Species composition and abundance of aphids in Australian hop gardens and their impact on spatiotemporal patterns of Carlavirus epidemics

Abstract

The species composition and abundance of aphids in commercial cv. Agate and cv. Super Pride hop gardens in Tasmania, Australia, were characterized over three seasons (1999–2001). Gunns Plains recorded 14 aphid species and Bushy Park 11 species, with nine of these common to both sites. The majority of aphids were trapped in the first 2 months (October and November) of active hop growth in all three seasons. Cultivar and geographical location had significant effects on the abundance of total aphids (species pooled) trapped and several individual aphid species in the three seasons. In general, significantly more aphids (total and individual species) were trapped in cv. Agate than cv. Super Pride gardens, and higher numbers were trapped at Bushy Park than at Gunns Plains. This coincided with a higher incidence of plants infected by carlaviruses in cv. Agate gardens at both locations. Differences in the spatiotemporal dynamics of Carlavirus epidemics were described by fitting a stochastic model to the data, with parameters for local spread within the garden (contagion) and background infection (disease increase unrelated to infected plants within the gardens). Local spread of Hop latent virus (HpLV) and Hop mosaic virus (HpMV) was indicated within all gardens. For HpMV in cv. Agate at Gunns Plains, however, infections caused by immigrant viruliferous aphids were also apparent. Using join‐count statistics, spatial aggregation of both virus diseases was found for all years, except for the initial year (1999) when incidence was low. Clusters of diseased plants extended to greater distances for HpLV than for HpMV. Based on spatial and spatiotemporal analyses, local spread (mechanical transmission and/or aphid movement within the garden) appears to be the dominant factor in the epidemics of HpLV. Aphid immigration from outside the crop over time may play a more significant role for HpMV epidemics, at least for one location.