MERCURY IN THE ENVIRONMENT Daniel J. Jacob

Electronic structure of mercury Mass number = 80: 1s2 2s2 2p6 3s2 3p6 3d10 4s2 4p6 4d10 4f14 5s2 5p6 5d10 6s2 Complete filling of subshells gives Hg(0) a low melting point, volatility Two stable oxidation states: Hg(0) and Hg(II)

BIOGEOCHEMICAL CYCLING OF MERCURY ATMOSPHERE Hg (gas) combustion industry mining volcanoes erosion deposition re-emission SOIL OCEAN burial SEDIMENTS DEEP EARTH

RISING MERCURY IN THE ENVIRONMENT Global mercury deposition has roughly tripled since preindustrial times Dietz et al. [2009]

HUMAN EXPOSURE TO MERCURY IS MAINLY FROM FISH CONSUMPTION Tuna is the #1 contributor Mercury biomagnification factor State fish consumption advisories EPA reference dose (RfD) is 0.1 μg kg-1 d-1 (about 2 fish meals per week)

MERCURY CYCLING INVOLVES CHEMICAL TRANSFORMATIONS elemental mercury VOLATILE mercuric compounds WATER-SOLUBLE ATMOSPHERE Hg(0) Hg(II) oxidation deposition re-emission Hg(0) Hg(II) reduction SURFACE RESERVOIRS (Ocean, Land) microbes MeHg Methylmercury TOXIC

Atmospheric transport of Hg(0) takes place on global scale Implies global-scale transport of anthropogenic emissions Anthropogenic Hg emission (2006) Mean Hg(0) concentration in surface air: circles = observed, background = GEOS-Chem model Transport around northern mid-latitudes: 1 month Hg(0) lifetime = 0.5-1 year Transport to southern hemisphere: 1 year Streets et al. [2009]; Soerensen et al. [2010]

LOCAL POLLUTION INFLUENCE FROM EMISSION OF Hg(II) High-temperature combustion emits both Hg(0) and Hg(II) 60% Hg(0) GLOBAL MERCURY POOL photoreduction 40% Hg(II) NEAR-FIELD WET DEPOSITION Hg(II) concentrations in surface air: circles = observed, background=model MERCURY DEPOSITION “HOT SPOT” Large variability of Hg(II) implies atmospheric lifetime of only days against deposition Thus mercury is BOTH a global and a local pollutant! Selin et al. [2007]

Atmospheric redox chemistry of mercury Older models X X OH, O3, Cl, Br Hg(0) Hg(II) X ? HO2(aq) Oxidation of Hg(0) by OH or O3 is endothermic Oxidation by Cl and Br may be important: No viable mechanism identified for atmospheric reduction of Hg(II) Goodsite et al., 2004; Calvert and Lindberg, 2005; Hynes et al., UNEP 2008; Ariya et al., UNEP 2008

Bromine chemistry in the atmosphere GOME-2 BrO columns Inorganic bromine (Bry) Thule O3 hv Br BrO BrNO3 Halons hv, NO OH HBr HOBr Stratospheric BrO: 2-10 ppt CH3Br Stratosphere Tropopause (8-18 km) Troposphere Tropospheric BrO: 0.5-2 ppt CHBr3 CH2Br2 OH Bry Satellite residual [Theys et al., 2011] debromination BrO column, 1013 cm-2 deposition Sea salt industry plankton

Mean tropospheric concentrations (ppt) TROPOSPHERIC BROMINE CHEMISTRY simulated in GEOS-Chem global chemical transport model GEOS-Chem Observed CHBr3 440 Gg a-1 CH2Br2 62 Gg a-1 Vertical profiles of short-lived bromocarbons at northern mid-latitudes CH3Br OH 1.1 years Mean tropospheric concentrations (ppt) 0.09 0.6 0.3 CH2 Br2 OH 91 days Br BrO BrNO3 CHBr3 hv, OH 14 days including HBr+HOBr on aerosols HBr HOBr Sea salt 1.4 0.9 debromination industry deposition plankton Parrella et al. [2012]

GEOS-Chem global model of mercury 3-D atmospheric simulation coupled to 2-D surface ocean and land reservoirs GEOS-Chem 3-D atmospheric chemical transport model (CTM) 2-D surface reservoirs ocean mixed layer vegetation Anthropogenic and natural emissions Hg(0)+Br ↔ Hg(I) → Hg(II) Long-lived reservoirs deep ocean soil

MERCURY WET DEPOSITION FLUXES, 2004-2005 Circles: observations Background: GEOS-Chem model tropopause Model contribution from North American anthropogenic sources Scavenging of Hg(II)-rich air from upper troposphere Model contribution from external sources updraft FLORIDA SCAVENGING BY DEEP CONVECTION Selin and Jacob [2008]

Historical inventory of global anthropogenic Hg emissions Large past (legacy) contribution from N. American and European emissions; Asian dominance is a recent phenomenon Streets et al. , 2011

1977-2010 surface air trend of Hg(0) over the Atlantic Ocean 1990-2010 data from ship cruises show 50% decrease over North Atlantic, no significant trend over South Atlantic Long-term observations at continental sites in N America and Europe also show 1990-2010 decrease though not as strong as over North Atlantic Sørensen et al., submitted

GEOS-Chem simulation of Hg(0) 1990-2010 trends in surface air Global 3-D atmospheric model coupled to 2-D surface ocean and land models Forced by Streets emission trends ng m-3 a-1 Forced by observed subsurface Atlantic trends Observations in the subsurface North Atlantic show a 80% decrease from 1990 to 2010 [Mason et al., 2012], which can explain the observed trends in surface air This must reflect a large decline in Hg inputs to the North Atlantic Ocean over the 1970-2010 period. Sørensen et al., submitted

Decreasing Hg input to subsurface North Atlantic, 1970-2000 1 Decreasing Hg input to subsurface North Atlantic, 1970-2000 1. Atmospheric deposition explanation 1970 2000 Hg(0) Hg(0) marine boundary layer Br Br Hg(0) Hg(II) Hg(0) Hg(II) fast slow ocean mixed layer subsurface ocean (down to thermocline) Hg deposition to ocean is driven by MBL oxidation of Hg(0) by Br atoms MBL ozone ~doubled during 1970-2000; Br concentrations would have correspondingly decreased (Br/BrO photochemical equilibrium) O3 Br BrO h Sørensen et al., submitted

Decreasing Hg input to subsurface North Atlantic, 1970-2000 2 Decreasing Hg input to subsurface North Atlantic, 1970-2000 2. Coastal margin explanation 1970 2000 Disposed Hg-containing commercial products incineration Hg(II) Hg Hg wastewater, leaching Hg Secondary wastewater treatment and phase-out of Hg from commercial products would have decreased the Hg input to the subsurface N Atlantic Sørensen et al., submitted

Disposal of Hg in commercial products: a missing component of the Hg biogeochemical cycle? Global source of commercial Hg peaked in 1970 Streets et al. [2011] and Hannah Horowitz (Harvard)

Global biogeochemical model for mercury 7-box model with 7 coupled ODEs dm/dt = s(t) – km where s is primary emission Loss rate constants k specified from best knowledge Primary emissions Model is initialized at natural steady state, forced with historical anthropogenic emissions for 2000 BC – present; % present-day enrichments are indicated Amos et al., submitted

Time scale for dissipation of an atmospheric emission pulse Reservoir fraction Pulse gets transferred to subsurface ocean within a few years and stays there ~100 years, maintaining a legacy in the surface ocean Amos et al., submitted

Global source contributions to Hg in present-day surface ocean emissions Human activity has increased 7x Hg content of the surface ocean Half of this human influence is from pre-1950 emissions N America, Europe and Asia share similar responsibilities for anthropogenic Hg in surface ocean Europe Asia N America S America former USSR ROW pre-1850 natural Amos et al., submitted

Looking toward the future: UNEP global treaty for Hg Negotiations to be completed by 2013 Effect of zeroing global anthropogenic emissions by 2015 Zeroing anthropogenic emissions would decrease ocean Hg by 30% by 2100, while keeping emissions constant would increase it by 40% Elevated Hg in surface ocean will take centuries to fix; the only thing we can do in short term is prevent it from getting worse. Amos et al., submitted