Mixture toxicity contribution of emerging and legacy contaminants in environmental monitoring data from Norway

Study 1 comprises 31 samples collected from brown trout, in which between 15 and 43 chemicals were analyzed and between 4 and 23 compounds were actually detected per sample at concentrations above their level of detection (LOD). Study 2 consists of 1149 samples, with up to 55 compounds included in the analytical profile and with up to 42 detected chemicals.. The mixtures were assessed using Concentration Addition based on internal toxic units, accompanied by an uncertainty assessment. Only for
approximately half the compounds actual experimental data on fish toxicity were found, the toxicity of the remainder was estimated using QSAR‐models. Non‐detects were considered in the exposure assessment by using a Kaplan‐Meier adjustment, instead of using artifical surrogate values. Mixures were in general dominate by only a very few compounds, even if up to 42 chemicals were present. This implies that (i) the error made by ignoring the competing
concept of Independent Action is negligible and (ii) that synergistic interactions are only relevant for the ‘’mixture risk drivers’’. None of the mixture had a sum of toxic units exceeding one, but several had toxic unit sums above 0.01, which might warrant further investgation, given the substantial uncertainties still present in the current assessment. (Relevant appendix to this report can be ordered at: postmottak@miljodir.no)