1a-3: Atmospheric mercury cycling and transformations: Insights from measurements and models

Microorganisms play a fundamental role in the mercury cycle on a global scale. In both terrestrial and aquatic environments, microorganisms modify Hg speciation and thereby change the contamination potential of Hg. Bacteria are able to magnify the toxicity of this compound via the methylation of inorganic forms of Hg (Jensen and Jernelv, Nature 1969) and they can act as detoxifiers through the reduction of divalent Hg species with the mer operon (Hamlett et al., J. Bacteriol 1992). The atmosphere is an important route for transporting Hg over long distances and is also where Hg is converted into soluble, labile species that are likely to interact with microorganisms. Yet, atmospheric studies tend to disregard the role of airborne microorganisms in the transformation of Hg species. Only recently have studies shown that up to 105 microbial cells can inhabit one cubic meter of air and that they might be metabolically active in the atmosphere (Klein et al., Front Microbiol 2016). Given the potential for microbial activity in the atmosphere coupled with the different forms of mercury, microorganisms might be transforming mercury in the atmosphere. INHALE (Investigation of tHe Atmosphere as a reaL Ecosystem, ANR) is one of the first projects to assess the Hg-microbial interaction in the atmosphere and to determine the potential role of airborne microorganisms in altering the speciation of Hg and its contamination potential at a global scale. We investigated air samples taken from several high altitude sites around the world involved in Hg geochemical cycling investigations. Hg cycling gene quantification was carried out following atmospheric DNA extraction from filters to assess the presence and abundance of these genes in the atmosphere. The global distribution of the genetic determinants involved in biological Hg cycling in atmospheric microbial communities was then determined. Based on these preliminary results, we proposed a potential role for airborne microorganisms in Hg transformations in the atmosphere. This atmospheric microbial activity may explain changes in Hg speciation and toxicity and help us to understand its potential impact on ecosystem and human health, and to improve global Hg cycling models.

Vast quantities of gaseous elemental mercury (GEM) are released to the atmosphere by fossil fuel combustion and waste incineration. The oxidation and precipitation of atmospheric GEM are principal processes governing the transfer of mercury to the ocean and soil environments. Molecular mechanisms of GEM transformation to gaseous oxidized mercury (GOM) and subsequent removal of GOM by surfaces are poorly understood. On the basis of recent experimental, field, and modeling studies, the reaction of GEM with atomic bromine has been suggested as the first step in GOM formation. We will present our work focused on the development of a flow reactor chemical ionization mass spectrometry approach for investigation of gas-phase and gas-surface reaction mechanisms of GEM and GOM. The implications of our work towards understanding of complete mercury oxidation mechanism and achieving chemically-resolved detection GOM in the atmosphere will be discussed.

The NOAA HYSPLIT-Hg model has been applied to simulate the fate and transport of mercury emitted to the atmosphere using Lagrangian (i.e., plume-oriented), Eulerian (i.e., gridded), and hybrid Lagrangian/Eulerian simulation methodologies. The hybrid approach has been implemented in several ways, but essentially utilizes a Lagrangian methodology to simulate local and regional fate/transport and an Eulerian framework at further distances from a given source. In each case, the model has been used to estimate: (a) the wet and dry deposition to selected receptors (e.g., the Laurentian Great Lakes), and (b) source-attribution for this deposition, including the relative importance of anthropogenic vs. other emissions sources, and among anthropogenic sources, the relative importance of different countrys emissions. In addition, concentrations and deposition at selected measurement sites ‒ e.g., Beltsville (Maryland, USA), Grand Bay (Mississippi, USA), and Mauna Loa (Hawaii, USA) ‒ have been estimated by the model and compared with observations at the sites. There are numerous computational and other tradeoffs among the different approaches. The different computational fate and transport frameworks can, in principle, yield equivalent results, if each was able to accurately capture all relevant phenomena. In practice, each has a number of limitations and inherent uncertainties. Further, uncertainties in emissions, atmospheric chemistry, and deposition processes can manifest differently in modeling using the different frameworks. All of these factors lead to variations in simulation results. Differences in deposition and source attribution results for key receptors are sometimes found among the different methodologies, as well as variations in the estimated concentrations and deposition at measurement sites. The relative importance of these discrepancies in different cases will be presented and discussed, along with estimates of the relative contributions of different computational factors to the dissimilarities found. These results provide useful insights into uncertainties in atmospheric mercury modeling due to the characterization of dispersion phenomena. Where the modeling results are compared with ambient measurements, the results show the relative ability of the different approaches -- in the cases and conditions studied to reproduce spatial, temporal, and chemical variations in observed mercury concentrations and deposition.

Goodsite et al. (2004) published the first mechanism for the Br-initiated oxidation of gaseous elemental mercury (Hg(0), aka GEM). Their mechanism considered BrHg radical reacting with OH or Br to form HgBr2 and BrHgOH as stable Hg(II) species (gaseous oxidized mercury, GOM). In 2012, Dibble et al. used quantum chemistry to verify the stability of a number of BrHgY species (Y=NO2, HOO, BrO, ClO). We noted that these radicals, Y, should react with BrHg with rate constants similar to those of BrHg + Br. Since these radicals, Y, are much more abundant than Br or OH, reaction of BrHg with Y should dominate over reaction with Br and OH throughout most of the atmosphere. Recent kinetic studies in our group confirm the high rate constant for BrHg reacting with NO2 and HOO to make BrHgONO and BrHgOOH.

Unfortunately for modelers, there is a great deal of additional chemistry that will likely need to be added to mechanisms of Br-initiated oxidation of mercury:

a) Near ground level, except in polar regions, BrHg tends to fall apart to Br + Hg before it can react with radicals. However, we find that BrHg may react with alkenes. As alkenes are much more abundant than radicals, reaction of BrHg with alkenes could enable efficient conversion of GEM to GOM. This would be of particular importance in the marine boundary layer influenced by continental air. We are computing rate constants for the reactions of BrHg with alkenes and will investigate the fate of the resulting radicals.

b) Recent findings suggest that BrHgONO will rapidly (~1 hour) photolyze to form BrHgO. Our results to date indicate that BrHgO will abstract hydrogen atoms from organic compounds in the gaseous atmosphere to form BrHgOH. Ironically, this was the major product of GEM oxidation in models based on the work of Goodsite et al. Competing reactions of BrHgO can lead to BrHgONO2 or regenerate BrHgONO.

These results point to significant interactions of oxidation pathways of atmospheric mercury with trace organic compounds. We hope that these results are incorporated into models and support the development of methods to identify Hg(II) compounds in the atmosphere.

Wet deposition of mercury (Hg) at an unpolluted site in windward northeastern Puerto Rico is comparable to the highest levels in the continental USA. Here we update the existing seven-year record after two years of measurements at a relocated station, provide estimates of dry Hg deposition based on throughfall and litterfall, and quantify methylmercury (MeHg) in wet and dry deposition. In 2006-2007 (the most complete record), annual wet Hg deposition averaged 27.9 g m-2. High deposition is attributed to scouring of global pool Hg from the upper free troposphere by rain near the tops of high convective clouds, as well as high rainfall amounts (2855 mm y-1). After years of logistical difficulty operating the station at a remote ridgetop tower, we relocated the station 11 km west to a NADP site in August 2014. This station has less rainfall due to its lower elevation (360 vs. 480 m) and the measurements coincided with a significant drought in 2015. Wet Hg deposition in calendar year 2015 was 14.3 g m-2 with 1777 mm rainfall. Weighted average Hg concentration decreased from 9.8 ng L-1 at the original site to 8.0 ng L-1 at the new site, perhaps because the drought was most intense during the summer months when rainfall Hg concentrations are typically highest. Preliminary wet and dry deposition measurements did not overlap well in time and space, but indicate that dry Hg deposition is important. For calendar year 2014, litterfall Hg deposition (representing primarily dry deposition of Hg0) at the NADP site was 38.0 g m-2. Throughfall Hg deposition (representing primarily dry deposition of Hg2+) for 9 months of overlapping measurements at the original site in 2006-2007 was 54% greater than wet Hg deposition. The percentage total Hg as MeHg was 0.06 % in rainfall, 0.56% in throughfall, and 0.31% in litterfall. Combined, these measurements suggest that dry Hg deposition is considerably greater than wet, and that relative input of (or conversion to) MeHg in the forest canopy is considerably less than published values.

Biomass Burning (BB) is an important source of mercury (Hg) in the atmosphere. A largest fraction of this Hg is released in the form of elemental Hg, however there are many experimental evidences that an important fraction (up to the 30%) is released bounded to particulate matter (PBM). The exact mechanisms and factors that control emission and speciation are highly uncertain. Indeed, Hg speciation is one of the most important factors affecting Hg deposition and the ratio local/non-local contribution. This work was aimed to investigate the speciation process during BB and its impact on Hg deposition by using the Global Fire Emissions Database (GFEDv4.1s) included into a global Hg chemistry transport model. Results showed that quantity and the geographical distribution of Hg species emitted by BB has a limited impact on a global scale. On the other hand, this impact is remarkable at local scale on those ecosystems close to the BB process. One of the most important consequence is the reduction of the Hg fraction amount deriving from BB, which deposits to the worlds oceans (from 71% to 62%). In addition, results of the model were compared with measurement made within GMOS network. Inclusion into the model of PBM released by BB showed a better agreement between modeled and observed PBM concentrations at remote sites. In light of the Minamata Convention and the progressive decline of Hg emissions from anthropogenic activities, this work shows the growing impact of BB contribution on Hg deposition.

East Asia is the largest emission region of anthropogenic mercury (Hg) in the world, with China as the biggest emitter. There have been concerns regarding the transport budget of mercury in the region. Earlier assessments of transport budget are based on relatively outdated emission estimates, which does not represent the regional emission appropriately. Since then, new estimates on anthropogenic (Wu et al., 2016, doi: 10.1021/acs.est.6b04308) and natural emissions (Wang et al., 2016, ACP, doi:10.5194/acp-16-11125-2016) in the region show substantial different from earlier estimate. Most notably, anthropogenic Hg emission in China gradually decreased since 2011 due to better emission control, with a much larger fraction of oxidized mercury than previously thought (56/43/3 for Hg0/HgII/Hgp in 2014). The natural release of elemental mercury vapor from soil, vegetation and water surfaces using new soil Hg data in China and updated model schemes shows a distinct spatial distribution of estimated mercury release compared to the previous estimate (Shetty et al., 2008, doi:10.1016/j.atmosenv.2008.08.026), despite a similar net natural release at ~460 Mg y-1 in China. Such a spatial distribution transition also has an impact on regional model results. In this study, we applied the updated Hg emission estimates to reassess the regional transport budget using CMAQ-Hg v5.1. The emission differences are compared and the model results detailing the emission, deposition and air Hg enrichment are presented. Total Hg deposition in East Asia is 746 Mg yr-1 (422 Mg yr-1 in China). Given the changes in anthropogenic emission, speciation and natural emissions spatial distribution, the transport budget in the East Asia region is 25% lower (631 Mg yr-1) than the previous estimate by Lin et al. (835 Mg yr-1, doi:10.5194/acp-10-1853-2010). The greater Hg mass accumulated within the regional domain also better explains the elevated atmospheric Hg concentrations observed in China recently. At least 60% of Hg deposition in China is caused by local anthropogenic emissions. The overall transport budget in China is 505 Mg yr-1, 55% of which is contributed by natural Hg emission. The Hg emission outflow from East Asia can contribute to 10-15% of Hg deposition in other regions of the world.

Continuous event-based rainfall samples were collected at three sites throughout the Pensacola airshed from 2005 - 2011. Samples were analyzed for total mercury (Hg), a suite of trace metals (TMs), and major ions to understand atmospheric Hg transport and estimate the contribution from regional coal-combustion on Hg wet deposition. Our findings show that summertime Hg rain concentrations are higher compared to other months despite higher rainfall amounts. In contrast, other measured pollutant TMs or ions did not show a consistent seasonal pattern. By incorporating Automated Surface Observing System data from nearby Pensacola Airport and WSR-88D (Nexrad) data from Eglin Air Force Base, we are able to classify the storm type (thunderstorms or non-thunderstorms) and analyze altitudes of hydrometeor formation for individual rain events. This showed that mid-altitude and high-altitude composite reflectivity Nexrad values were higher for both thunderstorm and non-thunderstorm warm season (May Sept) rain events compared to cool season (Oct Apr) events including cool season thunderstorms. Thus, warm season events can scavenge more soluble reactive gaseous Hg from the free troposphere. Two separate multiple linear regression analyses were conducted on log-transformed data using interaction and non-interaction terms to understand the relationship between precipitation depth, season, and storm-type on sample concentrations. The regressions without interaction terms showed that the washout coefficients for more volatile TMs like Hg and selenium were less pronounced compared to other pollution-type elements and were therefore less diluted for a given increase in precipitation depth, but otherwise displayed similar coefficients for season and storm-type. The regression analysis with interaction terms revealed a more interesting dynamic where thunderstorms caused enhanced rainfall Hg concentrations regardless of season while causing increased dilution for all other TMs compared to non-thunderstorm samples for a given precipitation depth. However, concentrations increased during thunderstorms for all non-Hg TMs with increasing sample depth. This suggests a vacuum cleaner effect was occurring such that for increasing storm strength, non-Hg aerosol TMs in the boundary layer are further entrained into a storm cell. With this understanding, a positive matrix factorization (PMF) analysis was conducted using the EPA PMF 5.0 software to estimate the contribution of different sources to Hg deposition. Preliminary results suggest that approximately 81% (71 92%; 95% CI) of Hg wet deposition along the northern Gulf of Mexico is due to long-range sources while 17% (0 18%; 95% CI) comes from regional coal-combustion.