A MODELING ANALYSIS OF THE SUCCESSFUL BIOLOGICAL-CONTROL OF SITONA-DISCOIDEUS (COLEOPTERA, CURCULIONIDAE) BY MICROCTONUS-AETHIOPOIDES (HYMENOPTERA, BRACONIDAE) IN NEW-ZEALAND

1. Simple models were developed to explore the interaction between the lucerne (Medicago sativa L.) pest Sitona discoideus Gyllenhal (Coleoptera: Curculionidae) and its introduced braconid parasite Microctonus aethiopoides Loan (Hymenoptera: Braconidae) in New Zealand. 2. A model for the population dynamics of S. discoideus demonstrated that observed levels of parasitism account for a decline in pest abundance, in spite of strong density-dependent mortality in the host. 3. An empirical model for changes in parasitism from year to year, spanning five to six parasitoid generations each year, suggested that control is likely to be maintained in the long term and that the interaction is stable. 4. Tentative values were established for the average searching efficiency of M. aethiopoides (approximately 0.25 m2 per parasitoid per generation) and for parameters governing pseudo-interference and a functional response to host density, based on analysis of discrete parasitoid generations. 5. Based on experimental data and field temperatures, a simulation model of daily parasitoid behaviour over summer, when generations overlap, supported the hypothesis that atypically developing parasitoids are those oviposited in newly emerged adult weevils. The model accounted for the observed atypical summer build-up of parasitism which is essential for the success of biological control. 6. The models suggested that biological control of S. discoideus is complementary to other forms of control, including other biological control agents, because subsequent levels of parasitism appear little affected by changes in pest density. 7. It is argued that quantitative analysis of specific biological control case studies, both successful and unsuccessful, is a vital prerequisite for improvements in general biocontrol theory and practice.