Attack of the Toxic Mist –

Ian D. Longley

School of Earth, Atmospheric & Environmental Sciences,University of Manchester

Why we study Urban Aerosol Dispersion

Urban air is cleaner than it was – isn’t it?

Urban smogs in the UK:December 1991

March 1996December 2001February 2003

Aftermath of the Clean Air Acts

0

100

200

300

400

500

600

1940 1950 1960 1970 1980 1990 2000 2010

traf

fic

(b

illio

n v

kt)

PM10

Mass concentration of particulate matter with an aerodynamic diameter < 10 m

>50% probability of penetrating into thorax

UK-wide Automated Urban and Rural Networks

Bury (M60) Piccadilly Gardens North-West

What PM10 levels do we experience?

weekly means 1999-2005 incl.

0

5

10

15

20

25

30

35

40

45

50

PM

10 /

g m

-3

EcclesPiccadillyShaw Heath

December 2001

0

50

100

150

200

01-Dec 06-Dec 11-Dec 16-Dec 21-Dec 26-Dec 31-Dec

PM

10 /

g

m-3

EcclesStockport

wind speed

0

2

4

6

8

10

12

14

16

01-Dec 06-Dec 11-Dec 16-Dec 21-Dec 26-Dec 31-Dec

U /

m s

-1

1999-2005 Means:

Piccadilly 27

Eccles 23

Stockport 22

PM10 episodes and health risk

• Rise of 10 gm-3 PM10 linked to ~0.5% rise in mortality

• Translates as ~8 000 deaths brought forward per year in UK urban population (COMEAP, 1998) – compared to 3 700 total deaths due to RTAs.

• PM10 also linked to increased morbidity

• No lower threshold!• Deaths from respiratory causes• Also cardiovascular causes• Deaths mostly of the ‘vulnerable’• Episode promotes an

exacerbation of pre-existing condition

• What is the cause?

Urban particle sources and sizes

Dp / m

0.001 0.01 0.1 1 10 100

dV/d

log(

Dp)

/ m

3 m

-3

0.01

0.1

1

10

100Vehicle emissions,

combustion

Long-range transport, secondary particles

Dust, wear products,

biological particles, minerals

Measured in Princess Street, Manchester

PM10

Urban Particle Size Distributions

Particle number concentrations dominated by ultrafines

Dp / m

0.001 0.01 0.1 1 10 100

dV/d

log(

Dp)

/ m

3 m

-3

0.01

0.1

1

10

100

Dp / m

0.001 0.01 0.1 1 10 100

dN/d

log(

Dp)

/ cm

-3

1e-1

1e+0

1e+1

1e+2

1e+3

1e+4

1e+5

1e+6

Above: mass size distribution from Manchester street canyon

Above: number size distribution from Manchester street canyon

Data: Longley et al., Atmos. Environ., 2003.

dM

/dlo

gD

a /

g m

-3

Aerodynamic Diameter (nm)

1.0

0.8

0.6

0.4

0.2

0.0

dM

/dlo

gD a

(µg

m-3

)

2 3 4 5 6 7 8 9100

2 3 4 5 6 7 8 91000

2 3

Aerodynamic Diameter (nm)

OrganicsSulphateNitrateAmmonium

Manchester winter

4

3

2

1

0

dM

/dlo

gD

a (µ

g m

-3)

2 3 4 5 6 7 8 9100

2 3 4 5 6 7 8 91000

2 3

Aerodynamic Diameter (nm)

SulphateNitrateOrganicsAmmonium

Manchester_Summer

3.0

2.5

2.0

1.5

1.0

0.5

0.0<20

0nm

Par

ticul

ate

Org

anic

s (µ

gm-3

)

14012010080604020NOx (ppb)

Urban particle speciated mass size distributions

UFP mostly organic compounds

Also:

•Black carbon

•Sulphuric acid

i.e. Traffic is the major source

Data: Allan et al., JGR 2002, Alfarra et al., Atmos. Environ., 2004

Ultrafine Particles in the body

•Ultrafines (UFP) efficiently deposit to alveolar walls

•Overload can cause chronic inflammation and irreversible damage to tissues and defences

•Inflammatory response triggers systemic reaction in cardiovascular system – increases in blood viscosity, formation and disruption of plaques, heart rate variability.

•Can lead to arrythmia, ischaemia and heart attack (immediately or in future)

Effects seen in ‘non-toxic’ particles –

is toxicity in size, surface area or composition???

UFP dispersion in urban areas

coagulation

condensation

Recirculation/sheltering

dilution

65000

70000

75000

80000

85000

90000

95000

100000

105000

110000

Ventilation of the urban canopy

0

10000

20000

30000

40000

50000

60000

00:00 06:00 12:00 18:00 00:00local time

FN /

cm

-2s-1

0

40

80

120

160

200

240

H /

W m

-2

particles

sensible heat

Regular emission cycles

Sheltering, recirculation, deposition, inversions

0

10000

20000

30000

40000

50000

60000

70000

00:00 06:00 12:00 18:00 00:00

N2

5 /

cm

-3

0

2000

4000

6000

8000

10000

12000

14000

N2 /

cm

-3

25 m

2 m

Mean diurnal fluxes at 90 m above Manchester

Diurnal mean concentrations at 2 and 25 m

NOTE: ventilation is suppressed during morning emission peak

Data: NERC CityFlux project, Longley et al., 2006

Long-term PM10 exposure and health risk

• For long-term exposure relative risk is doubled to ~1.0% (Dockery et al., 1993, Pope et al., 1995, Kunzli et al., 2001)

• Suggests long-term exposure (to mean or repeated episodes?) causes or increases vulnerability

• Episodes NOT followed by ‘harvesting’ for cardiovascular deaths – episodes create a newly vulnerable cohort

Consequences:

Need to consider complete history of personal exposure, especially UFP

Fixed PM10 monitors versus personal UFP exposure

PM10 monitors are supposed to be ‘representative’

UFP has high spatial gradients, fixed monitors expensive and unrepresentative(?)

Daily/Hourly PM10 data biases concept of episode to day/hour durations

Personal UFP exposure dominated by short (< 1 hour) very high ‘excursions’

Need to consider residential and workplace, but especially commuting exposures

commuting exposure, 23 Nov 05

0

50000

100000

150000

200000

250000

300000

350000

400000

8:38 8:45 8:52 9:00 9:07

N /

cm-3

Walk to stop At bus stop On bus

Street canyon UFP number size distribution in channelled flow

Dp / nm

1 10 100 1000dN

/dlo

g(D

p) /

cm-3

103

104

105

106

channelled

Street canyon UFP number size distribution in recirculating flow

Dp / nm

1 10 100 1000dN

/dlo

g(D

p) /

cm-3

103

104

105

106

channelledrecirculation

Recirculation caused by perpendicular approach flow (>40 deg from canyon axis)

Extra particles in ‘fresh exhaust’ size range

Data: Longley et al., Atmos. Environ., 2003, CityFlux: Longley et al., 20060

10000

20000

30000

40000

50000

60000

70000

00:00 06:00 12:00 18:00 00:00

N25

/ c

m-3

0

2000

4000

6000

8000

10000

12000

14000

N2

/ cm

-3

25 m

2 m

Parameterising street canyon turbulencei

2 = (AiU)2 + 22

Where i = u, v, w

Ai, 2 = f(z)

U / m s-1

0 1 2 3 4

w/U

0.0

0.5

1.0

1.5

2.0

3.5 m3.5 m

3.5m: w2 = (0.17 U)2 + ((0.62 T / 3600) + 0.01)

Longley et al. 2004, Atmos. Env. 38, 69-79

Longley et al. 2004, Atmos. Env. 38, 4589-4592

Au, v, w

0.0 0.1 0.2 0.3 0.4 0.5 0.6

z / m

0

2

4

6

8

10

12

14

16

18

20

Au

Av

Aw

Au, v, w

0.0 0.1 0.2 0.3 0.4 0.5

z / m

1

2

3

4

5

6

7

8

9

Au

Av

Aw

Explicit numerical modelling of dispersion hampered by sub-grid

scale processes

Thus, empirical modelling presents practical

alternative

Outstanding problems

• Rate of dispersal from street canyons – advection - turbulent diffusion – deposition

• Generalising dispersion into the neighbourhood

• Coagulation/condensation/nucleation/reactions – require timescales

• How indoor exposure is related to outdoor concentrations

• Rapid growth of megcities – a challenge for the 21st century

Thanks for your attention

Statistical variation in ultra-fine concentrations

N0.1/cm-3

1x103 10x103 100x103 1x106

n i/n

tota

l

0.00

0.05

0.10

0.15

0.20

channelledrecirculated

N0.1 / cm-3

1000 10000 100000

Cum

f

0.0

0.2

0.4

0.6

0.8

1.0channelledrecirculatedbackground

Spatial variation in urban PM10

PM10 heavily influenced by

•Resuspended dusts

•Long-range transport of secondary PM

… but does this represent the variation in exposure potential?

Emission/dispersion modelling leads to…

Edinburgh measurements (SASUA, 1999-2001)

SASUA diurnal particle number flux (May)

Below: sensible surface heat flux

0

5000

10000

15000

20000

25000

0:00 6:00 12:00 18:00 0:00Time

N (

cm-3

)

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

FN (

cm-2

s-1

)

ConcentrationFlux

-50

0

50

100

150

200

0:00 6:00 12:00 18:00 0:00

Time

H (

W m

-2)

Oct/NovMay

`

Dorsey et al., 2002, Atmos. Environ. 36, 791-800.

•Diurnal cycle in urban ventilation related to heat flux cycle

UFP dispersion in road corridors

Factors affecting dispersion

Ptrak data

Visualisation of road corridor concept