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…

Mobile device to induce heat-stress on grapevine berries

Studying heat stress response of grapevine berries in the field often relies on weather conditions during the growing season. We constructed a mobile heating device, able to induce controlled heat stress on grapes in vineyards. The heater consisted of six 150 W infrared lamps mounted in a profile frame. Heating power of the lamps could be controlled individually by a control unit consisting of a single board computer and six temperature sensors to reach a pre-set temperature. The heat energy applied to individual berries within a cluster decreases by the squared distance to the heat source, enabling the establishment of temperature profiles within individual clusters. These profiles can be measured by infrared thermography once a steady state has been reached. Radiant flux density received by a berry depending on the distance was calculated based on a view factor and measured lamp surface temperature and resulted to 665 Wm-2 at 7cm. Infrared thermography of the fruit surface was in good agreement with measurements conducted with a thermocouple inserted at epidermis level. In combination with infrared thermography, the presented device offers possibilities for a wide range of applications like phenotyping for heat tolerance in the field to proceed in the understanding of the complex response of plants to heat stress. Sunburn necrosis symptoms were artificially induced with the aid of the device for cv. Bacchus and cv. Sylvaner in the 2020 and 2021 growing season. Threshold temperatures for sunburn induction (LT5030min) were derived from temperature data of single berries and visual sunburn assessment, applying logistic regression. A comparison of threshold temperatures for the occurrence of sunburn necrosis confirmed the higher susceptibility of cv. Bacchus. The lower susceptibility of cv. Sylvaner did not seem to be related to its phenolic composition, rendering a thermoprotective role of berry phenolic compounds unlikely.

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.

Mapping and tracking canopy size with VitiCanopy

Understanding vineyard variability to target management strategies, apply inputs efficiently and deliver consistent grape quality to the winery is essential. However, despite inherent vineyard variability, the majority are managed as if they are uniform. VitiCanopy is a simple, grower-friendly tool for precision/digital viticulture that allows users to collect and interpret objective spatial information about vineyard performance. After four years of field and market research, an upgraded VitiCanopy has been created to achieve a more streamlined, technology-assisted vine monitoring tool that provides users with a set of superior new features, which could significantly improve the way users monitor their grapevines. These new features include:
• New user interface
• User authentication
• Batch analysis of multiple images
• Ease the learning curve through enhanced help features
• Reporting via the creation of colour maps that will allow users to assess the spatial differences in canopies within a vineyard.
Use-case examples are presented to demonstrate the quantification and mapping of vineyard variability through objective canopy measurements, ground-truthing of remotely sensed measurements, monitoring of crop conditions, implementation of disease and water management decisions as well as creating a history of each site to forecast quality. This intelligent tool allows users to manage grapevines and make informed management choices to achieve the desired production targets and remain profitable.

Climate change impacts on Douro Region viticulture and adaptation measures

Climate has a significant impact in the success of any agricultural system, with a direct influence on the crops suitability to a given region, interfering on yield and quality and also with the economic sustainability of the productive activity. In the Douro Demarcated Region (RDD), as in most regions of the Mediterranean climate, the scarce precipitation (33% has less than 600 mm per year), and your high variability, associated with high rates of evapotranspiration during the summer, is usually one of the fundamental factors that limit the grapevine development, as well as the production and quality of the harvest. Thus, facing the scenario in temperature changes for the next decades (1.5-2.5°C) and confirming the predictions of precipitation decreases and/or great variability in the occurrence of heat waves and intense rainfall, the consequences for slope stability in mountain viticulture and sustainability of all operations involved, are risks to be taken into account. In this way, a deepest and sustained knowledge regarding the adaptation measures to adverse environmental conditions is of a crucial importance, enabling a more efficient adaptation of plant growth conditions and the optimization of production and quality of the grapevines. The development of this work, carried out in two commercial vineyards, one located in Soutelo do Douro, São João da Pesqueira, Cima Corgo sub-region, and another located in Numão, Vila Nova de Foz Côa, Douro Superior sub-region, it seeks to establish a relationship between climatic elements and physiological, productive and qualitative parameters, as well as to evaluate the effectiveness of adaptation measures, including different types of deficit irrigation (2002-2019) and the application of shading nets (2019-2020) in the physiological, viticultural and oenological behavior in the Touriga Nacional and Moscatel Galego Branco varieties, respectively. The results showed that the application of deficit irrigation allowed to significantly reduce the impact of the adverse weather conditions at key moments in the development of the grapevine, particularly in the period immediately before veráison and maturation, reducing the negative effects on the physiological processes and productivity, without compromise the must quality parameters. On the other hand, the application of shading nets significantly reduced de leaves temperature, allowing to increase the water potential, stomatal conductance and photosynthetic rate of grapes, which was reflected in the yield increase in the 2nd year of the study. For the maturation indicators, higher levels of total acidity, malic acid and assimilable nitrogen were obtained. The last measure presents a huge potential, being essential to carry out more years of trials to obtain stronger conclusions in terms of production parameters, but also in characteristics as important as the grape ripening components and the organoleptic characteristics of wines.

A multidisciplinary approach to evaluate the effects of the training system on the performance of “Aglianico del Vulture” vineyards

Vineyards are complex agro-ecosystems with high spatial and temporal variability. An efficient training system may counteract the adverse effects of this variability. Moreover, considering the climate change issues, choosing an efficient training system that enhances water use and protects the vines from radiative thermal stress has become a priority for the farmers. A multidisciplinary approach that assesses the soil-crop-yield-wine relationships of vineyards in a distributed and holistic way could bring added knowledge on the behavior of the different training systems. This ongoing research aimed to implement a multidisciplinary approach to study the behavior of “Aglianico del Vulture” grapevines trained with two different systems: a spurred cordon (SC) and an “Alberello in parete” (AL), grown in a high-quality wine production area of Basilicata region (Italy). The approach merged several methods and scales of soil, ecophysiology, must/wine quality, and spectral data collection to assess the influence of the training system. Homogeneous zones (HZs) in both training systems were defined through a procedure based on geomorphological classification, unmanned aerial vehicles (UAV) images analysis, and a traditional soil survey supported by geophysical scanning. During the 2021 season, TDR probes monitored soil water content, while grapevine health status was assessed using eco-physiological measurements (LWP, chlorophyll content, PSII photosynthetic efficiency, LAI, and point-based field spectroscopy). These grapevine in-vivo measurements validated the spectral vegetation indexes (NDVI, RENDVI, CVI, and TVI) derived from the UAV multispectral imagery, which monitored the grapevine status in a distributed and non-invasive way. Grape yield, quality of berries, must and wine were measured to assess the effects of the training systems. The first experimental year results showed the variability of the vineyards and revealed relationships among soil parameters, crop characteristics, and vegetation indices of the SC and AL training systems. This multidisciplinary study could bring new insights into the vineyard training system’s effects on grape yield and wine quality.