Sockeye passage shifts offset part of the reconstructed rise in hot-day passage at Bonneville Dam, 1960–1989 to 1996–2025

We tested how river temperature and passage dates changed the fraction of adult fish counts on hot days at Bonneville Dam. Hot days have reconstructed water temperature at or above 20°C. We compared April–October counts in 1960–1989 and 1996–2025 for Chinook, sockeye and steelhead. An exact calculation separates temperature, passage-date and same-year association components. The planned comparison has 29 years per period and 208 common dates. Sockeye passage shifts reduce overlap by 3.9–4.1 percentage points across two temperature series. A check with all 214 dates and 28 years per period gives −3.6 points in both series. Its 95% year-block intervals are −5.3 to −2.0 and −5.6 to −1.9 points. The temperature component adds 9.0–11.3 points. Total sockeye hot-day overlap increases by 6.3–7.7 points, but one total-change interval includes zero. Chinook passage shifts add 4.3–4.5 points in that check. The steelhead passage-date component is small and uncertain. The sockeye passage component stays negative across thresholds, count weights, period changes and a later Bonneville-only temperature check. Source records, calibration code, results, figures, coverage decisions and offline reproduction instructions are preserved. This is a descriptive count index. Mixed stocks, missing count cells, changes in temperature sources and fixed seasonal limits restrict interpretation. It does not measure fish body temperature, cumulative heat dose, survival or adaptation.

fish migration · sockeye salmon · Chinook salmon · steelhead · Bonneville Dam · river temperature · phenology · thermal overlap · reproducible analysis