Wine learning – Study of a multi-disciplinary experimental model for the characterization of the “fermentation” system

Introduction – Paolo Capra, Ever Srl

Presentation of the experimental model, from the characterization of musts (chemistry, trace elements, aromatic precursors), to that of wines (chemistry and aromatics), through real-time monitoring of fermentations – Paolo Antoniali, Italiana Biotecnologie Srl

THiols: Methodology in UPLC-MS – Study and implementation of the method for the determination and quantification of thiols in wines – Sara Bogialli / Marco Roverso, University of Padua

MASUT DA RIVE: case study – Presentation of results after three years of predictive harvesting – Nicola Biasi, Nicola Biasi Consulting

“Learning from wine” means doing innovation and research together with winemakers and producers; collaborating and sharing knowledge to preserve and improve the quality of Wine every day.

After ten years of the Y-TEAM program for the development of the quality of yeast and its derivatives for oenological fermentation, Ever presents WINE-LEARNING: an innovative, multi-disciplinary, experimental model for the characterization of the “grape-fermentation-wine” system.

Predicting the progress of a fermentation is extremely useful and allows appropriate corrective action to be taken where necessary to ensure that winemaking ends successfully, whether it is a primary fermentation or a refermentation for sparkling wine making.
In order to develop a robust, flexible and reliable predictive model of wine fermentation, it is necessary to have a large database that includes data inherent in fermentations of a variety of grape varieties, from different areas and vintages; information related to the treatment of musts or bases, the yeast strain used and the nutrition employed and, last but not least, the operating conditions adopted.
Several mathematical models focusing on must composition and yeast physiology have been proposed to predict the kinetics of alcoholic fermentation. These models require the estimation of a large number of parameters, which can make them difficult to apply in an industrial winery situation and, in particular, during harvest.
The WINE LEARNING project, rather than defining a predictive model of wine fermentation, outlines an empirical diagnostic system that helps, through a hands-on approach and experimental observation, to characterize the state of the “grape-fermentation-wine” system and to define thoughtful operational choices prior to fermentations.