Spectral reflectance in the detection of Persian lime (Citrus latifolia Tanaka) leaves with Huanglongbing under fertilization treatments

Abstract

Objective: Evaluate the spectral reflectance method to determine whether Persian lime leaves, under fertilization treatments, present wavelengths associated with Huanglongbing (HLB) symptoms.

Design/methodology/approach: A completely randomized experiment with different fertilization treatments was performed to observe their effect on reflectance changes, chlorophyll content, the severity of HLB symptoms, and the quantity of bacteria on Persian lime leaves. Previously inoculated plants by grafting buds from a diseased tree were maintained under greenhouse conditions for 12 months.

Results: Leaves of diseased plants showed significantly higher reflectance in the visible spectrum (VIS) and lower chlorophyll content than healthy plant leaves. An inverse trend was observed in the near-infrared (NIR) spectrum, although it was less consistent. Fertilization treatments were not significant in reflectance, chlorophyll content, severity of symptoms, or bacteria quantity.

Limitations on study/implications: It is suggested that the evaluation of spectral reflectance be expanded to susceptible varieties that develop visible symptoms more quickly. The spectrometer's use should be complemented with multispectral and machine learning techniques for early detection of asymptomatic diseased plants.

Findings/conclusions: Reflectance in the visible spectrum may differentiate effectively between leaves from healthy plants and leaves with dim HLB symptoms, being associated with a greater reflectance and a lower chlorophyll content in diseased leaves, even with management practices such as fertilization that might masc the symptom expression in diseased plants.