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