A physiologically-based tritrophic metapopulation model of the African cassava food web

Abstract

The metapopulation dynamics of the African cassava food web is explored using a physiologically based tritrophic model. The interacting species are cassava, cassava mealybug and its natural enemies (two parasitoids, a coccinellid predator and a fungal pathogen), and the cassava greenmite and its natural enemies (two predators and a fungal pathogen). The metapopulation model is based on a single patch age-structured population dynamics model reported by Gutierrez et al. (Gutierrez, A.P., Wermelinger, B., Schulthess, F., Baumgärtner, J.U., Herren, H.R., Ellis, C.K., Yaninek, J.S., 1988b. Analysis of biological control of cassava pests in Africa: I. Simulation of carbon nitrogen and water dynamics in cassava. J. Appl. Ecol. 25, 901-920; Gutierrez, A.P., Neuenschwander, P. van Alphen, J.J.M., 1993. Factors affecting the establishment of natural enemies: biological control of the cassava mealybug in West Africa by introduced parasitoids. J. Appl. Ecol. 30, 706-721). The same model simulates the mass number dynamics of each plant or animal species in each patch and the movement of animals between patches. Movement is based on species specific supply–demand relations. The pathogen mortality rate is a simple function of rainfall intensity. The within-patch species composition, their initial densities, and the initial values of edaphic variables may be assigned stochastically. Sensitivity, graphical and multiple linear regression analyses are used to summarize the effects of spatial and resource heterogeneity on species dynamics. Important plant level effects on higher trophic levels are demonstrated, and recommendations are made as to the appropriate model for different ecological studies.