A Bayesian Network to Prevent Mite Infestations in Rose Greenhouses
2019
Abbal, Philippe | Boll, Roger | Drouineau, Ange | Paris, Bruno | Latrille, Eric | Sciences Pour l'Oenologie (SPO) ; Université Montpellier 1 (UM1)-Institut National de la Recherche Agronomique (INRA)-Université de Montpellier (UM)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro) | Institut Sophia Agrobiotech (ISA) ; Institut National de la Recherche Agronomique (INRA)-Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS) | ASTREDHOR Méditerranée (SCRADH) ; Association nationale des structures d'expérimentation et de démonstration en ... et de démonstration en horticulture ornementale (ASTREDHOR) | Association nationale des structures d'expérimentation et de démonstration en ... et de démonstration en horticulture ornementale (ASTREDHOR) | Chambre d'Agriculture des Alpes Maritimes (CA 06) | Laboratoire de Biotechnologie de l'Environnement [Narbonne] (LBE) ; Institut National de la Recherche Agronomique (INRA)-Institut national d’études supérieures agronomiques de Montpellier (Montpellier SupAgro)
Mite infestation is a big threat for rose greenhouses. It is much easier to halt and destroy them if their future development can be predicted. Taking into account temperature, humidity, cropping practices, plant vigour and some other influent parameters, an expert is able to predict the future development of the mites. Unfortunately, not all greenhouses can call on an expert permanently to help them in their fight against mites. To help them we have developed a novel model to assess and anticipate mite invasions in greenhouses in the short term. The model, based on a Bayesian network, takes into account the environment and the parameters defining invasion status with their interactions Data have been collected using knowledge from horticultural experts. The model has been validated in real situations emanating from the field. We obtained a good correlation between forecasts and expert predictions for the 18 cases used in this study. Thus, using this model should help the growers to protect against mite outbreaks. It constitutes a framework for studies of other harmful pest invasions.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by Institut national de la recherche agronomique
