Assessment of large-scale spatial variation in age-specific survival and age at first breeding in a long-lived species.

Age-dependent survival and age at first breeding are key demographic parameters that drive population dynamics. Although crucial for understanding population trends and informing region-specific conservation strategies, the large-scale spatial variability of these demographic parameters, as well as how temporal fluctuations are correlated across space (i.e. spatial synchrony), are poorly known. This is particularly the case in long-lived species characterised by complex, age-dependent life cycles. White storks (Ciconia ciconia) from Germany provide an opportunity to study spatial variation in age-dependent survival and age at first breeding as they are long-lived and extensively monitored through long-term ringing programmes across the whole country. The contrasting population trends observed between western (increase) and eastern Germany (slight decrease), between which migratory flyways differ, suggest substantial spatial variation in demographic parameters. We analysed ringing, resighting and recovery data of 92,251 individuals collected from 2000 to 2023 across Germany, using the federal states (Bundesländer) as spatial units. We fitted three spatial multistate capture-recapture-recovery models that differed in whether and how spatial autocorrelation among spatial units was incorporated. The spatial patterns of the demographic parameters were similar across the three fitted models, but the model that used flyway information for the structure of spatial autocorrelation performed best. Survival in all age classes was higher and age at first breeding lower in western than in eastern Germany, consistent with the respective population trends. Spatial variability in survival was higher in juveniles than in other age classes, and synchrony across space was found in juvenile survival, but not in the survival of older age classes. The developed models with alternative formulations of spatial autocorrelation were useful to assess the structure of spatial variation in survival and age at first breeding. Our results highlight substantial large-scale spatio-temporal variation in demographic parameters within a long-lived species, helping to explain spatial differences in population dynamics across the study area.