Examination of the effect of aphid vector population composition on the spatial dynamics of citrus tristeza virus spread by stochastic modeling.

ABSTRACT Aphid vector species population composition is known to affect the spatial patterns of citrus tristeza virus (CTV) and the changes in these patterns over time. However, the biological processes that are associated with virus spread have not been well defined. The spatiotemporal dynamics of CTV were examined using data collected from research plots in the Dominican Republic and Costa Rica, where the brown citrus aphid (BCA), Toxoptera citricida, was the predominant species, and in Florida, where the BCA was absent and the melon aphid, Aphis gossypii, was the predominant vector. Data were analyzed using a spatiotemporal stochastic model for disease spread, and parameter values were evaluated using Markov chain Monte Carlo stochastic integration methods. Where the melon aphid was the dominant species, the model parameter likelihood values supported the hypothesis that the disease was spread through a combination of random background transmission (transmission originating from inoculum sources outside the plot) and a local interaction (transmission from inoculum sources within the plot) operating over short distances. Conversely, when BCA was present, results often suggested a local short-range transmission interaction that was not restricted to nearest-neighbor interactions and that the presence of background infection was not necessary to explain the observations.