Factors That Influence the Downstream Migration Rates of Juvenile Salmon and Steelhead through the Hydroelectric System in the Mid-Columbia River Basin

We investigated the extent to which key factors influenced the migration rate of the smolts of Pacific salmon Oncorhynchus spp. through impounded portions of the mid-Columbia River, during the years 1989–1995. Actively migrating chinook salmon O. tshawytscha (oceantype and stream-type forms), sockeye salmon O. nerka, and steelhead O. mykiss were analyzed by bivariate and multiple-regression methods. The dependent variable was the rate (km/d) at which uniquely coded PIT-tagged (passive integrated transponder tags) smolts migrated between Rock Island Dam and McNary Dam. Predictor variables consisted of indices of river discharge volume (flow), water temperature, release date of tagged fish, and fish size. The variable of key interest was flow because water management strategies are in place to increase water velocity through flow augmentation, with the intention of increasing smolt migration rate to decrease smolt mortality. For spring-migrating sockeye salmon, hatchery steelhead, and wild steelhead, flow was the primary predictor variable entering the models, and the bivariate models explained 42, 36 and 31% of the observed variation in migration rate for those species, respectively. Yearling chinook salmon migration rate was not correlated with any variable. Summer-migrating ocean-type chinook salmon showed no response to flow over a broad range of discharge (1,500–5,000 m3/s). However, there was a positive relationship between migration rate and fish length at the time of tagging for oceantype chinook salmon; r2 in the bivariate model = 0.59. Implications of these findings to water management strategies are discussed.