Summertime tropospheric ozone photochemistry over the Western North Atlantic Ocean.

Abstract

This research addresses the impact of pollution from the North American continent on the distribution and budget of ozone over the Western North Atlantic Ocean. The issues we investigate include: (1) the photochemical production and loss rates of ozone in air masses reaching Sable Island, Nova Scotia during the North Atlantic Regional Experiment (NARE) 1993 summer intensive; (2) the frequency and chemical characteristics of different types of air masses impacting Sable Island; and (3) the chemical and physical history of air masses traveling over the Western North Atlantic Ocean. The techniques we use to address these issues involve: (1) a photochemical modeling study of instantaneous ozone production and loss rates; (2) an extensive statistical correlation analysis of chemical measurements at Sable Island; and (3) a modeling study of a polluted air mass transported from the Northeastern United States over the North Atlantic Ocean, with an emphasis on the impact of fog episodes on the loss of reactive oxidized nitrogen species. The results of this research establish that there is on average net ozone production at Sable Island (average equal to 3.6 ppbv/d) during the NARE 1993 summer intensive. The dominant ambient variables controlling the photochemical production and loss rates of ozone at the site during the measurement campaign appear to be levels of nitrogen oxides, ozone, non-methane hydrocarbons, and solar intensity determined by cloud cover. Chemical correlations of O₃ with CO, CO with NO_y, O₃ with NO_y, and O₃ with NO_z categorized according to air mass origin indicate the influence of continental pollution as well as potential impact from regional sources such as ship plumes. Weak correlations of CO and O₃ with reactive nitrogen species suggest significant loss processes for NO_z during transport from the continent, potentially attributable to wet deposition. Finally, we investigate recent suggestions that PAN is lost through fog processes [<italic>Roberts et al</italic>., 1996], concluding that unrealistic assumptions are necessary to achieve a significant loss of PAN through fog chemistry and instead proposing a scenario involving gas phase chemistry only and variations in the temperature histories of air masses reaching the site.