Thermal adaptation affects the temperature-dependent toxicity of the insecticide imidacloprid to soil invertebrates

Abstract

Terrestrial ectotherms are vulnerable to climate change since their biological rates depend on the ambient temperature. As temperature may interact with toxicant exposure, climate change may cause unpredictable responses to toxic stress. A population's thermal adaptation will impact its response to temperature change, but also to interactive effects from temperature and toxicants, but these effects are still not fully understood. Here, we assessed the combined effects of exposure to the insecticide imidacloprid across the temperatures 10–25 °C of two populations of the Collembola Hypogastrura viatica (Tullberg, 1872), by determining their responses in multiple life history traits. The con-specific populations differ considerably in thermal adaptations; one (arctic) is a temperature generalist, while the other (temperate) is a warm-adapted specialist. For both populations, the sub-lethal concentrations of imidacloprid became lethal with increasing temperature. Although the thermal maximum is higher for the warm-adapted population, the reduction in survival was stronger. Growth was reduced by imidacloprid in a temperature-dependent manner, but only at the adult life stage. The decrease in adult body size combined with the absence of an effect on the age at first reproduction suggests a selection on the timing of maturation. Egg production was reduced by imidacloprid in both populations, but the negative effect was only dependent on temperature in the warm-adapted population, with no effect at 10 °C, and decreases of 41 % at 15 °C, and 74 % at 20 °C. For several key traits, the population best adapted to utilize high temperatures was also the most sensitive to toxic stress at higher temperatures. It could be that by allocating more energy to faster growth, development, and reproduction at higher temperatures, the population had less energy for maintenance, making it more sensitive to toxic stress. Our findings demonstrate the need to take into account a population's thermal adaptation when assessing the interactive effects between temperature and other stressors.