Parsons, M., Mitchell, I., Butler, A., Ratcliffe, N., Frederiksen, M., Foster, S., and Reid, J. B. 2008. Seabirds as indicators of the marine environment. – ICES Journal of Marine Science, 65: 1520–1526. We report on the development of seabird indicators that support the Scottish Biodiversity Strategy. The application of high-quality monitoring data on breeding abundance and productivity in Scotland was explored in three ways: as indicators of seabird status in its own right, as indicators of the “health” of the marine environment, and as indicators of the food supply of vertebrate predators. Data on breeding productivity of seabirds, which responds more immediately to environmental variation than adult abundance, provided a novel supplement to indicators based solely on abundance trends. Grouping of species according to ecological guilds provided indicators of change in particular aspects of the marine environment. The role of seabird indicators in relation to policy frameworks is discussed, with a look to further developments at the UK and regional scales.

No abstract

Many species of bird recognize acoustic and visual cues given by their predators and have complex defence adaptations to reduce predation risk. Recognition of threats posed by specific predators and specialized anti-predation behaviours are common. In this study we investigated predator recognition and anti-predation behaviours in a pelagic seabird, Leach's Storm-petrel Oceanodroma leucorhoa, at a site where predation risk from Great Skuas Stercorarius skua is exceptionally high. Leach's Storm-petrels breed in burrows and come on land only at night. Counter-predator adaptations were investigated correlatively in relation to changing natural light levels at night, and experimentally in relation to nocturnal visual and acoustic signals from Great Skuas. Colony attendance by Leach's Storm-petrels was attuned to changes in light conditions at night and was highest when nights were darkest. This behaviour is likely to reduce predation risk on land; however, specific recognition of Great Skuas and specialized defence behaviours were not found. Leach's Storm-petrels, in particular apparently non-breeding individuals, were entirely naïve to the threat posed by Great Skuas and were captured easily in a variety of different ways, on the ground and in the air. Lack of specialized behavioural adaptations in Leach's Storm-petrels against Great Skuas may be because spatial overlap of breeding distributions of these species appears to be a rare and recent phenomenon.

Many of the breeding seabird populations in Britain and Ireland are of international importance; consequently, there is a statutory duty to protect these populations, as part of national biodiversity strategies and under Article 4 of the EU's Directive on the Conservation of Wild Birds (EC/79/409). As part of this process, populations have been monitored annually at a sample of colonies since the mid-1980s and (near) complete surveys have been undertaken twice. Results of this monitoring are currently reported regionally, in an effort to reflect the impact of spatially varying environmental drivers of change; however, there is concern that these regions reflect policy requirements rather than ecological relevance, particularly for mobile species. We used the monitoring data to identify a series of ecologically coherent regions in which trends in abundance and breeding success varied in a consistent fashion and examined how closely the annually sampled data matched the change quantified by the whole population surveys. The number of ecologically coherent regions identified varied from 2 for the northern gannet Morus bassanus and common guillemot Uria aalge to 7 for the great cormorant Phalacrocorax carbo. Trends imputed for ecologically coherent regions more closely matched those observed between whole population censuses and were more consistent than those identified for more policy-driven monitoring regions. By accounting for ecology in the design of monitoring regions, population variation in mobile species can be more accurately represented, leading to the design of more realistic monitoring regions.