Member States of the European Community are required to protect bird species listed in Annex I to the Birds Directive, and similarly to protect regularly occurring migratory species of birds by identifying Special Protection Areas (SPAs) for these species. To date in the UK terrestrial and coastal SPAs have been classified, and work is proceeding to identify marine SPAs. Identification of marine SPAs in the UK comprises three complementary components, one of which is the extension of existing seabird breeding colony SPAs into the marine environment. This report presents work carried out to inform the identification of possible generic boundaries for seaward extensions to existing terrestrial seabird colony SPAs.

 

Between 10 and 27 June 2001, the JNCC systematically surveyed seabirds in the waters immediately adjacent (up to approximately 5 km from mean low water [MLW]) to six seabird colonies hosting nationally and internationally important numbers of seabird species. These breeding season surveys were conducted from chartered vessels using a strip-transect method of counting.

 

Breeding seabirds spend much of their time either foraging at sea or attending the nest. When not engaged in these activities, seabirds, particularly auks, spend time on the water adjacent to colonies preening, bathing, and displaying (termed "active" behaviours here). Unlike the distribution of feeding birds, the distribution of birds engaged in "active" behaviours is largely independent of the physical or oceanographic characteristics of the colony or adjacent waters. For species with large enough sample sizes, "active" seabird density and small-scale distribution (up to 5 km from the colony MLW) were analysed using geostatistical modelling (variography and kriging) and distance band analysis.

 

The data analyses here allowed geostatistical and distance band analysis of the densities and distributions of four seabird species engaged in "active" behaviours, namely common guillemot (Uria aalge), Atlantic puffin (Fratercula arctica), razorbill (Alca torda) and northern gannet (Morus bassanus). Kriged density contours and mean modelled densities showed that modelled densities decreased with increasing distance from the colony. This pattern of decreasing density at greater distances from the colony was similar for all four species at all six colonies. Distance band analyses of modelled densities indicate that the highest densities of all three auk species engaged in "active" behaviours were observed within 1 km from the colony shore. The highest densities of "active" gannets were found within 2 km of the colony shore. Observations of "active" individuals of other seabird species were few.

 

There were several qualifying seabird species at many of these colonies that were observed at low density and with limited distribution in the waters adjacent to the colony. Spatial modelling could not be performed on these type of data due to lack of spatial autocorrelation and/or low sample sizes. However, distribution maps of observed densities revealed that the northern fulmar (Fulmarus glacialis), black-legged kittiwake (Rissa tridactyla) and European shag (Phalacrocorax aristotelis) consistently used the waters around colonies, with highest densities within 1 km of the colony. Despite this, the paucity of data make it impossible to recommend extensions to existing breeding colonies of northern fulmar, European shag and black-legged kittiwake.

 

Low sample sizes of some species' observations may have arisen due to a number of factors. Some species may aggregate outside the spatial scale of this study (>5 km from the colony) or perhaps simply may not use the water adjacent to the colony for "active" behaviour. Others may form aggregations in the pre-breeding period or at night. It is possible that use of waters adjacent to colonies is short-lived with high daily turnover of individuals making the observed densities, in the short snap-shot of time on a survey, relatively low. Further fieldwork may be necessary to investigate distribution patterns of species falling under these categories.

 

This is the first time that detailed surveys of the small-scale distributions of seabirds around British seabird colonies have been undertaken. Quantification of the spatial patterns of seabird species has hitherto not occurred at this scale and at such high resolution. As the general distribution patterns identified are not site-specific, we recommend that a generic approach to defining seaward extensions to classified SPAs for common guillemot, razorbill, Atlantic puffin and northern gannet will provide appropriate protection for these species in the waters immediately adjacent to their colonies.

 

We recommend that the boundaries of existing common guillemot, razorbill and Atlantic puffin colony SPAs be extended by 1 km from mean low water (mean low water springs in Scotland) into the marine environment. Similar boundaries extending 2 km are recommended for gannet colony SPAs. We recommend that the boundary of a seaward extension to an existing coastal or island seabird colony SPA should be defined by a rectilinear polygon drawn along parallels of latitude and meridians of longitude using a minimum number of lines. Polygon vertices should be defined in degrees and minutes to two decimal places. Other simple shapes and alignments also may be used where practical. The land area within this polygon that is not included in the existing SPA should not be included in the extension.