OBSERVATION OF ATMOSPHERIC COMPOSITION FROM SPACE

1
OBSERVATION OF ATMOSPHERIC COMPOSITION FROM SPACE
Colette L. Heald ATS 737, October 15, 2008
With material from Daniel J. Jacob (Harvard),
Andreas Richter (Bremen), Cathy Clerbaux (Service
dAéronomie)
2
WHAT IS THE EFFECT OF ATMOSPHERIC COMPOSITION ON
RADIATION?

• OBSERVED RADIATION includes
• Reflection (solar, UV-visible)
• Emission (Earth/atmosphere, IR)
• Absorption (by gases and particles)
• Scattering (by gases and particles)

Absorption and emission spectra provide a means
of identifying and measuring the composition of
the atmosphere. Radiation interacts with gases
via (1) Ionization-dissociation
(UV-visible) (2) Electronic transitions
(UV-visible) (3) Vibrational transitions
(IR) (4) Rotational transitions (far IR and
microwave) ? IR spectra of many molecules is a
combination of (3) and (4)

• Instead of discrete lines, transitions are
  observed in a whole wavelength region.
• natural line broadening (upper stratosphere,
  mesosphere)
• Doppler broadening (upper atmosphere gt 40 km)
• pressure broadening (lower atmosphere lt 40 km)

Convolution Voigt lines
3
EXAMPLES OF ABSORPTION SPECTRA
4
ALL TOGETHER NOW
5
STRATOSPHERIC OZONE HAS BEEN MEASURED FROM SPACE
SINCE 1979
Method UV solar backscatter
l2
l1
Ozone layer
Scattering by Earth surface and atmosphere
Ozone absorption spectrum
l1
l2
6
SATELLITE OBSERVATIONS REVEAL THE MECHANISM FOR
POLAR OZONE LOSS AND HELP US TRACK OZONE RECOVERY
DU
Southern hemisphere ozone column seen from TOMS,
October
MLS ClO
TOMS O3
Polar ozone depletion driven by halocarbon
break-down (source of ClO)
1 Dobson Unit (DU) 0.01 mm O3 STP 2.69x1016
molecules cm-2
7
ATMOSPHERIC COMPOSITION RESEARCH IS NOW MORE
DIRECTED TOWARD THE TROPOSPHERE
Air quality, climate change, ecosystem issues
but tropospheric composition measurements from
space are difficult optical interferences from
water vapor, clouds, aerosols, surface, ozone
layer
but tropospheric composition measurements from
space are difficult optical interferences from
water vapor, clouds, aerosols, surface, ozone
layer
8
OBSERVING TROPOSPHERIC COMPOSITION
Observing system is presently very sparse
satellites will change this
SATELLITES
AIRCRAFT CAMPAIGNS
SURFACE SITES
Long-term monitoring at the surface
Chemical characterization throughout the
troposphere
Continuous, global measurements
9
WHY OBSERVE TROPOSPHERIC COMPOSITION FROM SPACE?
Global/continuous measurement capability
important for range of issues
Monitoring and forecasting of air quality ozone,
aerosols
Long-range transport of pollution
Monitoring of sources pollution and
greenhouse gases
Radiative forcing

• solar backscatter
• thermal emission
• solar occultation
• lidar

FOUR OBSERVATION METHODS
10
SOLAR BACKSCATTER MEASUREMENTS (UV to near-IR)
Examples TOMS, GOME, SCIAMACHY, MODIS, MISR,
OMI, OCO
absorption
l1
l2
z
l1
l2
wavelength
Retrieved column in scattering atmosphere depends
on vertical profile need chemical transport and
radiative transfer models
Scattering by Earth surface and by atmosphere
concentration

• Daytime only
• Column only
• Interference from stratosphere

• sensitivity to lower troposphere
• small field of view (nadir)

Pros
Cons
11
THERMAL EMISSION MEASUREMENTS (IR, mwave)
Examples MLS, IMG, MOPITT, MIPAS, TES, HIRDLS,
IASI
NADIR VIEW
LIMB VIEW
elIl(T1)
T1
Absorbing gas

• versatility (many species)
• small field of view (nadir)
• vertical profiling

Pros
Il(To)
To
EARTH SURFACE

• low S/N in lower troposphere
• water vapor interferences
• cannot see through clouds

Cons
12
OCCULTATION MEASUREMENTS (UV to near-IR)
Examples SAGE, POAM, GOMOS
satellite sunrise
Tangent point retrieve vertical profile of
concentrations
EARTH

• sparse data, limited coverage
• upper troposphere only
• low horizontal resolution

• large signal/noise
• vertical profiling

Pros
Cons
13
LIDAR MEASUREMENTS (UV to near-IR)
Examples LITE, GLAS, CALIPSO
Pros

• High vertical resolution

Laser pulse

• Aerosols only (so far)
• Limited coverage

Cons
Intensity of return vs. time lag measures
vertical profile
backscatter by atmosphere
EARTH SURFACE
14
ALL ATMOSPHERIC COMPOSITION DATA SO FAR HAVE BEEN
FROM LOW-ELEVATION, SUN-SYNCHRONOUS POLAR
ORBITERS

• Altitude 1,000 km
• Observation at same time of day everywhere
• Period 90 min.
• Coverage is global but sparse

15
TROPOSPHERIC COMPOSITION FROM SPACEplatforms,
instruments, species
Platform multiple multiple ERS-2 ADEOS Terra Terra Envisat Envisat Aqua Space station Aura Aura Aura Aura MetOp-A
Sensor TOMS AVHRR/SeaWIFS GOME IMG MOPITT MODIS/ MISR SCIAMACHY MIPAS AIRS SAGE-3 TES OMI MLS HIRDLS CALIPSO IASI OCO
Launch 1979 1995 1996 1999 1999 2002 2002 2002 2004 2004 2004 2004 2004 2004 2007 2009
O3 X X X X X X X X X
CO X X X X X X X
CO2 X X X
NO X
NO2 X X X X
HNO3 X X X
CH4 X X X
HCHO X X X
SO2 X X X X
BrO X X X
CH3CN X
aerosol X X X X X X X
16
OBSERVING TROPOSPHERIC OZONE AND ITS SOURCES FROM
SPACE
Nitrogen oxide radicals NOx NO NO2 Sources
combustion, soils, lightning Methane Sources
wetlands, livestock, natural gas Nonmethane VOCs
(volatile organic compounds) Sources vegetation,
combustion CO (carbon monoxide) Sources
combustion, VOC oxidation
Tropospheric ozone precursors
17
A NEEDLE IN A HAYSTACK DERIVING TROPOSPHERIC
OZONE

• Issues
• high uncertainty
• seasonal averages only
• does not extend to high latitudes

Fishman and Larson, 1987 Fishman et al., 2008
18
FIRST REMOTE MEASUREMENTS OF CO MAPS ABOARD THE
SPACE SHUTTLE
Gas-correlation radiometer (IR 4.7 ?m) flew 4
times between 1981 and 1994
APR 1994
OCT 1994
Connors et al., 1999 Reichle et al., 1999
19
RETRIEVALS IN THE IR THE STANDARD INVERSE PROBLEM
Characteristic absorption features in the
IR. Use a known T profile to estimate the
constituents
INVERSE PROBLEM solution is not
unique! SOLUTION maximum a posteriori
Averaging kernel (A) describes the relative
weighting of the true mixing ratio (x) at each
level to the retrieved value ( )
Typical MOPITT Averaging Kernel
20
MOPITT FIRST SATELLITE INSTRUMENT TARGETTING
TROPOSPHERIC POLLUTION
Spring 2001
MOPITT CO Column
CO Column over the NE Pacific in Spring 2001
MOPITT solid Model dotted
MOPITT Model
Observations used to track transpacific transport
of pollution
Comparison indicates that emission inventories
may be inaccurate
Heald et al., 2004
21
POLLUTION AND BIOMASS BURNING OUTFLOW DURING
ICARTT AIRCRAFT MISSION (Jul-Aug 2004)
NEAR-REAL-TIME DATA FOR CO COLUMNS ON JULY 18
AIRS
GEOS-Chem Model
Alaskan fires
U.S. pollution
Asian pollution
Wallace McMillan (UMBC)
Turquety et al., 2006
22
USING MODIS TO MAP FIRESAND MOPITT CO TO OBSERVE
EMISSIONS
Bottom-up emission inventory (Tg CO) for North
American fires in Jul-Aug 2004
From above-ground vegetation
From peat
9 Tg CO
18 Tg CO
MOPITT CO Summer 2004
GEOS-Chem CO x MOPITT AK
with peat burning
without peat burning
MOPITT data support large peat burning source,
pyro-convective injection to upper troposphere
Turquety et al., 2006
23
USING ADJOINTS OF CHEMICAL TRANSPORT MODELS TO
INVERT FOR EMISSIONS WITH HIGH RESOLUTION
MOPITT daily CO columns (Mar-Apr 2001)
Correction to model sources of CO
Inverse of atmospheric model
A priori emissions from Streets et al. 2003
and Heald et al. 2003
Kopacz et al., 2008
24
CONSTRAINING NOx AND REACTIVE VOC EMISSIONS
USING SOLAR BACKSCATTER MEASUREMENTSOF
TROPOSPHERIC NO2 AND FORMALDEHYDE (HCHO)
GOME 320x40 km2 SCIAMACHY 60x30 km2 OMI 24x13
km2
Tropospheric NO2 column ENOx Tropospheric HCHO
column EVOC
2 km
hn (420 nm)
hn (340 nm)
BOUNDARY LAYER
NO2
NO
HCHO
OH
CO
hours
O3, RO2
hours
VOC
1 day
HNO3
Emission
Emission
Deposition
VOLATILE ORGANIC COMPOUNDS (VOC)
NITROGEN OXIDES (NOx)
25
DIFFERENTIAL OPTICAL ABSORPTION SPECTROSCOPY
Pioneered for stratospheric ozone, used for
detection in UV-visible

• Use multiple wavelengths to characterize optical
  absorption of a species.
• ? determine the amount of absorber along the
  light path (slant column, ?s)

Scattering by Earth surface and by atmosphere
Vertical column
Air mass factor (AMF) depends on the viewing
geometry, the scattering properties of the
atmosphere, and the vertical distribution of the
absorber
Or alternate of DOAS direct fit of GOME
backscattered spectrum in 338-356 nm HCHO band
Requires an RT model and a CTM
Chance et al. 2000
26
AMF FORMULATION FOR A SCATTERING ATMOSPHERE
w(z) GOME sensitivity (scattering weight),
determined from LIDORT radiative transfer model
including clouds and aerosols S(z) normalized
mixing ratio (shape factor) from GEOS-Chem
CTM AMFG geometric air mass factor (no scatter)
AMFG 2.08 actual AMF 0.71
Palmer et al., 2001
27
GOME CONSTRAINTS ON NOx EMISSIONS
GEOS-CHEM model (GEIA)
Tropospheric NO2 Columns
GOME
JJA 1997
r 0.75 bias5
1015 molecules cm-2
Error weighting
Martin et al. 2003
A priori emissions (GEIA)
A posteriori emissions
Difference
28
HIGHER SPATIAL RESOLUTION FROM SCIAMACHY
Launched in March 2002 aboard Envisat
60x30 km2
320x40 km2
Potential for finer resolution of sources, but
need to account for transport will complicate
the inversion
29
TROPOSPHERIC NO2 FROM OMI CONSTRAINT ON NOx
SOURCES
October 2004
K. Folkert Boersma (KNMI)
30
NOX MEASUREMENTS REVEAL TRENDS IN DOMESTIC
EMISSIONS
NO2 emissions in US, EU and Japan decline
while emissions growing in China
East-Central China
Importance of long-term record!
Richter et al., 2005 Fishman et al., 2008
31
FORMALDEHYDE COLUMNS MEASURED BY GOME (JULY 1996)
2.5x1016 molecules cm-2
2
1.5
1
detection limit
0.5
South Atlantic Anomaly (disregard)
0
-0.5
High HCHO regions reflect VOC emissions from
fires, biosphere, human activity
32
RELATING HCHO COLUMNS TO VOC EMISSION
hn (340 nm), OH
oxn.
VOCi
HCHO
k 0.5 h-1
yield yi
Emission Ei
smearing, displacement
In absence of horizontal wind, mass balance for
HCHO column WHCHO
Local linear relationship between HCHO and E
but wind smears this local relationship between
WHCHO and Ei depending on the lifetime of the
parent VOC with respect to HCHO production
Isoprene
WHCHO
a-pinene
propane
detection limit
Distance downwind
100 km
VOC source
33
SEASONAL VARIATION OF GOME FORMALDEHYDE COLUMNS
reflects seasonal variation of biogenic isoprene
emissions
GOME GEOS-Chem (GEIA)
GOME GEOS-Chem (GEIA)
JUL
MAR
AUG
APR
SEP
MAY
JUN
OCT
Abbot et al., 2003
34
AEROSOLS FROM SPACE

• MIE SCATTERING
• scattering on large particles (aerosols,
  droplets, suspended matter in liquids)
• explained by coherent scattering from many
  individual particles
• for spherical particles, Mie scattering can be
  computed from the refractive index using the
  Maxwell equations
• wavelength of incoming radiation is not changed
• angular distribution is changed
• depending on ?, forward scattering is strongly
  favoured
• effectiveness of Mie scattering is proportional
  to ?sMie (?) ? ?-1 ... ?-1.5
• in general, Mie scattering is not polarising

Usually in visible
Extinction Scattering Absorption
To retrieve aerosol optical depth need aerosol
properties (size distribution, index of
refraction). Can use wavelength dependence to
get idea of composition/size ISSUE Need to
characterize Rayleigh scattering and surface
reflectance (including sun glint) ? thus easier
over oceans (dark surfaces)
MODIS
MISR

• Depending on the ratio of the size of the
  scattering particle (r) to the wavelength (?) of
  the light
• Mie parameter ? 2? r / ?,
• different regimes of atmospheric scattering can
  be distinguished.

MULTI-ANGLE 9 cameras (visible)
MULTI-SPECTRAL 7 bands from 0.4 2.1 µm
35
TRANSPACIFIC TRANSPORT OF ASIAN AEROSOL POLLUTION
AS SEEN BY MODIS
Detectable sulfate pollution signal correlated
with MOPITT CO
Heald et al., 2006
36
MAPPING SURFACE PM2.5 USING MISR (2001 data)
MISR AOD (annual mean)
Validation with AERONET
R20.80 Slope0.88
Convert AOD to surface PM2.5 using GEOS-CHEM
GOCART scaling factors
MISR PM2.5
EPA (FRMSTN) PM2.5
Evaluate against EPA station data R 0.78,
Slope 0.91
Liu et al.,2004
37
NASA AURA SATELLITE (launched July 2004)
Polar orbit four passive instruments observing
same air mass within 14 minutes
Tropospheric measurement capabilities

• OMI UV/Vis solar backscatter
• NO2, HCHO. ozone, BrO columns
• TES high spectral resolution thermal IR
  emission
• nadir ozone, CO
• limb ozone, CO, HNO3
• MLS microwave emission
• limb ozone, CO (upper troposphere)
• HIRDLS high vertical resolution thermal IR
  emission
• ozone in upper troposphere/lower stratosphere

38
TROPOSPHERIC OZONE OBSERVED FROM SPACE
IR emission measurement from TES
UV backscatter measurement from GOME
GOME JJA 1997 tropospheric columns (Dobson Units)
Coincident CO measurements from TES
Coincidental observations of CO and O3 with TES
allows us to look at ozone production
Liu et al., 2006
Zhang et al., 2006
39
OBSERVING CO2 FROM SPACEOrbiting Carbon
Observatory (OCO) to be launched in 2009
Polar-orbiting solar backscatter instrument,
measures CO2 absorption at 1.61 and 2.06 mm, O2
absorption (surface pressure) at 0.76 mm global
mapping of CO2 column mixing ratio with 0.3
precision
Pressure (hPa)
Averaging kernel
(sensitivity)
OCO will provide powerful constraints on regional
carbon fluxes
40
LOOKING TOWARD THE FUTURE GEOSTATIONARY ORBIT

• UV-IR sensors would provide continuous
  high-resolution mapping (1 km)
• on continental scale boon for air quality
  monitoring and forecasting

NRC Decadal Survey Recommendation GEO-CAPE in
2013-2016, with Aura-like GACM in 2016-2020 (also
ACE for aerosols 2013-2016)
NRC, 2007