By Justin-Hervé Noubissi , defended in 2019 under the supervision of Christophe Cambier , Jean-Claude Kamgang and Januarius Asongu

The eradication of malaria is a major concern for computer scientists, mathematicians, epidemiologists, entomologists, physicians, and many others. Proposals range from curing patients to the complete elimination of the disease. However, the often inefficient collaboration between these scientists leads to incomplete prototypes or an underutilization of the results obtained. Environmental and climatic factors are among those elements generally overlooked by computer scientists and mathematicians when modeling malaria transmission dynamics. The tropical countries most affected by the disease are also mostly underdeveloped or developing countries, where statistical data is often nonexistent or difficult to access. Populations, constantly on the move, traverse diverse ecosystems with environmental and climatic conditions that sometimes differ from one region to another. In this thesis, we propose a model that integrates these migratory, spatial, and temporal elements, while ensuring mathematical stability. We study existing approaches and then propose three types of models, which we compare: a meta-population model without considering climatic factors, a meta-population model with consideration of climatic factors at the time of human-mosquito contact, and a meta-population model with consideration of climatic factors throughout the mosquito's life cycle; the latter emerged as the most realistic.