Calibration of a laboratory-based chronic toxicity model to nickel effects on stream invertebrates in the field

Predicting effects of metals on stream invertebrate communities can be hindered by spatial and temporal variation in toxicity-modifying factors, a paucity of laboratory toxicity data for stream taxa (mainly insects), and variation in benthic invertebrate community structure related to habitat and factors other than the stressor of interest. We addressed these challenges by combining laboratory-based chronic toxicity data with field-based biological monitoring data to build a lab-to-field stressor-response relationship. A laboratory-based toxicity model for Ceriodaphnia dubia was used to translate field nickel (Ni) concentrations into predicted toxicity, and a quantile regression model of field data from a mine-affected watershed was then used to describe the limiting effect of Ni toxicity on the benthic invertebrate community. Many taxa showed no evidence of Ni effects up to the highest studied exposure (30 µg/L dissolved Ni, or 92% effect to C. dubia reproduction). The most sensitive metric was percent Ephemeroptera, reflecting declines in abundance of some sensitive mayflies (e.g., Baetis, Epeorus) and concurrent increases in abundance of some tolerant non-mayfly taxa (e.g., Rheocricotopus, Eukiefferiella) across the gradient of Ni toxicity in the field. The field 10% effect concentration for percent Ephemeroptera occurred at a 22% effect of Ni to C. dubia reproduction. This finding supports previous estimates of 20% as a critical effect size in laboratory test organisms that could be predictive of discernible effects on sensitive invertebrates in the field.