Population genetic analysis of Chehalis River watershed coho salmon (Oncorhynchus kisutch)

Understanding the population structure of wild salmon and steelhead (Oncorhynchus spp.) in the Chehalis Basin is an important part of the Chehalis Flood and Aquatic Species Project and contributes to the Chehalis Basin Flood Hazard Project and Aquatic Species Enhancement Plan. Current predictive models (Ecosystem Diagnostic Treatment, NOAA Watershed Assessment) partition species into geospatial units that have an unknown relationship to actual population structure. Here, we examined the genetic population structure of coho salmon (O. kisutch) in the Chehalis River basin. Specifically, our objectives were to determine the genetic relationship of Chehalis Basin coho with available reference baseline coho collections in Washington State, to examine the spatial and temporal (among cohort) genetic structure of coho among the Chehalis River sub-basins, and to evaluate genetic relationships of coho with early and late return timing.

Samples were collected in 2017 and 2018 from throughout the Chehalis Basin, including fish from all major spawning tributaries and from the mainstem Chehalis River upstream of the proposed dam site at river mile 108.2. These samples represented fish that spawned mainly in 2014 and 2015. Genetic diversity was generally high, though late coho hatchery programs had lower estimated effective population sizes (Ne) and weak indication of Hardy-Weinberg and linkage disequilibrium typical of smaller (i.e., fewer broodstock) hatchery programs observed in other salmonid species.

Genetic data revealed that Chehalis Basin coho salmon are very different from Puget Sound coho salmon (the only available reference collections). This difference allowed us to detect two Puget Sound coho, one each in the Skookumchuck and Newaukum rivers. These fish likely represented unintentional releases of Puget Sound coho, which are temporarily reared at the Skookumchuck Hatchery.

Within the Chehalis Basin, population structure was complex with Chehalis coho displaying genetic structure related to cohort (brood year), spawning location, and run timing. Hatchery practices likely greatly influenced observed structure, especially for populations where hatchery broodstock were collected and direct plants of hatchery produced fish occurred. Most populations had differences among cohorts, i.e., 2017 collections were different from 2018 collections. This suggests that few jacks return to the Chehalis, that jacks have poor reproductive success, or both, and in the case of populations with hatchery programs, it suggests that those hatchery programs spawn few jacks as broodstock. Indeed, the jack fraction of hatchery broodstock for all programs was, on average, less than the target 10% for recent years (including 2014 and 2015, which produced the adults sampled in 2017 and 2018), but it was not zero for programs with available data. Spatially, most spawning tributaries were different from other spawning tributaries, including the upper Chehalis coho, which were collected from upstream of the proposed dam site. Other researchers have also noted that "upper Chehalis coho" were different from other Chehalis coho, though the collection location of their samples was either unknown or from tributaries downstream of the proposed dam site. Exceptions to the general pattern of spatial structure were mainly the late coho in rivers with late coho hatchery programs, i.e., the Humptulips, Satsop, and Skookumchuck rivers. In those locations, the late coho were genetically similar, likely a legacy of long-term releases and propagation of Satsop late coho at the Humptulips and Skookumchuck hatcheries. These locations, along with the Wishkah River, were also the only locations Seamons et al. Chehalis River coho salmon genetic structure where early and late timed coho were consistently genetically different, again, likely a legacy of hatchery practices. In the Wishkah River, the source broodstock of the early hatchery program is unknown, but could include fish from Puget Sound or the north Washington coast. Results hint of an out of Chehalis basin influence, especially for the 2018 cohort, but our ability to provide a more precise test was limited by a lack of a comprehensive reference baseline.

This study represents a comprehensive survey and genetic analysis of Chehalis Basin coho salmon populations, but some improvements could be made in the future. Sample sizes could be increased, particularly from smaller spawning tributaries. Given that among cohort differences were observed, sampling the third cohort for all locations could be required, especially if the future use of genetic stock identification was needed for within the Chehalis Basin. Improvements could be made in the spatial coverage of collections as well, particularly from tributaries of South Grays Harbor and mid-Chehalis tributaries. The Washington Department of Fish and Wildlife Molecular Genetics Laboratory is in the process of developing a statewide coho reference baseline. Once this reference baseline is expanded beyond Puget Sound and the Chehalis Basin, particularly to other systems near the Chehalis Basin (e.g., Quinault River to the north and Willapa River to the south), reanalysis of Chehalis Basin coho with the reference baseline would improve our understanding of how coho are related on a broader spatial scale and in particular how Chehalis Basin coho are related to other coho in Washington State. In addition, future analyses could incorporate power analysis to determine the ability to use genetic stock identification to assign coho of unknown origin to their source population among populations within the Chehalis Basin, but also among major systems or on a regional scale.