Using a grape compositional model to predict harvest time and influence wine style

Linking wine composition to fruit composition is difficult due to the numerous biochemical pathways and substrate transformations that occur during fermentation. Grape composition regulates the production and final concentrations of most wine aroma compounds, as exemplified by methoxypyrazine and rotundone concentrations in wine being confidently predicted from the corresponding grape concentration. However, the final concentrations of many compounds in wines (aromatic and non-aromatic) are substantially dependent on the winemaking process.
The aim of this study was to better understand grape flavour evolution in relation to wine composition and subsequent wine style using sequential harvests (n=3). To achieve this goal, Shiraz was chosen as a model variety across two different climatic regions (warm-hot and cool-temperate) in New South Wales, Australia. The objective was not to compare the two regions but to assess the consistency of grape flavour evolution over the ripening period.

Irrespective of the region, a clear separation of samples was noted according to the harvest stage. Shiraz wines from the first harvest (H1) were associated with red fruit descriptors and higher acidity. Wines from the third harvest (H3) were correlated with dark fruit characters and a higher perception of alcohol. Higher concentrations of some higher alcohol acetates, dimethyl sulfide and lower concentrations of Z-3-hexenol, ethyl isobutyrate and ethyl leucate were measured in H3 wines.
Irrespective of the environment, this study demonstrated that in Shiraz, a common evolution of grape flavours exists, influencing the final wine sensory properties. Furthermore, during the late ripening stage, no direct nexus was observed between sugar concentration and grape and wine flavour evolution.