Arctic Haze and climate change
Arctic Haze Phenomenon, sources & transport Spatial-temporal characteristics 2 examples Effects on climate & modeling Arctic Haze & climate change
Arctic Haze: • sulphate, particulate organic matter, nitrogen compounds, dust, black carbon • median particle diameter ≤0.2 μm
Visibility, 1951-93, North America Summer
Winter
Hoff, 1994
Emissions of SO2 and NOx, year 2000
from NILU
• Few pollution sources within the Arctic itself • Arctic remote from major pollution sources Æ Long range transport plays an important role
Prevailing winds spread the pollution plume from the Norilsk smelters.
Transport of pollution into the Arctic
European tracer columns 2–4 days and 8–10 days after emission (composites represent the three Dec, Jan, Feb months with the lowest and highest NAO index within 1979–1993)
after 8-10 days
after 2-4 days NAO-
NAO+ Eckhardt et al., 2003 mg/m2
Æ NAO impacts transport of Arctic air pollution NAO+: more tracer is found in Arctic & poleward transport is faster dependence is strongest for European tracer
3 main mechanisms that contribute to Arctic Haze formation: • long-range transport from pollution sources outside the Arctic • borderline between Arctic and mid-latitude air masses southward extended • strong and persistent temperature inversions • removal processes very slow or absent Æ Combination of these facors result in transport of precursor gases and particulates to the Arctic and the trapping of pollutant haze for up to 15-30 days
Mean geopotential height at 500 hPa (in gpdam) January
July
Position of polar front 40-70°N; seasonal movement
Transport of pollution into the Arctic according to the polar front position
from AMAP
Mean temperature profiles, Arctic stations, Feb. 1987 Mean monthly median inversion top, base, and strength
Serreze et al., 1992
Overland et al., 1997
Monitoring stations
Barrow
Alert Nord Zeppelin Svanvik Karasjok
Janiskoski Oulanka
Monthly averaged sulphate concentration [μgS/m3] Subarctic
Arctic
-3
1.0
40 20 0
nss SO4
NO3
-
=
60x10
Alert (82.47篘 )
0.8 0.6 0.4 0.2 0.0 1/82 1/84 1/86 1/88 1/90 1/92 1/94 1/96 1/98 1/00 1/02
Quinn et al., 2007
trends in the aerosol chemical composition
Monthly averaged sulphate concentrations [μgS/m3], March & April 1.5
3.0
a) Alert
e) Karasjok
1.0
-63%
2.0
0.5
1.0
0.0 0.8
0.0 1.2
0.6
April
b) Nord
-63%
0.8
-48%
f) Svanvik +50%
0.4 0.4
0.2 0.0 0.6
c) Zeppelin 0.4 0.2
-27%
0.0 1.5
=
g) Janiskoski 1.0 0.5
April SO4
0.0 0.4 0.3
= March SO4 LT April Trend ST April Trend LT March Trend ST March Trend
0.0 1.2
h) Oulanka 0.8
-56%
0.2 0.1
0.4
d) Barrow
0.0
0.0 1975 1980 1985 1990 1995 2000 2005
1975 1980 1985 1990 1995 2000 2005
Quinn et al., 2007
trends in the aerosol optical properties
Monthly averaged light scattering (10-6m-1) at 550 nm for sub-10 micron aerosol
March 1982-1996: -63% April 1982-1996: -56%
March 1997-2006: +46%
AMAP, 2006
Temporal-spatial characteristics of Arctic Haze: • Seasonality: Winter-Spring max. Summer-Fall min. • Episodicity Events lasting from 1-10 days • Height: primarily in lowest 5 km, peak in lowest 2 km vertical layering, also well above ABL • Highly inhomogenous vertically (meters to 1 km thick) spatially (20-200 km horizontal extent) • Trends decreasing trend throughout the 1990s cause of some recent increase is not known
2 recent haze events in years 2000 & 2006 measured in Ny Alesund/Spitsbergen Aerosol optical depth, 535 nm
Clean atmosphere
March/April, 2000
May 2, 2006
AWI press release, 2006
Event in 2000
Variation of tropospheric aerosol optical depth at 532 nm
23.3. Haze
26.3. Background
DFG report, ASTAR 2000
Event in 2000
23.3.2000 12UTC 0
102
L
20
10
1020
°N
20
20
°N
H
°N
20
L
20
°N
20
L
10
L
26.3.2000 12UTC
10
20 10
00 10
L
H
H
0
101
L
H
0
0
100
H
H
1010
00
10
H
H
L
40°N
1020
20
°N
°N
20
H
H
H
H
10
10
1020
°N
20
H
H
H
10
L L
40°N
L
1010
20
L
H
L H
L
10
0
H
L
10
H
60°N
990
101
H
L
H
990
L
20
10
H
10
10
60°N
L
10
L L
L
10
1020
H
L
80°N
H
0
20
L
H
L 100
L
10
1010
H
L
H
H 1010
L
80°N
H
H
1010
H 1020
H
1020
0
2 10
H
L
L
L
10
10
L
1020
H
102
00
L
H
L H
H
10
10
10
1010
H L
H
L
L
0
102
H
10
10
1010
1020
L
H L
L
990
L
L
L
1020
H
L
H
0
101
H
°N
0
H
20
1000
20
10
H 103
1030
10
10
10
L
L
10
0 102
20
40°W
20°W
L 0°
20°E
40°E
from ECMWF
scattering coefficient (σs) km
-1
Event in 2000 Daily mean values of selected aerosol parameters, 26 March March 2000, Ny Alesund/Spitsbergen 20 March
- Zeppelin - Rabben
0.02
0.01
500
particle number (N) cm
-3
23 March 100 50
10 5
copper (Cu) ng m
-3
4 3 2 1 0 15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Treffeisen, pers. comm.
Event in 2000 Rel. humidity [%]
Extinction coefficient [1/km]
Wind direction
Height [km]
by LIDAR, 532 nm
Temperature [˚C] Wind speed [m/s] [km]
[km]
22. M鋜z (Background)
12:30
13:00
13:30 Zeit [UTC]
14:00
12. April (erh鰄 t)
0.050 0.030
14:30
0.050 0.030
0.020
0.020
0.015
0.015
0.010
0.010
0.008
0.008
0.006
0.006
14:00
15:00
16:00
17:00
Zeit [UTC]
DFG report, ASTAR 2000
Event in 2000
Height [m]
Relative frequency of different aerosol components
Relative frequency [%]
DFG report, ASTAR 2000
Event in 2006
View from Zeppelin station, Spitsbergen
26. 4.
2. 5.
Stohl et al., 2007
Event in 2006
3.5.2006 0 UTC
Trajectory analysis for May 2, 2006
0
102
L °N
°N
20
20
H
10
L 100
8
H
10
20
L
20
H
996 1020
984
L
L
H
100
8
996
HL 08
L
H
10
L H
L
L
L
L
H
1020
H
L
H
H 10
20
H
L H
L
L
H
L
H
L
H
8
100
H
L
L H
60°N
H
1008
6
99
H
H L
1020
L
L
32
H H
L
L
10
80°N
1032
H
H L
H H
H
L
20
10
L
20
°N
20
°N
40°N
H H
L 40°W
20°W
0°
20°E
L
L 40°E
from ECMWF
Treffeisen et al., 2007
Event in 2006
MODIS fire detection, 21.4.-5.5.
Daily number of fire detection & estimated burning area (north 40˚N, 20-60˚E)
Stohl et al., 2007
Event in 2006 Particle number concentration,
Zeppelin station
Relative contribution of different aerosol components to the total aerosol mass, Zeppelin station
Aerosol optical depth, 500 nm,
Ny Alesund
EM: elemental carbon OM: organic carbon SIA: secondary inorganic aerosol (SO42-, NO3-,NH4+) SS: sea salt (Na+, Cl-, Mg2+) Pot.+Calc.: K+, Ca2+
clean atmosphere
Stohl et al., 2007
Climatic effect of aerosol: direct & indirect effects
impact on atmospheric circulation and storm stracks impact on snow albedo
IPCC, 2007
Components of aerosol radiative forcing, TOA, global mean
IPCC, 2007
A modeling study with the regional climate model HIRHAM (prescribed aerosol) 1)
Specification of aerosol from Global Aerosol Data Set (GADS)
2)
Input from GADS into climate model: for each grid point in each vertical level: aerosol mass mixing ratio optical aerosol properties for short- and longwave spectral intervals f(RH)
3)
Climate model run with and without aerosol Æ aerosol radiative forcing
Global Aerosol Data Set (GADS); Koepke et al., 1997
Æ Arctic Haze: WASO, SOOT, SSAM
Consideration of aerosol in the climate model HIRHAM -direct effect-
Basic model’s thermodynamic equation: Temperature:
∂T/∂t = - ∙T + QT
; QT: radiation, convection, phase changes
Δ
Radiation code:
QTrad=(∂T/∂t)rad=g/cp ∂Fnetrad/∂p Radiative transfer equation within radiation code
Fnetrad (p)=Fsw(p)+Flw(p) =radiancesw(zenit angle,optical depth)+radiancelw(transmiss.,emiss.) Optical properties of aerosol and cloud particles
Direct effect of Arctic Haze “Aerosol run minus Control run”, March ensemble 2m temperature change
x W1
x C2
5
5
4
4 Höhe [km] Height [km]
Height [km]
x W2 x C1
Temperature profiles at selected points
Height-latitude temperature change
3 2 1
[°C]
0
70
75
80
Geographical latitude 0.60
[°C]0.50 ∆Fsrfc= 5 to –3 W/m2 1d radiative model studies: ∆Fsrfc=-0.2 to -6 W/m2
3 2 1
65
0.40 0.30 0.20 0.10 0.00 -0.10 -0.20 -0.30
85
W1
C2 C1
W2
0
-3
-2 -1 0 1 2 3 Temperature change [˚C]
1990
Fortmann, 2004
Direct climatic effect of Arctic aerosols in climate model HIRHAM via specified aerosol from GADS u(x,y,z) v(x,y,z) ps(x,y) T(x,y,z) q(x,y,z) qw(x,y,z) α(x,y) μ(x,y)
Effective aerosol distribution as function of (x,y,z)
Direct aerosol forcing in the vertical column
Additional diabatic heating source Qadd = Qsolar + QIR
Aerosol – Radiation Dynamical changes: Δu(x,y,z) Δv (x,y,z) Δps(x,y) ΔT(x,y,z) Δq(x,y,z) Δqw(x,y,z)
New effective aerosol distribution due to 8 humidity classes in the aerosol block
Circulation - Feedback
Direct effect of Arctic Haze “Aerosol run minus Control run”, March ensemble Sea level pressure change
500 hPa height change
[hPa]
[m]
mean SLP changes of ±3 hPa and 500 hPa height changes of ±15 m aerosol modifies the development and paths of cyclones (shown for North Atlantic and Pacific)
Rinke et. al., 2004
Cloud Radiative Forcing (CRF)
Clouds, radiation, Arctic Haze Arctic cloud radiative forcing (CRF) & Arctic haze
- CRF > 0 : Arctic clouds warm the surface (LW > SW)
- CRF depends also on cloud radiative properties (SW: reflect., LW: emiss.) which are changed by anthropogenic aerosol - Arctic stratus often thin (ε
Arctic Haze: • sulphate, particulate organic matter, nitrogen compounds, dust, black carbon • median particle diameter ≤0.2 μm
Visibility, 1951-93, North America Summer
Winter
Hoff, 1994
Emissions of SO2 and NOx, year 2000
from NILU
• Few pollution sources within the Arctic itself • Arctic remote from major pollution sources Æ Long range transport plays an important role
Prevailing winds spread the pollution plume from the Norilsk smelters.
Transport of pollution into the Arctic
European tracer columns 2–4 days and 8–10 days after emission (composites represent the three Dec, Jan, Feb months with the lowest and highest NAO index within 1979–1993)
after 8-10 days
after 2-4 days NAO-
NAO+ Eckhardt et al., 2003 mg/m2
Æ NAO impacts transport of Arctic air pollution NAO+: more tracer is found in Arctic & poleward transport is faster dependence is strongest for European tracer
3 main mechanisms that contribute to Arctic Haze formation: • long-range transport from pollution sources outside the Arctic • borderline between Arctic and mid-latitude air masses southward extended • strong and persistent temperature inversions • removal processes very slow or absent Æ Combination of these facors result in transport of precursor gases and particulates to the Arctic and the trapping of pollutant haze for up to 15-30 days
Mean geopotential height at 500 hPa (in gpdam) January
July
Position of polar front 40-70°N; seasonal movement
Transport of pollution into the Arctic according to the polar front position
from AMAP
Mean temperature profiles, Arctic stations, Feb. 1987 Mean monthly median inversion top, base, and strength
Serreze et al., 1992
Overland et al., 1997
Monitoring stations
Barrow
Alert Nord Zeppelin Svanvik Karasjok
Janiskoski Oulanka
Monthly averaged sulphate concentration [μgS/m3] Subarctic
Arctic
-3
1.0
40 20 0
nss SO4
NO3
-
=
60x10
Alert (82.47篘 )
0.8 0.6 0.4 0.2 0.0 1/82 1/84 1/86 1/88 1/90 1/92 1/94 1/96 1/98 1/00 1/02
Quinn et al., 2007
trends in the aerosol chemical composition
Monthly averaged sulphate concentrations [μgS/m3], March & April 1.5
3.0
a) Alert
e) Karasjok
1.0
-63%
2.0
0.5
1.0
0.0 0.8
0.0 1.2
0.6
April
b) Nord
-63%
0.8
-48%
f) Svanvik +50%
0.4 0.4
0.2 0.0 0.6
c) Zeppelin 0.4 0.2
-27%
0.0 1.5
=
g) Janiskoski 1.0 0.5
April SO4
0.0 0.4 0.3
= March SO4 LT April Trend ST April Trend LT March Trend ST March Trend
0.0 1.2
h) Oulanka 0.8
-56%
0.2 0.1
0.4
d) Barrow
0.0
0.0 1975 1980 1985 1990 1995 2000 2005
1975 1980 1985 1990 1995 2000 2005
Quinn et al., 2007
trends in the aerosol optical properties
Monthly averaged light scattering (10-6m-1) at 550 nm for sub-10 micron aerosol
March 1982-1996: -63% April 1982-1996: -56%
March 1997-2006: +46%
AMAP, 2006
Temporal-spatial characteristics of Arctic Haze: • Seasonality: Winter-Spring max. Summer-Fall min. • Episodicity Events lasting from 1-10 days • Height: primarily in lowest 5 km, peak in lowest 2 km vertical layering, also well above ABL • Highly inhomogenous vertically (meters to 1 km thick) spatially (20-200 km horizontal extent) • Trends decreasing trend throughout the 1990s cause of some recent increase is not known
2 recent haze events in years 2000 & 2006 measured in Ny Alesund/Spitsbergen Aerosol optical depth, 535 nm
Clean atmosphere
March/April, 2000
May 2, 2006
AWI press release, 2006
Event in 2000
Variation of tropospheric aerosol optical depth at 532 nm
23.3. Haze
26.3. Background
DFG report, ASTAR 2000
Event in 2000
23.3.2000 12UTC 0
102
L
20
10
1020
°N
20
20
°N
H
°N
20
L
20
°N
20
L
10
L
26.3.2000 12UTC
10
20 10
00 10
L
H
H
0
101
L
H
0
0
100
H
H
1010
00
10
H
H
L
40°N
1020
20
°N
°N
20
H
H
H
H
10
10
1020
°N
20
H
H
H
10
L L
40°N
L
1010
20
L
H
L H
L
10
0
H
L
10
H
60°N
990
101
H
L
H
990
L
20
10
H
10
10
60°N
L
10
L L
L
10
1020
H
L
80°N
H
0
20
L
H
L 100
L
10
1010
H
L
H
H 1010
L
80°N
H
H
1010
H 1020
H
1020
0
2 10
H
L
L
L
10
10
L
1020
H
102
00
L
H
L H
H
10
10
10
1010
H L
H
L
L
0
102
H
10
10
1010
1020
L
H L
L
990
L
L
L
1020
H
L
H
0
101
H
°N
0
H
20
1000
20
10
H 103
1030
10
10
10
L
L
10
0 102
20
40°W
20°W
L 0°
20°E
40°E
from ECMWF
scattering coefficient (σs) km
-1
Event in 2000 Daily mean values of selected aerosol parameters, 26 March March 2000, Ny Alesund/Spitsbergen 20 March
- Zeppelin - Rabben
0.02
0.01
500
particle number (N) cm
-3
23 March 100 50
10 5
copper (Cu) ng m
-3
4 3 2 1 0 15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Treffeisen, pers. comm.
Event in 2000 Rel. humidity [%]
Extinction coefficient [1/km]
Wind direction
Height [km]
by LIDAR, 532 nm
Temperature [˚C] Wind speed [m/s] [km]
[km]
22. M鋜z (Background)
12:30
13:00
13:30 Zeit [UTC]
14:00
12. April (erh鰄 t)
0.050 0.030
14:30
0.050 0.030
0.020
0.020
0.015
0.015
0.010
0.010
0.008
0.008
0.006
0.006
14:00
15:00
16:00
17:00
Zeit [UTC]
DFG report, ASTAR 2000
Event in 2000
Height [m]
Relative frequency of different aerosol components
Relative frequency [%]
DFG report, ASTAR 2000
Event in 2006
View from Zeppelin station, Spitsbergen
26. 4.
2. 5.
Stohl et al., 2007
Event in 2006
3.5.2006 0 UTC
Trajectory analysis for May 2, 2006
0
102
L °N
°N
20
20
H
10
L 100
8
H
10
20
L
20
H
996 1020
984
L
L
H
100
8
996
HL 08
L
H
10
L H
L
L
L
L
H
1020
H
L
H
H 10
20
H
L H
L
L
H
L
H
L
H
8
100
H
L
L H
60°N
H
1008
6
99
H
H L
1020
L
L
32
H H
L
L
10
80°N
1032
H
H L
H H
H
L
20
10
L
20
°N
20
°N
40°N
H H
L 40°W
20°W
0°
20°E
L
L 40°E
from ECMWF
Treffeisen et al., 2007
Event in 2006
MODIS fire detection, 21.4.-5.5.
Daily number of fire detection & estimated burning area (north 40˚N, 20-60˚E)
Stohl et al., 2007
Event in 2006 Particle number concentration,
Zeppelin station
Relative contribution of different aerosol components to the total aerosol mass, Zeppelin station
Aerosol optical depth, 500 nm,
Ny Alesund
EM: elemental carbon OM: organic carbon SIA: secondary inorganic aerosol (SO42-, NO3-,NH4+) SS: sea salt (Na+, Cl-, Mg2+) Pot.+Calc.: K+, Ca2+
clean atmosphere
Stohl et al., 2007
Climatic effect of aerosol: direct & indirect effects
impact on atmospheric circulation and storm stracks impact on snow albedo
IPCC, 2007
Components of aerosol radiative forcing, TOA, global mean
IPCC, 2007
A modeling study with the regional climate model HIRHAM (prescribed aerosol) 1)
Specification of aerosol from Global Aerosol Data Set (GADS)
2)
Input from GADS into climate model: for each grid point in each vertical level: aerosol mass mixing ratio optical aerosol properties for short- and longwave spectral intervals f(RH)
3)
Climate model run with and without aerosol Æ aerosol radiative forcing
Global Aerosol Data Set (GADS); Koepke et al., 1997
Æ Arctic Haze: WASO, SOOT, SSAM
Consideration of aerosol in the climate model HIRHAM -direct effect-
Basic model’s thermodynamic equation: Temperature:
∂T/∂t = - ∙T + QT
; QT: radiation, convection, phase changes
Δ
Radiation code:
QTrad=(∂T/∂t)rad=g/cp ∂Fnetrad/∂p Radiative transfer equation within radiation code
Fnetrad (p)=Fsw(p)+Flw(p) =radiancesw(zenit angle,optical depth)+radiancelw(transmiss.,emiss.) Optical properties of aerosol and cloud particles
Direct effect of Arctic Haze “Aerosol run minus Control run”, March ensemble 2m temperature change
x W1
x C2
5
5
4
4 Höhe [km] Height [km]
Height [km]
x W2 x C1
Temperature profiles at selected points
Height-latitude temperature change
3 2 1
[°C]
0
70
75
80
Geographical latitude 0.60
[°C]0.50 ∆Fsrfc= 5 to –3 W/m2 1d radiative model studies: ∆Fsrfc=-0.2 to -6 W/m2
3 2 1
65
0.40 0.30 0.20 0.10 0.00 -0.10 -0.20 -0.30
85
W1
C2 C1
W2
0
-3
-2 -1 0 1 2 3 Temperature change [˚C]
1990
Fortmann, 2004
Direct climatic effect of Arctic aerosols in climate model HIRHAM via specified aerosol from GADS u(x,y,z) v(x,y,z) ps(x,y) T(x,y,z) q(x,y,z) qw(x,y,z) α(x,y) μ(x,y)
Effective aerosol distribution as function of (x,y,z)
Direct aerosol forcing in the vertical column
Additional diabatic heating source Qadd = Qsolar + QIR
Aerosol – Radiation Dynamical changes: Δu(x,y,z) Δv (x,y,z) Δps(x,y) ΔT(x,y,z) Δq(x,y,z) Δqw(x,y,z)
New effective aerosol distribution due to 8 humidity classes in the aerosol block
Circulation - Feedback
Direct effect of Arctic Haze “Aerosol run minus Control run”, March ensemble Sea level pressure change
500 hPa height change
[hPa]
[m]
mean SLP changes of ±3 hPa and 500 hPa height changes of ±15 m aerosol modifies the development and paths of cyclones (shown for North Atlantic and Pacific)
Rinke et. al., 2004
Cloud Radiative Forcing (CRF)
Clouds, radiation, Arctic Haze Arctic cloud radiative forcing (CRF) & Arctic haze
- CRF > 0 : Arctic clouds warm the surface (LW > SW)
- CRF depends also on cloud radiative properties (SW: reflect., LW: emiss.) which are changed by anthropogenic aerosol - Arctic stratus often thin (ε
