Post-hydropower system delayed mortality of transported Snake River stream-type Chinook salmon: Unraveling the mystery

Past research indicates that on an annual basis, smolts of stream-type Chinook salmon Oncorhynchus tshawytscha collected at Snake River dams and transported by barge to below Bonneville Dam have greater post-hydropower system mortality than smolts that migrate in-river. To date, this difference has most commonly been attributed to stress from collection and transportation, leading to decreased disease resistance or predator avoidance ability. Using both hatchery and wild passive integrated transponder (PIT) tagged Chinook salmon, we explored two mechanisms that either separately or jointly contributed to an alternative explanation: Altered timing of ocean entry and lost growth opportunity leading to size-selective predation. Based on weekly estimates of in-river survival and adult return rates of smolts that were transported or that migrated in-river between Lower Granite and Bonneville dams, we found greater post-hydropower system mortality for smolts transported early in the season but greater mortality for in-river migrating smolts later in the season. Migrants took 2–4 weeks to travel between the two dams, while transported fish took less than 2 d. Thus, fish leaving Lower Granite Dam under the two transit modes encountered different conditions downstream from Bonneville Dam. Further, wild and hatchery migrants grew 6–8 and 5–6 mm, respectively, while transported fish had no apparent growth in the less than 2-d barge ride. Using length data and regression equations of size selectivity, we found that transported smolts were more vulnerable to predation by northern pikeminnow Ptychocheilus oregonensis (freshwater) and Pacific hake Merluccius productus (marine) than were migrants; this was particularly true for the smaller wild smolts transported early in the season. We concluded that the most parsimonious explanation for differential post-hydropower system mortality of transported Chinook salmon smolts related not to effects of stress but to differential size and timing of ocean entry