Vector Preference and Disease Dynamics: A Study of Barley Yellow Dwarf Virus

Using simple analytical models of the probability of disease transmission and a spatially explicit computer simulation of the spread of the aphid—transmitted barley yellow dwarf virus, we examined the effect of vector preference for diseased or healthy hosts on the spread of an economically important plant pathogen. Our analytical models indicate that the effect of vector preference for diseased plants on the probability of disease spread depends on the frequency of diseased plants in the population. In a non—spatial environment with a high frequency of diseased plants, disease spread is favored by vectors preferring healthy plants. With a low frequency of diseased plants, disease spread is favored by vectors preferring diseased plants. The effect of vector preference depends on the amount of persistence exhibited by the disease. For persistently transmitted diseases, the vector remains infective for a long period after visiting a diseased host. Persistence increases the rate of spread for a vector preferring healthy hosts more than it increases the rate of spread for a vector preferring diseased hosts. Using a Markov chain model of disease transmission, we have shown that an increase in the spatial patchiness of the disease can lead to a decrease in the rate of disease spread by a vector capable of moving only limited distances. The effect of spatial disease structure depends on the preference behavior exhibited by the vector. Our spatially explicit computer simulation explored the effect of frequency, persistence, and spatial structure in a dynamic model. All of these factors were shown to be important in describing the impact of vector disease preference on epidemiology. Many of our results contrast with the assumption found in the agricultural literature that a preference for diseased plants leads to an increase in disease spread. These results may have implications for the evolution of pathogen—modified vector behavior and/or host attractiveness. Explicit knowledge of the interaction between spatial dynamics and vector preference will improve our ability to model epidemics and predict the spread of infectious diseases.