Time-weighted conversion of acute to chronic equivalent endpoints for derivation of chronic ecotoxicity threshold values of six neonicotinoids in freshwater.

Abstract

Neonicotinoid insecticides pose a risk to aquatic invertebrates through their unique selective mode of action and time-cumulative toxicity. Ecotoxicity threshold values (ETVs) are guideline water concentrations for chemical toxicants, concentrations above which represent an unacceptable risk to aquatic environments. Currently, there are no ETVs for neonicotinoids endorsed for use in Australia, mainly due to a lack of suitable chronic ecotoxicity data. Furthermore, the current method for the conversion of acute to chronic equivalent ecotoxicity data using an acute-to-chronic ratio (ACR) is inappropriate for use on neonicotinoids due to their time-cumulative toxicity. The aim of this study was to derive chronic ETVs for the protection of 80%, 90%, 95% and 99% of aquatic species for six neonicotinoids approved for agricultural use in Australia. This was achieved using a novel time-weighted log-log linear regression scaling method for the conversion of acute ecotoxicity data to their 28-day chronic equivalent coupled with the most recent developments in species sensitivity distributions modeling, including model averaging. Ecotoxicity threshold values for six neonicotinoids were derived from compound specific data sets composed of 22-44 individual species' ecotoxicity endpoint data. Chronic data made up approximately 29% of the data; the remaining 71% was composed of 28-day chronic equivalent (i.e., acute converted) ecotoxicity data. Aquatic species were most sensitive to thiacloprid (95% ETV = 0.031 μg L-1), acetamiprid (95% ETV = 0.055 μg L-1) and imidacloprid (95% ETV = 0.109 μg L-1), followed by clothianidin (95% ETV = 0.303 μg L-1) and were least sensitive to thiamethoxam (95% ETV = 0.566 μg L-1) and dinotefuran (95% ETV = 0.655 μg L-1). Compared with the ACR method of converting the same acute to chronic equivalent ecotoxicity endpoints, the ETVs derived here using the time-weighted 28-day chronic equivalent method were within 40%-200% of the ETVs derived using the more traditional ACR approach.