New frontiers in quantifying smoke taint in vineyards

by | Oct 31, 2019 | Winetech Scan

The aim of this projects was to start development of a non-destructive, in field method to detect what the extent of smoke taint is in grapevine canopies, after exposure to smoke from fires.

PROJECT LAYOUT:

Seven cultivars were exposed to smoke under controlled conditions in two sites over two seasons. Sauvignon Blanc, Pinot Gris, Chardonnay and Pinot Noir in a commercial vineyard in Adelaide Hills region was used and Shiraz, Cabernet Sauvignon and Merlot in a vineyard in Adelaide.

In the first part, physiological measurements were taken inside the canopy to determine degree of smoke contamination after exposure to smoke:

  • Stomatal conductance was measured and infrared thermal images were acquired. Using machine learning modelling based on pattern recognition, which is based on predictable changes in stomatal conductance patterns as determined from infrared thermal image analysis, a model was developed to detect the degree of smoke contamination in canopies.

In the second part, smoke taint was assessed in berries at harvest and the final wine:

  • Full berries were scanned using a spectrophotometer. Wine was made on a small scale and chemical analyses (including volatile phenols and guaiacol glycoconjugates) done on the wine. A model was developed to predict smoke taint in berries and wine, by quantifying levels of smoke taint related compounds, using near-infrared spectroscopy (NIR), as inputs for machine learning fitting modeling.

Statistical analyses were done on the data.

RESULTS:

  • The results showed that the pattern recognition model to detect smoke contamination of canopies, from the first part of the experiment, was 96% accurate.
  • The second model to predict smoke taint compounds in berries and wine fit the near infrared spectroscopy data with a correlation coefficient (R) of 0.97.

SIGNIFICANCE OF THE STUDY:

Currently, there are no practical, in-field methods to measure the extent of contamination or smoke taint on grapes. The models developed in this study, combined with affordable geo-referenced NIR spectroscopy measurements of berries can provide a rapid, non-destructive and reliable screening tool. This technology has the potential to allow producers to map the severity of smoke taint in vineyards which will aid in decision making at harvest, to manage smoke taint in wines.

Link to article: https://www.mdpi.com/1424-8220/19/15/3335


Image created by Stephany Baard combining Kevin Crause and Shutterstock images

Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors