Weighting functions (Box AMFs) for Limb measurements of stratospheric trace species
Space-Based Measurements of Stratospheric … Measurements of Stratospheric Aerosols Larry W....
Transcript of Space-Based Measurements of Stratospheric … Measurements of Stratospheric Aerosols Larry W....
6/17/2003 L. Thomason 1
Spaced-Based Measurements
of Stratospheric Aerosols
Larry W. ThomasonNASA Langley Research CenterHampton, Virginia USA
6/17/2003 L. Thomason 2
Measurement by Extinction of Solar Radiation
• Stratospheric aerosols are highly variable (102) on the decadal time scale
• Since 1978, variability has been dominated by a few significant volcanic events
• Loading in 2002 was at the lowest levels ever observed
• Focus of the presentation:– How are stratospheric
aerosol measured from space?
– What are the strengths and limitations of these measurements?
SAM II, SAGE I & II Stratospheric Aerosol V6.10
1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
80S
60S
40S
20S
EQ
20N
40N
60N
80N
St. HelensEl Chichon
RuizKelut
PinatuboHudson1000-nm Optical Depth
<6.x10-4 10-3 10-2 10-1 >2x10-1
6/17/2003 L. Thomason 3
Solar Occultation Measurement Strategy
• Measure line-of-sight transmission during each sunrise/sunset encountered by space craft
• Use exo-atmospheric measurements to normalize measurements
• Use wavelength dependence to infer vertical profiles of target species
• Good dynamic range (>1000) & vertical resolution (down to 0.7 km)
• Subject to ‘saturation’ effects during extreme events
Solar Occultation Geometry
Line of Sight
Tangent Point
Orbit
6/17/2003 L. Thomason 4
Retrieval Strategy
• Gas species retrieved based on their spectral variations using a variety of techniques (MLR, OE, LS, etc.)
• Aerosol can be derived as a residual or using a model to constrain its spectral shape
6/17/2003 L. Thomason 5
Aerosol Spectra (Vis)
• SAGE II/III algorithms derive aerosol as a residual
• Spectra are not constrained to fit a predetermined shape
• Spectral artifacts are possible where measurement modeling is deficient
400 600 800 1000 1200 1400 1600Wavelength (nm)
10-6
10-5
10-4
10-3
Ext
inct
ion
(km
-1)
10.0 km12.5 km
17.5 km20.0 km
25.0 km
30.0 km
15.0 km
22.5 km
27.5 km
6/17/2003 L. Thomason 6
Jan 1 Apr 1 Jul 1 Oct 1 Jan 1-90
-60
-30
0
30
60
90
METEOR-3M
SAGE II & III SOLAR OCCULTATION COVERAGE (~2002)
Latitu
de
Date
SAGE II Sunrise SAGE II Sunset
Occultation Coverage
• Occultation events occur twice an orbit or about 30/day (data rates are low)
• Latitude coverage orbit dependent– Sun-Synchronous orbits
yield high latitude measurements
– Mid-inclination orbits cover low and latitudes (30-40 day period)
6/17/2003 L. Thomason 7
Measurement by Limb Emission (CLAES/HRDLS)
• Aerosol measured by IR emission observed through the Earth’s limb
• Measured continuously through an orbit
• ‘Saturation’ effects are about the same as for solar transmission measurements.
6/17/2003 L. Thomason 8
Lidar Observations of Stratospheric Aerosol
0
5
10
15
20
25
30
LITE Orbit 115 - September 17, 1994
0
5
10
15
20
25
30UT
LATLON
50.0 -51.0
1: 2
40.0 -34.2
1: 6
30.0 -24.1
1:10
20.0 -16.9
1:13
10.0 -10.6
1:16
0.0 -4.6
1:19
-10.0 1.3
1:22
-20.0 7.6
1:25
-30.0 14.9
1:28
-40.0 25.0
1:31
-50.0 41.2
1:35
.01 .10 1.0 10.
532 nm Scattering Ratio - 1
6/17/2003 L. Thomason 9
Modeling Stratospheric Aerosol
• Is a fairly well-behaved parameter– Tends to behave well in θ−potential vorticity space
as aerosol extinction ratio (relative to Rayleigh)– Composed primarily of sulfate
• Complicating factors– Measured quantities are often not the most
important – Condensates (e.g., PSCs) – Second order humidity/temperature effects– Irreversible processes
• Post-volcanic aerosol sedimentation• PSC formation and sedimentation
6/17/2003 L. Thomason 10
Observing Volcanic Effects - Initial
-4.0
-4.0
-4.0
-4.0
-4.0
-4.0
-3.0-3.0
-3.0
-3.0
-3.0
-3.0
-3.0
-3.0
-3.0
-2.0-2.0
-2. 0
-2.0
-180 -100 0 100 180Longitude
30
25
20
15
10
1020-nm Extinction; Lats.: 6.1-14.4; 10/21/2002
• Initial dispersion as a tracer of transport• Observations of small volcanic events
6/17/2003 L. Thomason 11
Observing Volcanic Effects- Long Term
1/e-folding time of ~1 year
1/e-folding time of ~1.5 years
• Events are rare (5 in SAGE II record; 4 in SAM II/SAGE)• Effects persist for many years (Pinatubo ~10 years; El Chichon/Ruiz
~8 years)• Aerosol ‘size’ appears to maintain persistent memory of volcanic
events– Processes that control aerosol s. d. must be slow
Ruiz
Kelut
Pinatubo/Hudson
6/17/2003 L. Thomason 12
Non-Volcanic Aerosol Processes
• Annual cycle in aerosol observed between 16 and 20 km. The differences in the phasing between 525 and 1020 nm extinction suggests that the cycle reflects the introduction of small aerosol.
• Also, PSC-related preferential loss of large aerosol
1996 1997 1998 1999 2000Year
0
5
10
15
20
25
6/17/2003 L. Thomason 13
Long-Term Trends in the Stratospheric Background• Stratospheric record
has several ‘clean’ periods: 1979, 1989, 1998-present
• Relative to the nominal 1979 ‘background’ period, the current period is ~30% lower except in the vicinity of the tropopause -50 0 50
5
10
15
20
25
30
-50 0 505
10
15
20
25
30
0.500.60
0.700.70
0.700.70
0.70
0.80
0.80
0.80
0.80
0.80
0.80
0.900.90
0.90
Alt
itu
de
(km
)
Latitude
SAGE II 1020 nm aerosol in Spring 2000 relate to SAGE 1000 nm inSpring 1979
6/17/2003 L. Thomason 14
Inferring Non-Measured Properties
• Most Desirable Properties:– Integral properties
• Surface area density (SAD)• Volume density or mass• Effective radius • Full aerosol extinction spectra
– Size distribution– Composition
• Estimation of non-measured properties of aerosol generally requires a non-linear retrieval algorithm
6/17/2003 L. Thomason 15
The Extinction Measurement
0.01 0.10 1.00 10.00Particle Radius (Microns)
0
2
4
6
8
10
120.385, 0.453, 0.525, 1.02, 2.45, 2.80, 3.40, 3.45, 5.26, 7.12, 8.27
4λQ(mλ,r)/3r
6/17/2003 L. Thomason 16
Size/Composition Sensitivity• Visible wavelength aerosol is
~10 times larger than that in the IR
• Infrared measurements tend to scale with composition and total volume but have little sensitivity to size
• Visible measurements tend to be more size sensitive but very insensitive to composition and aerosol with radii less than 0.1 µm.
• A system with both IR and visible extinction measurements would be a considerably stronger measurement suite than either wavelength set alone 0.01 0.10 1.00 10.00
Radius (microns)
0.385, 0.453, 0.525, 1.02, 2.45, 2.80, 3.40, 3.45, 5.26, 7.12, 8.27
4λQ(mλ,r)/3r dV(r)/dr
6/17/2003 L. Thomason 17
Problems in Computing SAD
FCA
S Su
rfac
e A
rea
Den
sity
(µm
2 /cm
3 )
1
2
3
4
5
SAGE II Surface Area Density (µm2/cm3)
0 1 2 3 4 5
FCAS - University of DenverSAGE II - NASA Langely Research Center
Slope = 1.0
Slope = 2.5
300 600 900 12000
5
10
15
20
25 Wyoming OPC SAGE II Version 6.0
Surf
ace
area
, µm
2 cm
-3
Days after Pinatubo eruption
• SAGE II, OPC, HALOE SAD values usually agree with 20-30% • Agreement with in situ systems with high size discrimination for
smaller (e.g., FCAS) can be rather poor. • Many retrieval processes are dependent on assumed size
distributions (e.g., log-normal) that can have an enormous impact on retrieval results
6/17/2003 L. Thomason 18
Stratospheric Aerosol Climatologies
• Extensive data sets for extinction, and some integral properties are available
• Data sets are mostly instrument-based
• Data sets are not spatially/temporally continuous
• Need for combined data set with data gap filling for both El Chichon and Pinatubo– Lidar data sets– SME
6/17/2003 L. Thomason 19
Assessment of Stratospheric Aerosol Properties (ASAP)
• ASAP is an on-going SPARC examination of stratospheric aerosols
• Topics include:– Aerosol precursors– Measurements &
climatologies• ‘Filled’ data set for 1979-
present
– “Trends”– Modeling
Reconstruction of NH aerosolextinction at 1020 nm during El Chichon from NASA LaRC 48-inch lidar system