Speaker
Description
Disease management is a key aspect of modern agriculture, but farmers must equally face the ever-growing concerns and increasingly strict governmental restrictions related to the use of pesticides. Unfortunately, farmers cannot afford to wait for symptoms to appear on their crops to know when to treat because, by that time, the disease has already settled in. Hence, decision-making tools have been developed to help farmers know when to treat their crops. Current tools are based on a variety of techniques but generally lack either the spatial or temporal resolution required to efficiently protect the harvest. Commonly, forecast techniques based on meteorological conditions as well as the lifecycle of the plant are used to determine an infection risk factor relevant to the area and the disease type, however, it lacks the essential knowledge as to whether the diseases are present in the fields and does not consider any interventions made. To this effect, holographic detectors were developed and placed in vineyards in various regions of Switzerland and France over five years with the objective of detecting and identifying airborne spores of downy and powdery mildew, two common grapevine diseases, before infection. The data is analysed using image processing techniques and artificial intelligence to correctly classify the disease from which the spores are released and identify any patterns representative of the infection risk. This method provides essential information on the quantitative development of fungal diseases and has been successfully used to identify the primary infection of downy mildew which was confirmed by a visual evaluation of symptoms within a control parcel. Furthermore, the real-time knowledge of the presence of spores in the air before the appearance of any symptoms, has allowed for the determination of the optimal time to deploy the implemented treatments as well as an evaluation of their effectiveness, resulting in a reduction, up to 30%, of fungicide use. This data coupled with the current risk prediction models enable farmers to optimise strategies in the management of grapevine diseases.
