Effect of temperature on the toxicity of imidacloprid to Eisenia andrei and Folsomia candida in tropical soils

Abstract

The influence of temperature on the chronic toxicity and risk of imidacloprid to soil non-target species was assessed in tropical soils. Earthworms Eisenia andrei and collembolans Folsomia candida were exposed to a tropical artificial soil (TAS) and two natural tropical soils from Brazil (Entisol and Oxisol) with increasing concentrations of imidacloprid under atmospheric temperatures of 20, 25 and 28 °C. The effect of temperature on the reproduction of both species was assessed through the number of juveniles and earthworm’s growth, and the risk associated was estimated through the Toxicity-Exposure Ratio (TER). Toxicity of imidacloprid increased with temperature in all tested soils, being generally lower in TAS soil (EC50s of 1.48, 0.66 and 0.40 mg kg−1 for E. andrei and 0.3, 0.2 and 0.06 mg kg−1 for F. candida at 20, 25 and 28 °C, respectively) compared to Entisol (EC50s of 0.19, 0.03 and 0.14 mg kg−1 for E. andrei and 0.04, 0.02, 0.01 mg kg−1 for F. candida at 20, 25 and 28 °C, respectively) and Oxisol (EC50s of 0.21, 0.07, 0.06 mg kg−1 for E. andrei and 0.16, 0.09, 0.06 mg kg−1 for F. candida at 20, 25 and 28 °C, respectively) within each temperature for both species. These values indicate that properties of TAS may not be representative of natural/local soils to adequately estimate the toxicity of pesticides to non-target soil species. At higher temperatures, the variability of imidacloprid toxicity between soils was lower, which suggests that the influence of soil properties on imidacloprid toxicity was overshadowed by temperature. TER values revealed that risk is also greater at higher temperatures. Data reported enforce the need for the inclusion of more realistic conditions in single-species tests in prospective risk assessment of pesticides to avoid underestimation of risk to non-target species.