Weirs: An effective tool to reduce hatchery–wild interactions on the spawning grounds?

Abstract

AbstractObjectiveHatchery‐produced salmon and steelhead Oncorhynchus mykiss mitigate for habitat loss and support valuable commercial, recreational, and tribal fisheries. However, excessive interactions between hatchery and wild fish can also cause harm to wild populations. Weirs have been implemented as a management tool to remove returning hatchery‐origin Chinook Salmon O. tshawytcha adults to reduce the proportion of hatchery‐origin spawners (pHOS) in spawning areas used by their wild counterparts. Using a decade of data since new weirs were implemented, our objective was to evaluate the efficacy of this management approach.MethodsWeirs were installed in six tributaries of the lower Columbia River, Washington. Using data from both weir operations and spawning ground surveys, we quantified weir efficiency (the proportion of upstream migrants captured by the weir) and resultant pHOS in wild spawning areas and evaluated evidence for unintended effects of weirs on naturally spawning salmon in years with (2011–2017) and without weirs (2002–2010) in one population by evaluating weir effects on spawner apparent residence time, distribution, and recruitment to the juvenile stage.ResultWeir efficiencies were highly variable across years and locations (7.3–100.0%). Estimates of pHOS were variable among populations and years (interannual population mean range: 11.9% to 75.2%) but were higher in all but 2 of the 43 population‐year combinations than needed to maintain recovery‐plan‐targeted levels of population viability using quantitative genetic model‐based standards identified by the Hatchery Scientific Review Group. Several factors contributed to the proportion of hatchery spawners exceeding pHOS standards at the population level, including fish passing weirs during freshets, spawning below weir sites and in tributaries without weirs, and imperfect marking of hatchery‐released juveniles. In the sole watershed with sufficient pre‐weir data to evaluate before–after changes, we documented a downstream shift in the spatial distribution of redds, lower apparent residence time of spawners, and lower production of parr in years following weir installation.ConclusionOur results indicate that using weirs as a tool to reduce hatchery–wild interactions on the spawning grounds is challenging. Weir implementation reduced pHOS; however, in most cases, resulting pHOS was not low enough to facilitate fit and viable wild populations. Weirs also were associated with unintended negative impacts to the single population where we had adequate before and after weir implementation data to evaluate effects. These impacts may be harmful to recovery efforts and offset the intended benefits of weirs to wild populations. Our study highlights the need for adaptive management to realize the goals intended for weirs, underscores the need for experimental design approaches, and identifies opportunities for future work.