research > mercury
Exchange of mercury between the atmosphere and continental surfaces
Figure 1. Drop of liquid mercury
Mercury (Hg), neurotoxic element, is naturally found in the environment. Various human activities, however, increase the atmospheric emissions of this metal for 150 years [1]: coal and oil combustion, mining, waste treatment... The anthropogenic emissions reach the remote areas of the planet. The average atmospheric concentration is higher in the Northern Hemisphere (1.5 ng∙m−3) compared to the Southern Hemisphere (1.3 ng∙m−3).

This metal is liquid at room temperature (Figure 1); it is therefore semi-volatile, and so, facilitating its reemission to the atmosphere. There are various chemical forms (elemental, divalent, particulate, and methyled) and various chemical reactions (redox, adsorption, and methylation). This makes its biogeochemical cycle relatively complex, encouraging many scientific research on its understanding.
Mercury exchanges between the atmosphere and continental surfaces are measured using two different approaches: dynamic chambers and micrometeorlogical methods. We compiled and synthetized measurements found in the literature in a global database [2]. The range showed both emission (positive fluxes) and deposition (negative fluxes) more or less high depending on the class characterized by both atmosphere and soil mercury concentrations (Figure 2). In background sites, bidirectional exchanges showed different trends in the ecosystem: emission of Hg from soils, while vegetation plays a role of sink of Hg. 
Several environmental variables, such as temperature, solar radiation, and humidity, are known to influence Hg exchange in a given study site [3]. Statistic correlations decreased or disappeared when different sites were taken into account. This showed that the inter-site variability offset the controlling factors. Only the soil Hg concentration influence our data set in enriched sites despite the natural variability.
Figure 2. Synthesis of mercury fluxes measured for each class of contamination
Figure 3. Mercury exchange estimates by land cover on different scales
Global-scale estimates are frequently done using parametrizations based on correlations more or less adapted on field measurements. We calculated estimates using all data compiled in the database applying surface area for each land cover on both global and the U.S. scales (Figure 3). Vegetation was included, except for forest environments where whole-ecosystem measurements are complex. We therefore combined both forest floor exchange and vegetation measurements. The last measurements using bag chambers are complex and highly uncertain.

Overall, we estimated an emission of 607 Mg of Hg per year, including 129 Mg∙a−1 for background sites and 478 Mg∙a−1 for enriched ones. This is lower than previous estimates, while global models evaluate a higher contribution from contaminated areas.
In polar environments, Hg dynamic follows to seasonal cycles linked to snow cover and atmospheric chemistry [4]. During springtime, concentrations decrease in the atmospheric boundary layer because of Hg oxidation. Tundra soils are rich in organic matter and play an important role in Hg cycling.

We monitored for two full years Hg, ozone (O3), and nitrogen oxides (NOx) in the atmosphere and gas phase of snow and soil in the Alaskan tundra (Toolik Field StationFigure 4). Complementary chemical analyzes were performed on snow, soil, and vegetation. Preliminary results showed that soil and vegetation play a role of Hg sink, especially during the winter season. Gas transfer is clearly identified from the atmosphere to soil, through the snowpack, without Hg reemission by oxidoreduction reactions as observed in temperate sites.
Figure 4. Landscape of arctic tundra at Toolik Field Station
[1] Lindqvist O. – Atmospheric mercury - a review. Tellus serie B 37: 136–159 (1985) 
[2] Agnan Y. et al.New constraints on terrestrial surface-atmosphere fluxes of gaseous elemental mercury using a global database. Environmental Science & Technology 50 (2): 507–524 (2016) 
[3] Ericksen J. A. et al.Air-soil exchange od mercury from background soils in the United States. Science of the Total Environment 366 (2–3): 851–863 (2006) 
[4] Steffen A. et al.A synthesis of atmospheric mercury depletion event chemistry in the atmosphere and snow. Atmospheric Chemistry and Physics 8: 1445–1482 (2008)