DOUGLAS H. LOWENTHAL, JUDITH C. CHOW, and DAVID M. MAZZERA, Desert Research Institute, Reno, Nevada 89506
The National Science Foundation (NSF) Environmental Research Program "support(s) research that can help reduce the environmental footprint of NSF's activities in Antarctica and establish a baseline for future measurements." PM10 (atmospheric aerosol particles with diameters less than 10 microns) is a regulated air pollutant in the United States. A study was undertaken to determine the natural and anthropogenic sources of PM10 at McMurdo Station (77°51'S 166°41'E), whose austral summer population exceeds 1,000 and whose reported summer diesel fuel consumption (Antarctic Support Associates 1994) for power generation, heating and water production, and equipment operation is 6.1 million liters. An understanding of the nature and sources of atmospheric particulates at McMurdo Station will provide a basis for evaluating their effects on the antarctic environment.
Air-sampling was initiated at McMurdo Station during austral summer 1995-1996. Emissions were sampled from the major sources: power generators, space heating, surface vehicles, helicopters, and C-130 aircraft. Local soils from around the station were also collected. PM10 monitoring was begun at Hut Point, located 600 meters (m) northwest and downwind of McMurdo center. Radar Sat, located on a ridge northeast of the station, was used as a background site. Forty-eight-hour-duration PM10 samples were collected continuously at both sites from November through January during the austral summers of 1995-1996 and 1996-1997.
Source and ambient samples were analyzed for particle mass, ion, element, and elemental and organic carbon (EC and OC) concentrations. Source profiles (fractional abundances of each chemical component with respect to emitted mass) were determined for the major sources. Source apportionment is accomplished using the chemical mass balance (CMB) model (Watson et al. 1990), which compares the source profiles and ambient PM10 chemical concentrations. The results are source contributions to PM10 mass and its chemical components.
Average concentrations (µg/m3, micrograms per cubic meter of air) of PM10 and selected chemical species at Hut Point and Radar Sat are given in the table for austral summer 1995-1996. Samples affected by emissions from ships docked at the ice pier near Hut Point exhibited much higher EC and OC concentrations than those not so affected. The average EC concentration at Hut Point (no ships) (0.171 µg/m3) was 100-200 times higher than that reported for the South Pole during austral summer (Bodhaine 1995). EC concentrations at Radar Sat were roughly 3 times lower than those at Hut Point, suggesting local contamination, perhaps by motor vehicles working in the area or visiting the site. The higher silicon concentrations (indicative of resuspended geological dust) at Radar Sat may also reflect disturbance of the exposed ground surface by vehicles. Note that average PM10 mass concentrations (excluding ships at Hut Point) were more than a factor of 10 lower than the U.S. Environmental Protection Agency annual average PM10 standard of 50 µg/m3.
Source contributions to PM10 at Hut Point were estimated for samples unaffected by ships. The average source contributions are shown in figure 1. Because most nitrate and sulfate in the atmosphere is secondary (formed by gas-to-particle conversion during transport from sources), source profiles representing pure ammonium nitrate, ammonium sulfate, ammonium bisulfate, and sulfuric acid were included in the CMB. Figure 1 indicates that the largest PM10 component was geological dust (46 percent). This material probably would not be present to the same extent in the absence of human activity. The next largest source is sea salt (21 percent), originating from nearby McMurdo Sound and the open ocean some 40 kilometers (km) north of McMurdo. Combustion sources accounted for 0.51 µg/m3, or 17 percent.
Ammonium nitrate accounts for only 2 percent of PM10. It is likely, however, that nitrate at McMurdo is associated with large sea salt particles and not ammonium (Mamane and Gottlieb 1992). Nitrate concentrations at McMurdo (table) were similar to levels (0.043 µg/m3) measured at coastal Mawson, Antarctica, during austral summer (Savoie et al. 1992), suggesting that McMurdo nitrogen oxide emissions had not been converted to particle nitrate in the short distance between McMurdo and Hut Point.
Secondary sulfate is a significant component of PM10 (15 percent). This amount represents sulfate not associated with geological dust, sea salt, or primary (emitted directly as particles) combustion emissions. We refer to it as "excess" sulfate (xSO4). It appears to be acidic, e.g., sulfuric acid and ammonium bisulfate. The equivalent concentration of xSO4 ion is 0.40 µg/m3. This quantity is higher than non-sea salt sulfate (NSS) concentrations (0.21 µg/m3) reported for Mawson during austral summer (Savoie et al. 1992) but lower than summer NSS concentrations (0.65 µg/m3) measured near the northern tip of Ross Island (Wylie et al. 1993). McMurdo sulfur dioxide emissions have apparently not been converted to sulfate en route to Hut Point.
Some sulfate in the marine boundary layer has a biogenic origin. Dimethylsulfide emitted by phytoplankton oxidizes to sulfate and methanesulfonate (MSA). The ratio of MSA to NSS shows a latitudinal dependence that is related to air temperature (Bates, Calhoun, and Quinn 1992). It has been hypothesized that variations in the MSA/NSS ratio in antarctic snow and ice are related to large-scale climatic and meteorological factors (Legrand and Feniet-Saigne 1991).
Based on the relationship described by Bates et al. (1992) (MSA/NSS (molar ratio) = -0.015(T°C)+0.422), the expected ratio at McMurdo, with a summer average temperature of -2.4°C, is 0.46. The average MSA/average xSO4 molar ratio at Hut Point was 0.28. Summer ratios in aerosol at Mawson (Savoie et al. 1992) and on the east antarctic plateau (De Mora, Wylie, and Dick 1997) were 0.18 and 0.20, respectively. The ratio in east antarctic plateau surface snow was 0.08 (De Mora et al. 1997). Low ratios on the plateau may be related to transport from low latitudes and/or to loss of larger particle MSA during transport.
Excess sulfate at Hut Point is not necessarily biogenic. Figure 2 shows the relationship between MSA and xSO4 at Hut Point (no ships). The regression equation in figure 2 implies an MSA/biogenic sulfate ratio of 0.56 and a nonbiogenic xSO4 background of 0.2 µg/m3.
Mount Erebus has been discounted as a significant source of sulfur on the antarctic plateau (Delmas 1992). Mount Erebus may be a significant source of sulfur in the Ross Island area, where down-slope flow from the volcano may enhance surface sulfate aerosol concentrations and may explain why NSS and xSO4 concentrations are higher on Ross Island than at Mawson. Figure 3 presents a sulfate source budget for Hut Point derived from the above analyses. Sea salt and resuspended geological dust together account for 25 percent. Combustion and biogenic emissions account for 13 and 37 percent, respectively. The remaining "other" 37 percent may represent a hemispheric background and/or emissions from Mount Erebus.
It is clear from the elevated elemental carbon concentrations at Hut Point that McMurdo has at least perturbed the atmosphere around the station. The potential impact of associated air pollutants such as polycyclic aromatic hydrocarbons and oxides of nitrogen and sulfur on downwind ecosystems is of great interest. As part of this study, meteorological modeling will be used to estimate the areal extent of the McMurdo plume.
This research was supported by National Science Foundation grant OPP 94-17829. We thank John Watson and William Dippel for their field support during the first year.
Antarctic Support Associates. 1994. Master permit application. Prepared by Consoer Townsend & Associates AECOM Technology Corporation, 2750 Prosperity Avenue, Suite 230, Fairfax, Virginia 22031 for Antarctic Support Associates, 61 Inverness Drive, Suite 300, Englewood, Colorado 80112.
Bates, T.S., J.A. Calhoun, and P.K. Quinn. 1992. Variations in the methanesulfonate to sulfate molar ratio in submicrometer marine aerosol particles over the South Pacific ocean. Journal of Geophysical Research , 97, 9859-9865.
Bodhaine, B.A. 1995. Aerosol absorption measurements at Barrow, Mauna Loa and the South Pole. Journal of Geophysical Research , 100, 8967-8975.
Delmas, R.J. 1992. Environmental information from ice cores. Reviews of Geophysics , 30, 1-21.
De Mora, S.J., D.J. Wylie, and A.L. Dick. 1997. Methanesulphonate and non-sea salt sulphate in aerosol, snow, and ice on the east antarctic plateau. Antarctic Science , 9, 46-55.
Legrand, M., and C. Feniet-Saigne. 1991. Methanesulfonic acid in South Polar snow layers: A record of strong El Niño? Geophysical Research Letters , 18, 187-190.
Mamane, Y., and J. Gottlieb. 1992. Nitrate formation on sea-salt and mineral particles—A single particle approach. Atmospheric Environment , 26A, 1763-1769.
Savoie, D.L., J.M. Prospero, R.J. Larsen, and E.S. Saltzman. 1992. Nitrogen and sulfur species in aerosols at Mawson, Antarctica, and their relationship to natural radionuclides. Journal of Atmospheric Chemistry , 14, 181-204.
Watson, J.G., N.F. Robinson, J.C. Chow, R.C. Henry, B.M. Kim, T.G. Pace, E.L. Meyer, and Q. Nguyen. 1990. The USEPA/DRI chemical mass balance receptor model, CMB 7.0. Environmental Software , 5(1), 38-49.
Wylie, D.J., M.J. Harvey, S.J. DeMora, I.S. Boyd, and J.B. Liley. 1993. Dimethylsulfide measurements at Ross Island, Antarctica. In G. Restelli and G. Angeletti (Eds.), Dimethylsulfide, oceans, atmosphere and climate . Dordrecht: Kluwer Academic Publishers.