Haze and other aerosol components in late winter Arctic Alaska, 1986

Three coarse and five fine aerosol components of different elemental compositions were identified at Barrow, Alaska, from March 17 to April 21, 1986, resolved by absolute principal component analyses of element concentrations in 280 sequential coarse and fine size fraction time step samples. In the coarse (> 2.5 μm), two components C-1 and C-2 had abundant Si, S, Cl, K, and Ca, but no Al, and together contained 85% of coarse S. Their compositions resembled expected products of carbonaceous fuel combustion, with Si being volatilized by carbon reduction and other metals volatilized perhaps as chloride salts. C-1, with high trace metal contents, might be from nonferrous smelting, whereas C-2, with high Fe, might be associated with conventional coal combustion. Both appeared semi-aged with respect to acidic gas uptake because the S chemical equivalents were less than those of metals contributing to alkalinity. When combined with Cl, S was close to the metal equivalents, indicating complete acid-base titration. A strong concentration rise of C-1 and C-2 occurred from March 25 to April 2 during a haze event, although C-1 was also present at other times. Air trajectories showed that air masses arrived at Barrow during the haze event from eastern or northern Europe. The third component C-3 was a dust aerosol rich in Al that contained high S but low Cl, suggesting saturation with H2SO4 and therefore aged and regional aerosols perhaps typical of the late winter Arctic. No major change in its concentration was observed to correspond to synoptic events. In the fine (< 2.5 μm), five components represent a sea-salt aerosol, an S-rich aerosol with some Si, K, Ca, and Fe, a trace metal aerosol, an Al-rich dust, and a marine product with Br, S, and Cl. The sea-salt was found only in three plumes when synoptic meteorology and air trajectories suggested origins in the North Pacific. The S-rich aerosol, accounting for 73% of S and 40% of Si, was enhanced during the haze event by 75%, and existed over the entire period at rather high levels and might be due to SO2 oxidation and condensation on fine Si-rich particles. The trace metal aerosol was always present and could have been generated from nonferrous smelters at lower latitudes. The Al-rich dust particles contained 7% of S and were similar to component C-3. The marine aerosol contained 74% of Br, 69% of Cl, and 20% of S, and could be from oxidation of marine biogenic Br and S gases. All but the sea-salt were present at Barrow throughout the period, with no clear relation to air trajectories or synoptic meteorology. The results show that certain winter meteorological conditions favor pollutant transport from lower latitudes to the Arctic. But while haze is related to industrial pollutants, other nonpollution products are present in the winter Arctic and may be important constituents of haze. They also show that by careful data reduction and meteorological analysis, the sources and transport pathways of haze may be better understood.