Abstract
Estuary dependence to the survival of Chinook salmon was investigated at the Campbell River estuary, British Columbia, Canada. Replicate batches of marked smolts were transferred from a hatchery and released in 1983, 1984 and 1985 at four ecosystems, two that ensured estuarine experience (river, estuary) and two seawards of the estuary (transition, marine). Juveniles were tracked with beach seine collections, providing residency times in the four ecosystems. Marked adult salmon were enumerated in the fishery and spawning grounds for 6 years. Recoveries from the marine and transition ecosystem releases comprised 35.1% of the total recoveries (smolt‐to‐adult survival 0.34%), whereas 64.9% were from river and estuary ecosystems (smolt‐to‐adult survival 0.60%). Fish released in the marine ecosystem in 1985 were recovered at higher rates than in 1983 or 1984, creating a statistical interaction between year and release location that implicated differences in annual ocean conditions. Results are related to present‐day salmon conservation challenges and restoration consequences. In essence, transfer experiments complement other approaches to studies of salmon estuary dependence but require considerable field labour efforts and attention to temporal replication to achieve statistical rigour. Experimental interpretation can be improved with investigations expanded to include several populations and estuaries and a more extensive consideration of annual variability in river and ocean conditions.