Does wine expertise influence semantic categorization of wine odors?

A large varietal collection including over 1700 varieties was maintained in Conegliano, ITA, since the 1950s. Phenological data on a subset of 400 grape varieties including wine grapes, table grapes, and raisins were acquired at bud break, flowering, veraison, and ripening since 1964. Despite the efforts in maintaining and acquiring data over such an extensive collection, the data set has varying degrees of missing cases depending on the variety and the year. This is ubiquitous in phenology datasets with significant size and length. In this work, we evaluated four state-of-the-art methods to estimate missing values in this phenological series: k-Nearest Neighbour (kNN), Multivariate Imputation by Chained Equations (mice), MissForest, and Bidirectional Recurrent Imputation for Time Series (BRITS). For each phenological stage, we evaluated the performance of the methods in two ways. 1) On the full dataset, we randomly hold-out 10% of the true values for use as a test set and repeated the process 1000 times (Monte Carlo cross-validation). 2) On a reduced and almost complete subset of varieties, we varied the percentage of missing values from 10% to 70% by random deletion. In all cases, we evaluated the performance on the original values using normalized root mean squared error. For the full dataset we also obtained performance statistics by variety and by year. MissForest provided average errors of 17% (3 days) at budbreak, 14% (4 days) at flowering, 14.5% (7 days) at veraison, and 17% (3 days) at maturity. We completed the imputations of the Conegliano dataset, one of the world’s most extensive and varied phenological time series and a steppingstone for future climate change studies in grapes. The dataset is now ready for further analysis, and a rigorous evaluation of imputation errors is included.

Choosing the rootstock, the scion variety and the training system best suited to the local soil and climate are the key elements for an economically sustainable production of wine. The choice of the rootstock/scion variety best adapted to the characteristics of the soil is essential but, by changing climatic conditions, ongoing climate change disrupts the fine-tuned local equilibrium. Higher temperatures induce shifts in developmental stages, with on the one hand increasing fears of spring frost damages and, on the other hand, ripening during the warmest periods in summer. Expected higher water demand and longer and more frequent drought events are also major concerns. The genetic control of the phenotypes, by genomic information but also by the epigenetic control of gene expression, offers a lot of opportunities for adapting the plant material to the future. For complex traits, genomic selection is also a promising method for predicting phenotypes. However, ecophysiological modelling is necessary to better anticipate the phenotypes in unexplored climatic conditions Genetic approaches applied on parameters of ecophysiological models rather than raw observed data are more than ever the basis for finding, or building, the ideal varieties of the future.

The increasingly frequent heat waves during grape ripening pose challenges for high quality wine grape production. Defoliation is a common practice that can improve the control of diseases in bunches, but also it increases the exposure to sunlight. Grapes exposed to solar radiation reach temperatures over the optimum for berry development and maturation. This makes the development of irrigation and canopy management techniques of great importance to maximize yield and grape quality. A field experiment was carried out during 2021 using Manto negro wine grapes to study the effect of applied irrigation and different light exposure levels on grape quality. Two irrigation treatments were imposed based on the frequency and amount of water doses in a four-block experimental vineyard at Bodega Ribas (Mallorca). Three light exposure treatments were randomly applied in each irrigation plot. The light treatments included exposed clusters from pea size, non-exposed clusters, and shaded clusters after softening. Leaf area index and canopy porosity was estimated every 2 weeks. Midday leaf water potential was measured weekly. Additionally, apparent electrical conductivity was measured between rows to estimate the soil water content variability. Light and temperature sensors were installed at the bunch level to quantify the differences in bunch temperature and light intensity among treatments. The effect of irrigation and cluster light exposure on berry weight, TSS, TA, malic acid, tartaric acid, K+, and pH were analysed at 5 moments along grape ripening. During different heat waves, the natural shading technique decreased the maximum bunch temperature around 10 °C respect to the exposed bunches in both irrigation strategies. The combination of defoliation and shading techniques after softening decreased TSS at harvest and affected most of the quality parameters during the last stages of ripening, showing an interesting technique to delay ripening in warm viticulture areas.

In viticulture, a warming climate can have a very significant impact on grapevine development and therefore on the quality and characteristics of wines across different spatial scales, ranging from global to local. In order to adapt wine-growing to climate change, global climate models can be used to define future scenarios, but only at the scale of major wine regions. Despite the huge progress made over the last ten years in terms of the spatial resolution of climate models (now downscaled to a few square kilometres), they are not yet sufficiently precise to account for the local climate variability associated with such parameters as local topography, in spite of these parameters being decisive for vine and wine characteristics. This study describes a method to downscale future climate scenarios to vineyard scale. Networks of data loggers have been used to collect air temperature at canopy level in the Waipara winegrowing region (New Zealand) over five growing seasons. These measurements allow the creation of fine-scale geostatistical models and maps of temperature (at 100 m resolution) for the growing season. In order to model climate change at pilot site scale, these geostatistical models have been combined with regional climate change predictions for the periods 2031-2050 and 2081-2100 based on the RCP8.5 climate change scenario. The integration of local climate variability with regionalized climate change simulations allows assessment of the impacts of climate change at the vineyard scale. The improved knowledge gained using this methodology results from the increased horizontal resolution that better addresses the concerns of winegrowers. The results provide the local winegrowers with information necessary to understand current processes, as well as historical and future viticulture trends at the scale of their site, thereby facilitating decisions about future response strategies.

The main innovation in the VINIoT service is the joint use of two technologies that are currently used separately: vineyard monitoring using multispectral imaging and deployed terrain sensors. One part of the system is based on the development of artificial intelligence algorithms that are feed on the images of the multispectral camera and IoT sensors, high-level information on water stress, grape ripening status and the presence of diseases. In order to obtain algorithms to determine the state of ripening of the grapes and avoid losing information due to the diversity of the grape berries, it was decided to work along the first year 2020 at berry scale in the laboratory, during the second year at the cluster scale and on the last year at plot scale. Different varieties of white and red grapes were used; in the case of Galicia we worked with the white grape variety Treixadura and the red variety Mencía. During the 2020 and 2021 campaigns, multispectral images were taken in the visible and infrared range of: 1) sets of 100 grapes classifying them by means of densimetric baths, 2) individual bunches. The images taken with the laboratory analysis of the ripening stage were correlated. Technological maturity, pH, probable degree, malic acid content, tartaric acid content and parameters for assessing phenolic maturity, IPT, anthocyanin content were determined. It has been calculated for each single image the mean value of each spectral band (only taking into account the pixels of interest) and a correlation study of these values with laboratory data has been carried out. These studies are still provisional and it will be necessary to continue with them, jointly with the training of the machine learning algorithms. Processed data will allow to determine the sensitivity of the multispectral images and select bands of interest in maturation.