Use of NOMADSS Observations to Improve Our Understanding of the Land and Ocean Fluxes of Mercury

Abstract:

We use measurements during the 2013 Nitrogen Oxidants Mercury and Aerosol Distributions Sources and Sinks (NOMADSS) campaign to constrain the land and ocean emissions of mercury, using a combination of forward and inverse modeling. Mercury emissions from land and ocean drive the global cycling of mercury. A recent bottom-up assessment of global mercury emissions estimated that the land and ocean sources release about 1500 and 2700 Mg yr^-1 of mercury into the atmosphere, respectively. Thus, the average emission per unit area from the land is thought to exceed the ocean (Pirrone et al., Atmos. Chem. Phys., 10, 5951-5964, 2010). Considering that there are additional mercury sources from the land (e.g., anthropogenic emission and biomass burning), this emission difference should lead to higher observed mercury levels over the land compared to over the ocean. However, several very recent measurement and modeling studies challenge such understanding. The NOMADSS campaign provided a first opportunity to measure speciated mercury concentrations over both land and ocean surfaces. As shown in Table 1, the median Total Mercury (TM) concentration in 0-1 km (within the planetary boundary layer) for over-ocean flights is 0.11 ng m^-3 higher (p < 0.05) than for over-land flights, whereas the over-ocean flights have lower TM concentrations in > 1 km than the over-land flights. We combine analysis of NOMADDS mercury measurements with the GEOS-Chem model, to better constrain land and ocean mercury sources. The standard GEOS-Chem model cannot reproduce the mercury concentration differences between over the land and the ocean. We improve the GEOS-Chem model’s ability to reproduce these observations by implementing updated air-sea and air-land exchange mechanisms. We find that increased ocean emissions are necessary to explain this discrepancy. These results are consistent with the inverse analysis of global GEM concentrations.