Eruptive population dynamics of forest insect species regularly attract the interest of ecologists but also often evoke debates among stakeholders concerning impacts that outbreaks or pest control activities have on forest communities including insectivorous birds. Lymantria dispar is the most serious native defoliator in mixed oak-broadleaf forests in Central Europe. To investigate the treatment effects of L. dispar density, aerial application of the insecticide Mimic (tebufenozide) and nest box accessibility for L. dispar caterpillars on nest box success of cavity nesting birds, we deployed 352 nest boxes in a full factorial experimental design with 11 replicates. L. dispar density was predicted by egg mass counts made in the year before the study. The accessibility of nest boxes for L. dispar caterpillars was manipulated in all study sites by placing nest boxes on trunks for easy access to caterpillars and on branches for more difficult access. Successful nest boxes (binary metric that combines failed breeding attempt and empty nest box) and nest productivity (number of fledged chicks in successful boxes) were monitored in the peak year of a L. dispar outbreak, during which half of the plots were treated with Mimic, as well as in the following year when L. dispar populations had widely collapsed, and no further treatments were applied. L. dispar and non-target caterpillar biomass was sampled by canopy fogging to quantify prey abundance. Caterpillar biomass was significantly reduced by Mimic in the treatment year and remained low in the post-treatment year. Despite the reduction of caterpillars, our zero-inflation model found no effect of any tested variables on nest box success of the early brood. However, in the late broods the number of successful nest boxes was reduced in Mimic treated plots in the first year by 42%, while unaffected by initial L. dispar density. The number of successful nest boxes in the second year was lower in areas with initially high L. dispar densities and there was a marginal carry-over effect of insecticide treatment. L. dispar caterpillars entered nest boxes on the trunk in higher numbers than boxes on the branch, but nest box position had no effect on nest box success of birds.We conclude that despite its strong negative impact on caterpillar prey, aerial application of Mimic targeting L. dispar caused an increase in the number of failed boxes in late broods in sprayed plots, particularly in the year of application and to some extent in the year after. We presume this delayed effect of Mimic only on late broods to be caused by Mimic-induced caterpillar shortages during first broods leading to an increased parental effort and a subsequent reduced number of second broods. The generally limited impact of Mimic might be explained by the overall extensive availability of alternate prey in oak forests in May during first broods and the ability of tits to increase their parental effort in times of caterpillar shortage. In unsprayed sites, L. dispar densities can create legacies of impact in following years, but caterpillars do not directly affect birds nest box success despite entering cavities.
Read full abstract