Dams without fish passage facilities block access to much of the historic spawning habitat of spring Chinook salmon (Oncorhynchus tshawytscha) in Oregon's Willamette River basin. Adult salmon are routinely outplanted above the dams to supplement natural production, but many die before spawning despite extensive suitable habitat. In 2004-2007, we examined prespawn mortality patterns using live detection and carcass recovery data for 242 radio-tagged outplants. Total prespawn mortality was 48%, but variability was high, ranging from 0% to 93% for individual release groups. Prespawn mortality was strongly condition dependent, consistently higher for females than males and higher for early release groups. Across years, warm water temperature in the migration corridor and at the collection site was associated with sharply higher mortality. Results highlight a need for better evaluations of the effects of adult mortality on population reintroduction and recovery and relationships among prespawn mortality, dam-related temperature change and salmon life history and behaviour.

High-head dams in Oregon's Willamette River basin inhibit seaward migration and present significant mortality risks to ESA-listed juvenile Chinook salmon (Oncorhynchus tshawytscha). Over 7 years, we passively collected 13,365 downstream-migrating juvenile salmon in rivers above and below Willamette dams. Most salmon emigrated from upstream sites in February-June, but passed dams in November-February when reservoirs were drawn down near annual lows, and access to deep-water passage routes improved. Samples collected above reservoirs were dominated by subyearlings, whereas below-dam samples were a phenotypically diverse mix of subyearling, yearling and older salmon. The life history data indicated that Willamette reservoirs seasonally entrap many salmon and some sea-ready smolts probably residualise. Annual dam-passage mortality estimates were 8-59% (mean = 26%). Individual salmon mortality risk increased significantly with body length and varied with reservoir elevation and discharge. Operational changes that allow timely volitional emigration and development of less hazardous passage routes would benefit this threatened population.

Many high-head dams in Oregon's Willamette River basin were constructed without fish passage facilities for downstream migrants. Instead, fish pass dams via hydroelectric turbines, surface spillways or deep-water regulating outlets. The availability of these routes varies seasonally with dam operations and reservoir depth, which can fluctuate by tens of meters.To assess how dam and reservoir operations affect fish movement timing and survival, we used rotary screw traps below three Willamette basin dams and at two riverine sites above reservoirs. Traps were operated 2950 days over 8 years, and >195 000 fish were collected. Samples above reservoirs were primarily native salmonids (Oncorhynchus spp.), daces (Rhinichthys spp.) and sculpins (Cottus spp.), while those below dams were often dominated by non-native Centrarchidae. Capture rates at riverine sites were highest from late winter to early summer, coincident with juvenile Chinook salmon emigration. Conversely, collection below dams was largely restricted to late fall and winter when reservoirs were drawn down to annual lows and discharge was high. We hypothesize that winter operations facilitated fish access to dam turbines and regulating outlets, whereas spring-summer operations entrapped fish in reservoirs and restricted volitional downstream passage.Total fish mortality was ≤2% at riverine sites and was 36-69% below dams. Estimates were highest for non-native species and juvenile Chinook salmon. Fatal injuries were consistent with traumas related to pressure, shear and contact and there were size-related and morphology-related risk differences. Mitigation opportunities include fish bypass system development, retrofits for existing routes and seasonally appropriate reservoir draw down to allow fish passage.