A time-dependent, size-structured, physiologically based krill growth model was used in conjunction with a circulation model to test the hypothesis that Antarctic krill (Euphausia superba) populations at South Georgia are sustained by import of individuals from upstream regions. Surface phytoplankton concentrations along the simulated drifter trajectories were extracted from historical Coastal Zone Color Scanner (CZCS) measurements and sea ice biota concentrations were calculated from sea ice concentration and extent extracted along drifter trajectories from Special Sensor Microwave/Imager measurements. As additional food sources, a time series of heterotrophic food was constructed from historical data, and time series of detritus concentrations were calculated using phytoplankton concentrations extracied from CZCS measurements together with measured particulate organic carbon to chlorophyll a ratios. These food resources along specified drifter trajectories were then input to the krill growth model to determine the size and viability of krill during transport from the source region to South Georgia. The krill growth model simulations showed that no single food source can support continuous growth of krill during the 58-306 days needed for transport to South Georgia. However, under the current assumptions results indicate that combinations of food sources during the transport time enhanced krill survival, with heterotrophic food and detritus being particularly important during periods of low phytoplankton concentrations. The growth model simulations also showed that larval and juvenile krill originating along the western Antarctic Peninsula can grow to 1+ (14-36mm) and 2+ (26-45 mm) age and size classes observed at South Georgia during the time needed for transport to this region. Krill originating in the Weddell Sea need 20 months for transport, which allows retention in a potentially high food environment, provided by sea ice, for almost 1 year. Krill then complete transport to South Georgia in the following year and larval and juvenile krill grow to 2+ (26-45 mm) and 3+ (35-60 mm) age and size classes during transport. The results of this study show that the successful transport of krill to South Georgia depends on a multitude of factors, such as the location of the spawning area and timing of spawning, food concentrations during transport, predation, and variations in the location of the Southern Antarctic Circumpolar Current Front (SACCF) and in sea ice extent. (c) 2006 Elsevier Ltd. All rights reserved.