Challenges in Global Mercury Modeling Ashu Dastoor Meteorological Service of Canada Environment Canada Acknowledgements: Didier Davignon and Arturo Quintanar.

Challenges in Global Mercury Modeling

Ashu Dastoor

Meteorological Service of Canada

Environment Canada

Acknowledgements: Didier Davignon and Arturo Quintanar

Atmospheric Mercury Cycling

Gas-phase chemistryGEM↔RGM↔TPMO3, OH, H2O2, Halogens: tropospheric chemistry Q: reactions, rates and products?

Heterogeneous Chemistry ?

Aerosol dynamicsQ: TPM size distribution? Planetary Boundary Layer

Turbulent mixingmet model

Anthropogenic emissionsGEM, RGM, TPMQ: emission speciation andplume chemistry?

Flue gas chemistry?

Point sourcesPlume rise

Volcanic emissionsQ: inventory?

Bio-mass burningemissionsQ: inventory?

Surface natural and re-emission:soils, vegetation water bodies, snow, oceansQ: inventory/processes?

Wet depositionMet modelQ: precipitation scavenging?

EvaporationMet model

Gas/liquidexchange

TransportMet model

Transport

Dry depositionQ: deposition velocities?

Cloud propertiesMet models

Snow/Ice dynamicsIce model

Evaporation

Area emissions

Mercury transformation processesQ: residence time and revolatilization rate and species?

GEM↔RGM↔TPMO3, OH, HO2, (Cl), SO3,aerosols: tropospheric chemistryQ: reactions, rates and products?

Goals for a Global Mercury Model towards estimating Mercury in Lake Ontario

• What is the atmospheric flux of Hg that arrives to Lake Ontario from global anthropogenic sources?

• What is the atmospheric flux of Hg that arrives to lake Ontario from global natural and recycled sources?

• What are the sources of this distant mercury?

• What is the speciation of transported mercury?

• What is the time-space distribution of this flux?

• How are the fluxes changing with changing emissions?

• What is the contribution of distant mercury to the deposition in Lake Ontario basin? This question can be better answered by a regional model using the atmospheric background and boundary flux information from global models.

• Preliminary results from N. American model intercomparison study indicate significant impact of trans-boundary flows of mercury to the regional deposition. The study compares impact of boundary fluxes from three different global models.

Major Challenges faced by a global model in addressing the goals

• Emissions: Natural and re-emissions from land and ocean surfaces (very few flux measurements and no emissions inventory)

• Mercury Chemistry: Chemical reactions- products, rates and phase

• Mercury dry deposition process• What is the trend in natural and re-

emissions of mercury relative to the anthropogenic emissions?

THE MERCURY CYCLE: CURRENT

Wet & DryDeposition 3500

ATMOSPHERE5000

SURFACE SOILS1,000,000 OCEAN

288,000

Wet & DryDeposition3100

Oceanic Evasion

2600

Net burial200

Land emissions1600

Quantities in Mg/yearUncertainty ranges in parenthesesAdapted from Mason & Sheu, 2002

AnthropogenicEmissions 2400

Extraction from deep reservoirs2400

River200

(700-3500)

(1680-3120)

(700-3500)

Ocean flux distribution

0 100 200 300 kg

July ocean flux

Jan. ocean flux

latitude

Flux

(ng/

m2 )

Jan.July

• Higher flux in tropics due to high temperature and radiation

• High flux in regions of high deposition

• Seasonality, spatial variation due to temperature, npp, radiation, and mixed layer depth

• Diurnal variation: photochemistry

Hg0

1.7 ng/m3

GaseousPhase

AqueousPhase

Hg0

Henry’s Constant 0.11 M/atm

Particulate Phase

Oxidation

Hg2+

Products and phase

unclear

10-200 pg/m3

HgP

1-100 pg/m3

Hg2+

k=8.7(+/-2.8) x 10-14 cm3 s-1 (Sommar et al. 2001)k=9.3(+/-1.3) x 10-14 cm3 s-1(Pal & Ariya 2004)

Probably unimportant reaction (Goodsite et al. 2004)

k=3(+/-2) x 10-20 cm3 s-1 (Hall 1995) k=7.5(+/-0.9) x 10-19 cm3 s-1 (Pal and Ariya 2004)

Longer lifetime suggested (Calvert & Lindberg 05) Henry’s Constant 1.4x106 M/atm

OH

O3

Oxidation

HO2

??Reduction

SO3

k=1.1-1.7 x 104 M-1 s-1 (Pehkonen & Lin 1998)Shouldn’t occur (Gårdfeldt & Jonsson 2003)

k=0.0106 (+/- 0.0009) s-1

(vanLoon et al. 2000)Occurs only where high sulfur, low chlorine

Oxalate?

Is there Hope?

• Global model is a closed atmospheric system therefore observations can be used to constrain uncertainties.

• Observations available: atmospheric mercury concentrations, wet deposition, terrestrial and aquatic fluxes of mercury and measurements of long-range transport of mercury.

ATMOSPHERE

Hg0 Hg(II)

Via OH:10236Via OH:10236

Dry Deposition

Ocean Emissions

Land (Natural) Emissions

Anthropogenic Emissions

Land Re-emissions

Hg(P)

775775 204204

Via O3: 2377Via O3: 2377

1500150014461446

500500

20002000

Dry DepositionWet Deposition

Wet Deposition

10411041

53275327

191191

1111

MERCURY BUDGET IN GEOS-CHEM

Inventories in MgRates in Mg/yr

k=8.7 x 10-14 cm3 s-1

k=3 x 10-20 cm3 s-1

τ = 0.77 yr τ = 7 days τ = 3.5 daysNet ox: 5489

Reduction7124

How are we addressing the goals?

• Develop a well constrained global model with known chemistry and emissions.

• Use the constrained model to address the goals.• Perturb the system using emerging chemical mechanism for

mercury on the global balance of mercury and provide possible solutions.

• Assess the impact of emerging mercury chemistry on long range transport of mercury and trans-boundary fluxes of mercury for regional models.

• Evaluate the accuracy of trend in anthropogenic emissions inventory by modeling the global budgets and verify the changes against the observations.

• Develop detailed processes such as mercury evasion from snow, soil, vegetation etc and including aerosol dynamics.

90S 60S 30S Eq. 30N 60N 90N

0

0.5

1

1.5

2

2.5

TGM

ng/

m3

Latitude

Observed (left; Lamborg et al., 2002) and modeled (right)

Inter-hemispheric gradient of TGM Observed and from GRAHM simulation

Date (Julian Date)

0 100 200 300

Mer

cury

con

cent

ratio

n (n

g/m

3)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Mercury depletion

Mercury reduction

Mean mercury concentration in the Arctic

Elemental mercury vapor concentration at Alert for the whole year 1995

Rapid, near-complete depletion of mercury observed during spring in the Arctic,sub-Arctic and Antarctic is which is also correlated with ozone depletion.

Questions: Which halogen gases are responsible and what is the impact on the Arctic and global mercury deposition?

Mercury deposition without MDEs Mercury deposition with MDEs

Main conclusions:

• Br atoms (~.4 day) and BrO (~1 day) radicals are the most effective halogens driving mercury

oxidation to more hygroscopic species which are readily deposited and could be incorporated

in the biota.

• MDEs in the Arctic increase the net deposition into the Arctic by 100 tons/yr.

• Net accumulation in the Arctic 325 tons/yr

The Arctic: a sink for mercury

Slow oxidation of Hg0 by O3 in the troposphere – 1-2 years life timeFast oxidation of Hg0 by halogens in the polar regions and marine boundary layer – hours - 10 days life time

Hg0 HgII

Surface ocean

Evasion deposition

transport

reduction

Hg0 HgII

Marine Boundary Layer

Fast oxidation

Loss

• Reduction proportional to radiation and net primary productivity

transport

Sea-salt aerosolHalogen activation

Cl2, Br2, Cl, Br, BrO

Impact of mbl chemistry on transboundary flow from Asia

Surface air mercury concentrations(ng/m3)

MBL chemistry

No MBL chemistry

Impact of mbl chemistry on total mercury deposition (ug/m3)

MBL Chemistry

No MBL Chemistry

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

J F M A M J J A S O N D

Canada

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

J F M A M J J A S O N D

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

J F M A M J J A S O N D

0.1

0.6

1.1

1.6

2.1

2.6

3.1

3.6

J F M A M J J A S O N D

0.1

0.6

1.1

1.6

2.1

2.6

J F M A M J J A S O N D

0.1

1.1

2.1

3.1

4.1

5.1

6.1

7.1

J F M A M J J A S O N D

USA

Arctic

Surface air Hgo concentrations Total deposition in ug/m2/year

No MBL

MBL

Spring 2004 Experiment: Simultaneous Observations at Mt.Bachelor and Okinawa ( Jaffe et al. 2005)

MBO

Okinawa

Okinawa: Hg0,RGM, PHg, CO, O3, aerosols, etc.

MBO: Total Hg0 , CO, O3, aerosols, etc.

Mauna Loa

1

2

3 4

5

6

Jaffe et al. 2005

Vector winds700 mb, April mean

Hg transport to North America: April 25th, 2004

ΔHg/ΔCO=0.50 ng/m3/100 ppbv

Good tracer of Asian air masses

Suggests much larger Asian emissions? (Jaffe et al. 2005)

Possible causes for this discrepancy:• Underestimate of the industrial or domestic Hg emissions;• Natural emissions;• Re-emission of previously deposited Hg;• Too low a ratio of Hg0/total Hg in the inventory;• Conversion of RGM to Hg0 during transport;

Explanations/Hypotheses (Jaffe et al. 2005)

Using the observed Hg0/CO ratio, and the known CO emissions, Jaffe et al. calculate Hg0 emissions from Asia of 1460 mt/year (+/-30%);

This can be compared to 770 mt/year in the Pacyna et al., 2003 inventory.

Anthropogenic Air Emissions of Anthropogenic Air Emissions of Mercury: Distribution by Region in Mercury: Distribution by Region in

1990 and 20001990 and 2000

Total: 1,881 metric tons/yr Total: 2,269 metric tons/yr

Asia and Africa account for about 70% of global emissions and show steady, significant increases due to industrialization.

Based on Pacyna, J., Munthe J., Presentation at Workshop on Mercury: Brussels, March 29-30, 2004Slide courtesy Grace Howland, Air Pollution Prevention Directorate, Environment Canada

20001990Africa 9%

Asia38%

Australia 3%

Europe33%

North America14%

South America3%

Africa 18% Asia

52%

Australia 6%Europe

11%

North America9%

South America4%

GRAHM simulation with year 2000 Global mercury emissions

January GEM ng/m3 July GEM ng/m3

JanuaryJanuary JulyJuly

GEM surface air concentrations ng/m**3 GEM surface air concentrations ng/m**3

GRAHM simulation with year 1990 global emissions

Impact of mercury emission reductions

• Impact of proposed Canada wide standards for coal-fired power plants-

Reduction of 1,224 kg/yr will result in ~ 580 kg/yr reduction in total mercury deposition over Canada

• Impact of proposed US mercury rule for coal-fired power plants- Reduction of 30,000 kg/yr will result in ~ 2,600 kg/yr reduction in

total mercury deposition over Canada

• Asia contributes ~ 24,700 (21,650 with mbl chemistry) kg/yr mercury deposition into Canada of which ~ 15,300 kg/yr comes from China.