RESUMEN
The objective of this study was to assess the predictive capacity of the biotic ligand model (BLM) for acute copper toxicity to daphnids as applied to a number of freshwaters from Chile and to synthetic laboratory-prepared waters. Thirty-seven freshwater bodies were sampled, chemically characterized, and used to determine the copper concentration associated with the 50% of mortality (LC50) for Daphnia magna, Daphnia pulex, and Daphnia obtusa (native to Chile). The data were then used to run three versions of the acute copper BLM, and the predicted LC50s were compared to the observed ones. The same was done with synthetic assay media at various hardness and dissolved organic carbon (DOC) levels. The BLM versions differed in the affinity constants for some biotic ligand-ion pairs, stability constants for inorganic Cu complexes, and assumptions regarding Cu binding to DOC. All three versions showed a high degree of predictive performance, mostly within a twofold range of observed toxicity values. The D. obtusa data set was used to compare water quality criteria (WQC) derived from the observed toxicity values with those derived from either the BLM or the U.S. Environmental Protection Agency (U.S. EPA) procedure. For most low DOC waters, the three procedures generated similar WQCs. For the high-DOC waters, the EPA-derived criteria were significantly lower, that is, greatly overprotective. The results are also discussed in terms of the validation of the BLM for regulatory use.