2.Here, we use an integrated Bayesian hidden process model framework for a comparative analysis of long term (1978-2005) capture-recapture data using different combinations of marking methods for individual female grey seals (Halichoerus grypus) at 2 UK colonies, the Isle of May (IoM) and North Rona (NR). The method takes account of ‘missing data’ e.g. where the status of marks is not known for some animals in some years.
3.Seals were identified using 3 methods: numbered plastic flipper tags, alphanumeric brands, or natural pelage markings. Animals identified by brands or natural markings were resighted more often than those with tags; the difference between brands and tags was greatest at IoM; and tags were more easily re-sighted at NR than at IoM.
4.Flipper tag loss rates were higher at NR than at IoM and were non-independent at NR.
5.Evidence of transient animals was found for NR, but not at the IoM.
6.Apparent survival was higher and more consistent at the IoM, and lower and more variable on NR. This is consistent with, but does not completely explain, the differences in observed trends in colony dynamics between the 2 sites.