Modeling the bioavailability of nickel and zinc to Ceriodaphnia dubia and Neocloeon triangulifer in toxicity tests with Natural Waters
We studied biotic ligand model (BLM) predictions of toxicity of nickel (Ni) and zinc (Zn) in natural waters from Illinois and Minnesota USA which had combinations of pH, hardness, and dissolved organic carbon (DOC) more extreme than 99.7% of waters in a nationwide database. We conducted 7-d chronic tests with Ceriodaphnia dubia, and 96-hr acute test and 14-d chronic tests with Neocloeon triangulifer, and estimated LC50s and EC20s for both species. Toxicity of Ni and Zn to both species differed among test waters by factors from 8 (Zn tests with C. dubia ) to 35 (Zn tests with N. triangulifer ). For both species and metals, tests with Minnesota waters (low pH and hardness, high DOC) showed lower toxicity than Illinois waters (high pH, high hardness, low DOC). Recalibration of the Ni BLM to be more responsive to pH-related changes improved predictions of Ni toxicity, especially for C. dubia . We compared several input data scenarios for the Zn BLM, which generally had minor effects on Model Performance Scores (MPS). A scenario that included inputs of modeled dissolved inorganic carbon and measured Al and Fe(III) produced highest MPS values for tests with both C. dubia and N. triangulifer . Overall, the BLM framework successfully modeled variation in toxicity for both Zn and Ni across wide ranges of water chemistry in tests with both standard and novel test organisms.