Selecting non-target species for risk assessment of entomophagous biological control agents: Evaluation of the PRONTI decision-support tool

Abstract

The release of entomophagous biological control agents can pose risks to non-target invertebrate species in the release area and beyond. Pre-release risk assessment of these agents often involves tests with non-target species; however, selecting appropriate test species can be difficult when there is a large number to choose from. The PRONTI (priority ranking of non-target invertebrates) tool has been developed to aid this selection process. This automated tool prioritises species for testing using five criteria: (1) direct and indirect hazards posed by the agent, (2) likelihood of exposure to the hazards, (3) ecological impacts that may result from that exposure, (4) species’ anthropocentric value and (5) testability. Criteria (1) and (2) produce a risk estimate that drives the ranking process. In a test of PRONTI’s ability to identify non-target species at most risk from a proposed biological control agent, we used a generalist predator already present in New Zealand, the Asian paper wasp Polistes chinensis, as if it were the agent in a hypothetical biocontrol programme aimed at lepidopteran pests in New Zealand kiwifruit orchards. A ranked list of 340 invertebrate taxa known to occur in kiwifruit orchards was produced. To validate the risk estimates for a direct attack by P. chinensis on each taxon, wasps were introduced to kiwifruit orchards and prey taxa identified. Risk estimates were accurate except where identified prey taxa had not previously been recorded from kiwifruit orchards.