Untangling the influence of air-mass

origin in interpreting composition trends

Zoë FlemingPaul Monks, Roland Leigh (Leicester University)

Brian Bandy, Stuart Penkett (UEA)

Lucy Carpenter, James Lee, Sarah Moller, Katie Read (York University)

Alistair Manning (Met office)

7th July 2010

The measurement stations

The NAME model

• Originally NAME stood for Nuclear Accident ModEl

• Lagrangian transport/dispersion model

Numerical

Atmospheric Dispersion

Modelling

Environment

• Uses Unified model met data and can run with

chemistry or for neutral inert particle releases

Run conditions

• 3 hourly release of particles backwards ten

days in time (1 or 5 days for London)

• Run at 2 or 3 height ranges (most interested in

low level to show interaction with surface

emissions)

Monthly footprints

Weybourne Cape Verde BT tower

Yearly averaged

Campaigns: SOS, Cape Verde and

TORCH2, Weybourne

Individual days

Campaign averaged

Weybourne air mass sectors

Weybourne regional distribution

100

80

60

40

20

0

% d

istr

ibu

tion

of se

cto

rs

01

/10

/20

06

01

/01

/20

07

01

/04

/20

07

01

/07

/20

07

01

/10

/20

07

01

/01

/20

08

01

/04

/20

08

01

/07

/20

08

01

/10

/20

08

01

/01

/20

09

01

/04

/20

09

01

/07

/20

09

01

/10

/20

09

America Europe Atlantic Scandinavia Greenland and Iceland UK North Sea Arctic

Weybourne regional variations

100

80

60

40

20

0

% d

istr

ibu

tio

n o

f se

cto

rs

13

/10

/20

07

00

:00

14

/10

/20

07

00

:00

15

/10

/20

07

00

:00

16

/10

/20

07

00

:00

17

/10

/20

07

00

:00

18

/10

/20

07

00

:00

19

/10

/20

07

00

:00

20

/10

/20

07

00

:00

21

/10

/20

07

00

:00

22

/10

/20

07

00

:00

23

/10

/20

07

00

:00

24

/10

/20

07

00

:00

25

/10

/20

07

00

:00

26

/10

/20

07

00

:00

27

/10

/20

07

00

:00

28

/10

/20

07

00

:00

29

/10

/20

07

00

:00

30

/10

/20

07

00

:00

31

/10

/20

07

00

:00

01

/11

/20

07

00

:00

0.8

0.6

0.4

0.2

0.0

America Europe Atlantic Scandinavia Greenland_and_Iceland UK North_Sea Arctic pure_Arctic

100

80

60

40

20

0

% d

istr

ibu

tio

n o

f se

cto

rs

10

/02

/20

07

02

/03

/20

07

22

/03

/20

07

11

/04

/20

07

01

/05

/20

07

21

/05

/20

07

10

/06

/20

07

30

/06

/20

07

America Europe Atlantic Scandinavia Greenland and Iceland UK North Sea Arctic

From Regional division to trajectory type

• Derive threshold for each region to decide

whether significant influence or proportion of

total particle journey

• Combination of which regions covered in

transport decides which trajectory type

4x10-3

3

2

1

0

part

icle

concentr

ation/ g s

/ m

3

01/01/2007 01/07/2007 01/01/2008 01/07/2008 01/01/2009 01/07/2009

Dust 'thresh0.0005' th_Af

NO2: 2006-2008 sectors

0

2

4

6

8

10

12N

O2

/ p

pb

v

2006

2007

2008

CN and SO2

0

20000

40000

60000

80000

100000

120000

140000

CN

2006

2007

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

SO2

/ p

pb

v

2008

2009

Arctic air masses

• Arctic and Scandinavian ozone

highest

• Run on global model domain to

track back 20 days to Asia up to

5000 m elevation

0

5

10

15

20

25

30

35

40

45

50

O3

O3

/ p

pb

v2006-2009 average

Arctic

Arctic + Europe

Atlantic

Europe

local

scand,greenland

America

Cape Verde air mass division

Cape Verde

100

80

60

40

20

0

% c

ontr

ibutio

n o

f e

ach

sect

or

01/1

1/2

006

01/0

1/2

007

01/0

3/2

007

01/0

5/2

007

01/0

7/2

007

01/0

9/2

007

01/1

1/2

007

01/0

1/2

008

01/0

3/2

008

01/0

5/2

008

01/0

7/2

008

01/0

9/2

008

01/1

1/2

008

01/0

1/2

009

01/0

3/2

009

01/0

5/2

009

01/0

7/2

009

01/0

9/2

009

Coastal African

Polluted marine (Europe)

Saharan Africa (Dust)

Atlantic continental (North America)

Atlantic marine

Cape Verde air mass division

African European

Coastal African

American

Classification into 7 major trajectory types

Atlantic and African coast

American, Atlantic and African coast

Saharan and no Europe

Saharan and Europe

Polluted and marine

Atlantic MarineAmerican Continental and Atlantic

Local air (flag for contamination or

slow transport)100

80

60

40

20

0

% o

f tim

e s

pen

t in

each

are

a

01/1

1/2

00

6

01/0

1/2

00

7

01/0

3/2

00

7

01/0

5/2

00

7

01/0

7/2

00

7

01/0

9/2

00

7

01/1

1/2

00

7

01/0

1/2

00

8

01/0

3/2

00

8

01/0

5/2

00

8

01/0

7/2

00

8

01/0

9/2

00

8

01/1

1/2

00

8

01/0

1/2

00

9

01/0

3/2

00

9

01/0

5/2

00

9

01/0

7/2

00

9

01/0

9/2

00

9

CoastalAfrican

Polluted_marine

Dust

Atlantic_cont

Atlantic_marine

site_ne

site_nw

site_sw

site_se

Relating to back to chemistry

(harmonised timeseries of 56

parameters)

01/1

1/2

006

01/0

1/2

007

01/0

3/2

007

01/0

5/2

007

01/0

7/2

007

01/0

9/2

007

01/1

1/2

007

01/0

1/2

008

01/0

3/2

008

01/0

5/2

008

01/0

7/2

008

01/0

9/2

008

01/1

1/2

008

01/0

1/2

009

01/0

3/2

00

9

01/0

5/2

009

01/0

7/2

009

01/0

9/2

009

01/1

1/2

009

01/0

1/2

010

01/0

3/2

010

2000

1500

1000

500

0

NO

xy a

nd V

OC

/ pptv

50

40

30

20

10

0

O3/p

pbv

220

200

180

160

140

120

100

80

60

CO

/ ppbv

3000

2500

2000

1500

1000

500

aero

sol n

um

ber

NO2

O3

CO ethane benzene isoprene ethene Aerosol number Temperature

Averaging/binning by trajectory type

-10

-5

0

5

10

15

20

Atlantic and

African coast

Atlantic

Marine only

American

and Atlantic

American,

Atlantic and African coast

European Africa (no

European)

Africa and

Europe

iso

pre

ne

/ p

ptv

isoprene 3 year averages

0

20

40

60

80

100

120

140

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and

African coast

European Africa (no European)

Africa and Europe

CO

/ p

pb

v

CO 3 year averages

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and African coast

European Africa (no European)

Africa and Europe

CH

Br3

/ p

ptv

CHBr3 3 year averages

0

100

200

300

400

500

600

700

800

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and

African coast

European Africa (no European)

Africa and Europe

NO

y/

pp

tv

NOy filtered 3 year averages

VOCs

0

5

10

15

20

25

30

35

Atlantic and African

coast

Atlantic Marine

only

American and

Atlantic

American, Atlantic

and African coast

European Africa (no European)

Africa and Europe

VO

C/

pp

tv

isobutane

isopentane

DMS

ethene

Ozone monthly averaging and

de-seasonalisation60

50

40

30

20

10

ozon

e/ p

pbv

350300250200150100500

3 year time series divided in 360 degrees

Sin wave equation y0 =31.491 ± 0.073 A =-8.1982 ± 0.101 f =0.05431 ± 0.000123 phi =1.8367 ± 0.0244

O3

polynomial to power 10 O3 sine function mean ozone ozone deseaonalised (from difference) ozone deseas ratio

0

5

10

15

20

25

30

35

40

45

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and African coast

European Africa (no European)

Africa and Europe

O3

/ p

pb

v

O3 deseasonalised 3 year averages

0

5

10

15

20

25

30

35

40

45

50

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and African coast

European Africa (no European)

Africa and Europe

O3

/ p

pb

v

O3 3 year averages

Aerosol monthly averages

0

100

200

300

400

500

600

700

800

900

Atlantic and African coast

Atlantic Marine only

American and Atlantic

American, Atlantic and

African coast

European Africa (no European)

Africa and Europe

aero

sol n

um

be

r

aerosol number 3 year averages

• Saharan and European have highest aerosol loading

Ratio analysis between trajectory types

• O-VOC ratios different

between marine and

terrestrial sources

5000

4000

3000

2000

1000

aceto

ne

/p

ptv

70006000500040003000200010000

methanol/ pptv

Atlantic and African coast Atlantic marine only American and Atlantic American, Atlantic and African coast Polluted and marine African and no Europe African and Europe

y = 0.4093x - 10.098 (R² = 0.4578)y = 0.4241x - 10.736 (R² = 0.5334)y = 0.3251x - 0.6948 (R² = 0.4688)y = 0.3685x - 3.6453 (R² = 0.6019)y = 0.4266x - 10.109 (R² = 0.5943)y = 0.3231x - 0.5268 (R² = 0.4171)y = 0.4635x + 289.71 (R² = 0.0998)

50

40

30

20

10

O3/

pp

bv

14012010080

CO/ ppbv

African and Europe

African and no Europe

American and Atlantic

American, Atlantic and African coast

Polluted and marine

Atlantic and African coast

Atlantic marine only

y = 0.4093x - 10.098 (R² = 0.4578)y = 0.4241x - 10.736 (R² = 0.5334)y = 0.3251x - 0.6948 (R² = 0.4688)y = 0.3685x - 3.6453 (R² = 0.6019)y = 0.4266x - 10.109 (R² = 0.5943)y = 0.3231x - 0.5268 (R² = 0.4171)y = 0.313x + 0.0773 (R² = 0.5026)

• O3 : CO ratios show a large

spread but higher gradients

for European and African

Mace Head, Harwell and Marylebone

road and Eltham 1 day backwards

• Marylebone rd• Eltham

• Harwell• Mace Head

BT tower 1 day back runs

• 5 day backwards split into sectors

• 1 day backwards split the UK and North

Sea, northern France/ Benelux into 10

regions

100

80

60

40

20

0

% c

ontr

ibution o

f geogra

phic

al secto

rs to a

ir a

rriv

ing a

t B

T tow

er

01/02/2006 01/03/2006 01/04/2006 01/05/2006 01/06/2006 01/07/2006

Arctic

Scandinavia

Europe

Atlantic

Greenland_and_Iceland

America

UK

North_Sea

Running forwards from London

• Forward run combined with NAEI emission

inventory to track chemistry emissions from

London affecting the surrounding areas

• Run forwards from surrounding areas towards

BT tower

Conclusions

• Have built up a library of successive 3 hourly back

trajectories from NAME for 3 sites

• Analysis of regional distribution of transport shows

how quickly and when air mass direction changes

• Combination of regional influences can define

trajectory types

• Can provide footprints or detailed 3 hourly influence

regional distribution for campaigns or one off events

• Can look at long term changes / seasonal changes or

variations in chemistry between trajectory types

• Future project to run NAME forwards with chemistry

and combining with emission inventories