Responsable: Stéphane Dupas
Équipe: DEEIT
Description du sujet:
The student will work in a project of integrative modeling of insect pests environmental dynamics in Africa (InSPRed, BASC). We use abundance and population genetic data to infer spatial and environmental response of the insect and propose probabilistic scenario for insects, based on climate, land use and vegetation information. The fitting and predictive power of the different versions of the model (including or not age class structure, coupling or not with land use model, coupling or not with vegetation model) are compared in a Bayesian inference framework. The specific task of the student will be to calibrate under the supervision of Juan Fernandez (ESE, U-Psud) and Stéphane Dupas (EGCE, CNRS) the land use version of the model based on available land cover data. The student will calibrate a land use model from times series data in Kenya using Dynamica ego software. The land use data may thereafter be included as independent variable in the Bayesian model, and the land-use-coupled and non-land-use versions will be compared.
Niveau : M1
Formations recherchées: écologie, statistiques, modélisation, informatique, géographie
Période: 2016
Durée 3 mois