INTEGRATING FUNGICIDE RESISTANCE MONITORING AND FORECASTING MODELS FOR THE PRECISION MANAGEMENT OF GRAPEVINE DOWNY MILDEW

The cultivation of grapevine (Vitis vinifera L.) represents one of the most important cropping systems worldwide, particularly within the European Union, where vineyards account for a substantial proportion of the global wine-growing area. Among the diseases affecting grapevine, downy mildew, caused by the biotrophic oomycete Plasmopara viticola, is one of the most destructive, with the potential to cause severe yield losses under favourable environmental conditions. In Europe, the widespread cultivation of a highly susceptible host and climatic conditions characterised by moderate rainfall and mild temperatures create optimal conditions for pathogen establishment and epidemic development. As a consequence, disease management still relies heavily on repeated applications of fungicides, which remain the most effective and widely adopted means of control. However, this high level of chemical input is associated with increasing concerns regarding environmental and human health impacts and has led to the progressive implementation of more restrictive regulatory frameworks promoting sustainable and resilient agricultural systems. Within this context, fungicide resistance has become a key constraint to the sustainable management of grapevine diseases, particularly for P. viticola, which is classified as a high-risk pathogen for resistance development due to its biological characteristics. The repeated and intensive use of single-site fungicides exerts strong selection pressure on pathogen populations, favouring the emergence and spread of resistant strains and potentially leading to reduced fungicide efficacy. Since the development of fungicide resistance cannot be fully avoided, resistance management has become an integral component of Integrated Pest Management (IPM), based on the principles of prediction, detection, and reduction of selection pressure. In this framework, disease forecasting models and resistance monitoring represent key tools to support rational decision-making and optimise fungicide use by aligning chemical interventions with actual infection risk. The aim of this PhD thesis was to improve the management of grapevine downy mildew in an IPM context by advancing current knowledge on fungicide resistance development and spread in P. viticola populations, while supporting the identification of strategies that optimise fungicide spray programmes and limit the emergence and persistence of resistance. To achieve this objective, the research combined quantitative monitoring of fungicide resistance under field conditions, the assessment of fitness and competitive ability of resistant strains, and the evaluation of disease forecasting models to support risk-based fungicide strategies. The results demonstrated that fungicide resistance is widespread in field populations and characterised by complex and dynamic patterns influenced by fungicide use history and mode of action. Resistance to CAA fungicides was fully established, oxathiapiprolin showed an intermediate resistance status, and populations remained largely sensitive to QiI fungicides. Despite the presence of resistant isolates, often displaying multiple resistance profiles, appropriately designed anti-resistance strategies ensured effective disease control even under high epidemic pressure, highlighting both the effectiveness and the limitations of fungicide-based approaches. The evaluation of the biological performance of resistant isolates showed that resistance to mandipropamid is associated with measurable but limited fitness costs, mainly affecting early vegetative development, while reproductive traits remained comparable to those of sensitive genotypes. Competitive interactions under fungicide-free conditions resulted in a decline in the frequency of resistant alleles, indicating a selective disadvantage in the absence of selection pressure. These findings suggest that resistance frequencies may decrease when fungicide applications are discontinued, although this process may be slow and reversible. The integration of epidemiological modelling with biological indicators demonstrated that the EPIcure forecasting model is able to accurately describe disease epidemic development and identify periods favourable to infection. The inclusion of empirical information on oospore germination dynamics allowed the identification of the temporal window of primary inoculum availability, improving the biological interpretation of infection risk and supporting more rational fungicide application strategies. Overall, the results of this thesis indicate that anti-resistance strategies based solely on fungicide use are insufficient to prevent the selection and persistence of resistant P. viticola populations, although they can ensure effective disease control when properly implemented. Sustainable management of grapevine downy mildew requires moving beyond fungicide-centred approaches towards integrated systems combining resistance monitoring, knowledge of pathogen fitness and population dynamics, and improved risk assessment through forecasting models. This integrated approach supports the rationalisation of fungicide use, reduces selection pressure, and contributes to the development of more sustainable and resilient viticultural systems.