We classified juvenile hatchery summer steelhead Oncorhynchus mykiss released from Wells Hatchery, Washington, from three brood years (2002)(2003)(2004) as those that migrated volitionally (VM) from rearing ponds or as those that were forced out of rearing ponds after volitional migration concluded (NM). Fish were implanted with PIT tags prior to release and we used recreational angling equipment to recapture tagged fish to estimate the relative contribution rates of VM and NM release groups to the stream-resident population of juvenile hatchery summer steelhead in the Twisp River. We also evaluated the survival for each group from release to McNary Dam, and from release to adult return (SAR) at Bonneville and Wells dams on the Columbia River. Overall, we estimated that 82% of stream-resident hatchery juvenile summer steelhead originated from releases of NM fish. The probability of survival from release to McNary Dam was significantly greater for VM groups (mean, 0.4817; SE, 0.023) than for NM groups (mean, 0.2182; SE, 0.021) within each year. The mean SAR to Bonneville and Wells dams was 1.54% and 1.26%, respectively, for VM fish and 0.37% and 0.32%, respectively, for NM fish; the differences were significant between groups within each year. As an index of release strategy performance, VM releases resulted in one stream-resident fish recaptured for every 7.8 adults returned, while NM releases produced one stream-resident fish recaptured for every 0.48 adults returned. These results suggest that managers employ a volitional release strategy to significantly reduce the abundance of stream-resident juvenile hatchery steelhead by not releasing NM fish into waters inhabited by anadromous fishes, thereby reducing negative ecological interactions between hatchery residual steelhead and wild salmonids at little cost to adult returns.Juvenile steelhead Oncorhynchus mykiss (anadromous Rainbow Trout) exhibit a range of life history strategies that when forced under an artificial propagation regime can result in unintended consequences from an ecological or management perspective. In the upper Columbia River basin, wild juvenile steelhead obtain the critical threshold size that initiates smoltification between 1 and 7 years of age with most fish migrating at age 3 (45%) and age 2 (44%; Peven et al. 1994). In most hatcheries however, constraints on rearing space and water quantity prohibit the rearing of multiple age-classes of steelhead

Survival of juvenile anadromous Pacific salmonids from their earliest age of seaward movement ("outmigration") through the tributaries that connect their rearing grounds to larger-order rivers ("cohort survival") is an important yet often unmonitored factor in the complex life history of these species. Populations with variable age at out-migration (e.g., steelhead Oncorhynchus mykiss) or multiple juvenile rearing strategies (e.g., Chinook Salmon O. tshawytscha) raise particular challenges in survival monitoring. A multiple-state release-recapture model is presented to estimate cohort survival for salmonid species with variable age at out-migration and demonstrated in two case studies. Annual releases of fish tagged with passive integrated transponder (PIT) tags from the same brood year (cohort) and their detections by instream PIT-tag detection arrays in tributaries and at hydroelectric dams are used to estimate survival through tributaries and the age composition of juvenile migrants reaching downstream sites. The 2010 cohort of Chinook Salmon from the Chiwawa River in Washington State had estimated survival of 0.59 to the middle Wenatchee River array in Tumwater Dam Reservoir (river distance of 30 km) and 0.12 to McNary Dam (364 km) on the Columbia River. The age composition shifted from 60.1% subyearlings at the Wenatchee River array to 100% yearlings at McNary Dam, reflecting the use of the Wenatchee River as a nonnatal rearing area, whereas McNary Dam is in the migratory corridor. The 2010 cohort of steelhead from the Twisp River in Washington State had estimated survival from the yearling stage of 0.30 to Rocky Reach Dam (146 km from the Twisp River mouth) on the Columbia River and 0.17 to McNary Dam (439 km). The low survival to the Columbia River resulted from both mortality and adoption of a resident life history and reflects the diverse population structure in the Twisp River.

Summer‐run Chinook Salmon Oncorhynchus tshawytscha migrating over Wells Dam, Washington, enter a habitat characterized by blocked upstream access, high tributary water temperature regimes, and robust tribal and recreational fisheries. In 2011, we initiated a 2‐year radiotelemetry study to identify population‐specific run timing, movement, and mortality of naturally produced fish passing the dam. Five hundred seventeen salmon were radio‐tagged at Wells Dam over 2 years of study. The highest proportion (44%) of tagged fish escaped to the Okanogan River, but spawning populations from the Methow River (16%), Wenatchee River (6%), Entiat River (5%), and the Columbia River upstream (14%) and downstream of Wells Dam (14%) were also represented. In general, tributary‐spawning fish had significantly earlier run timing than did main‐stem–spawning fish. We observed very little movement among spawning tributaries, but a significant proportion of fish (~30%) were detected holding in the tailrace of Chief Joseph Dam on the Columbia River prior to spawning, including fish from populations many kilometers downstream of Wells Dam. Fallback was common in each year of the study, and we calculated that passage at Wells Dam was overestimated by 27.5% in 2011 and 32.0% in 2012 due to fallback and re‐ascension. Of the fish that remained upstream, 16% in 2011 and 22% in 2012 were estimated to have died prior to spawning, excluding fish that were known or suspected to have been harvested. In 2012, warm water temperatures (>20°C) in the Okanogan River resulted in a thermal barrier that delayed migration but did not affect spawning distribution within the river compared with 2011, when no significant thermal barrier was detected. Our results highlight some of the complex migration and distribution patterns of natural‐origin Chinook Salmon passing Wells Dam and should assist managers in upstream population and fishery modeling efforts.

Mountain Whitefish Prosopium williamsoni is a native salmonid distributed throughout the Columbia River basin, but little is known about the life history traits of tributary-specific populations. We captured and tagged 129 Mountain Whitefish with passive integrated transponder (PIT) tags between 2010 and 2015 at a weir on the Twisp River, Washington, to assess migration patterns. Detection records suggested that two populations of Mountain Whitefish overlap in the Twisp River during spring: resident fish that spawn in the Twisp River (42% of fish) and migrant fish that spawn in the Okanogan River (58% of fish). The median entry day into the Twisp River was similar for resident (April 17) and migrant (April 16) populations, but median exit day was significantly earlier for migrant (July 2) than for resident fish (November 6). Migrant fish exited the Methow basin and traveled at 14.1 river kilometers per day, passing the lower Okanogan River PIT antenna with a median date of September 17. After spawning and overwintering in the Okanogan River with a median exit date of March 16, these fish returned to the Methow basin traveling at 22.1 river kilometers per day. These results improve our understanding of Mountain Whitefish life history and habitat use in tributaries upstream of Wells Dam, Washington.