Date of Award


Publication Type

Master Thesis

Degree Name



Great Lakes Institute for Environmental Research

First Advisor

Semeniuk, Christina

Second Advisor

Love, Oliver


Aquaculture, Behaviour, Cortisol, Integrated phenotype, Salmon




Across evolutionary time, correlational selection can cause the co-dependencies of multiple functionally related traits, leading to tightly correlated suites of traits known as integrated phenotypes. While it is conceptually appreciated that integrated phenotypes should benefit fitness through increased functional performance, few studies have shown evidence for this empirically. Further, within applied fields such as aquaculture, single-trait predictors of performance and fitness have been predominantly targeted without accounting for multi-trait contributions to desired phenotypes, potentially failing to maximize growth across generations in captivity. Throughout this thesis, I holistically examine the causes and consequences for the co-variation between functionally related physiological (cortisol) and behavioural traits (foraging, sociality and disturbance-sensitivity) amongst seven outcrossed (Wild x Domestic) and one fully domesticated populations of Chinook salmon (Oncorhynchus tshawytscha). I take a hierarchical approach progressing through levels of complexity (single- to multi-trait models) to determine whether integrated phenotypes can be identified at the family and/or population levels across two environmental conditions (fresh and salt water) found in salmonid aquaculture. I determined that the process of outcrossing can generate important inter-population variation in cortisol and behavioural traits, particularly within the freshwater environment. In one study, I found that at a family level, the tight integration of nighttime cortisol exposure (6:00pm-6:00am) and behavioural phenotype, with intermediate rather than extreme values interacting to generate a hormetic relationship that maximized body size at a population level. Additionally, I determined in a second study that a three-way interaction between baseline cortisol (7:30-10:30am), foraging and sociality behaviours predicted performance (biomass and body mass), again within the freshwater environment. In this same study across environments, I determined that the interaction between foraging and sociality in fresh water (juveniles) could predict body mass in salt water (adults). However, at no level of investigation (i.e., family or population) did I observe integrated phenotypes amongst saltwater-expressed traits and further, freshwater performance was not a simple predictor of saltwater performance. Overall, interactions between multiple traits within our models were consistently the best predictors of freshwater performance supporting predictions that functional performance can be increased through phenotypic integration. This work illuminates the role of phenotypic integration in optimizing relationships between hormones and behavioural traits to increase functional performance, and can be applied within aquaculture production to promote more effective and robust selection of ideal phenotypes.