Analysis of Spatiotemporal Dynamics of Virus Spread in an Australian Hop Garden by Stochastic Modeling

The dynamics of spread of Hop latent virus (HpLV), Hop mosaic virus (HpMV), and Apple mosaic virus (ApMV) in an Australian hop garden were characterized by fitting a stochastic spatiotemporal model, using the Monte Carlo Markov Chain (MCMC) method, to the changes in patterns of virus-infected plants. The model has parameters for background infection rate (from virus reservoirs outside the gardens) and local infection rate (spread from infected plants to virus-free plants). The HpLV epidemic could be explained by very short range spread within the garden, suggesting mechanical transmission through cultural operations. This is consistent with the absence of the only known aphid vector of HpLV from Australia. The HpMV epidemic could be explained by local spread coupled with background infection. This suggests that HpMV may be introduced by aphids, and subsequent localized spread may occur by aphids or mechanical means. The ApMV epidemic was also consistent with localized spread coupled with background infection, but the possibility of longer-range spread could not be rejected for this virus. Although intra-garden transmission of ApMV may occur through contact or cultural operations, these results suggest that other mechanisms could contribute to transmission of ApMV in Australian hop gardens. Join-count statistics for the patterns of infected plants supported the conclusion from the spatiotemporal modeling that short-range spread was the primary determinant of disease increase.