Studies increasingly show that social connectedness plays a key role in determining survival, in addition to natural and anthropogenic environmental factors. Few studies, however, integrated social, non-social, and demographic data to elucidate what components of an animal’s socio-ecological environment are most important to their survival. Female giraffes (Giraffa camelopardalis) form structured societies with highly dynamic group membership but stable long-term associations. We examined the relative contributions of sociability (relationship strength, gregariousness, and betweenness), together with those of the natural (food sources and vegetation types) and anthropogenic environment (distance from human settlements), to adult female giraffe survival. We tested predictions about the influence of sociability and natural and human factors at two social levels: the individual and the social community. Survival was primarily driven by individual- rather than community-level social factors. Gregariousness (the number of other females each individual was observed with on average) was most important in explaining variation in female adult survival, more than other social traits and any natural or anthropogenic environmental factors. For adult female giraffes, grouping with more other females, even as group membership frequently changes, is correlated with better survival, and this sociability appears to be more important than several attributes of their non-social environment.

Individual giraffe data were collected over 5 years (Jan 2012- Oct 2016) in the Tarangire Ecosystem of northern Tanzania using photographic capture-resight techniques. Surveys were conducted 6 times per year using a robust design. All tracks were driven once per survey. Individuals were identified using Wild-ID pattern recognition software. Social network analysis was used to calculate individual social metrics edge weight, gregariousness, and betweenness, and to detect social communities. The dataset Giraffe_Community_Covariates contains data on community-level values for environmental covariates for input into a design matrix in Program MARK. The dataset Giraffe_Survival is an encounter history for 512 individual adult female giraffes and social and non-social covariates structured as an input file for Program MARK. For input into Program MARK, convert Giraffe_Survival to an .INP file using methodology outlined in Cooch EG, White GC. 2019 Program MARK: A gentle introduction. 19th edition.