Antarctic krill Euphausia superba spawned on the outer continental shelf of the west Antarctic Peninsula can be entrained into the Southern Front of the Antarctic Circumpolar Current and transported across the Scotia Sea to South Georgia. A time-dependent, size-structured, physiologically based krill growth model was used to assess the food resources that are needed to sustain Antarctic krill during transport across the Scotia Sea and to allow them to grow to a size observed at South Georgia. Initial Lagrangian simulations provide trajectories that are followed by particles released on the west Antarctic Peninsula shelf. Pelagic phytoplankton concentrations along these trajectories are extracted from historical Coastal Zone Color Scanner measurements from the Antarctic Peninsula-Scotia Sea region and are input to the growth model. The results of these simulations show that pelagic phytoplankton concentrations are not sufficient to support continuous growth of Antarctic krill during the 140 to 160 d needed for transport to South Georgia. The inclusion of a supplemental food source during part of the transport time, such as sea ice algae (up to 80 mg chl a m(-3)), does not significantly alter this result. Survival and growth of larval krill during modeled transport is, however, enhanced by encounters with mesoscale patches of high chlorophyll concentrations (1 mg m(-3)), while subadults and adults benefit less from these conditions. Further simulations show the importance of an additional food source, such as heterotrophic food, for the survival of subadult and adult Antarctic krill. For all planktonic food scenarios tested, krill that begin transport at the Antarctic Peninsula did not reach the smallest age group often observed at South Georgia, the 2+ group, during the 140 to 160 d of transport. Including the effect of increasing temperature across the Scotia Sea on krill growth rate does not significantly alter these results, since the maximum increase in growth due to increased temperature obtained in the simulations was 1.0 mm for both 2 and 22 mm Antarcic krill. These simulations suggest the possibility of alternative transport scenarios, such as Antarctic krill beginning transport at the Antarctic Peninsula in austral summer and overwintering under the sea ice that extends northward from the Weddell Sea into the Scotia Sea. Such an interrupted transport would allow the Antarctic krill to overwinter in a potentially better food environment and begin transport again the following year, growing to a size that is within the range observed for Antarctic krill populations at South Georgia.