GiESCO 2019 banner
IVES 9 IVES Conference Series 9 GiESCO 9 GiESCO 2019 9 Aroma and quality assessment for vertical vintages using machine learning modelling based on weather and management information

Aroma and quality assessment for vertical vintages using machine learning modelling based on weather and management information


Context and purpose of the study ‐ Wine quality traits are usually given by parameters such as aroma profile, total acidity, alcohol content, colour and phenolic content, among others. These are highly dependent on the weather conditions during the growing season and management strategies. Therefore, it is important to develop predictive models using machine learning (ML) algorithms to assess and predict wine quality traits before the winemaking process.

Material and methods ‐ Samples in duplicates of Pinot Noir wines from vertical vintages (2008 to 2013) of the same winery located in Macedon Ranges, Victoria, Australia were used to assess different chemical analytics such as i) aromas using gas chromatography – mass spectrometry, ii) color density, iii) color hue, iv) degree of red pigmentation, v) total red pigments, vi) total phenolics, vii) pH, viii) total acidity (TA), and ix) alcohol content. Data from weather conditions from the specific vintages were obtained both from the bureau of meteorology (BoM) and the Australian Wine Availability Project (AWAP) climate databases. Such data consisted of: i) solar exposure from veraison to harvest (V‐H), ii) solar exposure from September to harvest (S‐H), iii) maximum January solar exposure, iv) degree days from S‐H, v) maximum January evaporation, vi) mean maximum temperature from veraison to harvest, vii) mean minimum temperature from V‐H, viii) water balance from S‐H, ix) solar exposure from V‐H, x) degree hour accumulation with base 25 – 30 °C, xi) degree hour accumulation with base 25 °C, xii) degree hour accumulation with base 30 °C, xiii) degree hour accumulation with base 35 °C, and xiv) total cumulative degree days accumulation with base 10 °C. All data were used to develop two machine learning (ML) regression models using Matlab® R2018b. The best models obtained were using artificial neural networks (ANN) with the Levenberg‐Marquardt algorithm with 5 neurons for Model 1 and 9 neurons for Model 2. Model 1 was developed using the 14 parameters from the weather data as inputs to predict 21 aromas found in the wines from the six different vinatges. Model 2 was developed using the same 14 parameters from weather data and the eight chemical parameters as targets and outputs.

Results ‐ Both models obtained presented very high accuracy to predict wine quality trait parameters. Model 1 had an overall correlation coefficient R = 0.99 with a high performance based on the mean squared error (MSE = 0.01), while Model 2 had an overall correlation coefficient R = 0.98 with a high performance (MSE = 0.03). These models would aid in the prediction of wine quality traits before its production, which would give anticipated information to winemakers about the product they would obtain to make early decisions on wine style variations.


Publication date: June 22, 2020

Issue: GiESCO 2019

Type: Article



School of Agriculture and Food, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, VIC 3010, Australia

Contact the author


wine quality, machine learning, weather, aromas


GiESCO 2019 | IVES Conference Series


Related articles…

Physiological and growth reaction of Shiraz/101-14 Mgt to row orientation and soil water status

Advanced knowledge on grapevine row orientation is required to improve establishment, management and outcomes of vineyards on terroirs with different environmental conditions (climate, soil, topography) and in view of a future change to more extreme climatic conditions. The purpose of this study was to determine the combined effect of row orientation, plant water status and ripeness level on the physiological and viticultural reaction of Shiraz/101-14 Mgt.

Effects of mechanical leafing and deficit irrigation on Cabernet Sauvignon grown in warm climate of California

San Joaquin Valley accounts for 40% of wine grape acreage and produces 70% of wine grape in California. Fruit quality is one of most important factors which impact the economical sustainability of farming wine grapes in this region. Due to the recent drought and expected labor cost increase, the wine industry is thrilled to understand how to improve fruit quality while maintaining the yield with less water and labor input. The present study aims to study the interactive effects of mechanical leafing and deficit irrigation on yield and berry compositions of Cabernet Sauvignon grown in warm climate of California.

The effects of cane girdling on berry texture properties and the concentration of some aroma compounds in three table grape cultivars

The marketability of the table grapes is highly influenced by the consumer demand; therefore the market value of the table grapes is mainly characterized by its berry size, colour, taste and texture. Girdling could cause accumulation of several components in plants above the ringing of the phloem including clusters and resulting improved maturity. The aim of the experiments was to examine the effect of girdling on berry texture characteristics and aroma concentration.

Application of a fluorescence-based method to evaluate the ripening process and quality of Pinot Blanc grape

The chemical composition of grape berries at harvest is one of the most important factors that should be considered to produce high quality wines. Among the different chemical classes which characterize the grape juice, the polyphenolic compound, such as flavonoids, contribute to the final taste and color of wines. Recently, an innovative non-destructive method, based on chlorophyll fluorescence, was developed to estimate the phenolic maturity of red grape varieties through the evaluation of anthocyanins accumulated in the berry skin. To date, only few data are available about the application of this method on white grape varieties.

Different yield regulation strategies in semi-minimal-pruned hedge (SMPH) and impact on bunch architecture

Yields in the novel viticulture training system Semi-Minimal-Pruned Hedge (SMPH) are generally higher compared to the traditional Vertical Shoot Positioning (VSP). Excessive yields have a negative impact on the vine and wine quality, which can result in substantial losses in yield in subsequent vintages (alternate bearing) or penalties in fruit quality. Therefore yield regulation is essential. The bunch architecture in SMPH differs from VSP. Generally there is a higher amount but smaller bunches with lower single berry weights in SMPH compared to VSP.