The distribution of seabirds and pinnipeds in Marguerite Bay and their relationship to physical features during austral winter 2001

The distribution of seabirds and pinnipeds and their relationship to physical oceanographic variables were investigated as part of the US Southern Ocean Global Ocean Ecosystem Dynamics field program along a study grid centered around Marguerite Bay on the west Antarctic Peninsula during late fall (April-May) and winter (July-August), 2001. Sea-ice conditions during the cruises provided an opportunity to compare the relationship among physical oceanographic variables and species distributions before and after the development of pack ice. During the fall cruise before pack ice development, both sea-ice-affiliated species and open-water-affiliated were observed in the area. The most common ice-affiliated species observed at this time were snow petrel (Pagodroma nivea, 0.7 individuals km-2) and Antarctic petrel (Thalassoica antarctica, 0.2 individuals km-2) and the most common open-water-affiliated species were blue petrel (Halobaena caerulea, 0.4 individuals km-2), cape petrel (Daption capense, 0.2 individuals km-2), and southern fulmar (Fulmarus glacialoides, 0.1 individuals km-2). In addition, Antarctic fur seals (Arctocephalus gazella, 0.1 individuals km-2) and crabeater seals (Lobodon carcinophagus, 0.4 individuals km-2) were observed in low numbers. Akaike's information criterion was used to assess competing models that predicted predator distributions based on physical oceanographic variables proposed to structure predator distribution in previous research. These analyses indicated that predator distributions were primarily associated with water-mass structure and variability in bottom depth during the fall cruise. Crabeater seal, snow petrel, Antarctic petrel, and southern fulmar had higher densities in Inner Shelf Water, particularly near Alexander Island where a coastal current was present. Blue petrel, kelp gull (Larus dominicanus), and southern giant petrel (Macronectes giganteus) were positively associated with variability in bottom depth in April-May, suggesting that hydrographic processes influenced by bathymetry may have been important in structuring bird distributions. After the development of pack ice, during July and August, only sea-ice-affiliated species, including snow petrel (1.0 individuals km-2), Antarctic petrel (0.1 individuals km-2), Ade??lie penguin (Pygoscelis adeliae, 0.4 individuals km-2), and crabeater seal (0.3 individuals km-2), were observed. Seabirds were primarily associated with sea-ice characteristics (e.g. sea-ice concentration, sea-ice type) rather than the water-column environment later in the winter. Results from this study suggest that the timing and extent of sea-ice development in the fall may influence over-winter predation by seabirds and pinnipeds on zooplankton and fish on the western Antarctic Peninsula. Delays in sea-ice development may allow seabirds and pinnipeds access to biologically important areas such as the Inner Shelf Water for a longer period of time thereby increasing predation on zooplankton and fish. ?? 2004 Elsevier Ltd. All rights reserved.