Application of fuzzy logic for the simulation of Plasmopara viticola using agrometeorological variables*

A mechanistic model called PLASMO was developed earlier to simulate grapevine downy mildew (Plasmopara viticola) and has been applied in several viticultural areas of Italy since 1988 by the collaboration of several research institutions of Firenze. In this study, a new simulation model based on fuzzy logic has been developed for the same structure (biological cycle of P. viticola). This approach allows classical quantitative information to be used together with qualitative information. Vague concepts can also be handled. Agrometeorological data is used, with an hourly time step, starting from budbreak to the end of the growing season. Air temperature, relative humidity, rainfall and leaf wetness are required. The simulated processes are the growth of grapevine leaf area and the main phases of the biological cycle of the pathogen: incubation, sporulation, germination, spore survival and inoculation. The main epidemiological outputs are timing of infection events and disease intensity. The performance of the model is evaluated and the mechanistic and fuzzy logic approaches are compared.