ObjectiveAlthough many Chinook Salmon Oncorhynchus tshawytscha populations overlap in nearshore areas prior to spawning migrations, it is unclear how life history diversity influences physical condition and habitat use. Here, we explored multiple dimensions of Chinook Salmon marine ecology. First, does condition differ between immature and mature fish, among stocks, and between wild and hatchery individuals? Second, is abundance correlated with abiotic variables? Third, does habitat use consistently covary with life history stage, stock, and wild versus hatchery rearing history?MethodsWe collected data on Chinook Salmon stock identity, condition, and abundance using a fisheries‐independent troll survey along the west coast of Vancouver Island, British Columbia. We then fitted generalized additive models and geostatistical generalized linear models to quantify variability in condition, abundance, and spatial distribution.ResultFork length and lipid content varied seasonally, with maturation stage, and among stocks but did not differ with rearing history. Although immature fish were initially less lipid rich than mature fish, the lipid content of immature individuals ultimately exceeded that of mature individuals. Chinook Salmon abundance covaried with bottom depth, slope, and sampling date, while diel and tidal effects were weak. Abundance varied among ecological groups by up to an order of magnitude. Chinook Salmon habitat use differed among size‐classes and stocks but did not differ with rearing history. The spatial distributions of each size‐class changed over summer, consistent with ontogenetic dispersal and variation in the migration timing of spawners.ConclusionSeasonal changes in Chinook Salmon condition suggested that immature individuals transition from growth to lipid storage, emphasizing that prey availability may impact overwinter survival. Stock‐specific patterns in size, lipid content, and abundance highlighted ecological diversity during marine residence. Although distributions varied seasonally, abundance was greatest in high‐relief areas. Finally, our estimated spatial distributions suggest that responses to environmental conditions vary with ontogeny and among populations but not with rearing history.