Computational ecology of vector-borne disease: Spatially detailed simulations and analyses

Vector-borne diseases have altered ecosystems and changed the course of human history. Therefore, it is important to better understand and predict the impact of vector-borne diseases on populations. To improve prediction, a combined approach of modeling and simulations was utilized. This approach was applied to a range of problems and the results were analyzed.; The first part of the dissertation centers on the analyses of how spatial patterns in host distribution drive the epidemic dynamics of vector-borne diseases. Infection with a pathogen requires vector infestation and the vector spreads only between hosts occupying the same neighborhood. The results of simulating a spatially detailed model indicate that increased host spatial heterogeneity reduces pathogen prevalence. Host clumping can lead to the physical separation of pathogen and vector in the initial phase of the epidemic process.; Secondly, spatial autocorrelation analyses were performed on the incidence rates and cases of Lyme disease in New York State. Join-counts methods and Moran's I---a global spatial autocorrelation statistic---revealed a consistent pattern of spatial dependence. The correlation distance over which incidence rates covary positively was estimated to be near 120 km. The results of a local spatial analysis around a major disease focal point in NY State, showed that global correlation distance matched the extent of the most intense local clustering. The spatial autocorrelation analyses of the Lyme disease epidemic may provide a spatial scale for regional control of the disease.; Finally, a spatially detailed model of superinfection was used to study how processes generate patterns as pathogen strains differing in virulence compete for hosts. Methods of adaptive dynamics were applied to examine the effects of spatially structured disease transmission on evolved levels of virulence and patterns in strain coexistence. In the model, superinfection, a form of contest competition between pathogen strains, depends explicitly on the difference in virulence levels. The simulation results indicated that spatial structure reduces disease virulence and that larger infection-transmission neighborhoods favor more virulent strains. Between-strain coexistence also increased with neighborhood size. A greater probability of superinfection increases convergent-stable virulence levels, and constraints between-strain coexistence.