Is an assessment factor of 10 appropriate to account for the variation in chemical toxicity to freshwater ectotherms under different thermal conditions?

Ecotoxicity tests are often conducted following standard methods, and thus carried out at a fixed water temperature under controlled laboratory conditions. Yet, toxicity of a chemical contaminant may vary in a temperature-dependent manner, depending on the physiology of the test organism and physicochemical properties of the chemical. Although an assessment factor of 10 (AF10) is commonly adopted to account for variability in toxicity data related to temperature in the development of water quality guidelines and/or ecological risk assessment, no one has ever rigorously assessed the appropriateness of AF10 to account for potential variation in temperature-dependent chemical toxicity to aquatic organisms. This study, therefore, aims to address this issue through a meta-analysis by comparing median lethal concentration data for nine chemicals (cadmium, copper, nickel, lead, silver, zinc, arsenic, selenium and DDT) on a range of freshwater ectothermic animal species at different temperatures, and to assess whether AF10 is under- or over-protective for tropical and temperate freshwater ecosystems. Our results reveal varying extents of interaction between temperature and different chemicals on organisms and the complexity of these interactions. Applying AF10 sufficiently protects 90% of the animal species tested over a range of temperatures for cadmium, copper, nickel, silver, zinc and DDT in the tropics, but it is insufficient to adequately encompass a larger temperature variation for most studied chemicals in temperate regions. It is therefore important to set specific AFs for different climatic zones in order to achieve the desired level of ecosystem protection.