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…

Impact on leaf morphology of Vitis vinifera L. cvs Riesling and Cabernet Sauvignon under Free Air Carbon dioxide Enrichment (FACE)

Atmospheric carbon dioxide (CO2) concentration has continuously increased since pre-industrial times from 280 ppm in 1750, and is predicted to exceed 700 ppm by the end of 21st century. For most of C3 plant species elevated CO2 (eCO2) improve photosynthetic apparatus results in an increased plant biomass production. To investigate the effects of eCO2 on morphological leaf characteristics the two Vitis vinifera L. cultivars, Riesling and Cabernet Sauvignon, grown in the Geisenheim VineyardFACE (Free Air Carbon dioxide Enrichment) system were used. The FACE site is located at Geisenheim University (49° 59′ N, 7° 57′ E, 94 m above sea level), Germany and was implemented in 2014 comparing future atmospheric CO2-concentrations (eCO2, predicted for the mid-21st century) with current ambient CO2-conditions (aCO2). Experiments were conducted under rain-fed conditions for two consecutive years (2015 and 2016). Six leaves per repetition of the CO2 treatment were sampled in the field and immediately fixed in a FAA solution (ethanol, H2O, formaldehyde and glacial acetic acid). After 24 h leaf samples were transferred and stored in an ethanol solution. Subsequently, leaf tissue was dehydrated using ethanol series and embedded in paraffin. By using a rotary microtomesections of 5 µm were prepared and fixed on microscopic slides. Subsequent the samples were stained using consecutive staining and washing solutions. Afterwards pictures of the leaf cross-sections were taken using a light microscope and consecutive measurements were conducted with an open source image software. Differences found in leaf cross-sections of the two CO2 treatments were detected for the palisade parenchyma. Leaf thickness, upper and lower epidermis and spongy parenchyma remained less affected under eCO2 conditions. The observed results within grapevine leaf tissues can provide first insights to seasonal adaptation strategies of grapevines under future elevated CO2 concentrations.

Modeling island and coastal vineyards potential in the context of climate change

Climate change impacts regional and local climates, which in turn affects the world’s wine regions. In the short term, these modifications rises issues about maintaining quality and style of wine, and in a longer term about the suitability of grape varieties and the sustainability of traditional wine regions. Thus, adaptation to climate change represents a major challenge for viticulture. In this context, island and coastal vineyards could become coveted areas due to their specific climatic conditions. In regions subject to warming, the proximity of the sea can moderate extremes temperatures, which could be an advantage for wine. However, coastal and island areas are particular prized spaces and subject to multiple pressures that make the establishment or extension of viticulture complex.
In this perspective, it seems relevant to assess the potentialities of coastal and island areas for viticulture. This contribution will present a spatial optimization model that tends to characterize most suitable agroclimatic patterns in historical or emerging vineyards according to different scenarios. Thanks to an in-depth bibliography a global inventory of coastal and insular vineyards on a worldwide scale has been realized. Relevant criteria have been identified to describe the specificities of these vineyards. They are used as input data in the optimization process, which will optimize some objectives and spatial aspects. According to a predefined scenario, the objectives are set in three main categories associated with climatic characteristics, vineyards characteristics and management strategies. At the end of this optimization process, a series of maps presents the different spatial configurations that maximize the scenario objectives.

Drought effect on aromatic and phenolic potential of seven recovered grapevine varieties in Castilla-La Mancha region (Spain)

The effects of climate change are seriously affecting the quality of wine grapes. High temperatures and drought cause imbalances in the chemical composition of grapes. The result is overripe grapes with low acidity and high sugar content, which produce wines with excessive alcohol content, lacking in freshness and not very aromatic. As a consequence, the search of varieties with capacity of produce quality grapes in adverse climate conditions is a good alternative to preserve the sustainability of vineyards. In this work, quality parameters of seven Vitis vinifera L. cultivars (five whites and two reds) recently recovered from extinction and grown under two different hydric regimes (rainfed and irrigated) were analyzed during the 2020 vintage. At harvest time, weight of 100 berries, must physicochemical parameters (brix degree, total acidity, malic acid, pH), and carbon and oxygen isotope ratios (δ13C, δ18O) were determined. Subsequently, varietal aroma potential index (IPAv) and total polyphenol index (TPI) were analyzed. Quality parameters, IPAv and TPI, showed significant differences between varieties and water regimes. Both red varieties, Moribel and Tinto Fragoso, stood out for their high aromatic and phenolic potential, which was higher under rainfed regime. Regarding to white varieties, Montonera del Casar and Jarrosuelto stood out in terms of varietal aroma potential. Montonera del Casar high acidity in its musts and Jarrosuelto showed the highest berry weights.

Under-vine management effects on grapevine production, soil properties and plant communities in South Australia

Under-vine (UV) management has traditionally consisted of synthetic herbicide use to limit competition between weeds and grapevines. With growing global interest towards non-synthetic chemical use, this study aimed to capture the effects of alternative UV management at two commercial Shiraz vineyards in South Australia, where the sole management variables were UV management since 2016. In adjacent treatment blocks, cultivation (CU) was compared to spontaneous vegetation (SV) in McLaren Vale (MV), and herbicide was compared to SV in Eden Valley (EV). Soil water infiltration rates were slower and grapevine stem water potential was lower in CU compared to SV in MV, with the latter having a plant community dominated by soursob (Oxalis pes-caprae) during winter; while in EV, there was little separation between the treatments. Yields were affected at both sites, with SV being higher in MV and HE being higher in EV. In MV, the only effect on grape must was a lower 13C:12C isotope ratio in CU, indicating greater grapevine water stress. In the grape must at EV, SV had higher total soluble solids, total phenolics, anthocyanins, and yeast available nitrogen; and lower pH and titratable acidity. Pruning weights were not affected by the treatments in MV, while they were higher in HE at EV. Assessments revealed that the differing soil types at the two sites were likely the main determinants of the opposing production outcomes associated with UV management. In the silty loam soil of MV, the higher yields in SV were likely due to more plant-available water, as a potential result of the continuous soil bio-pores formed by winter UV vegetation. Conversely, in the loamy sand soils of EV with a lower cation exchange capacity, the lower yields and pruning weights in SV suggest the UV vegetation competed significantly with the grapevines for available water and nutrients.

A blueprint for managing vine physiological balance at different spatial and temporal scales in Champagne

In Champagne, the vine adaptation to different climatic and technical changes during these last 20 years can be seen through physiological balance disruptions. These disruptions emphasize the general grapevine decline. Since the 2000s, among other nitrogen stress indicators, the must nitrogen has been decreasing. The combination of restricted mineral fertilizers and herbicide use, the growing variability of spring rainfall, the increasing thermal stress as well as the soil type heterogeneity are only a few underlying factors that trigger loss of physiological balance in the vineyards. It is important to weigh and quantify the impact of these factors on the vine. In order to do so, the Comité Champagne uses two key-tools: networking and modelization. The use of quantitative and harmonized ecophysiological indicators is necessary, especially in large spatial scales such as the Champagne appellation. A working group with different professional structures of Champagne has been launched by the Comité Champagne in order to create a common ecophysiology protocol and thus monitor the vine physiology, yearly, around 100 plots, with various cultural practices and types of soil. The use of crop modelling to follow the vine physiological balance within different pedoclimatic conditions enables to understand the present balance but also predict the possible disruptions to come in future climatic scenarios. The physiological references created each year through the working group, benefit the calibration of the STICS model used in Champagne. In return, the model delivers ecophysiology indicators, on a daily scale and can be used on very different types of soils. This study will present the bottom-up method used to give accurate information on the impacts of soil, climate and cultural practices on vine physiology.