This dissertation discusses three facets of Pacific Northwest (PNW) Chinook salmon, Oncorhynchus tshawytscha, ecology that are relevant to current management models used for this
species: 1) life history and migration pathways, 2) the impacts of spatial and temporal variability on PNW Chinook survival and maturation as determined using coded wire tags, and 3) the relationship between recruitment variability and environmental variation, explored via development of a new age-structured model.

Chinook salmon are commonly categorized as ocean-type (migrating to the ocean in their first year of life) or stream-type (migrating after a full year in freshwater). To test the hypothesis that stream-type and ocean-type Chinook have different migration pathways, I used a pair-wise design based on coded wire tagging data to compare the marine distributions of stream- and ocean-type Chinook salmon from a suite of rivers producing both forms. In cases where there is minimal gene-flow between both types of Chinook, the hypothesis is true, but in other cases it is not.

For the second element of the study, I examined PNW Chinook survival and maturation as estimated using coded wire tag data and related them to environmental conditions. PNW Chinook
variability in survival and maturation is summarized for twenty three different PNW stocks and analyzed using multivariate data analysis techniques. Results indicate that survival is grouped into eight distinct regional clusters. Age at maturation did not exhibit regional patterns. Environmental data indicate that PNW salmon survival is only weakly correlated with ocean conditions and co-vary at distances of 500 kms, while age at maturation appears to have no linkage to ocean conditions.

Finally, in order to demonstrate how direct measures of environmental conditions can be integrated into Chinook salmon stock assessments I developed a catch-at-age model of Chinook salmon life history which fits estimated parameters to observed data on ocean catches, terminal catches, and terminal escapement by age. The model quantifies uncertainty so that managers can use precautionary approaches to managing these stocks that would help protect the resource in the long run, and is tested for one of the stocks used in management.