A North American inventory of anthropogenic mercury emissions

A source-specific, spatially resolved mercury emission inventory is a key input to an air-quality model that is used to understand relationships between mercury emissions and observed deposition. Previously we simulated the transport and deposition of anthropogenic mercury emissions in the contiguous United States. Due to the long residence time of elemental mercury, i.e., 1 year, it is possible that mercury deposition observed at receptors in the US could be due to long-range transport of emissions across political boundaries. We have recently added mercury emission estimates of anthropogenic sources in Canada and Mexico, to our US inventory. This combined inventory will be used with a regional model to understand source–receptor relationships of atmospheric mercury emissions in North America. Our estimate of 272 Mg of mercury emissions from various anthropogenic sources in North America is comparable to estimates in global inventories (ranging from 240 to 333 Mg). In this paper we discuss the differences in the three regional inventories, i.e., US, Canada, and Mexico, for the year 1990. The dominant source categories of mercury emissions in 1990 in the three regions are different. While combustion of municipal waste and coal accounts for 56% of the 176 Mg mercury emissions in the US in 1990, smelting of nonferrous metals represents the largest source of mercury emissions in Canada (=24 Mg) and Mexico (=31 Mg) in 1990. The main source of uncertainty in the emission estimates is due to the emission factors used in estimation. The emission factors are derived from measurements at a limited number of sources and could potentially not be representative of the operating conditions at other sources for which they are used. In addition, the inventories could still be missing potentially important source categories due to lack of data to develop reliable emission factors. These sources of uncertainties are common features of all three inventories and clearly point to additional research needs.