The aerosol at Barrow, Alaska: long-term trends and source locations

• Published in: Atmospheric environment (1994) 1999-07, Vol.33 (16), p.2441-2458
• Main Authors: Polissar, A.V., Hopke, P.K., Paatero, P., Kaufmann, Y.J., Hall, D.K., Bodhaine, B.A., Dutton, E.G., Harris, J.M.
• Format: Article
• Language: English
• Subjects: Arctic haze; Atmospheric aerosol; Earth, ocean, space; Exact sciences and technology; External geophysics; Factor analysis; Meteorology; Particles and aerosols; Potential source contribution function; Trends
• ISSN: 1352-2310; 1873-2844
• DOI: 10.1016/S1352-2310(98)00423-3

Aerosol data consisting of condensation nuclei (CN) counts, black carbon (BC) mass, aerosol light scattering (SC), and aerosol optical depth (AOD) measured at Barrow, Alaska from 1977 to 1994 have been analyzed by three-way positive matrix factorization (PMF3) by pooling all of the different data into one large three-way array. The PMF3 analysis identified four factors that indicate four different combinations of aerosol sources active throughout the year in Alaska. Two of the factors (F1, F2) represent Arctic haze. The first Arctic haze have factor F1 is dominant in January–February while the second factor F2 is dominant in March–April. They appear to be material that is generally ascribed to long-range transported anthropogenic particles. A lower ratio of condensation nuclei to scattering coefficient loadings is obtained for F2 indicating larger particles. Factor F3 is related to condensation nuclei. It has an annual cycle with two maxima, March and July–August indicating some involvement of marine biogenic sources. The fourth factor F4 represents the contribution to the stratospheric aerosol from the eruptions of El Chichon and Mt. Pinatubo. No significant long-term trend for F1 was detected while F2 shows a negative trend over the period from 1982 to 1994 but not over the whole measurement period. A positive trend of F3 over the whole period has been observed. This trend may be related to increased biogenic sulfur production caused by reductions in the sea-ice cover in the Arctic and/or an air temperature increase in the vicinity of Barrow. Potential source contribution function (PSCF) analysis showed that in winter and spring during 1989 to 1993 regions in Eurasia and North America are the sources of particles measured at barrow. In contrast to this, large areas in the North Pacific Ocean and the Arctic Ocean was contributed to observed high concentrations of CN in the summer season. Three-way positive matrix factorization was an effective method to extract time-series information contained in the measured quantities. PSCF was useful for the identification possible source areas and the potential pathways for the Barrow aerosol. The effects of long-distance transport, photochemical aerosol production, emissions from biogenic activities in the ocean, volcanic eruptions on the aerosol measurements made at Barrow were extracted using this combined methodology.