Multi-spectral sensors monitoring of the epidemic of Xylella Fastidiosa in the Apulia Region

In October 2013, the Apulia Region in Italy, unexpectedly, had to find itself to face the epidemic of Xylella Fastidiosa which in less than 10 years has decimated millions of olive trees, endangering the 40% of the regional olive-growing heritage. Taking into account the fact that there is not a definitive cure discovered yet, the only solution is the prevention linked to constant monitoring in the so-called buffer areas with diligent efforts to delimitate the phenomenon on a very local scale. Back then, remote sensing and, in particular, multispectral and hyperspectral sensors, seemed to be the key to depicting the problem to a multidimensional extent and coming up with the most adequate ways to tackle it. This research work focuses on the analysis of 4 olive groves (of the Ogliarola cultivar) in the province of Brindisi (Italy), with different symptomatic states, using a multispectral sensor mounted on a UAV system. Among a number of different vegetation indexes (IVs) calculated, the GNDVI and the BNDVI outperform to identify the effects of the Xylella Fastidiosa infection, which was further confirmed by the laboratory analyses. This confirmation has affirmed the validity of the approach used aiming at identifying "anomalies" on non-symptomatic trees and establishing a first early warning system for a subsequent more in-depth investigation in the field. Further developments will investigate the implementation of a segmentation algorithm, according to the threshold values of the IVs defined in this work, and the use of hyperspectral sensors, in order to identify anomalies on the foliage, attributable to a potential Xylella Fastidiosa infection on the trees.