Discriminating Xylella fastidiosa from Verticillium dahliae infections in olive trees using thermal- and hyperspectral-based plant traits

Globally, olive ( Olea europaea L.) productivity is threatened by plant pathogens, particularly the fungus Verticillium dahliae Vd ) and bacterium Xylella fastidiosa Xf ). Infections these pathogens restrict water nutrient flow through xylem, producing a similar set of symptoms that can also be confounded with stress. Conventional in situ monitoring techniques are time consuming expensive, necessitating development large-scale detection methods. Airborne hyperspectral thermal imagery have been successfully used to detect both infection independently, i.e., when only one two diseases present. Nevertheless, discrimination from infections contexts where present has not addressed date. This study proposes three-stage machine learning algorithm distinguish infections, using series datasets 27 orchards affected outbreaks Italy Spain between 2011 2017. Plant traits were derived airborne imagery, including physiological indices radiative transfer model inversion, Solar-induced Fluorescence emission (SIF @760 ), Crop Water Stress Index (CWSI), selection narrow–band indices. Several distinct spectral discriminated infections. The method generated false-positive rate 9%, an overall accuracy (OA) 98%, kappa coefficient (κ) 0.7 identifying mixed + dataset. When was 4%, OA 92%, κ 0.8. These results indicate discriminate caused xylem–limited trigger visual symptoms.