A mechanistic model simulating ascosporic infections by Erysiphe necator, the powdery mildew fungus of grapevine

A new dynamic model for Erysiphe necator ascosporic infections on grapevine was developed. Between budbreak of vines and the time when the pool of ascospores is depleted, the model uses weather data for calculating, at daily intervals: curve of ascospore maturation; ascospore discharge events and relative proportion of the discharged ascospores; infection periods and their relative infection severity; and progress of latency period and time when secondary infections should begin. The model was validated over a 4-year period (2005-2008) in 26 vineyards in Italy by comparing model predictions with actual observations of the first seasonal symptoms of powdery mildew. The model showed high sensitivity, specificity and accuracy. Proportions of true and false positive predictions were TPP = 0 center dot 94 and FPP = 0 center dot 26, respectively. Because a proportion of predicted infection periods did not result in actual disease onset, confidence was higher for prediction of non-infections than for prediction of infections. Most of the false positive predictions occurred in the earlier growth stages of the host, when the surface area of susceptible tissue may be very small so that the probability that ejected ascospores land on susceptible tissue is low. An equation was then developed to describe the probability that a predicted infection period results in disease onset as a function of the growth stage of vines at the time of prediction. The new model should improve early season powdery mildew management by helping vineyard managers schedule fungicide sprays or schedule the scouting of the vineyard for detection of first disease signs