Estimates of Lower Columbia River Steelhead pHOS: A Report to NOAA Fisheries

Hatchery steelhead provide valuable angling opportunities and can be used to supplement natural-origin populations but may also pose risks to natural-origin populations. As a result, the proportion of hatchery-origin spawners (pHOS) is often monitored as a measure of genetic and ecological interactions between hatchery-origin and natural-origin steelhead. We used observations of hatchery- and natural-origin spawners during spawning ground surveys (winter steelhead) and snorkel surveys (summer steelhead) to develop estimates of pHOS for steelhead populations in the Washington State portion of the lower Columbia River. Three models were used to estimate pHOS: 1) a multivariate state-space model fit to the count data using a logit link function and a binomial response using STAN rstan, 2) a Generalized Additive Model fit to the count data using a logit link function and a binomial response where a thin plate spline is fit independently to each population’s data using the R package mgcv, and 3) method of moment estimates where pHOS is calculated annually using the count data from each watershed for each population in years data are available. Results from each method were compared to determine the most appropriate estimates to report. The multivariate state-space model produced the most robust estimates. The five-year mean pHOS (spawn years 2019-23) was within Mitchell Act Biological Opinion (MA BiOp) standards for all populations except for Washougal Winter Steelhead and pHOS was unknown for Salmon Creek Winter Steelhead. Estimates for Kalama and NF Toutle winter run populations are reported both for below-dam subpopulations as well as the whole populations. Results currently exclude populations within the Cowlitz and Lewis basins for which estimates are developed as part of other reporting processes associated with dam operator licenses. Our results suggest that actions to reduce pHOS (gene banks, reductions in hatchery plants, changes in release locations of hatchery plants, and changes to recycling protocols) appear to have largely achieved pHOS targets identified in the MA BiOp.