Disentangling the influence of demographic parameters and the role of density dependence on species’ population dynamics may be challenging, especially when there are fractions of the population that are unobservable (e.g. dispersers, non-breeders). Additionally, due to the difficulty of gathering data at large spatial scales, most studies ignore the global dynamic of a species, which would integrate heterogeneity in local dynamics and remove the noise of dispersal. We developed a global scale, integrated population model (IPM), to disentangle the main demographic drivers of population dynamics in a long-lived colonial seabird. We used 28 years of Audouin’s gull demographic data encompassing 69 local patches (90% of the world population). Importantly, we took into account the unobservable fraction of non-breeders and also assessed the strength of density dependence for this fraction of the population. As predicted by life histories of long-lived organisms, temporal random variation in survival was highest for immatures (), and lowest for adults (). Large temporal fluctuations in the probability of taking a sabbatical from reproduction would partly explain constancy in adult survival, adults probably refraining from breeding when environmental conditions were harsh. Thus, skipping breeding would be a bet hedging strategy for long-lived organisms having many reproductive chances during their lifespan. Immature survival and fertility were the main drivers of population dynamics during the study period (r2= 0.83 (0.77-0.87) and 0.73 (0.63-0.79) respectively). We found strong evidence of density dependence, not only due to the number of breeders (r2= -0.34 (-0.43 -0.24)) but also to that of sabbaticals (r2= -0.18 (-0.33 -0.01)). Even though population dynamics of long-lived organisms are very sensitive to changes in adult survival, we show here that in the absence of strong environmental perturbations affecting this life history trait, fluctuations in population density in those species may be driven by variations of immature survival and fertility. The species is showing a 5% of annual global decrease during the last ten years, and we propose an updated species conservation status. Integrated models based on long term monitoring at a global scale may enhance our ecological and evolutionary understanding of how demographic drivers influence population dynamics.