Untangle berry shrivel environmental risk factors and quantify symptoms with AI – GeomAbs meets BAISIQ

Abstract

Berry Shrivel (BS, Traubenwelke) is a sugar accumulation disorder of grapevine of unknown causes, having a great negative impact on grape quality and incalculable risks for yield losses, and for which no reliable curative practices are available. The Austrian autochthonous red wine cultivar Blauer Zweigelt is heavily affected by BS, but the intensity of the damage can change from year to year making it unpredictable and hard to quantify. The induction of BS may include environmental factors and imaging technologies in combination with AI may facilitate a yield loss prediction. These questions are the aims of two projects starting 2025. In GeomaBS, we are going to identify the potential underlying environmental factors influencing BS abundance in vineyards to develop BS risk factors. Our experimental approach involves a) to link BS information with spatial and temporal maps representing geology, soil, management and climate, b) to determine the consequences of BS grapes on wine aroma profiles, and c) to extract BS risk factors of a study region and to translate the obtained information also to other viticulture risks. A reliable and standardized, easy and cost-effective, method to quantify BS in vineyards would be very helpful to support growers’ decisions. The project BAISIQ aims to establish an image-base quantification method of BS in vineyards using explainable AI. The BAISIQ partnership includes experts in the fields of grapevine biology working on BS, explainable artificial intelligence and data science, and image acquiring on the proximal and remote scale and their processing. Our results will close current knowledge gaps on BS, establish methods for BS quantification, develop a database of BS, combine multi-sensor data with machine learning methods for yield quantification, further develop algorithm for image analyses, and provide tools and recommendations for wine growers to cope with BS.