Detecting Grapevine Virus Infections in Red and White Winegrape Canopies Using Proximal Hyperspectral Sensing

Area of Science:

• Agricultural science
• Plant pathology
• Remote sensing technology

Background:

• Grapevine leafroll disease (GLD) poses a significant global threat to grapevine health.
• Current diagnostic methods for GLD are either expensive laboratory tests or unreliable visual assessments.
• Hyperspectral sensing offers a promising non-destructive and rapid alternative for plant disease detection.

Purpose of the Study:

• To evaluate the efficacy of proximal hyperspectral sensing for detecting grapevine leafroll disease (GLD) in Pinot Noir and Chardonnay grapevines.
• To identify the optimal timepoint within the growing season for accurate GLD detection using spectral data.
• To assess the potential of hyperspectral technology for large-scale vineyard disease surveillance.

Main Methods:

• Proximal hyperspectral sensing was employed to collect leaf reflectance spectra from infected and healthy grapevines.
• Spectral data were gathered at six timepoints throughout the growing season for both Pinot Noir and Chardonnay cultivars.
• Partial least squares-discriminant analysis (PLS-DA) was utilized to develop predictive models for GLD presence or absence.

Main Results:

• The harvest timepoint demonstrated the highest prediction accuracy for GLD detection.
• Prediction accuracies reached 96% for Pinot Noir and 76% for Chardonnay.
• Canopy spectral reflectance showed significant temporal changes indicative of virus infection.

Conclusions:

• Hyperspectral sensing is a viable and effective method for non-destructive GLD detection in grapevines.
• The optimal timing for GLD detection using this technology is around harvest.
• The hyperspectral approach can be integrated with mobile platforms (ground vehicles, UAVs) for efficient vineyard disease surveillance.