Morphological differences between adult wild and first-generation hatchery upper yakima river spring Chinook salmon

Four-year-old adult wild spring Chinook salmon Oncorhynchus tshawytscha at a supplementation hatchery were compared morphologically with their first-generation hatchery counterparts over three consecutive brood years (BYs) using thin-plate spline analysis on 12 digitized landmarks and an analysis of 27 truss characters based on the landmarks. Canonical discriminant analysis (CDA) of sex-specific partial warp scores correctly classified females to origin (hatchery or wild) with 75% accuracy (up to 84% accuracy for one BY) and males to origin with 65% accuracy (up to 89% accuracy for one BY). Classification to BY using sex- and BY-specific partial warps was 62% accurate for both females and males. The results of the truss analysis were very close to those from the thin-plate spline analysis. Sex-specific CDAs correctly classified 75% of females and 68% of males to origin. Correct classification to BY was 61% for females and 58% for males. Consensus shapes based on partial warps suggested that hatchery fish of both sexes have longer and deeper heads and shallower midbodies than wild fish. They also appear to be somewhat shorter in the posterior body. Analysis of variance of individual truss characters led to the same general conclusion but also provided evidence that hatchery fish have wider anal fins. There was no evidence of sex-specific differences between hatchery and wild fish. Body proportion differences between hatchery and wild fish at the eight most diagnostic truss characters averaged 0.31 standard deviations in females and 0.35 standard deviations in males. In terms of actual measurement, these differences amounted to an average of 1.4% of the wild mean, providing little discriminatory power.