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…

Influence of polysaccharide extracts from wine by-products on the volatile composition of sparkling white wines

In the production of sparkling wines, during the second fermentation, mannoproteins are released by yeast autolysis, which affect the quality of the wines. The effect of mannoproteins has been extensively studied, and may affect aroma and foam quality. However, there are no studies on the effect of other polysaccharides such as those from grapes. Considering the large production of waste from the wine industry, it was proposed to obtain polysaccharide-rich extracts from some of these by-products[1].

Determination of titratable acidity, sugar and organic acid content in red and white wine grape cultivars during ripening by VIS–NIR hy¬perspectral imaging

Grape harvest time is one of the most fundamental aspects that affect grape quality and thus wine quality. Many factors influence the decision of harvest; among them technological and phenolic maturity of grape. Technological ripeness is mainly related to sugar concentration, titratable acidity and pH. Conventional methods for chemical analysis of grapes are normally sample-destructive, time-consuming, include laborious sample preparation steps, and generate chemical waste, thereby limiting their utility in online/in-line quality monitoring. Moreover, destructive analyses can be performed only on a limited number of fruit pieces and, thus, their statistical relevance could be limited. This study evaluated the ability of a lab-scale hyperspectral imaging (HYP-IM) technique to predict titratable acidity, organic acid and sugar content of grapes. Samples of Cabernet franc and Chenin blanc grapes were consecutively collected six times at weekly intervals after veraison. The images were recorded thanks to the hyperspectral imaging camera Pica L (Resonon) in a spectral range from 400 to 1000 nm. Statistics were performed using Microsoft Xlstat software. Successively, the berries were analyzed for their sugar (glucose and fructose) and organic acid (malic and tartaric acid) content and titratable acidity according to usual methods.

Biodiversity in the vineyard agroecosystem: exploring systemic approaches

Biodiversity conservation and restoration are essential for guarantee the provision of ecosystem services associated to vineyard agroecosystem such as climate regulation trough carbon sequestration and control of pests and diseases. Most of published research dealing with the complexity of the vineyard agroecosystems emphasizes the necessity of innovative approaches, including the integration of information at different temporal and spatial scales and development of systemic analysis based on modelling. A biodiversity survey was conducted in the Franciacorta wine-growing area (Lombardy, Italy), one of the most important Italian wine-growing regions for sparkling wine production, considering a portion of the territory of 112 ha. The area was divided into several Environmental Units (EUs), defined as a whole vineyard or portion of vineyard homogenous in terms of four agronomic characteristics: planting year, planting density, cultivar, and training system. In each EU a set of compartments was identified and characterised by specific variables. The compartments are meteorology, morphology (altitude, slope, aspect, row orientation, and solar irradiance), ecological infrastructures and management. The landscape surrounding EU was also characterised in terms of land-use in a buffer zone of 500 m. For each component a specific methodology was identified and applied. Different statistical approaches were used to evaluate the method to integrate the information related to different compartments within the EU and related to the buffer zone. These approaches were also preliminarily evaluated for their ability to describe the contribution of biodiversity and landscape components to ecosystem services. This methodological exploration provides useful indication for the development of a fully systemic approach to structural and functional biodiversity in vineyard agroecosystems, contributing to promote a multifunctional perspective for the all wine-growing sector.

Leveraging the grapevine drought response to increase vineyard sustainability

In this video recording of the IVES science meeting 2024, Silvina Dayer (PhD in Agronomy, Les Sanctuaires du Mirazur-Groupe Mauro Colagreco, Menton, France) speaks about grapevine drought response to increase vineyard sustainability. This presentation is based on an original article accessible for free on IVES Technical Reviews.

Interest and impact of PVP/PVI (Polyvinylpyrrolidone/ Polyvinylimidazole) on winemaking and final quality of wines

Céline Sparrow a, Christophe Morge a, a SOFRALAB SAS, 79, av. A.A. Thévenet – CS 11031 – 51530 Magenta, France Consumers’ health and security force authorities to limit, in wine as in others food industry products, the concentration in « dangerous » molecules. Therefore the legal limit in heavy metals keeps on decreasing. As per proof EU regulation just decrease the stain concentration in wine from 0,2 to 0,15 mg/l. Certain changes , such as sodium arsenite treatment in vines, disappearance of brass in wineries to the benefit of stainless steel, limit even more the concentration of heavy metals in wines. But the use of copper derivates in vines treatments is difficult to replace. In the case of wine and its elaboration, the problem is even more complex. Indeed, regulation forces the wine producers to control the concentration of certain heavy metals in final wines.