Nontarget effects of orchard pesticides on natural enemies: Lessons from the field and laboratory

The nontarget effects of insecticide programs used to control codling moth, Cydia pomonella L. (Lepidoptera: Tortricidae), were studied in large-plot field trials in apples, pears, and walnuts in the western United States. We assessed the health of the natural enemy community by sampling the abundance of natural enemies and by monitoring for outbreaks of secondary pests. The insecticides used in the field tests overlapped those tested in laboratory bioassays. Using these parallel lab and field studies, we examined two hypotheses: 1) pesticides found to have negative effects on natural enemy fitness in laboratory bioassays will predict reductions in natural enemy densities in the field, and 2) reductions in natural enemy densities in the field will result in outbreaks of secondary pests. We found only one clear instance, Forficula auricularia (L.) (Dermaptera: Forficulidae), where laboratory results documenting negative effects corresponded to a significant reduction in field studies (apple). This same instance was the only case where a reduction in a natural enemy population was associated with a significantly increased density of a secondary pest, Eriosoma lanigerum (Hausmann) (Hemiptera: Aphididae). There were several instances where secondary pest outbreaks were associated with unchanged or even increased natural enemy densities. The limited number of field trials, variability in field trial conditions among years and sites, duration of the negative effect relative to sampling interval, sampling efficiency, plot size/inter-plot movement, and compensation by other natural enemies attracted to a large host/prey resource may all have contributed to the poor predictive success. Overall, the laboratory bioassays predicted a far greater negative impact than was found in the field trials