Arctic Aerosol Sources and Continental Organic Aerosol Hygroscopicity

Abstract

Atmospheric particles can affect climate directly, by scattering solar radiation, or indirectly, by acting as the seed upon which cloud droplets form. These clouds can then cool the earth's surface by reflecting incoming sunlight. In order to constrain the large uncertainties in predicting the ultimate effect of aerosol on climate, the sources of atmospheric particles and their subsequent ability to turn into cloud droplets needs to be better understood. This thesis addresses two parts of this issue: the sources of Arctic aerosol and the hygroscopicity of continental organic aerosol. Small particles were observed in Baffin Bay during September 2008 that coincided with high atmospheric and ocean surface dimethyl sulphide (DMS) concentrations suggesting that the aerosol formed from oceanic sources. An aerosol microphysics box model confirmed that local DMS could have produced the observed particles. In addition, the particle chemical composition was measured using aerosol mass spectrometry in the central Arctic Ocean in August 2008 and particles were found to be 43% organic and 46% sulphate. Factor analysis further apportioned the aerosol mass to marine biogenic and continental sources 33% and 36% of the time, respectively, with the source of the remaining mass unidentified. The second part of the study parameterises the hygroscopicity of the ambient organic aerosol fraction (κorg) at Egbert, Ontario and Whistler, British Columbia. This was done using two methods: 1) by assuming that the oxygenated organic component was hygroscopic and that the unoxygenated organic component was non-hygroscopic, κ of the oxygenated component was found to be 0.22 ± 0.04, and 2) by assuming that κorg varied linearly with the atomic oxygen to atomic carbon ratio, it could be parameterised as κorg = (0.29 ± 0.05) × (O/C). Calculations predict that knowing κorg is important in urban, semi-urban, and remote locations whenever the inorganic mass fraction is low.