Assessing the diet of great skuas, <i>Catharacta skua</i>, using five different techniques

Several different techniques have been used to study the diet composition of skuas and gulls in polar regions. In this study, we assessed the diet of great skuas using five techniques (pellets, prey remains, spontaneous regurgitates, observed feeds and water off-loading). The estimates of diet derived from each sampling technique during a single breeding season were generally similar. Although it is easy to collect large samples of pellets and remains, these over-estimate the amount of indigestible material in the diet. Prey remains provide a high degree of taxonomic accuracy, but may under-estimate prey swallowed whole. Water off-loading provides unbiased estimates of chick diet but only during a short period in the breeding season. It may also stress birds, and is labour-intensive. Spontaneous regurgitates are useful for assessing chick and adult diet but are difficult to collect systematically, and differing prey digestibility may bias results. Observed feeds are time consuming to record and over-represent easily identifiable prey. Technique-dependent biases highlight that sampling methods should be selected on the basis of sample sizes, time, taxonomic detail and age of study birds. Biomass may be estimated using pellets with correction factors, and also spontaneous regurgitates and water off-loading, but prey remains and observed feeds may be more inaccurat