Chemical characteristics of aerosols over Arabian Sea & Bay of Bengal: Impact of Anthropogenic Sources

Manmohan Sarin

Physical Research LaboratoryAhmedabad-380 009, India

GESAMP – 36 Annual Meeting: WMO, Geneva

sarin@prl.res.in

29th April’09

Rationale:Rationale:

Climate Change: One of the most important societal concerns for the 21st

century

Atmospheric Chemistry: Plays a critical role

-Abundances & distributions of natural and anthropogenic constituents

Greenhouse gases, aerosols and clouds

-Influence incoming and/or outgoing solarradiation, temperature and precipitation

Climate Forcing by Anthropogenic Aerosols

Key issue : Role of atmospheric chemistry in

amplifying/damping climate change

Questions : What are the effects of aerosols on clouds,

their optical properties, precipitation and

regional hydrologic cycle?

African drought

Rosenfeld et al [2001]:

Correlation between increasing dust frequency&

Rainfall in Sahelian region (African Monsoon)(from 1950s through 1980s)

Conclusion:Dust aerosols act as cloud condensation nuclei (CCN)

Increased dust frequency decreased precipitation(cloud life-time effect)

Cause of

•

Atmospheric Dust, Chemistry and Transport

Variety of individual minerals(quartz, calcite, gypsum, clays, etc.)(different physical/chemical properties)

Information on dust emission sources, transport, particle size and compositional data

Requires quantification of its global impacts

• Mineral Dust

Atmospheric chemistry is over-simplified in current models(Global transport/chemistry model & dust model)

Considered as single entity aerosol with single kinetic parameter for chemical reactions

Ozone decomposition on mineral surface (2O3 3O2)( mediated via POC, Organic complexes, Fe & Mn)

Mineral dust & Atmospheric Chemistry

Alkaline nature favours uptake of SO2 and NOx

NO3

DUST

O3

HNO3OH

NO2

SO2

NO

Changes surface properties: Hydrophobic Hydrophilic

Mineral dust : Recent work

*

Provide stoichiometric/catalytic reaction centers -

to perturb important gas-phase cycles

*

Enhances kinetics of SO2

/NO2

oxidation

*

Perturbation of photo-oxidation cycles of: O3

, HNO3 & HO2

*

--

radicals

*

Heterogeneous loss of organic compounds: Acetaldehyde, acetone ….

Comparable to loss by direct photolysis

Influence of Dust on Sulphate

Asian regions: High SO2

Overestimate the forcing during high dust season

Present estimates of SO4

cooling ignore reactions with mineral aerosols

CaCO3

+ H2

SO4

→ CaSO4

+ H2

O +CO2

(fine) (coarse)

Longitude (oE)

65 70 75 80 85 90 95

Latit

ude

(oN

)

-5

0

5

10

15

20

25

Chennai

I N D I A

I N D I A N O C E A N

Goa

A r

a b

i a

n

S e

a

B a y of B e n g a l

Maldives

18-27 Feb 2001

2-22 Mar 2001

15 Sept- 12 Oct2002

19-27 Feb 2003

ORV Sagar Kanya Cruise Tracks

BOB Mean Wind Streamlines(March 2001)

EC (BC), OC, IC, K+, NO3

-

and SO4

2-

and

trace elements

MODIS AOD for Dec. month

Objectives

1. Characterize the temporal and spatial variability in chemical and isotopic composition of aerosols.

2. How this variability is related to atmospheric transportprocesses (physical and chemical).

3. Deposition fluxes of natural and anthropogenic constituentsacross the air-sea interface.

IMPLICATIONS:

Long-range transport of air pollutants from the south & south-east Asia towards the Indian Ocean.

Experiment :

Sampling : Aerosols samples collected on quartz filters(using high-volume air samplers)

Analysis :Water soluble constituents : NH4

+ , Na+ , K+ , Mg2+ , Ca2+ , SO42- , Cl- , NO3

-

Acid soluble constituents :Al, Fe, Ca, Mg, Mn, Cu, Zn, Pb

Atmospheric nuclides :7Be (53.3 d) & 210Pb (22.3 yr)

Total Ca2+ (μg m-3)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

nss-

Ca2

+ ( μg

m-3

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Total Ca2+ (μg m-3)

0 1 2 3 4

nss-

Ca2+

( μg

m-3

)

0

1

2

3

4

• Arabian Sea• Indian Ocean

• Feb-March 2001• Feb 2003

1:11:1

March 2001

Bay of Bengal

Dust & Biogenic Sources

Total SO42- (μg m-3)

2 3 4 5 6 7

nss-

SO42-

( μg

m-3

)

2

3

4

5

6

7

Total SO42-(μg m-3)

0 3 6 9 12

nss-

SO42-

( μg

m-3

)

0

3

6

9

12

• Arabian Sea• Indian Ocean

1:11:1

March 2001

Bay of Bengal

• Feb-March 2001• Feb 2003

Dominance of Anthropogenic Sources

Mg2+NO3-

8.0

Σ = 9.2 (μg m-3)

60%

1.9

8.3 2.97.2

2.0

9.8

Na+

Cl- K+NH4+

Mg2+

NO3-

Ca2+

SO42-

Feb , 2003Feb , 2001

66%

12 2.65.9

4.85.3

2.01.5

Σ = 7.6 (μg m-3)

Na+ Cl-

K+

NH4+

Ca2+

SO42-

SO42-

NO3-

Mg2+

K+

Ca2+

Cl-

Na+

16%

1.43.8

1.42.445%

30%

Σ = 12 (μg m-3)Sep , 2002

Average Composition: Water-soluble Constituents

Bay of Bengal

SO42- in marine boundary layer (MBL)

Sea-salts

Biogenic: DMS SO2 SO42-

Anthropogenic : SO2 SO42-

Dust , CaSO4 (gypsum)(cement industries)

Biomass Burning ?

Volcanic (transient)

SO42- / Na+ in sea water = 0.25

oxioxi

oxi

nss-SO42- = ΣSO4

2- 0.25Na+

(non-sea-salt)

Cd Pb Zn

E.F.

(w.r.

t. Fe

)

1

10

100

1000

2001(Feb)2001(Mar)2003(Feb)

Bay of Bengal

Cd Pb Zn

E. F

. (w

.r.t.

Fe)

1

10

100

1000

Bay of Bengal (Feb)Bay of Bengal (Mar)Indian OceanArabian Sea

Bay of B

engal

Indian O

cean

Arabian

SeaPac

ific O

cean

Av.

Con

c. (n

g m

-3)

0

100

200

300

400 817762

FeAl

(SEAREX data

)

The cruise transects carried out in the Bay of Bengal and Arabian Sea for collection of bulk-aerosol samples during spring inter-

monsoon (March-May)

Arabian Sea

47%

Na+

SO42-

18%15%

Cl-

11 %

Ca2+

K+1.5%Mg2+2%

NO3-

5%

Σ = 11 μg m-3

Bay of Bengal

SO42-

68%

Na+

9% Ca2+

K+

NH4+

10%

3%

Mg2+NO3

-

2%1%

7.5%

Σ = 9 μg m-3

Average water-soluble composition of aerosols over Bay of Bengal and Arabian Sea, reflecting dominance of SO4

2-

Kumar, AKS, MMS (JESS 2008)

Ca (μg m-3)

0.0 0.5 1.0 1.5 2.0 2.5 3.0

nss-

Ca2+

( μg

m-3

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0r2 = 0.91, m = 0.87, ARSr2 = 0.97, m = 0.97, BOB

1:1 line

Enhanced solubility of Calcium as dust (CaCO3) undergoes neutralization process with acidic species

Unequivocal evidence: Acid uptake by mineral dust

CaCO3 + H2 SO4 CaSO4 + CO2 + H2 O

CaCO3 + 2HNO3 Ca(NO3 )2 + CO2 + H2 O

Excess Acid/(nss-Ca2+ + nss-Mg2+)0.0 0.3 0.6 0.9

HC

O3- /(n

ss-C

a2+ +

nss

-Mg2+

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2Ahmedabad Nov-Feb

Mar-Apr

Atmos. Environ, 2005a(neutralization reactions)

pH5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5

Num

ber o

f sam

ples

0

10

20

30

40

n = 91

Rainwater pH

Rainwater

Avg. pH = 6.7 ± 0.5 indicating “Alkaline” nature of rainwater

Atmos. Environ, 2005b

Excess Acid/(nss-Ca2+ + nss-Mg2+)0.0 0.3 0.6 0.9

HC

O3- /(n

ss-C

a2+ +

nss

-Mg2+

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2Ahmedabad Nov-Feb

Mar-Apr

Atmos. Environ, 2005a

Present-day models need to account for such chemical reactions to infer climate cooling due to Sulphate aerosols

IMPLICATIONS:

Change in size distribution of SO42- and NO3

- aerosols from fine to coarse mode

Decrease in Cloud Condensation Nuclei

Consequence of decreasing aerosol pH as

droplets evaporate

HCl(aq.) reaches saturation

HCl (g) Aqueous phase gas phase

Chloride depletion in MBLComplete degassing of HCl from NaCl

(Deliquescent sea-salt aerosols)

97--100% RH, buffered pH ~8

2NaCl + H2SO4 2HCl + Na2SO4

NaCl + HNO3 HCl + NaNO3

----------------oxi.------------------nss-SO4

2- SO2 (S +4) SO42- (S +6)

---------------------------------------

nss-SO42-(%)

40 60 80 100

Cl- de

ficit

(%)

0

20

40

60

80

100

80 85 90 95 10040

60

80

100

Feb 2003

Feb 2001Sept 2002

SO42- -rich aerosols: Implications to Cl- loss

Implications :

(SO2) SO42- limit CCN

Enhanced anthropogenic Not adequate(SO2 ) SO4

2-

Sea-salt

Sink

Increase Ozone loss :

1. O3 + Siv (SO2 ) SO42- + O2 [major O3 sink]

(at pH : ~6-8 , Rate constant 4 x 1011)

2. O3 + Cl- ClO- + O2 [slow rate]

oxi.

oxi.

(Unaccounted O3 sink !)HCl degassingfrom sea-salt

Year-2007

03 J

an04

Jan

07 J

an10

Jan

11 J

an12

Jan

16 J

an20

Jan

29 J

an01

Feb

08 F

eb11

Feb

16 F

eb17

Feb

26 F

eb27

Feb

05 M

ar10

Mar

15 M

ar20

Mar

25 M

ar01

Apr

05 A

pr09

Apr

12 A

pr17

Apr

21 A

pr23

Apr

25 A

pr29

Apr

03 M

ay07

May

11 M

ay15

May

17 M

ay19

May

24 M

ay28

May

02 N

ov07

Nov

10 N

ov20

Nov

27 N

ov03

Dec

08 D

ec13

Dec

18 D

ec23

Dec

28 D

ec

WS-

Fe (%

)

02468

1012141618

Year-2007

03 J

an04

Jan

07 J

an10

Jan

11 J

an12

Jan

16 J

an20

Jan

29 J

an01

Feb

08 F

eb11

Feb

16 F

eb17

Feb

26 F

eb27

Feb

05 M

ar10

Mar

15 M

ar20

Mar

25 M

ar01

Apr

05 A

pr09

Apr

12 A

pr17

Apr

21 A

pr23

Apr

25 A

pr29

Apr

03 M

ay07

May

11 M

ay15

May

17 M

ay19

May

24 M

ay28

May

02 N

ov07

Nov

10 N

ov20

Nov

27 N

ov03

Dec

08 D

ec13

Dec

18 D

ec23

Dec

28 D

ec

FeA

(ng

m-3

)

0

200

400

600

800

1000

Winter Summer Winter

High-altitude Stn: Temporal variability of aerosol iron (FeA ) and water-soluble iron (WS-Fe)

Bio-available Fe & surface ocean bio-geochemistry

Ahmedabad, 49 m asl, a megacity Mt Abu, ~1700 m asl, clean site

Rainfall ~650 mm: Jun-Aug

Arabian SeaBay of Bengal

SW-winds

(May-A

ug)

NE-W

inds

(Nov

-Feb

)

Study sites

nss-SO42- (neq m-3)

0 50 100 150 200 250 300

NH

4+ (neq

m-3

)0

50

100

150

200

250

300

SummerWinter

m = 1.1r = 0.95

nss-SO42- (μg m-3)

0 2 4 6 8 10 12 14

WS-

Fe (%

)

02468

1012141618

Solubility of aerosol iron and Direct emission (fossil fuel combustion)

Cd/

Al (

x100

0)

0

2

4

6

8

10

FeA (ng m-3)

0 200 400 600 800 1000

WS-

Fe (%

)

02468

1012141618

SummerWinter

FeA (ng m-3)

0 200 400 600 800

WS-

Fe (%

)

02468

10121416

WS-Fe = 0.06 % and FeA = 915 ng m-3

during summer season (as end member);

WS-Fe of 16.1 % and FeA = 50 ng m-3

(2nd end member) during winter season.

Fe solubility in semi-arid western India is controlled by composition and chemical nature of aerosols rather

than chemical processing during long-range transport.

Kumar and SarinTellus ‘B’, 2009(Under Review)

Nanoparticles: Photochemical mechanism

+

C2 O42-

HSO3-

CO2

SO42-

By product: Fe3+ Fe2+Dissolution

(cloud droplet)

HematiteFe3+

Mineral Aerosols

Oxidation

Reduction

e-

Generate electron- hole pairs

Bay of Bengal (March-April’

06)

70 75 80 85 90 95 1000

5

10

15

20

25La

titud

e(0 N

)

Longitude(0E)

onboard ORV sagarkanyaICARB - Campaign

5.27

4.31 6.214.28

6.59 4.153.24

3.8 3.7811.977.246.85

5.64 13.59

2.955.18

11.1213.093.472.56

2.352.58

6.97

Spatial distributions of Water SolubleFe(%)

Chennai

Cochin Bay of Bengal

March -2006

55 60 65 70 75 80 85 900

5

10

15

20

25

30

0.050.03

0.02

0.040.10

0.040.020.090.030.11

0.04 0.08

Latit

ude(

0 N)

Longitude(0E)

G oa

C ochin

0.05

0.430.080.12

0.03

0.07 0.02

0.19

Spatial distribution of WSFe(%) during April2006

onboard ORV Sagara KanyaICARB- Cam paign

Arabian Sea

70 75 80 85 90 95 1000

5

10

15

20

25La

titud

e(0 N

)

Longitude(0E)

Enrichment .Factors. of Cu from Mar-Apr06

ICARB-Campaign

13.8

14.6 17.819.0

12.5 28.9

12.420.4

14.9 12.325.116.233.5

44.2 48.1

105.4162.2

142.9200.584.728.2

24.5

58.6 Bay of BengalVisakhapatnam

1. Long-range transport of dust, biogenic and pollution-derived components dominate the aerosol composition in the MABL of Bay of Bengal & Arabian Sea.

2. On average, non-sea-salt-SO42- accounts for nearly 60% of the

water soluble components indicating highly acidic environment (MABL of BOB).

3. Enhanced Fe deposition fluxes over Bay of Bengal and Arabian Sea could serve as dominant source for surface waters.

Conclusions :