Nous avons communiqué les informations ci-après à la société CHRISAR de la Seyne-sur-Mer lauréate d’un appel à projet de l’écologie portant sur le même sujet.
Nous avions évoqué ces nouvelles recherches très porteuses dans un poste du mois de janvier. http://www.sauvonsnospalmiers.fr/spip.php?article1355
L’occasion pour SNP de suggérer dans le cahier des charges de ces appels à projet que les bénéficiaires soient dans l’obligation de faire un communiqué de presse deux fois par an.
ISHS Acta Horticulturae 1099 : II International Symposium on Horticulture in Europe
ELECTRONIC NOSE FOR THE EARLY DETECTION OF RED PALM WEEVIL (RHYNCHOPHORUS FERRUGINEOUS OLIVIER) INFESTATION IN PALMS : PRELIMINARY RESULTS
Authors : A. Rizzolo, G. Bianchi, P. Lucido, B. Cangelosi, L. Pozzi, G. Villa, F. Clematis, C. Pasini, P. Curir
Keywords : larva, RPW, e-nose, pattern recognition techniques, discrimination
DOI : 10.17660/ActaHortic.2015.1099.40
The red palm weevil (RPW, Rhynchophorus ferrugineus Olivier) is one of the worst pests for palms and when the infestation of this insect is discovered, usually it is too late for recovering the plant. In the Mediterranean area RPW mainly attacks the Phoenix canariensis species, and in Italy it was recorded since 2004. Various methods have been applied to control and manage RPW, but at the moment none of them successfully prevented its spread out. The aim of this work was to study whether RPW pest in palm trees could be detected at early stages using a commercial electronic nose (PEN3) by monitoring both the air surrounding the crown and the detached leaves from healthy, drilled, and infested palms with an increasing number of RPW larvae (2, 4 and 8) (2 palms per treatment) at 8, 15 (only air) and 23 days from infestation. Normalized sensors signals of air and leaves samples were separately processed using pattern recognition techniques (Principal Component Analysis (PCA) and Discriminant Analysis (DA)). With PCA of air samples, RPW palms could be distinguished along PC2 axis by the healthy and drilled ones, whereas PC1 increased with the increase of time from infestation. With PCA of leaves samples, opposite PC1 trends with time from infestation were observed for healthy trees (increase) and for drilled and RPW ones (decrease). The performance of classification models based on RPW infestation whatever the number of larvae ranged from 76.7% for air samples to 100% for leaves sample, whereas those based on the treatments ranged from 60% for air samples collected at day 23 to 100% for leaves. This preliminary study demonstrates that the E-nose technology has potential for use as an effective RPW pest monitoring method.
Article - full text (enhanced PDF format, 406978 bytes)