Abstract. Nighttime chemistry in polluted regions is dominated by the nitrate radical (NO₃) including its direct reaction with natural and anthropogenic hydrocarbons, its reaction with NO₂ to form N₂O₅, and subsequent reactions of N₂O₅ to form HNO₃ and chlorine containing photolabile species. We report nighttime measurements of NO₃, NO₂, and O₃, in the polluted marine boundary layer southwest of Vancouver, BC during a three week study in the summer of 2005. The concentration of N₂O₅ was calculated using the well known equilibrium, NO₃+NO₂↔N₂O₅. Median overnight mixing ratios of NO₃, N₂O₅ and NO₂ were 10.3 ppt, 122 ppt and 8.3 ppb with median N₂O₅/NO₃ molar ratios of 13.1 and median nocturnal partitioning of 4.9%. Due to the high levels of NO₂ that can inhibit approach to steady-state, we use a method for calculating NO₃ lifetimes that does not assume the steady-state approximation. Median and average lifetimes of NO₃ in the NO₃-N₂O₅ nighttime reservoir were 1.1–2.3 min. We have determined nocturnal profiles of the pseudo first order loss coefficient of NO₃ and the first order loss coefficients of N₂O₅ by regression of the NO₃ inverse lifetimes with the [N₂O₅]/[NO₃] ratio. Direct losses of NO₃ are highest early in the night, tapering off as the night proceeds. The magnitude of the first order loss coefficient of N₂O₅ is consistent with, but not verification of, recommended homogeneous rate coefficients for reaction of N₂O₅ with water vapor early in the night, but increases significantly in the latter part of the night when relative humidity increases beyond 75%, consistent with heterogeneous reactions of N₂O₅ with aerosols with a rate constant k_het=(1.2±0.4)×10⁻³ s⁻¹−(1.6±0.4)×10⁻³ s⁻¹. Analysis indicates that a correlation exists between overnight integrated N₂O₅ concentrations in the marine boundary layer, a surrogate for the accumulation of chlorine containing photolabile species, and maximum 1-h average O₃ at stations in the Lower Fraser Valley the next day when there is clear evidence of a sea breeze transporting marine air into the valley. The range of maximum 1-h average O₃ increase attributable to the correlation is ΔO₃=+1.1 to +8.3 ppb throughout the study for the average of 20 stations, although higher increases are seen for stations far downwind of the coastal urban area. The correlation is still statistically significant on the second day after a nighttime accumulation, but with a different spatial pattern favouring increased O₃ at the coastal urban stations, consistent with transport of polluted air back to the coast.