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

Abstract

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.

DOI:

Publication date: June 22, 2020

Issue: GiESCO 2019

Type: Article

Authors

Sigfredo FUENTES, Claudia GONZALEZ VIEJO, Xiaoyi WANG, Damir D. TORRICO

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

Contact the author

Keywords

wine quality, machine learning, weather, aromas

Tags

GiESCO 2019 | IVES Conference Series

Citation

Related articles…

Teasing apart terroir: the influence of management style on native yeast communities within Oregon wineries and vineyards

Newer sequencing technologies have allowed for the addition of microbes to the story of terroir. The same environmental factors that influence the phenotypic expression of a crop also shape the composition of the microbial communities found on that crop. For fermented goods, such as wine, that microbial community ultimately influences the organoleptic properties of the final product that is delivered to customers. Recent studies have begun to study the biogeography of wine-associated microbes within different growing regions, finding that communities are distinct across landscapes. Despite this new knowledge, there are still many questions about what factors drive these differences. Our goal was to quantify differences in yeast communities due to management style between seven pairs of conventional and biodynamic vineyards (14 in total) throughout Oregon, USA. We wanted to answer the following questions: 1) are yeast communities distinct between biodynamic vineyards and conventional vineyards? 2) are these differences consistent across a large geographic region? 3) can differences in yeast communities be tied to differences in metabolite profiles of the bottled wine? To collect our data we took soil, bark, leaf, and grape samples from within each vineyard from five different vines of pinot noir. We also collected must and a 10º brix sample from each winery. Using these samples, we performed 18S amplicon sequencing to identify the yeast present. We then used metabolomics to characterize the organoleptic compounds present in the bottled wine from the blocks the year that we sampled. We are actively in the process of analysing our data from this study.

Influence of a spontaneous cover crop on the vineyard and soil erosion under Mediterranean climate

Sixty five % of the agricultural area of the Basque Country located in the DO Ca Rioja corresponds to vineyards. More than 40% of it has an average slope greater than 10%, which makes it sensitive to erosive processes. Furthermore, it is foreseeable that extreme weather events (storms, hail, extreme heat and cold, etc.) will be favored due to climate change. Cover cropping can mitigate this risk, and therefore the objective of this work is to evaluate the impact that a vegetable cover has on the agronomic behavior of the vineyard, the quality of the grape and soil erosion. For this, a trial has been carried out with a Graciano variety vineyard with a slope between 10% -20% during the years 2020 and 2021. Conventional tillage management in the area has been compared (4-6 passes per year of tillage machinery) versus spontaneous vegetation cover management in the vineyard. This implies not tilling and allowing the grass of the land to colonize the range between the lines of vines, controlling their height through 1-3 mowing passes per year, always trying to affect the surface of the land as little as possible. The vegetative growth, yield and quality of the grape and wine was measured. Furthermore, erosion has been measured using Gerlasch boxes. The yield was lower in the second year of the trial in the cover crop treatment, but erosion was significantly reduced.

Grapevine yield estimation in a context of climate change: the GraY model

Grapevine yield is a key indicator to assess the impacts of climate change and the relevance of adaptation strategies in a vineyard landscape. At this scale, a yield model should use a number of parameters and input data in relation to the information available and be able to reproduce vineyard management decisions (e.g. soil and canopy management, irrigation). In this study, we used data from six experimental sites in Southern France (cv. Syrah) to calibrate a model of grapevine yield limited by water constraint (GraY). Each yield component (bud fertility, number of berries per bunch, berry weight) was calculated as a function of the soil water availability simulated by the WaLIS water balance model at critical phenological phases. The model was then evaluated in 10 grapegrowers’ plots, covering a diversity of biophysical and technical contexts (soil type, canopy size, irrigation, cover crop). We identified three critical periods for yield formation: after flowering on the previous year for the number of bunches and berries, around pre-veraison and post-veraison of the same year for mean berry weight. Yields were simulated with a model efficiency (EF) of 0.62 (NRMSE = 0.28). Bud fertility and number of berries per bunch were more accurately simulated (EF = 0.90 and 0.77, NRMSE = 0.06 and 0.10, respectively) than berry weight (EF = -0.31, NRMSE = 0.17). Model efficiency on the on-farm plots reached 0.71 (NRMSE = 0.37) simulating yields from 1 to 8 kg/plant. The GraY model is an original model estimating grapevine yield evolution on the basis of water availability under future climatic conditions.  It allows to evaluate the effects of various adaptation levers such as planting density, cover crop management, fruit/leaf ratio, shading and irrigation, in various production contexts.

The concept of terroir: what place for microbiota?

Microbes play key roles on crop nutrient availability via biogeochemical cycles, rhizosphere interactions with roots as well as on plant growth and health. Recent advances in technologies, such as High Throughput Sequencing Techniques, allowed to gain deeper insight on the structure of bacterial and fungal communities associated with soil, rhizosphere and plant phyllosphere. Over the past 10 years, numerous scientific studies have been carried out on the microbial component of the vineyard. Whether the soil or grape compartments have been taken into account, many studies agree on the evidence of regional delineations of microbial communities, that may contribute to regional wine characteristics and typicity. Some authors proposed the term “microbial terroir” including “yeast terroir” for grapes to describe the connection between microbial biogeography and regional wine characteristics. Many factors are involved in terroir including climate, soil, cultivar and human practices as well as their interactions. Studies considering “microbial terroir” greatly contributed to improve our knowledge on factors that shape the vineyard microbial structure and diversity. However, the potential impact of “microbial terroir” on wine composition has yet not received strong scientific evidence and many questions remain to be addressed, related to the functional characterization of the microbial community and its impact on plant physiology and grape composition, the origins and interannual stability of vineyard microbiota, as well as their impact on wine sensorial attributes. The presentation will give an overview on the role of microbiota as a terroir component and will highlight future perspectives and challenges on this key subject for the wine industry.

Differential responses of red and white grape cultivars trained to a single trellis system – the VSP

Commercial grape production relies on training grapevine cultivars onto a variety of trellis systems. Training allows for well-lit leaves and clusters, maximizing fruit quality in addition to facilitating cultivation, harvesting, and diseases control. Although grapevines can be trained onto an infinite variety of trellis systems, most red and white cultivars are trained to the standard VSP (Vertical Shoot Positioning) system. However, red and white cultivars respond differently to VSP in fruit composition and growth characteristics, which are yet to be fully understood. Therefore, the objective of this study was to examine the influence of the VSP trellis system on fruit composition of three red, Cabernet Sauvignon, Merlot and Syrah, and three white, Chardonnay, Riesling, and Gewurztraminer cultivars grown under uniform growing conditions in the same vineyard. All cultivars were monitored for maturity and harvested at their physiologically maximum possible sugar concentration to compare various fruit quality attributes such as Brix, pH, TA, malic and tartaric acids, glucose and fructose, potassium, YAN, and phenolic compounds including total anthocyanins, anthocyanin profile, and tannins. A distinct pattern in fruit composition was observed in each cultivar. In regards to growth characteristics, Syrah grew vigorously with the highest cluster weight. Although all cultivars developed pyriform seeds, the seed size and weight varied among all cultivars. Also varied were mesocarp cell viability, brush morphology, and cane structure. This knowledge of the canopy architectural characteristics assessed by the widely employed fruit compositional attributes and growth characteristics will aid the growers in better management of the vines in varied situations.