Incorporating ecologically relevant measures of pesticide effect for estimating the compatibility of pesticides and biocontrol agents.

The compatibility of biological control agents with pesticides is a central concern in integrated pest management programs. The most common assessments of compatibility consist of simple comparisons of acute toxicity among pest species and select biocontrol agents. A more sophisticated approach, developed by the International Organisation of Biological Control (IOBC), is based on a tiered hierarchy made up of threshold values for mortality and sublethal effects that is used to determine the compatibility of pesticides and biological control agents. However, this method is unable to capture longer term population dynamics, which is often critical to the success of biological control and pest suppression. In this article, we used the delay in population growth index, a measure of population recovery, to investigate the potential impacts that the threshold values for levels of lethal and sublethal effects developed by the IOBC had on three biocontrol agents: sevenspotted lady beetle, Coccinella septempunctata L.; the aphid parasitoid Diaeretiella rapae (M'Intosh), and Fopius arisanus (Sonan), a parasitoid of tephritid flies. Based on life histories of these economically important natural enemies, we established a delay of 1-generation time interval as sufficient to disrupt biological control success. We found that delays equivalent to 1-generation time interval were caused by mortality as low as 50% or reductions of offspring as low as 58%, both values in line with thresholds developed by the IOBC. However, combinations of mortality and reduction of offspring lower than these values (from 32 to 43% each) over a simulated 4-mo period caused significant population delays. Furthermore, the species used in these simulations reacted differently to the same levels of effect. The parasitoid D. rapae was the most susceptible species, followed by F. arisanus and C. septempunctata. Our results indicate that it is not possible to generalize about potential long-term impacts of pesticides on biocontrol agents because susceptibility is influenced by differences in life history variables. Additionally, populations of biocontrol agents may undergo significant damage when mortality approaches 50% or when there is mortality of -30% and a 30% reduction in offspring caused by a sublethal effect. Our results suggest that more ecologically relevant measures of effect such as delays in population growth may advance our knowledge of pesticide impacts on populations of beneficial species.