Many taxa from across the animal kingdom undergo some form of seasonal or annual migrations. These interesting behaviors are generally thought to convey energetic advantages and reduce exposure to harsh environments, both of which tend to increase survival and reproduction. Thus, the timing of migration can prove critical for animals aiming to maximize any potential benefits while minimizing associated risks. We studied how the timing of seaward migration by juvenile anadromous Pacific salmon (Oncorhynchus spp.) relates to their subsequent survival to adulthood, concentrating on two species from the Columbia River basin in the northwestern United States currently listed as threatened under the Endangered Species Act: Snake River Spring/Summer Chinook salmon (O. tshawytscha) and Snake River steelhead (O. mykiss). Using logistic regression to analyze data from more than 45,000 individual fish tagged over a 6-year period (1998-2003), we examined the effects of year, date of arrival in the estuary, water temperature, and coastal ocean upwelling on the proportion of juveniles surviving to return as adults.

Results/Conclusions

Models with terms for year, day, and day² performed much better than models with terms for temperature and upwelling (although day and temperature were highly correlated). For both Chinook salmon and steelhead, we found that juveniles migrating in early May survived nearly 6 times greater than those migrating in early June, although the mean annual survival also varied considerably among years. As expected, the estimated peak in survival shifted somewhat across years (i.e. a year-by-day interaction), reflecting interannual variation in the nearshore ocean food web and physical environment. Our results would suggest that a relatively strong selection gradient exists, favoring earlier migrants, but that we must view these results in light of more recent anthropogenic changes to the river environment including altered hydrology and changes in food web interactions.