Effects of smolt release timing and size on the survival of hatchery-origin coho salmon in the Strait of Georgia

Altering release sizes and timings of coho salmon smolts from hatcheries in the Strait of Georgia will not reverse the precipitous survival declines of the past three decades. We modeled the effects on survival of ocean entry year, mean smolt size (weight), and release day. Ocean entry year was by far the most important. During 1979-2006, smolt to adult survivals declined similarly for hatchery and wild coho salmon, although wild salmon consistently survived at higher rates. Best models differed among hatcheries, implying location-specific differences in the optimal size and timing of release. At four of five hatcheries, heavier smolts survived significantly better than lighter smolts. At one hatchery, a significant interaction between ocean entry year and smolt weight reflected an increased positive effect of weight later in the time series. At two Vancouver Island hatcheries, early release groups appeared to survive better than later releases in early years, while later release groups survived best in recent years. We recommend: (1) hatchery managers release coho salmon smolts throughout the outmigration period of higher surviving wild coho salmon smolts and (2) an experimental approach using hatcheries to evaluate density-dependent effects on coho salmon growth and survival.