Sea-air carbon dioxide (CO<sub>2</sub>) fluxes have been analyzed from recently-collected winter, summer and autumn surface ocean CO<sub>2</sub> partial pressure (<em>p</em>CO<sub>2</sub>) data spanning a large portion of the western Canadian coastal ocean, and historical underway <em>p</em>CO<sub>2</sub> measurements from the southwest Vancouver Island shelf and the Strait of Juan de Fuca. Sea-air CO<sub>2</sub> fluxes from the recent data for specific subregions of the coastal ocean, selected based on geography or bathymetry, were used to make seasonal area-specific estimates of CO<sub>2</sub> exchange. These show significant differences between subregions, which have important consequences for estimating seasonal area-weighted fluxes on the margin. Climatologies of sea-air CO<sub>2</sub> flux were calculated from the historical data using two approaches: One based on fluxes calculated from temporally-averaged values of sea-air <em>p</em>CO<sub>2</sub> differences, solubility and gas transfer velocities, and the other from temporally-averaged instantaneous flux estimates. Seasonal flux estimates from our recently-collected data are consistent with the climatological estimates, in that both show stronger outgassing of CO<sub>2</sub> in autumn relative to winter, and both reveal straits as important atmospheric CO<sub>2</sub> source regions. Taken together, both analyses of recent and historical data suggest that the transition seasons (spring and autumn) contain the largest (positive and negative) fluxes because of the coincidence of high gas transfer velocities and large surface seawater disequilibria with the atmosphere. By combining the results from these analyses, and making some assumptions where data are missing, we estimate moderate net annual sea-air CO<sub>2</sub> influx in the western Canadian coastal ocean of −6 mmol m<sup>−2</sup> d<sup>−1</sup>.