Technical Yearbook 2024

However, traditional sensory analysis methods have limitations: • Human senses, including the sense of smell, can be influenced by physical and mental conditions and external factors. • Panel sensory analyses are prone to subjectivity and inter-panel variability. • Sensory analysis often requires significant training and resources. • Continuous exposure to aromas during a tasting session can lead to sensory fatigue among panellists, affecting their ability to detect and differentiate aromas accurately over time. • In many cases, panel tastings are qualitative, making determining the concentration of specific compounds in wine difficult. • The matrix effect in wine can mask subtle aromas, making them difficult for human tasters to identify. Analytical techniques such as gas chromatography can mitigate some of these limitations. However, such analytical tools require costly equipment, significant time, intricate sample preparation and expertise. The modern wine industry requires swift and straightforward quality assessment tools. E-noses offer a promising alternative, leveraging advanced sensor technology to analyse the aromatic composition of wine rapidly and objectively. Their key advantage lies in their ability to emulate the biological sense of smell. They convert volatile compounds in wine into detectable electric signals, often in digital form. These signals are then analysed to derive meaningful patterns relevant to the specific analysis. By circumventing the constraints of human perception, E-noses provide a practical and efficient means of assessing sensory properties, particularly aroma description, often related to wine quality. Practical use of e-noses E-noses equipped with different sensors have been used for the following applications:

• Detection of after-bottling wine defects, such as TCA and cork taint. • Detection of aromatic defects associated with Brettanomyces spoilage. • Evaluation of grape and wine quality. • Discrimination between musts coming from different grape varieties or picked at different levels of ripeness. • Identification of grape varieties used to make wine. • Assessment of the effect of added enzymes on aroma. • Monitoring carbonic maceration. • Detection of Botrytis aromas. • Monitoring aromatic differences between sparkling wines produced by traditional and Charmat methods. • Monitoring the effect of leaf removal on grape aromatic composition. • Monitoring varietal characteristics of specific grape varieties. • Authentication and quantification of red wines. • Detection of smoke taint in wines. The future of wine production The future holds the possibility of developing affordable and portable devices that can be used directly in the field to monitor possible problems, such as pathogen development. Integrating E-noses with artificial intelligence promises to further revolutionise the wine industry. Machine learning algorithms can enhance aroma analysis models, enabling producers to optimise winemaking techniques based on complex correlations between chemical composition and aroma profile. In conclusion, the marriage of E-noses with AI represents a new frontier of innovation in oenology, offering unprecedented levels of precision, efficiency and sophistication in wine production. As these technologies continue to evolve, they will undoubtedly shape the future of how we perceive, evaluate and appreciate the sensory delights of the vine. 

Reference https://www.wineland.co.za/vine-and-wine-innovation-watch-electronic-noses/

For more information, contact Karien O’Kennedy at karien@sawine.co.za.

97

TECHNICAL YEARBOOK 2024