Submitted to: International Research Conference on Huanglongbing
Publication Type: Abstract Only
Publication Acceptance Date: December 5, 2009
Publication Date: N/A
Technical Abstract: Researchers have used various hyperspectral systems, covering several areas of the electromagnetic spectrum to investigate all types of disease/plant interactions. The purpose of this research was to investigate using visible and near-infrared (400-1100nm) spectroscopy to differentiate HLB infected citrus leaves from uninfected leaves. Leaves were collected from various citrus plants on the United States Horticultural Research Laboratory's research farm that showed the various symptoms of HLB infection, including HLB-negative controls. The leaf samples were analyzed using an NIRSystems 6500 monochromator, and the data analyzed for partial least squares (PLS) regression in comparison to real-time PCR results for the leaves. The PLS model correctly predicted 95% and 64% of the leaves that were positive and negative for HLB, respectively. The PLS loadings indicated absorption bands related to chlorophyll, and bands associated with chlorophyll absorption decreased in infected leaves. However, some HLB negative leaves also showed decreased chlorophyll absorption which resulted in an inaccurate prediction. Based on the results from this study, it is unlikely that the visible near-infrared (400-1000 nm) portion of the electromagnetic spectrum alone will be able to differentiate HLB infected plants from plants that are compromised by some other biostressor.