Precision viticultural strategy for managing intra-vineyard variability in grape aroma using UAV-based vigour indices

Abstract

In several cultivars, such as Gewürztraminer and Riesling, grape and wine aromas are determined by volatile terpenoids. Intra-vineyard variations in vine nutritional state, vegetative growth, physiological performance, and yield might result in variations in volatile terpenoid concentration and overall grape quality. Precision viticulture strategies such as zonal management can deal with these intra-vineyard variations allowing to separate grapes of different quality, and produce wines of distinct quality from the same vineyards.

We investigated the intra-vineyard variation in canopy growth, nutritional state, physiological performance, and yield, and grape technological quality and grape volatile terpenoids. We then evaluated the use of UAV-based indices to perform zone delineation to separate grapes with higher or lower volatile terpenoid concentrations.

The study was conducted in a Gewürztraminer vineyard for two seasons (2022, 2023) and in a Riesling vineyard for three seasons (2021, 2022, 2023). Both vineyards were located in the Okanagan Valley wine region (British Columbia, Canada). Plant water status (stem water potential), leaf area, and leaf gas exchanges (CO2 assimilation, transpiration, stomatal conductance) were measured at anthesis and veraison. At the same stages, normalized difference vegetation index (NDVI) and normalized difference red edge index (NDRE) were calculated from multispectral drone images to assess vine vigor. Vine nutritional status was determined by petiole analysis at veraison. Yield and berry technological quality (total soluble solids, pH, and titratable acidity) were determined at harvest. Grape volatile terpenoid concentration at harvest was assessed via SPME GC-MS analyses.

The results showed that grape volatile terpenoids positively correlated to petiole nitrogen content and yield in Gewürztraminer, and to leaf area, petiole nitrogen content, and yield in Riesling. NDVI and NDRE positively correlated to leaf area, petiole nitrogen content, yield, and volatile terpenoids in both vineyards. Our study showed that remote sensing-based NDVI and NDRE can serve for estimating vegetative growth and volatile terpenoids within vineyards. These indices allowed to segregate, using statistical approaches such as unsupervised k-means clustering analysis, zones of different volatile terpenoid concentrations within vineyards, which would allow the production of distinct wine styles under selective harvest.