Investigation of atmospheric chemistry of gaseous oxidized mercury at a coastal site in Atlantic Canada
Irene Cheng1, Leiming Zhang2, Huiting Mao3, Zhuyun Ye4, Robert Keenan5 and Rob Tordon6
Atmospheric mercury chemistry is an integral part of the environmental cycling of mercury; however the understanding of the chemical reaction mechanisms and kinetics remains incomplete. Atmospheric chemistry drives the cycling of mercury by converting gaseous elemental mercury (GEM) to speciated mercury, including gaseous oxidized mercury (GOM) and particle-bound mercury (PBM), which enters terrestrial and aquatic ecosystems through atmospheric deposition. In this study, a box model containing the most up-to-date gaseous and aqueous chemical reactions involving mercury, bromine, chlorine, iodine, and ozone was used to simulate the formation and dry deposition of GOM at the Kejimkujik National Park, Nova Scotia, Canada. The measurements available as initial input to the box model included GEM, O3, NO2, solar radiation, and temperature. The modeled results were evaluated through comparisons with GOM measurements at this site. For selected dry, clear sky days from 2009 to 2016, the mean and standard deviation of the GOM concentration was 1.3 ± 2.0 pg m-3 in the observed data and 1.8 ± 1.5 pg m-3 in the modeled results. The model was able to reproduce not only the observed GOM concentrations, but also the spring/summer variation and land/marine differences in GOM. The normalized mean bias of the model was +42% for GOM and -94% for PBM, indicating that the model overestimated the observed GOM and significantly underestimated the observed PBM. The model-measurement discrepancies in PBM suggest that gas-particle partitioning may not be accurately represented given a lack of particle size-resolved PBM data and uncertainties in the aqueous chemistry. PBM could also be originating from natural sources that have not been included in the box model. Thus, we find that the box model is most suitable for simulating GOM. Based on the GOM species from the model output, GOM was predominantly formed by GEM oxidation by O3 and OH (74%), H2O2 (17%), Br with NO2 in the second reaction step (6%), BrO (2%) and other oxidants (<1%). Future work will include a detailed comparison of model and measurement differences by stratifying the data into different seasons and transport patterns to examine how these factors affect atmospheric mercury oxidation.
1Environment and Climate Change Canada, irene.cheng@canada.ca 2Environment and Climate Change Canada, leiming.zhang@canada.ca 3State University of New York, hmao@esf.edu 4State University of New York, zye01@syr.edu 5Environment and Climate Change Canada, robert.keenan@canada.ca 6Environment and Climate Change Canada, rob.tordon@canada.ca