Walnut blight, caused by Xanthomonas arboricola pv. juglandis, is a disease of major economic importance in Tasmania, Australia. The current management strategy is based on weekly sprays of copper formulated with mancozeb for protecting susceptible fruit, which is not economically or environmentally sustainable. Using data from walnut blight epidemics at 10 site-years in Tasmania, moisture intensity, defined as the total daily rainfall divided by duration of surface wetness, accounted for 83% of the variance in the percentage of walnut fruits developing blight. Moisture intensity was used to develop a weather-based model that calculated a critical risk value for predicting the optimum time to apply crop protection. Two trials to test the model were conducted in commercial Vina orchards in northern (Forth) and eastern (Swansea) Tasmania. The temporal progression of disease was markedly different between locations with Gompertz and monomolecular growth models describing disease progression at Forth and Swansea respectively. At Forth, nearly 100% of non-treated fruits, and fruits treated with two bud-burst sprays only were diseased at harvest. Four bud-burst sprays significantly reduced disease incidence in comparison to two bud-burst sprays; however, disease control was inadequate with 66% incidence. Fewer than 40% of fruits at Forth had blight lesions with sprays timed according to the model or with the weekly-spray regime i.e., eight or nine sprays respectively. Crop yield was significantly increased at Forth through use of the model in comparison to bud-burst sprays and non-treatment. At Swansea, nearly 60% incidence was observed with non-treatment and bud-burst sprays only, whereas less than 15% incidence occurred with the model and weekly-spray regime. Five sprays were applied with the model at Swansea, three fewer than the weekly-regime.