Application of Biotic Ligand and Toxic Unit Modeling Approaches to Predict Improvements in Zooplankton Species Richness in Smelter- Damaged Lakes near Sudbury, Ontario Farhan R. Khan,* ,†,‡ W. (Bill) Keller, § Norman D. Yan, ∥ Paul G. Welsh, ⊥ Chris M. Wood, # and James C. McGeer* ,‡ ‡ Department of Biology, Wilfrid Laurier University, 75 University Avenue W, Waterloo, Ontario N2L 3C5, Canada § Cooperative Freshwater Ecology Unit, Laurentian University, Sudbury, Ontario P3E 2C6, Canada ∥ Department of Biology, York University, Toronto, Ontario M3J 1P3, Canada ⊥ Ontario Ministry of the Environment, Toronto, Ontario M4 V 1M2, Canada # Department of Biology, McMaster University, 1280 Main Street, Hamilton, Ontario, L8S 4K1, Canada * S Supporting Information ABSTRACT: Using a 30-year record of biological and water chemistry data collected from seven lakes near smelters in Sudbury (Ontario, Canada) we examined the link between reductions of Cu, Ni, and Zn concentrations and zooplankton species richness. The toxicity of the metal mixtures was assessed using an additive Toxic Unit (TU) approach. Four TU models were developed based on total metal concentrations (TM-TU); free ion concentrations (FI-TU); acute LC50s calculated from the Biotic Ligand Model (BLM-TU); and chronic LC50s (acute LC50s adjusted by metal- specific acute-to-chronic ratios, cBLM-TU). All models significantly correlated reductions in metal concentrations to increased zooplankton species richness over time (p < 0.01) with a rank based on r 2 values of cBLM-TU > BLM-TU = FI-TU > TM-TU. Lake-wise comparisons within each model showed that the BLM-TU and cBLM-TU models provided the best description of recovery across all seven lakes. These two models were used to calculate thresholds for chemical and biological recovery using data from reference lakes in the same region. A threshold value of TU = 1 derived from the cBLM-TU provided the most accurate description of recovery. Overall, BLM-based TU models that integrate site-specific water chemistry-derived estimates of toxicity offer a useful predictor of biological recovery. 1.0. INTRODUCTION Over 7000 lakes around Sudbury (Ontario, Canada) were affected by acidification and increased metal concentrations from historic industrial emissions. 1 As a result, many became inhospitable for aquatic life; however, subsequent emission controls improved water quality and many plant, invertebrate, and fish species have returned. Zooplankton community changes in lakes with increasing pH and decreasing metal concentrations have been previously discussed. 2−5 Metals, particularly Cu 2+ and Ni 2+ , have been implicated as potential factors limiting the recovery of zooplankton diversity to levels typically found in reference lakes but consideration has been on an individual metal basis. 3,5 Although metal speciation (i.e., free ion concentrations) has been considered, 4 the combined toxicity of metal mixtures has yet to be investigated. There are few studies linking the effects of metal mixtures to toxic impacts. Borgmann and colleagues 6,7 established a bioaccumulation modeling approach with Hyalella azteca but this requires knowledge of burden-to-effect relationships and has not been developed for metal concentrations in the exposure medium. One approach of combining metals is concentration or exposure additivity (as opposed to response addition) to produce a single linear variable, the “Toxic Unit” (TU 8 ). The TU approach normalizes the exposure concen- tration for each contaminant by expressing it as a proportion of a toxicity end point and then these are summed to estimate toxicity on a proportional basis. Of the summation methods (additivity, antagonism, and synergism) additivity is generally used in the absence of data to demonstrate synergistic or antagonistic effects. 9 In this study we varied both the exposure concentration (numerator of each TU proportion) and the toxicity end point (TU denominator) to provide different estimates of metal mixture impacts for contaminated lakes over time. Received: September 7, 2011 Revised: December 15, 2011 Accepted: December 21, 2011 Published: December 21, 2011 Article pubs.acs.org/est © 2011 American Chemical Society 1641 dx.doi.org/10.1021/es203135p | Environ. Sci. Technol. 2012, 46, 1641−1649