GiESCO 2019 banner
IVES 9 IVES Conference Series 9 GiESCO 9 Machines and fire: developing a rapid detection system for grapevine smoke contamination using NIR spectroscopy and machine learning modelling

Machines and fire: developing a rapid detection system for grapevine smoke contamination using NIR spectroscopy and machine learning modelling

Abstract

Context and purpose of the study – Bushfires are a common occurrence throughout Australia and their incidence is predicted to both rise and increase in severity due to climate change. Many of these bushfires occur in areas close to wine regions, which receive different levels of exposure to smoke. Wine produced from smoke-affected grapes are characterised by unpalatable smoky aromas such as “burning rubber”, “smoked meats” and “burnt wood”. These smoke tainted wines are unprofitable and result in significant financial losses for winegrowers. This study investigated the use of near-infrared (NIR) spectroscopy and machine learning (ML) modelling for the rapid and non-destructive detection of grapevine smoke exposure by analysing grapevine leaves and/or grape berries.

Materials and methods – The trial was conducted during the 2018/2019 season at the University of Adelaide’s Waite campus in Adelaide, South Australia (34° 58’ S, 138° 38’ E) and involved the application of five different smoke and water misting treatments to Cabernet Sauvignon grapevines at approximately seven days post-veraison. Treatment vines were exposed to straw-based smoke for one hour under experimental conditions described previously by Kennison et al. (2008) and Ristic et al. (2011). Near-infrared (NIR) measurements were then taken from berries and leaves a day after smoking using the microPHAZIR TM RX NIR Analyser (Thermo Fisher Scientific, Waltham, USA) which has a spectral range of 1600-2396 nm. The NIR spectra were then used as inputs to train different ML algorithms, which resulted in two artificial neural networks (ANNs) with the best classification performance for either berry or leaf readings according to the different smoke treatments.

Results – Both ANN models found were able to correctly classify the leaf and berry spectral readings with high accuracy. The leaf model had an overall accuracy of 95.2%, 97.7% accuracy during training with a mean square error (MSE) 0.0082, 90.9% during validation with a MSE of 0.0353 and 88.1% during the testing stage with a MSE of 0.0386, while the berry model had an overall accuracy of 91.7%, 95.2% accuracy during training with a MSE of 0.0173, 86.4% during validation with a MSE of 0.0560 and 80.2% during the testing stage with a MSE of 0.0560. These results showed the potential of developing a rapid, non-destructive, in-field detection system for assessing grapevine smoke contamination following a bushfire using NIR spectroscopy and artificial neural network modelling.

DOI:

Publication date: September 28, 2023

Issue: GiESCO 2019

Type: Poster

Authors

Vasiliki SUMMERSON, Claudia GONZALEZ VIEJO, Damir TORRICO, Sigfredo FUENTES*

The University of Melbourne, School of Agriculture and Food, Faculty of Veterinary and Agricultural Sciences, Parkville 3010, Victoria, Australia

Contact the author

Keywords

bushfires, machine learning, smoke taint, climate change, non-destructive

Tags

GiESCO | GiESCO 2019 | IVES Conference Series

Citation

Related articles…

Mechanization of pre-flowering leaf removal under the temperate-climate conditions of Switzerland

Grapevine leaf removal (LR) in the cluster area is typically done between fruit set and cluster closure to create an unfavorable microclimate for fungal diseases, such as Botrytis cinerea and powdery mildew. Grape growers are now turning their attention to pre-flowering LR, which has additional benefits under certain conditions. When applied before flowering, LR strongly affects fruit set and thus the number of berries per cluster. It is therefore a good yield control tool, replacing time-consuming manual cluster thinning (Poni et al. 2006). It also improves berry structure, that is, skin thickness, skin-to-pulp ratio, and berry composition (total soluble solids, titratable acidity, and polyphenols) (Palliotti et al. 2012; Komm and Moyer 2015). By exacerbating competition for assimilates between reproductive and vegetative organs, pre-flowering LR also poses some risks. Excessive yield loss at the same year’s harvest due to a too low fruit set rate is the main concern: intensive pre-flowering LR (100% of the cluster area) can induce up to 50% yield loss in potted vines (Poni et al. 2005). Other parameters, such as cool climatic conditions during flowering, also affect fruit set rate and make it difficult to predict potential yield at harvest. Repeated and overly intensive preflowering LR can have repercussions over time and induce a decline in bud fruiting and plant vigor (Risco et al. 2014).

Aromatic maturity is a cornerstone of terroir expression in red wine

Harvesting grapes at adequate maturity is key to the production of high-quality red wines. Enologists and wine makers define several types of maturity, including technical maturity, phenolic maturity and aromatic maturity. Technical maturity and phenolic maturity are relatively well documented in the scientific literature, while articles on aromatic maturity are scarcer. This is surprising, because aromatic maturity is, without a doubt, the most important of the three in determining wine quality and typicity (including terroir expression). Optimal terroir expression can be obtained when the different types of maturity are reached at the same time, or within a short time frame. This is more likely to occur when the ripening takes place under mild temperatures, neither too cool, nor too hot. Aromatic expression in wine can be driven, from low to high maturity, by green, herbal, fresh fruit, ripe fruit, jammy fruit, candied fruit or cooked fruit aromas. Green and cooked fruit aromas are not desirable in red wines, while the levels of other aromatic compounds contribute to the typicity of the wine in relation to its origin. Wines produced in cool climates, or on cool soils in temperate climates, are likely to express herbal or fresh fruit aromas; while wines produced under warm climates, or on warm soils in temperate climates, may express ripe fruit, jammy fruit or candied fruit aromas. Growers can optimize terroir expression through their choice of grapevine variety. Early ripening varieties perform better in cool climates and late ripening varieties in warm climates. Additionally, maturity can be advanced or delayed by different canopy management practices or training systems.

Terroirs and legal protection

Le concept AOC permet, par une délimitation précise, la mise en valeur de terroirs particulièrement adaptés à la viticulture. Seuls les terroirs ainsi identifiés peuvent produire des vins portant le nom de l’AOC. Le nom de cette AOC ne peut être utilisé que pour des vins issus de terroirs compris dans l’aire d’appellation, sous peine de sanctions pénales. La délimitation ainsi opérée participe à la protection du nom de l’AOC. A l’inverse, le terroir délimité n’est pas protégé.

Characterizing graft union formation in different scion/rootstock combinations of grapevine 

In most viticultural regions, grapevines are cultivated grafted, employing either hybrid or pure species of various American Vitis spp., such as V. berlandieri, V. rupestris, and V. riparia, as grapevine rootstocks. These rootstocks play a crucial role in providing resistance to the Phylloxera insect pest. Beyond Phylloxera resistance, it is desirable for grapevine rootstocks to exhibit resistance to other soil-borne pathogens and adaptability to abiotic stress conditions. The introduction of new rootstocks holds promise for adapting agriculture to climate change without altering the characteristics of the final harvested product.

Heat berry: the influence of abiotic factors on the composition of berries, must and wine in Vitis vinifera L. CV Riesling

Recurring heat and drought episodes during the growing season can produce adverse impacts on grape production in many wine regions around the world.