Speaker
Description
C. Riefolo1, S. Ruggieri1, M.R. Muolo2, C. Galeone3, N.A. Ranieri2, F. Nigro4
1 Research Centre for Agriculture and Environment, Council for Agricultural Research and Economics (CREA-AA), 70125 Bari, Italy;
2 Servizi di Informazione Territoriale S.r.l., 70015 Noci, Italy;
3 Water Research Institute, National Research Council (CNR-IRSA), 70125 Bari, Italy
4 Dipartimento di Scienze del Suolo, della Pianta e degli Alimenti, Università degli Studi Aldo Moro,
Via Amendola 165/A, 70126 Bari, Italy;
Xylella fastidiosa subspecies pauca (Xfp), a Gram-negative bacterium in the family Xanthomonadaceae (γ-proteobacteria), is one of the most dangerous plant pathogens worldwide. It colonizes the xylem of the host, and is transmitted by several xylem sap-feeding insect vectors (Homoptera, Auchenorrhyncha). Formerly restricted to the Americas, a very aggressive genotype, Xfp ST53, has been reported in Apulia as responsible for the Olive Quick Decline Syndrome (OQDS), a vascular disease causing the death of millions of young and centenarian olive trees. The monitoring of the infected area must be fast and reliable, thus, allowing an early diagnosis of the disease also when the symptoms are not yet visible. In fact, any delay would preclude the effectiveness of the mandatory phytosanitary measures to slow down the epidemic progression, thus, increasing the infection risk for the surrounding plants. The aim of this work was to assess whether the analysis of hyperspectral data, using different statistical methods, allow to select with sufficient accuracy, which plants to sample and test with qPCR, to save time and economic resources. Partial Least Square Regression (PLSR) and Canonical Discriminant Analysis (CDA) indicated that the most important bands were those related to the chlorophyll function, water, lignin content, as can also be seen from the wilting symptoms in plants infected by Xfp. The confusion matrix of CDA showed an overall accuracy of 0.67, but with a better capability to discriminate the infected plants. Finally, an unsupervised classification, using only spectral data, was able to discriminate the infected plants at a very early stage of infection. Then, testing by qPCR should be performed only on the plants predicted as infected from hyperspectral data, thus, saving time and financial resources.
