Articles | Volume 12, issue 4
Atmos. Chem. Phys., 12, 1903–1922, 2012
Atmos. Chem. Phys., 12, 1903–1922, 2012

Research article 17 Feb 2012

Research article | 17 Feb 2012

Source-receptor relationships for speciated atmospheric mercury at the remote Experimental Lakes Area, northwestern Ontario, Canada

I. Cheng1, L. Zhang2, P. Blanchard2, J. A. Graydon3, and V. L. St. Louis3 I. Cheng et al.
  • 1Independent researcher, 5785 Yonge St, Toronto, Ontario, M2M 4J2, Canada
  • 2Air Quality Research Division, Science and Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto, Ontario, M3H 5T4, Canada
  • 3Department of Biological Sciences, University of Alberta, Edmonton, Alberta, T6G 2E9, Canada

Abstract. Source-receptor relationships for speciated atmospheric mercury measured at the Experimental Lakes Area (ELA), northwestern Ontario, Canada were investigated using various receptor-based approaches. The data used in this study include gaseous elemental mercury (GEM), mercury bound to fine airborne particles (<2.5 μm) (PHg), reactive gaseous mercury (RGM), major inorganic ions, sulphur dioxide, nitric acid gas, ozone, and meteorological variables, all of which were measured between May 2005 and December 2006. The source origins identified were related to transport of industrial and combustion emissions (associated with elevated GEM), photochemical production of RGM (associated with elevated RGM), road-salt particles with absorption of gaseous Hg (associated with elevated PHg and RGM), crustal/soil emissions, and background pollution. Back trajectory modelling illustrated that a remote site, like ELA, is affected by distant Hg point sources in Canada and the United States. The sources identified from correlation analysis, principal components analysis and K-means cluster analysis were generally consistent. The discrepancies between the K-means and Hierarchical cluster analysis were the clusters related to transport of industrial/combustion emissions, photochemical production of RGM, and crustal/soil emissions. Although it was possible to assign the clusters to these source origins, the trajectory plots for the Hierarchical clusters were similar to some of the trajectories belonging to several K-means clusters. This likely occurred because the variables indicative of transport of industrial/combustion emissions were elevated in at least two or more of the clusters, which means this Hg source was well-represented in the data.

Final-revised paper