Introduced social wasps (Vespula vulgaris) reach high densities in some New Zealand beech forests, because honeydew provides an abundant high-energy food source. We manipulated wasp density to estimate an "ecological damage threshold" for large, free-living Lepidoptera larvae. There will be a continuum of ecological damage thresholds for wasp density depending on the prey species or habitat. Experimentally placed small caterpillars had a significantly higher survival rate than large caterpillars, and the survival rate of both groups decreased with increasing wasp density. Spring-occurring caterpillars have a probability of surviving of 0.90-0.95, assuming wasps are the only source of mortality. However, at the peak of the wasp season we predict caterpillars would have virtually no chance (probability of 10 to 10) of surviving to adults. Wasp abundance must be reduced by at least 88% to conserve the more vulnerable species of free-living caterpillars at wasp densities similar to those observed in our study sites. This equates to a damage threshold of 2.7 wasps per Malaise trap per day. It was exceeded for about 5 months of the year in non-poisoned sites. There are currently no biological or chemical control techniques available in New Zealand that will reduce wasp abundance below this damage threshold throughout the year. Our models show that most Lepidoptera with spring caterpillars will be able to persist, but species with caterpillars occurring in the peak wasp season will be eliminated.

No abstract

The number of Vespula wasp workers in a colony can be predicted from the mean of !-minute counts of wasps entering or leaving the colony (traffic rate). The predictive equation (Adjusted R 2 = 0.87) is: number of workers= 32.243 X traffic rate per minute and is independent of species, time of year, or location. However, we mainly measured mature colonies. The equation may overestimate the number of workers in spring colonies, because a greater proportion of workers forage then. Traffic rate can also predict the biomass of a V. vulgaris colony (Adjusted R 2 = 0.90) by allowing for season. Colonies measured after mid May generally had a larger biomass for any given traffic rate than those measured earlier. There were too few data from V. germanica colonies to evaluate the relationship between traffic r~ and biomass for this species.

Exotic social insects are a major threat to New Zealands natural and agroecosystems Social wasps (Vespula vulgaris V germanica) in particular have had devastating effects in native forests and have displaced indigenous fauna through their predatory and polyphagous feeding habits High nest densities also limit access to and use of wilderness areas Insecticidal baits can reduce wasp populations for the short term on a local scale but contain potent toxins that pose environmental risks Novel baiting technologies are urgently needed by DOC and Regional Councils for use in conservation areas Baits must be palatable to Vespula spp and enable prolonged delivery of an efficacious yet selective control agent to the nest Using appropriate preservatives and humectants the field life of a protein bait was extended to 4 days without compromising its acceptance by foraging wasps Incorporation of waspactive isolates of Metarhizium anisopliae and Beauveria bassiana into the baits did not affect their attractiveness to wasps; nest traffic rates were significantly reduced by the Metarhizium treatment and infected larvae were recovered from nests exposed to each fungal treatment demonstrating the potential utility of this approach in a wasp mitigation strategy