Spatiotemporal dynamics of insects in large-scale agricultural systems: Methodology and analysis with the silverleaf whitefly, Bemisia argentifolii (Bellows and Perring

This study focused on the development of methods for studying and analyzing insect dynamics in agricultural systems. Emphasis was placed on the formulation of a spatiotemporal model as a tool for studying these dynamics. The utility of the model was demonstrated with the silverleaf whitefly (SLW), Bemisia argentifolii, in two contrasting spatial systems.;The model developed is a grid-based model that integrates ideas from spatially explicit population models, coupled map lattices and integrodifference equations (IDEs). It combines an age-structured prey-predator population model with a resource map that describes the structure of the insect's habitat. Spatial movement is accomplished with two-dimensional IDEs. One IDE is used to couple prey reproduction with their dispersal, while another IDE redistributes adult natural enemies over the spatial region at each time step in the model. The couplings in these IDEs are convolutions that are solved efficiently with Fourier transforms.;The dynamics generated from simulations with the model on artificial systems were similar to those observed with other discrete-time models that used different dispersal rules. These dynamics included spiral waves of predator chasing prey, crystal lattice structures, and spatiotemporal chaos.;The utility of the model was first demonstrated with the SLW at a small-scale mixed-crop organic system on Pine Island, FL. After simulation results were matched with whitefly population trends from pan-sampling data, the model was used to explore several scenarios. Simulations with experimental resource maps created by altering the arrangement of crops in the original resource map showed that changes in the spatial structure of resources at the farm affected SLW dynamics.;SLW dynamics in a large-scale regional system, the Imperial Valley, CA were also explored with the model. Resource maps were constructed from Landsat thematic mapper data on the Valley. Qualitative whitefly trends from simulations were consistent with those observed for the insect in the Valley. Simulations also showed that among all SLW host plants, alfalfa had the greatest influence on whiteflies. Planting no alfalfa resulted in a 6.4-fold reduction in overall whitefly abundance. The importance of alfalfa in the management of whiteflies in the Valley was noted.