Detection of Annual Changes in the Numbers of Cliff-Nesting Seabirds in Orkney 1976-80

Abstract

(1) This paper reports on a 5-year project set up in Mainland, Orkney, Scotland to assess annual and longer-term changes in numbers of four seabird species against a pattern of temporal variability. (2) Between five and ten standardized counts of individual guillemots Uria aalge, razorbills Alca torda, kittiwakes Rissa tridactyla and fulmars Fulmarus glacialis and apparently occupied nests and sites of kittiwakes and fulmars were made in fixed plots at five colonies in the month of June from 1976 to 1980. (3) Analyses of variance were used to test for changes in numbers between years. The method's sensitivity using five or ten counts per year was assessed by calculating the power of the two-tailed t-test for specified proportionate changes in the mean counts. (4) Between 1976 and 1980 there were significant differences between annual colony means in all the species and categories counted. Interaction of colonies with years for all categories (except kittiwake nests) indicated unidirectional change but at different rates. For kittiwake nests, interaction indicated opposing trends in the colonies. All colonies showed mean annual increases in numbers of individual guillemots, razorbills and fulmars and fulmar sites (except Costa Head). There were mean annual decreases at all colonies (except Mull Head) in numbers of kittiwake nests and individuals. (5) Five counts per year were sufficient to detect proportionate changes of +30% for guillemots and kittiwakes (nests and individuals) and fulmar (individuals). To achieve this precision for razorbills, ten counts would need to be made in a shortened sampling period. Fulmar sites were borderline, probably partly because of the subjectivity involved in defining this category.