نبذة مختصرة : This thesis focuses on the study of social animal populations, with the main objective of including aspects of sociality in demographic models. Animals are termed social when individuals associate with other individuals of the same species not exclusively for reproduction. Some aspects of sociality are important in the different events of an individual's life, such as grooming, aggression, breastfeeding, etc. and other aspects at the population level, such as the number and size of groups, for example, in the case of animals with hierarchical structure, the number of cubs and the size of the harems, or the duration of association for cases of aggregations of animals. In this thesis, I study the components of sociality and its effects on population dynamics through capture-recapture (CR) models, which make it possible to estimate the main demographic parameters under natural conditions. We are interested in three species with different social structures. For a first species, the wolf (Canis lupus), an approach is developed to estimate the size and number of packs through a combination of spatially explicit capture-recapture models and hierarchical methods. For a second species, the Southern elephant seal (Mirounga leonina), we analyze the influence of the social structure on population dynamics in an integrated modeling framework. On a third case study, the Commerson's dolphin (Cephalorhynchus commersonii), we model the social network of animals and estimate the main characteristics of this network (e.g., degree, centrality) thanks to a state space formulation of CR models. Overall, my work illustrates the potential of modern population dynamics methods for studying the demography of social species under natural conditions. ; Cette thèse est centrée sur l’étude de la dynamique des populations animales, avec comme principal objectif d’inclure des aspects de socialité dans les modèles démographiques. On qualifie des animaux de sociaux lorsque des individus s’associent à d’autres individus de la même espèce pas ...
No Comments.