Behavior, passage, and downstream migration of juvenile Chinook salmon from Detroit Reservoir to Portland, Oregon, 2014–15

Abstract

An evaluation was conducted to estimate dam passage survival of juvenile Chinook salmon ( Oncorhynchus tshawytscha ) at Detroit Dam during a period of spill. To estimate dam passage survival, we used a paired-release recapture study design and released groups of tagged fish upstream (997 fish) and downstream (625 fish) of Detroit Dam. A total of 43 fish (6.8 percent) passed Detroit Dam from the upstream release group and passage occurred through regulating outlets (54.8 percent), spill bays (31.0 percent), and turbines (14.3 percent). We do not present dam passage survival estimates from 2014 because these estimates would have been highly uncertain due to the low number of fish that passed Detroit Dam during the study. Secondary objectives were addressed using data collected from tagged fish that were released at the downstream release site. Juvenile salmonids have multiple passage options at the Bennett Dam complex, which includes a series of dams and braided channels. A pair of Passive Integrated Transponder (PIT) monitoring arrays were installed at Upper Bennett Dam and in the Stayton Canal by the U.S. Army Corps of Engineers and the Oregon Department of Fish and Wildlife during 2014. We deployed acoustic telemetry hydrophones near these arrays to detect acoustic-tagged fish from our study and used these detections to quantify proportions of tagged fish passing through the two routes. About one-fourth (0.257) of the tagged fish that were released downstream of Big Cliff Dam were detected on the new PIT tag array while passing the Bennett Dam complex. A total of 402 acoustic-tagged fish were detected at the complex and many (248 fish; 62 percent) eventually entered the Stayton Canal. Median residence time in the canal was 6.5 hours, but 12.7 percent of the fish had extended residence times (7–37 days). Passage also was monitored at the Sullivan Project at Willamette Falls and about 40 percent (0.398) of the tagged fish passing the project were detected on the PIT tag array. A Cormack-Jolly-Seber mark-recapture model was developed to provide reach-specific survival estimates for juvenile Chinook salmon. A portion of the tagged population overwintered in the Willamette River Basin and outmigrated several months after release. As a result, survival estimates from the model would have been negatively biased by factors such as acoustic tag failure and tag loss. Data from laboratory studies were incorporated into the model to provide survival estimates that accounted for these factors. In the North Santiam River between Minto Dam and the Bennett Dam complex, a distance of 37.2 kilometers, survival was estimated to be 0.844 (95-percent confidence interval 0.795–0.893). The survival estimate for the 203.7 kilometer reach between the Bennett Dam complex and Portland, Oregon, was 0.279 (95-percent confidence interval 0.234–0.324), and included portions of the North Santiam, Santiam, and Willamette Rivers. The cumulative survival estimate in the 240.9 kilometer reach from the Minto Dam tailrace to Portland was 0.236 (95-percent confidence interval 0.197–0.275).