Diagnosing a decline in return rate of 1-year-old cormorants: mortality, emigration or delayed return?

1. In long-lived birds with delayed recruitment, variation in prebreeding population parameters is difficult to study, although potentially important. Cormorants (Phalacrocorax carbo sinensis) often return to their natal colony when 1 year old, enabling direct study of first-year survival and emigration. Density-dependent regulation mechanisms may be strong at this stage of the life cycle. 2. In the Danish cormorant colony Vorsø, 11 000 chicks have been colour-ringed between 1977 and 1997. Returning birds have been resighted in the colony and dead recoveries have been collected. Since 1990, the proportion of a cohort observed in the colony as 1-year-olds (return rate) has declined from 0·40 to 0·10. Possible explanations include increased mortality, increased emigration and a later age of first return to the colony. Breeding success has also declined strongly as a consequence of low food availability. 3. We used capture-recapture analysis of recovery and resighting data to investigate variation in first-year survival, emigration and resighting probabilities. Survival fluctuated widely (0·42-0·75, mean 0·58); emigration increased in the 1990s from 0·05 to 0·15; resighting probability declined from 0·75 to 0·20 after 1990. 4. First-year survival was particularly low in 1993 and 1996. The causes of the year-to-year variation were not clear. Survival may have been affected by food availability during the post-fledging period. 5. The declining return rate was caused mainly by decreasing resighting probability of 1-year-olds, although increasing emigration also contributed. The biological mechanism was that increasing numbers of cormorants did not return to the natal colony as 1-year-olds. We suggest that this was a consequence of low physiological condition among newly fledged young, caused by low food availability in the 1990s. 6. We conclude that declining food availability has had several consequences for colony dynamics. First-year emigration and the age of first return to the colony have increased and, in the worst years, survival has decreased. If the decline in food availability was due to cormorant predation, this would constitute an example of density-dependent regulation of immature colony attendance.