Yoram Kaufman,

1
Remote Sensing of Aerosols over land RS of
radiative forcing

Yoram Kaufman, NASA Goddard SFC
2
Remote Sensing of Aerosols over land radiative
forcing
Climate change --gt Which aerosol measurements
are needed? Radiative effects of aerosol ltgt
remote sensing two mirror sides of the
same issue Remote sensing from MODIS over the
land Black carbon absorption of sunlight why
is it important? Why is it complex? And how do we
remote sense it. Remote sensing of Radiative
effects
3
- Effect of aerosol on climate Cooling past
climates, possibly warming future climates -
Effect of aerosol on hydrologic cycle Less
evaporation from cooler land and ocean, more
stable atmosphere, less clouds and
precipitation. - Effect of aerosol on health
May be more important then ozone in causing
cancer and heart problems. - Effect on
agriculture, vegetation Shift of precipitation
away from polluted land, less sunlight to
vegetation
4
Aerosol, their sources and effects on climate
Properties Net effect Aerosol
type Main Source Reflect Cool
the earth Desert dust, dry lake
beds sunlight sulfate smog
industry Absorb Heat the earth
Black carbon biomass burning sunlight
atmosphere dirty
engines reduce cloudiness Cloud
brighter clouds sulfate smog
industry Condensation less precipitation
smoke fires Nuclei
5
Remote sensing of Aerosol Open questions -
Where does aerosol begin and cloud ends?
- Does aerosol in cloud free area
represent the aerosol that interacts with clouds?
- How to handle the spatial
and temporal variability of aerosol properties?
Haze layer
Clean atmosphere
Haze layer
Toxic trucks Washington Post 11/19/00
Trouble in the greenhouse Nature, Sept., 2000
6
Does Population cause Pollution ?
POLDER aerosol index Feb. 1997
population density (Kaufman, Tanré Boucher,
Nature 2002)
7
The Brown cloud
Fine aerosol optical thickness - smoke and
pollution
Coarse aerosol optical thickness - dust and sea
salt
8
Mottarone, Italy, June 2001
Towards equal distribution of Pollution around
the world
Marmin, Nepal March 2001 (V. Ramanathan)
Dust and pollution over Lago Magiore, Italy
9
Pollution over India The view from Tibet Sea of
pollution Over India Image from the Shuttle
10
On December 18, 1999 Terra was launched into
Earth Orbit. Monitor global aerosol, radiative
fluxes, fires, burn scars,land use, CO
1993-2001--gt Aerosol Robotic Network (AERONET) is
launched Monitor detailed properties of
aerosol
11
Which aerosol properties should be
measured? Climate Cooling by reflection to
space - spectral flux reflected to space -
DFlTOA Atmospheric warming - aerosol absorption
- wo gt tabs , DFlSUR Brightening clouds number
of hygroscopic particles gt0.05 µm radius, size
distribution - CCN (r) Hydrologic cycle Less
evaporation - change in temperature profile -
aerosol absorption profile - Dtabs (H) Shift in
precipitation - interplay of fine vs. coarse CCNs
- CCNfine , CCNcoarse Effect on health ?? -
Back carbon, sulfates, organics
12
Radiative effects of aerosol lt--gt remote sensing
13
What is radiative forcing by aerosols?
DFlTOA
DFlSUR
14
Remote sensing and radiative forcing of climate
at TOA two faces of the same radiative transfer
problem
DFlTOA
DFlSUR
Radiance measured by satellites is part of the
radiation field reflected to space the
radiative forcing at the top of the atmosphere
15
Remote sensing and radiative forcing Two faces of
the same radiative transfer problem (described
here using the single scattering approximation)
r
16
MODIS RS of aerosol over land
visible
rl rol(t) Tl(t)rsl In the visible
MODIS observes smoke and the surface In the mid
Infra-Red there is no smoke MODIS observes the
surface The difference the smoke
concentration or optical thickness (not as
sensitive to dust)
l(µm) 0.47 0.55 0.66 1.2 1.6 2.1
Fires in the North-West
mid IR
(Kaufman, Tanré Boucher, Nature 2002)
17
MODIS aerosol over the land
rl rol(t) Tl(t)rsl
18
Can surface reflectance at 2.1 µm predict the
surface reflectance in the visible? Empirical
measurements
19
Model analysis
20
rl rol(t) Tl(t)rsl rol(t) Pl(q)woltl
Pl(q) lt-- dV(r)/dlnr
21
Application to MODIS Dust and haze in East Asia
- MODIS March 20, 2001
22
MODIS Aerosol optical thickness
23
MODIS Aerosol optical thickness of coarse dust
and fine pollution
24
Validation MODIS vs. AERONET (Aug 2000 to Nov.
2001)
Levy Remer
25
MODIS Validation land
AERONET
AERONET
Allen Chu et al 2002
26
SAFARI-2000
MODIS underestimates AOT vs. AERONET during
SAFARI-2000 especially in Zambia with heavy fresh
smoke. Lower SSA (w0) suspected.
27
Validation of monthly averages MODIS 3x3 vs.
AERONET
AERONET measurements are in cloud free
conditions. We validate MODIS AOT weighted by the
cloud free area Sti(1-cfi)/S(1-cfi)
0.2t
28
Sti(1-cfi)/S(1-cfi)
Sti/Si
29
Aerosol examples in 2000
(a)
(b)
(c)
Asian dust outbreak air pollution (April 7)
Indian pollution (October 12)
African smoke dust (March 11)
Air pollution in the E. US (October 23)
Air pollution in W. Europe (March 19)
(d)
(e)
(f)
Central American Smoke (April 19)
30
Frequency maps of aerosol retrievals
(June - August 2001)
(March - May 2001)
(December 2000 - January 2001)
(September - November 2001)
31
Average optical thickness
32
Example pollution accumulating under the Alps
MODIS aerosol optical thickness
Ispra
Milan
Venice
MODIS RGB
Allen Chu, SSAI/GSFC
33
Example Los Angeles
100 km
3.0
UCLA
50 km
2.0
AOT at 0.55 ?m
1.0
0.0
34
Example Planet Earth
India
MODIS Monthly Mean Aerosol optical thickness
(0.55 mm) in 4 Regions
Europe
35
Additional information that helps define the
aerosol
Wild fires North-West US MODIS Aug. 23 2000
Fires smoke
MOPITT CO Aug 22-27
36
Aerosol transport 2/16 Africa --gt Europe
4/16 China--gtUSA 5/5 Mexico, Arabian
sea---gt7/12 dust,smoke Africa 8/26 smoke
from South America and Africa
37
(No Transcript)
38
Dust and pollution
Smoke/Pollution?
Saharan dust
Sea Salt
39
Smoke Pollution?
Saharan dust
Pollution/dust in India
Sea Salt
40
pollution
Smoke
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
West
East
46
Aug. 26
1
Ocean
47
Ocean
Ocean
48
April 16
Ocean
49
(No Transcript)
50
April 16
51
May
July
52
Aug.
Sept
Oct
Nov
53
Black carbon absorption of sunlight Why is it
important? Why is it complex? And how do we
remote sense it.
54
Aerosol absorption - Black Carbon growing impact
A scenario for additional climate forcing between
2000 and 2050. Reduction of BC moves the aerosol
forcing to lower, negative rather than positive
level (Hansen et al., 2000)
Pollution over Indian subcontinent - How do we
measure BC?
55
Large Variability in BC Absorption Properties!!
Microphysical Properties of Aerosol Particles
e.g. Light Absorption Efficiencies and BC
contents
0.27
0.54
aabs 1 10m2/g
aabs 1 25m2/g
0.78
aabs 2 30m2/g
Very Large Uncertainties in Climate Radiative
Forcing!!!
Martins et al., 1998
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
Aerosol climatology from AERONET (Dubovik et al
2002)
No absor-ption
Cooling Heating
25 absor-ption
63
(No Transcript)
64
(No Transcript)
65
MODIS - Distinguishes the (Lat., Long., Time) of
fine and coarse particles - Measures the effect
of aerosol on reflection of sunlight to
space AERONET - Measures the aerosol size
distribution - Measures the aerosol absorption of
sunlight - Measures the angular distribution of
sky radiation --gt effect of aerosol on
transmission of sunlight to the surface
Can we use this information to distinguish
measure the human impact on climate through
aerosol?
66
MODIS and models for September 2000
MODIS
Model M. Chin et al., 2002
Natural fine aerosol
Fine anthropogenic
Fine aerosol
0.0 0.2
0.4
Coarse aerosol
Coarse aerosol
67
Global results - September 2000 Modis monthly
composite aerosol fine and coarse mode aerosol
forcing of climate
Fine mode
Aerosol optical thickness
Coarse mode aerosol
68
Aerosol AERONET climatology (Dubovik et al 2001)
coarse
coarse
fine
fine

Aerosol radius
tfine /ttot.
1-wo ()
69
MODIS AERONET --gt forcing
Aerosol radius
70
ARERONET analysis of spectral flux at the surface
71
What was it all about? Climate, research
requires to know tfine, tcoarse, fluxes, Rfine,
Rcoarse, nr, wo , profile, CCNs
clouds MODIS lidars in
space Spectral polarization measurements A
ERONET (150 locations) Space observations of the
structure of fine and coarse aerosol AERONET
realistic models gt the anthropogenic
component Validation, new processes -gt
concentrated field experiments
72
Question If a box of corn-flakes is used to make
fine aerosol what area can it cover with the
average AOT of 0.2?
A) 1 cm2 B) 1 m2 C) 100 m2 D) 10,000 m2 E) 1
km2 F) 100 km2
1gr 1cm3 Q1, extinction is pr2 Box is 500 gr