Science implementation of a comprehensive mercury model in CMAQ framework to assess mercury fate in the atmosphere

The Community Multi-scale Air Quality (CMAQ) modeling system was designed to couple multiple atmospheric processes for "one atmosphere" modeling. It was also designed to have multi-scale capabilities so that separate models were not needed for urban and regional scale air quality modeling. Therefore, CMAQ is a useful tool to assess the transport and deposition of mercury. USEPA developed the mercury model coupled in CMAQ (CMAQ-Hg) for modeling of various atmospheric mercury species including GEM, RGM, and PHg. However, there exist issues caused by model assumptions/simplifications and incomplete understanding of mercury science. This research evaluates the model science commonly implemented in atmospheric mercury models, tests potential chemical mechanisms, develops and demonstrates an alternative dry deposition treatment of mercury species. I assessed the causes of the model uncertainties in terms of gas phase chemistry (oxidation rates, products, and potential reduction mechanism of mercury species), aqueous phase chemistry, aqueous phase speciation, aqueous phase sorption, dry deposition, wet deposition, initial and boundary conditions, emission inventory, domain grid resolution, and intercontinental mercury transport. The researcher selected the year 2001 for simulations. The study area comprises of contiguous Unites States (CONUS) domain, intercontinental transport, and climatic effects of air pollutants (ICAP) domain. The author utilized CONUS domain for the model uncertainty evaluation and testing, while the ICAP domain is used to study intercontinental mercury transport and its impact on the United States. The results show that assignment of mercury oxidation product is very sensitive to the concentration and depositions of various mercury species. More laboratory and field data regarding the oxidation speciation are needed to reduce model uncertainties. Aqueous Hg(II) reduction by HO 2· is the key reduction pathway for the current model configuration, although its occurrence in atmospheric water was questioned. In a sensitivity simulation for the July, 2001, the implemented rate constant needs to be reduced by an order of magnitude to offset the model overestimation of wet Hg deposition compared to the MDN observed data. In cold months CMAQ-Hg overestimates the wet deposition by 500%. From my calculation, this can be caused by the high Henry's law coefficient of HgCl2 at lower temperature. Because of the controversy of the aqueous HO2· or O2·- reduction pathway, I tested the two alternative reduction mechanisms, i.e., RGM photoreduction and RGM reduction by CO. In the sensitivity simulations, the researcher removed the HO 2· reduction and assigned the gas-phase mercury oxidation products by OH· and O3 to be 100% Hg(II). I found that a CO reduction rate of 5 x 10-18 cm 3 molecule-1 s-1 or a RGM photoreduction rate of 1 x 10-5 s-1 gives optimal model agreement to MDN data. When the rate constants are lower than 1 x 10-20 cm3 molecule-1 s-1 and 1 x 10-7 s-1, the model results will not be influenced. The proposed rate constants provide a preliminary estimate for further verification by more kinetic laboratory studies. Model results in the ICAP domain show that chemical forcing of background mercury is the most dominant factor leading to the dry and wet depositions in North America, contributing 65-95% of total mercury deposition at selected MDN sites. Incorporation of natural and/or re-emission of mercury in the emission inventory does not significantly affect mercury deposition (<2% in total deposition), while local anthropogenic emissions can have a significant contribution to the deposition in the U.S. (up to 34%). Although the direct contribution of Asian mercury emission to the dry and wet deposition in North America is lower than expected, its contribution to the global mercury background cannot be overlooked.;Keywords. Aqueous mercury sorption, atmospheric mercury, chemical mechanism, CMAQ, deposition, mercury reduction, mercury speciation, modeling.