Relationships between the distributions of marine avian predators and their prey, Euphausia superba, in Bransfield Strait and southern Drake Passage, Antarctica

Publisher Website
Google Scholar
Abstract
We investigated the joint distributions of seabirds and krill (primarily Euphausia superba) in Bransfield Strait and southern Drake Passage, Southern Ocean. We estimated seabird and krill densities, simultaneously and continuously using visual and acoustic techniques while traversing a transect grid standardized for the Second International BIOMASS Experiment (SIBEX) program. This is one of only 3 studies to find a correlation between the densities of seabirds and their prey. We defined 2 statistically independent components of distributional correlation: ''spatial concordance'', the tendency for predators and prey to co-occur in the same sampling unit, and ''numerical concordance'', the tendency for densities of spatially concordant predators and prey to covary. Within nautical-mile transect intervals, we found cape petrels Daption capensis and antarctic fulmars Fulmarus glacialoides spatially concordant, and cape petrels, and adelie penguins Pygoscelis adeliae numerically concordant with krill. A few large concentrations of seabirds and krill accounted for the majority of seabird and krill biomass. Such large concentrations were rarely found within the same mile intervals, but were found within a few miles of each other more often than expected by chance. Seabird-krill correlations increased with spatial scale, and attained their maxima over large areas defined by their seabird species composition or krill depth profiles. In general, species specializing on krill as a food resource showed the highest degree of distributional correlation with krill, and species using less than 50% krill in their diets showed little evidence of distributional correlation with krill. Although significant, correlations at small spatial scales were weak, which we suggest, that although exacerbated by sampling biases, was primarily due to difficulties seabirds have in locating and tracking krill swarms. Further, we suggest that these difficulties result in (1) most krill swarms being unexploited at any given time, (2) a few very large swarms being disproportionately important to seabirds, and (3) krill densities being a poorer predictor of seabird densities than vice versa.