Age- and sex-specific survival and dispersal are important components in the dynamics and genetic structure of bird populations. For many avian taxa survival rates at the adult and juvenile life stages differ, but in long-lived species juveniles' survival is logistically challenging to study. We present the first estimates of hatch-year annual survival rates for a sea duck, the King Eider (Somateria spectabilis), estimated from satellite telemetry. From 2006 to 2008 we equipped pre-fiedging King Eiders with satellite transmitters on breeding grounds in Alaska and estimated annual survival rates during their first 2 years of life with known-fate models. We compared those estimates to survival rates of adults marked in the same area from 2002 to 2008. Hatch-year survival varied by season during the first year of life, and model-averaged annual survival rate was 0.67 (95% CI: 0.48–0.80). We did not record any mortality during the second year and were therefore unable to estimate second-year survival rate. Adults' survival rate was constant through the year (0.94, 95% CI: 0.86–0.97). No birds appeared to breed during their second summer. While 88% of females with an active transmitter (n = 9) returned to their natal area at the age of 2 years, none of the 2-year old males (n = 3) did. This pattern indicates that females' natal philopatry is high and suggests that males' higher rates of dispersal may account for sex-specific differences in apparent survival rates of juvenile sea ducks when estimated with mark—recapture methods.