Analyzing and Modeling Temporal Disease Progress of Barley yellow dwarf virus Serotypes in Barley Fields

ABSTRACT Population dynamics of Padi avenae (PAV), Macrosiphum avenae (MAV), and Rhopalosiphum padi (RPV) virus serotypes of Barley yellow dwarf virus (BYDV) and of their main aphid vectors were studied in winter barley (Hordeum vulgare) fields for three successive years in western France. An epidemiological model of the spread of viruses in the field was developed based on vector populations as forcing variables and the population dynamics of each virus serotype. This model accurately simulated the kinetics of the epidemic for PAV serotypes, which are the most common ones. For RPV and to some extent for MAV, the results were less satisfactory. The occurrence and spread of PAV and MAV serotypes in the field was clearly and easily related to that of their main vector species. Conversely, the spread of RPV serotypes showed no consistent relationships with the dynamics of their vectors. Incidence of PAV in 1989 to 1990 and 1990 to 1991 showed a bimodal distribution, with maximums in fall (December) and spring (May) that were linked to fall infestations by R. padi and spring infestations by three (R. padi, Sitobion avenae, and Metopolophium dirhodum) or two (S. avenae and M. dirhodum) aphid species. In 1991 to 1992, the PAV infection curve was monomodal and mainly due to a primary spread of the virus by very large populations of alate R. padi. MAV incidence was low in fall and winter and reached a maximum in spring 1990 and 1991 related to the occurrence of S. avenae and M. dirhodum. RPV incidence was low every year, despite the abundance of its vector, R. padi. Mixed infections were more frequent than expected by chance and were assumed to be partly related to heterologous encapsidation. The occurrence of each serotype is discussed in relation to the time of crop infection and possible damage.