The interplay between culling and density-dependence in the great cormorant: a modelling approach

1. The population of great cormorants Phalacrocorax carbo sinensis breeding in northern Europe has increased from 5000 pairs around 1970 to c. 100 000 pairs in the late 1990s, leading to serious conflicts with fishery and aqu aculture interests. Management action, including widespread culling, has be en taken in several countries. 2. Since 1990, presumed density-dependent declines in demographic performan ce have appeared in cormorant populations. We employed an extended Leslie m atrix model to study the interaction between culls and density-dependence i n regulating breeding and autumn population sizes, with emphasis on evaluat ing the effects of culling. 3. During 1979-92, the breeding population of great cormorants in northern Europe increased by 18% year(-1), in accordance with observed life-cycle pa rameters before the appearance of density-dependent declines. 4. We modelled six scenarios with varying assumptions about the strength of density-dependence in adult survival and the proportions of breeding cormo rants. A series of cull estimates was also included. Scenarios with moderat e or strong levels of density-dependence provided predictions that fit the observed numbers of breeding pairs, whereas scenarios without density-depen dence in survival overestimated real population growth. 5. The most well-supported scenarios indicated that the effect of culls at the present level (1998-99: 17 000 cormorants shot) was limited (< 10% redu ction at equilibrium). Increasing the annual cull to 30 000 still had a lim ited effect, whereas shooting 50 000 birds year(-1) led to population extin ction within 20-40 years. Shooting a fixed proportion of the population exc eeding a threshold, through density-dependent culling, could eliminate diff erences among scenarios and stabilize the population. 6. We conclude that culls probably have had a limited effect on cormorant p opulations, but if carried out in a density-dependent way they could stabil ize numbers near a desired level. However, a reduction in the number of cor morants may not lead to a similar reduction in conflicts, and actions to co ntrol damage rather than cormorant populations are likely to be more cost-e ffective. If culling is to be continued, we recommend the adoption of an ad aptive and co-ordinated management strategy across Europe. We also advocate the need to account for density-dependent mechanisms in general culling st rategies.