Light scattering by fine particles during the Pittsburgh Air Quality Study: measurements and modeling

Cabada, Juan C

Khlystov, Andrey

Wittig, Ann E

Pilinis, Christodoulos

Pandis, Spyros N

2013-04-16T17:00:05Z

2004

{[}1] Light scattering by fine particulate matter was measured during the Pittsburgh Air Quality Study (PAQS) as close to ambient conditions as possible. Several approaches are used for the theoretical calculation of the scattering coefficient and the results are compared to the direct measurements. The first approach uses ambient high time and daily resolved PM2.5 composition concentrations to estimate the scattering coefficient assuming that the aerosol is an external mixture. The second approach uses a thermodynamic model and Mie theory to predict the scattering coefficient of aerosols from daily size composition distributions. The third approach introduces high time and daily resolved ambient aerosol water concentrations and concentrations of sulfate, nitrate, organic material, and soil with fixed scattering efficiencies. During the summer the first two approaches underestimate the measured scattering coefficient by around 20%. Agreement within experimental error is obtained between the measured scattering coefficient and the model, incorporating measured water aerosol concentrations. During the winter the first two approaches tend to overpredict the measured scattering by around 15%. This overprediction is weakly correlated to the organic mass. The modeling approaches suggest that sulfate and the associated water contribute 65 - 73% to the scattering coefficient during the summer, with organic material contributing 25 - 30%. During the winter, sulfate accounts for 35 - 43%, nitrate accounts for 24 - 32%, and organic material accounts for 30 - 40% of the scattering coefficient.

0148-0227

https://hdl.handle.net/10161/6546

American Geophysical Union (AGU)

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES

Light scattering by fine particles during the Pittsburgh Air Quality Study: measurements and modeling

Journal article

D16

Duke

Faculty

Published

109