LAUSR

Abstract Early detection of infected asymptomatic plants could prevent the dissemination of tomato bacterial canker throughout the crop and, thus, reduce associated economic losses. Reflectance spectroscopy has been proposed for the detection of diseases in plants, mainly caused by foliar fungal pathogens. In contrast, bacterial canker is a vascular disease that does not produce visible foliar symptoms until it is too late to avoid contagion to other plants. The effects of the disease on the conductivity of the xylem have an impact on the water content of the leaves; initially, a subtle effect that cannot be assessed with the naked eye but can be evaluated by reflectance spectroscopy. The objective of this work was to identify indices, based on the reflectance spectral signature of the plants, for the detection of tomato plants affected by bacterial canker prior to the appearance of symptoms. To verify that the proposed methodology is applicable to a broad range of tomato genotypes, two experiments were carried out with different tomato cultivars grown in pots and another on the ground in a greenhouse. Plants were inoculated with the pathogenic bacteria and the reflectance spectrum of the apical leaflet of two leaves per plant were obtained before symptoms expression; several spectral indices were calculated from the reflectance spectra. Three indices, of the shortwave-infrared zone, allowed the detection of bacterial canker inoculated plants in a fast and non-destructive way, up to one week before symptoms were visible: Normalized Difference Water Index (NDWI), Simple Ratio of Water Index (SRWI) and Water Index 1 180 (WI1180). Our research demonstrates, for the first time, the usefulness of spectral indices sensitive to water content for the identification of diseased plants. These indices arise as a promising tool for the early detection of vascular plant diseases, that could be integrated into sustainable management plans.