Importance of a global scale approach to using regional models in the assessment of source-receptor relationships for mercury

Regional atmospheric models simulate their pertinent processes over a limited portion of the global atmosphere. This portion of the atmosphere can be a large fraction, as in the case of continental-scale modeling, or a small fraction, as in the case of urban-scale modeling. Regional modeling of any air pollutant requires that the meteorological and chemical conditions at the boundaries of the model domain be taken into account, especially if the pollutants are long-lived. It was once a common practice for the boundary concentrations of mercury and its reactants to be specified using time-constant values based on limited historical observation. These values were often invariant in the horizontal and vertical dimensions too. This relatively simple procedure for establishing boundary concentrations could be justified based on the previous notion that mer-cury was a rather inert and long-lasting air pollutant. However, with the subsequent discovery of rapid physical and chemical transformations of atmospheric mercury and significant concentra-tions of oxidized mercury far removed from known emission sources, the presumption of mer-cury as an inert substance has generally disappeared. The effect of intercontinental transport is now treated with greater concern in regional atmospheric mercury modeling. Global mercury models are now commonly used to define boundary values for regional mercury modeling. How-ever, the global and regional models must use consistent information for emissions, surface physiology and meteorology to achieve consistent simulation results and associated source-receptor relationships. There is certainly a need for international InstitutionalAuthorNameoration on field research and numerical model development to supply the tools needed for confident assessment of source-receptor relationships for mercury on both global and regional scales.