Dry deposition of speciated mercury, i.e., gaseous oxidized mercury (GOM), particulate-bound mercury (PBM), and gaseous elemental mercury (GEM), was estimated for the year 2008–2009 at 19 monitoring locations in eastern and central North America. Dry deposition estimates were obtained by combining monitored two- to four-hourly speciated ambient concentrations with modeled hourly dry deposition velocities (<i>V</i>_d) calculated using forecasted meteorology. Annual dry deposition of GOM+PBM was estimated to be in the range of 0.4 to 8.1 μg m⁻² at these locations with GOM deposition being mostly five to ten times higher than PBM deposition, due to their different modeled <i>V</i>_d values. Net annual GEM dry deposition was estimated to be in the range of 5 to 26 μg m⁻² at 18 sites and 33 μg m⁻² at one site. The estimated dry deposition agrees very well with limited surrogate-surface dry deposition measurements of GOM and PBM, and also agrees with litterfall mercury measurements conducted at multiple locations in eastern and central North America. This study suggests that GEM contributes much more than GOM+PBM to the total dry deposition at the majority of the sites considered here; the only exception is at locations close to significant point sources where GEM and GOM+PBM contribute equally to the total dry deposition. The relative magnitude of the speciated dry deposition and their good comparisons with litterfall deposition suggest that mercury in litterfall originates primarily from GEM, which is consistent with the limited number of previous field studies. The study also supports previous analyses suggesting that total dry deposition of mercury is equal to, if not more important than, wet deposition of mercury on a regional scale in eastern North America

Abstract. We have developed a new nested-grid mercury (Hg) simulation over North America with a 1/2 • latitude by 2/3 • longitude horizontal resolution employing the GEOSChem global chemical transport model. Emissions, chemistry, deposition, and meteorology are self-consistent between the global and nested domains. Compared to the global model (4 • latitude by 5 • longitude), the nested model shows improved skill at capturing the high spatial and temporal variability of Hg wet deposition over North America observed by the Mercury Deposition Network (MDN) in [2008][2009]. The nested simulation resolves features such as higher deposition due to orographic precipitation, land/ocean contrast and and predicts more efficient convective rain scavenging of Hg over the southeast United States. However, the nested model overestimates Hg wet deposition over the Ohio River Valley region (ORV) by 27 %. We modify anthropogenic emission speciation profiles in the US EPA National Emission Inventory (NEI) to account for the rapid inplume reduction of reactive to elemental Hg (IPR simulation). This leads to a decrease in the model bias to −2.3 % over the ORV region. Over the contiguous US, the correlation coefficient (r) between MDN observations and our IPR simulation increases from 0.60 to 0.78. The IPR nested simulation generally reproduces the seasonal cycle in surface concentrations of speciated Hg from the Atmospheric Mercury Network (AMNet) and Canadian Atmospheric Mercury Network (CAMNet). In the IPR simulation, annual mean gaseous and particulate-bound Hg(II) are within 140 % and 11 % of observations, respectively. In contrast, the simulation with unmodified anthropogenic Hg speciation profiles overestimates these observations by factors of 4 and 2 for gaseous and particulate-bound Hg(II), respectively. The nested model shows improved skill at capturing the horizontal variability of Hg observed over California during the ARCTAS aircraft campaign. The nested model suggests that North American anthropogenic emissions account for 10-22 % of Hg wet deposition flux over the US, depending on the anthropogenic emissions speciation profile assumed. The modeled percent contribution can be as high as 60 % near large point sources in ORV. Our results indicate that the North American anthropogenic contribution to dry deposition is 13-20 %.

We have developed a new high-resolution (1/2° latitude by 2/3° longitude) nested-grid mercury (Hg) simulation over North America employing the GEOS-Chem global chemical transport model. Emissions, chemistry, deposition, and meteorology are self-consistent between the global and nested domains. Compared to the global model (4° latitude by 5° longitude), the nested model shows improved skill at capturing the high spatial and temporal variability of Hg wet deposition over North America observed by the Mercury Deposition Network (MDN) in 2008–2009. The nested simulation resolves features such as land/ocean contrast and higher deposition due to orographic precipitation, and predicts more efficient convective rain scavenging of Hg over the southeast United States. However, the nested model overestimates Hg wet deposition over the Ohio River Valley region (ORV) by 27%. We modify anthropogenic emission speciation profiles in the US EPA National Emission Inventory (NEI) to account for the rapid in-plume reduction of reactive to elemental Hg (IPR simulation). This leads to a decrease in the model bias to +3% over the ORV region. Over the contiguous US, the correlation coefficient (<i>r</i>) between MDN observations and our IPR simulation increases from 0.63 to 0.78. The IPR nested simulation generally reproduces the seasonal cycle in surface concentrations of speciated Hg from the Atmospheric Mercury Network (AMNet) and Canadian Atmospheric Mercury Network (CAMNet). In the IPR simulation, annual mean reactive gaseous and particulate-bound Hg are within 80% and 10% of observations, respectively. In contrast, the simulation with unmodified anthropogenic Hg speciation profiles overestimates these observations by factors of 2 to 4. The nested model shows improved skill at capturing the horizontal variability of Hg observed over California during the ARCTAS aircraft campaign. We find that North American anthropogenic emissions account for 10–22% of Hg wet deposition flux over the US, depending on the anthropogenic emissions speciation profile assumed. The percent contribution can be as high as 60% near large point emission sources in ORV. The contribution for the dry deposition is 13–20%

Dry deposition of speciated mercury, i.e., gaseous oxidized mercury (GOM), particulate bound mercury (PBM), and gaseous elemental mercury (GEM), was estimated for the year 2008–2009 at 19 monitoring locations in Eastern and Central North America. Dry deposition estimates were obtained by combining monitored 2–4 hourly speciated ambient concentration with modeled hourly dry deposition velocities (<i>V</i>_d) calculated using forecasted meteorology. Annual dry deposition of GOM + PBM was estimated to be in the range of 0.4 to 8.1 μg m⁻² at these locations with GOM deposition being mostly 5 to 10 times higher than PBM deposition, due to their different <i>V</i>_d values. Net annual GEM dry deposition was estimated to be in the range of 5 to 26 μg m⁻² at 18 sites and 33 μg m⁻² at one site. The estimated dry deposition agrees very well with limited surrogate-surface dry deposition measurements of GOM and PBM, and also agrees with litterfall mercury measurements conducted at multiple locations in Eastern and Central North America. This study suggests that GEM contributes much more than GOM + PBM to the total dry deposition at the majority of sites considered here; the only exception is at locations close to significant point sources where GEM and GOM + PBM contribute equally to the total dry deposition. The relative magnitude of the speciated dry deposition and their good comparison with litterfall deposition suggest that mercury in litterfall primarily originates from GEM, consistent with previous limited field studies. The study also supports previous analyses suggesting that total dry deposition of mercury is equally if not more important as wet deposition of mercury on a regional scale in Eastern North America

Two and a half years of data of ambient concentrations of elemental mercury (Hg(0)), reactive gaseous mercury (RGM), and particle-bound mercury (Hg(p)) were collected at measurement sites at Elizabeth, New Jersey and New Brunswick, New Jersey with Tekran sampling units. The data were processed, summarized, and analyzed from a variety of perspectives. Data quality control and quality assurance procedures are described. Wind direction and wind speed data for each of the sites were also collected. Significant temporal variations in concentrations of all three species were observed. Some significant directional variations were also seen. The sporadic nature of many of the temporal variations is consistent with and could reflect highly variable emissions patterns from anthropogenic mercury sources. Overall mean concentrations of all species were determined. These were, for Hg(0), Hg(p), and RGM respectively; 2.25 +/- 0.04 nanograms per cubic meter (ng/m(3)), 8.21 +/- 0.39 picograms per cubic meter (pg/m(3)), and 8.93 +/- 0.31 pg/m(3) (arithmetic means and 95% confidence intervals) at Elizabeth, and 2.15 +/- 0.02 ng/m(3), 10.73 +/- 0.45 pg/m(3), and 6.04 +/- 0.30 pg/m(3) at New Brunswick. Mean concentrations were determined for 16 different sectors representing wind directions. The impact of one known large source is suggested by these data. Reasons for some directional variations are not apparent and suggest a need for further investigation.