Precise homing by salmon to natal habitats is considered the primary mechanism in the evolution of population-specific traits, yet few studies have focused on this final phase of their spawning migration. We radio tagged 157 female sockeye salmon (Oncorhynchus nerka) as they entered Lake Clark, Alaska, and tracked them every 1-10 days to their spawning locations. Contrary to past research, no specific shoreline migration pattern was observed (e.g., clockwise) nor did fish enter a tributary unless they spawned in that tributary. Tributary spawning fish migrated faster (mean = 4.7 km??day-1, SD = 2.7, vs. 1.6 km??day-1, SD = 2.1) and more directly (mean linearity = 0.8, SD = 0.2, vs. 0.4, SD = 0.2) than Lake Clark beach spawning fish. Although radio-tagged salmon migrated to within 5 km of their final spawning location in an average of 21.2 days (SD = 13.2), some fish migrated five times the distance necessary and over 50 days to reach their spawning destination. These results demonstrate the dynamic nature of this final phase of migration and support studies indicating a higher degree of homing precision by tributary spawning fish. ?? Journal compilation 2007 Blackwell Munksgaard No claim to original US government works.
