Adaptation of the Forecasting Model DOWNCAST for Determination of Downy Mildew Epidemics of Oilseed Poppy in Tasmania, Australia

Peronospora cristata, causing downy mildew, has become an economically important pathogen of oilseed poppy (Papaver somniferum) since its first record in Tasmania in 1996. DOWNCAST, originally developed for downy mildew of onion, was evaluated for its ability to predict sporulation and infection of poppy by P. cristata. To assess the influence of the parameters within DOWNCAST, two modifications (POPCAST1 and POPCAST2) of the standard DOWNCAST model were also developed and tested. Initial accuracy of determination of sporulation events by DOWNCAST was 75%, while POPCAST1 (but not POPCAST2) was a significantly more accurate predictor of sporulation events (86%) than DOWNCAST (χ2 = 5.74, P = 0.025). This increase in accuracy was attributed to increasing both the rainfall threshold inhibitory to sporulation from 0.2 to 3.0 mm, and the relative humidity threshold for induction and development of sporulation from 95% to 96%. While DOWNCAST, POPCAST1 and POPCAST2 had accuracies of infection determination of 65%, 80% and 75%, respectively, chi-squared analysis indicated no significant difference between models. Preliminary field evaluation indicated that POPCAST1 and POPCAST2 were more sensitive in detecting early infection events than DOWNCAST. Field observations indicated that conditions are frequently conducive for downy mildew spread, therefore the ability to predict early season infection events would allow growers to reduce initial infection rates, and thus inoculum loads later in the season. POPCAST1 and POPCAST2 represent a significant step towards the development of a predictive model for downy mildew in poppy crops. © 2007 The Authors.