Thèse soutenue

Conception de modèles multi-échelles pour l’aide à la décision environnementale : application au contrôle des invasions de cicadelles brunes dans le Delta du Mékong (Vietnam)
FR
Accès à la thèse
Auteur / Autrice : Vinh Nguyen Nhi Gia
Direction : Alexis Drogoul
Type : Thèse de doctorat
Discipline(s) : Informatique
Date : Soutenance en 2013
Etablissement(s) : Paris 6

Résumé

FR  |  
EN

In South-East Asia, the problem of controlling the invasions of rice pests is a major economical problem, which is tackled by multiple scientific disciplines and is dealt with by several decisional scales. This PhD thesis contribute to the researches undertaken since 40 years on the eradication (or at least the control) of the rice pest named Brown plant hopper (BPH) by proposing a design methodology of dynamically scaling models as a foundation for decision support systems dedicated to the assessment of regional and local control policies. It has been applied to and validated on different scenarios of BPH migrations in the Vietnamese Mekong Delta, where four different geographical and political levels have been coupled in the same framework (village, commune, province, and region), each of them represented with its own dynamics (social, biological, ecological ones) but also with its relationships with the other levels. The main contribution of this research is an agent-based multi-level modeling system that allows modelers to simulate and evaluate different control policies against invasions of the Brown plant hoppers at different scales of time and space. This system couples and organizes several sub-models with separate functions: growth and migration models of BPH that take ecological and environmental processes into account at different spatial and temporal scales, social models of the various stakeholders (from farmers to political deciders), models dedicated to the up- and downscaling processes occurring between the levels of representation