Vine field monitoring using high resolution remote sensing images: segmentation and characterization of rows of vines

Artificial intelligence (AI) for winegrowers refers to robotics, smart sensor technology, and machine learning applied to solve climate change problems. Our research group has developed novel technology based on AI in the vineyard to monitor vineyard growth using computer vision analysis (VitiCanopy App) and grape maturity based on berry cell death to predict flavor and aroma profiles of berries and final wines.

Global warming is one of the biggest environmental, social, and economic threats. In the Douro Valley, change to the climate are expected in the coming years, namely an increase in average temperature and a decrease in annual precipitation. Since vine cultivation is extremely vulnerable and influenced by the climate, these changes are likely to have negative effects on the production and quality of wine.
Adaptation is a major challenge facing the viticulture sector where the choice of plant material plays an important role, particularly the rootstock as it is a driver for adaptation with a wide range of effects, the most important being phylloxera, nematode and salt, tolerance to drought and a complex set of interactions in the grafted plant.
In an experimental vineyard, established in the Douro Region in 1997, with four randomized blocs, with five varieties, Touriga Nacional, Tinta Barroca, Touriga Franca and Tinta Roriz, grafted in four rootstocks, Rupestris du Lot, R110, 196-17C, R99 and 1103P, data was collected consecutively over 20 years (2001-2020). Phenological observations were made two to three times a week, following established criteria, to determine the average dates of budbreak, flowering and veraison. During maturation, weekly berry samples were taken to study the dynamics of sugar accumulation, amongst other parameters. Climate data was collected from a weather station located near the vineyard parcel, with data classified through several climatic indices.
The results achieved show a very low coefficient of variations in the average date of the phenophases and an important contribution from the rootstock in the dynamic of the phenology, allowing a delay in the cycle of up to10-12 days for the different combinations. The Principal Component Analysis performed, evaluating trends in the physical-chemical parameters, highlighted the effect of the climate and rootstock on fruit quality by grape varieties.

The wine industry, a key sector for the European Union’s economy, exhibits significant energy consumption, amounting to approximately 1,750 million kWh annually within this geographic context, with major contributions from Italy, France, Spain, and Portugal (Fuentes Pila et al., 2015).

The implementation of food safety assurance systems in wineries involves ensuring that the wines produced do not pose a risk to consumer health and are therefore free from harmful substances, such as those that may be incorporated during the production process (pesticides, additives, etc.), allergens or mycotoxins.

Aim: To investigate the possibility of utilising simultaneous measurements of absorbance-transmittance and fluorescence excitation-emission matrix (A-TEEM) combined with chemometrics, as a robust method that gives rapid results for classification of wines from different regions of South Australia according to their Geographical Indication (GI), and to gain insight into the effect of terroir on inter regional variation.