Agrometeorology can be applied to the analysis and modelling of plant disease development and creating a decision-support system for the operational management of crop protection. An approach for the simulation of grapevine (Vitis vinifera) downy mildew (Plasmopara viticola) is described in this paper. Mathematical models for grapevine leaf area growth and downy mildew development and infection have been proposed and integrated. The inputs to the models are represented by the fundamental agrometeorological variables of air temperature (degrees C), relative humidity (%), rainfall (mm) and leaf wetness (0-1), and tuning parameters have been included to calibrate the models for any possible behaviour differences in the patho-system. The main outputs of the model are infection intensity during the growing season, and the timing of the different infection events. The research was conducted in Tuscany (Central Italy) where, since 1995, an experimental vineyard of Sangiovese variety has been monitored for the presence of the disease, and a standard station has been established for agrometeorological data collection. The model was calibrated on data collected in 1995 and 1996 and then applied and validated on an independent data set collected from 1998 to 2003. The results presented in this paper demonstrate how, based on the biology of both host and pathogen, the simulation calculated leaf area growth and the dynamics of disease infections during the season, giving an accurate description of the field situation.