperature, on H. azteca survival in a water column toxicity test. A model was developed to quantify and predict the independent and interacting effects of water quality variables on 10-day survival. The model allows simultaneous assessment of multiple potential predictors recorded during the tests. Data used for modeling came from 1089 tests performed on ambient water samples over a period of three years (2006-2008). The final model reflects significant effects of predictors and their two-way interactions. The effect of each level of all predictors on survival probability of H. azteca was examined by comparing levels of each predictor at a time, while holding all others at their lowest (reference) level. This study showed that predictors of survival in water column tests should not be evaluated in isolation in the interpretation of H. azteca water column tests. Our model provides a useful tool to predict expected control survival based on relevant water quality parameters, and thus enables the use of H. azteca tests for toxicity monitoring in estuaries with a wide range of water quality conditions.
Keywords: Ammonia; Hyalella azteca; Multivariate logistic regression; Piperonyl butoxide; Toxicity testing; Water quality.
Copyright © 2015 Elsevier Inc. All rights reserved.