Aerosol layers over Oklahoma as seen from Twin Otter on May 27, 2003 Photo courtesy of Roy Woods

Aerosol layers over Oklahoma as seen from Twin Otter on May 27, 2003

Photo courtesy of Roy Woods

How well can we measure the vertical profile of aerosol extinction?

0

1

2

3

4

5

6

0 0.1 0.2 0.3Aerosol Extinction (1/km)

Alti

tude

(km

)

Neph+PSAP (453 nm) 18:34-18:52 UT

B. Schmid [1], R. Ferrare [2], C. Flynn [3], R. Elleman [4],D. Covert [4], A. Strawa [5], E. Welton [6], J. Barnard [3],

M. Bartholomew [7], M. Clayton [8], J. Eilers [5], G. Hallar [5], B. Holben [6], H. Jonsson [9], J. Michalsky [10], J. Redemann [1],

K. Ricci [11], A. Smirnov [12], D. Turner [3]

[1] Bay Area Environmental Research Institute, Sonoma, CA[2] NASA Langley Research Center, Hampton, VA

[3] Pacific Northwest National Laboratory, Richland, WA[4] University of Washington, Seattle, WA

[5] NASA Ames Research Center, Moffett Field, CA[6] NASA GSFC, Greenbelt, MD

[7] Brookhaven National Laboratory, Upton, NY [8] SAIC/NASA Langley Research Center, Hampton, VA

[9] Center for Interdisciplinary Remotely-Piloted Aircraft Studies, Marina, CA[10] NOAA/CMDL, Boulder, CO

[11] Los Gatos Research, Mountain View, CA[12] GEST/UMBC/ NASA GSFC, Greenbelt, MD

Photo courtesy Yin-Nan Lee, BNL

How do we measure the vertical profile of aerosol extinction?

Airborne:• Nepehelometer+PSAP• Cavity-Ring-Down (Cadenza) • Sunphotometry (AATS-14)

Ground based:• Raman Lidar (CARL)• MPLNET• MPLARM

Airborne in-situ measurement of

Extinction =Absorption

• Particle Soot Absorption Photometer (PSAP)

=467, 530, 660 nm• low RH• Corrections (Bond et al.,

1999, Virkkula et al, 2004)– scattering contribution– loading correction– ambient p, T

Scattering• Nephelometers

– TSI: = 450, 550, 700 nm, low RH

– RR: = 540 nm, RH=20,60,85%

• Corrections (Anderson & Ogren, 1998) – light source– angular truncation (7°-170°)– ambient RH, p, T

+

L

cR L L Lext Ray absorber( )1 1

ext c aer 1 1 1

0( )

where L = cell lengthc = speed of lightR = reflectivity of mirrors = coefficient of extinction, Rayleigh, gaseous absorbers

where = ringdown timeaer = with aerosol0 = without aerosol

Iin(t)

ring-down cavity

Iout(t)

time

Extinction from Airborne Cavity Ring-Down System

Aerosol was sampled at low RH (comparable to TSI)

Used RR-Neph measured growth curve to adjust to ambient RH

NASA Ames Airborne Tracking SunphotometerAATS-14: 354-2139 nm

Aerosol Optical Depth and H2O columnAerosol Extinction and H2O density in suitable profiles

Photo courtesy Yin-Nan Lee, BNL

Raman Lidar

• Aerosol Extinction profiles =355 nm • Unnoticed loss of sensitivity leading up to AIOP• Automated algorithms had to be modified to reduce impact of sensitivity loss.• Significant upgrades/mods were performed after AIOP resulting in significantly better performance than during any other time.

•Backscatter Lidar •Deployed during AIOP (in SMART trailer) •Aerosol Extinction profiles =523 nm •AOD anchored to AERONET, BER constant with altitude•Level 2.0 (ext. prof. when valid AERONET AOD)•Periodic overheating problems

•Backscatter Lidar •Deployed permanently at SGP•Aerosol Extinction profiles =523 nm is a new product in -stage

•AOD anchored to NIMFR, BER constant with altitude

MPLARM

Focus on data obtained in vertical profiles over SGP CF

Altitudes: ~90 – 5600 m

CIRPAS Twin OtterARM Aerosol IOP – May 2003

x y # profiles # data pointsAATS-14 Neph+PSAP 26 3484AATS-14 Cadenza 26 2856

AATS-14 MPLNET 2.0 13 587AATS-14 MPLARM 19 2073

AATS-14 Raman Lidar 11 468

AATS-14 in-situ H2O 35 6705

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

340 nmbias= 0.022, 6%rms= 0.038, 10%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

380 nmbias= 0.022, 7%rms= 0.035, 11%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

440 nmbias= 0.022, 8%rms= 0.032, 12%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

501 nmbias= 0.016, 7%rms= 0.027, 11%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

675 nmbias= 0.008, 5%rms= 0.019, 11%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

870 nmbias= 0.002, 2%rms= 0.015, 12%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

1020 nmbias= 0.002, 2%rms= 0.015, 14%

1 3 6 9 12 15 18-0.1

-0.05

0

0.05

0.1

1638 nmbias= -0.006, -9%rms= 0.013, 20%

AO

D: A

ER

ON

ET

min

us A

ATS

-14

Observation #

1 3 6 9 12-0.1

-0.05

0

0.05

0.1

415 nmbias= -0.004, -1%rms= 0.012, 4%

1 3 6 9 12-0.1

-0.05

0

0.05

0.1

500 nmbias= 0.006, 2%rms= 0.010, 4%

1 3 6 9 12-0.1

-0.05

0

0.05

0.1

615 nmbias= 0.007, 4%rms= 0.013, 7%

1 3 6 9 12-0.1

-0.05

0

0.05

0.1

673 nmbias= 0.001, 1%rms= 0.009, 6%

1 3 6 9 12-0.1

-0.05

0

0.05

0.1

870 nmbias= -0.002, -1%rms= 0.007, 6%

AO

D: N

IMFR

min

us A

ATS

-14

Observation #

0 2 4 6 8 10 12 14 16

0

2

4

6

8

10

12

14

16

n= 6705 r2 = 0.958y = 1.021 x + -0.224rms= 0.628, 19.4 %

in-situ - H2O Density [g/m3]

AA

TS-1

4 - H

2O D

ensi

ty [g

/m3 ]

Extinction

-0.004

0.000

0.004

0.008

0.012

0.016

0.020

0.024

0.028

Raman

Neph+P

SAP

Neph+P

SAP

MPLNET

MPLARM

Neph+P

SAP

Caden

za

Caden

za

Bias

(1/k

m)

Extinction

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

Raman

Neph+P

SAP

Neph+P

SAP

MPLNET

MPLARM

Neph+P

SAP

Caden

za

Caden

zaRela

tive

Bias

Aerosol Optical Depth

-0.03-0.02-0.010.000.010.020.030.040.050.060.070.080.09

Raman

NIMFR

Cimel

Neph+P

SAPNIM

FRCim

el

MPLNET

MPLARM

Neph+P

SAPNIM

FRCim

el

Caden

za

Neph+P

SAPCim

el

Caden

za

Bias

Diff

eren

ce

Aerosol Optical Depth

-45%

-35%

-25%

-15%

-5%

5%

15%

25%

35%

45%

55%

Raman

NIMFR

Cimel

Neph+P

SAPNIM

FRCim

el

MPLNET

MPLARM

Neph+P

SAPNIM

FRCim

el

Caden

za

Neph+P

SAPCim

el

Caden

zaBias

Diff

eren

ce

How well did past campaigns do on AATS vs Neph+PSAP comparisons?

U.S. East Coast, 1996, Bias -14% (Hegg et al., 1997, Hartley et al., 2000)

Canary Islands, 1997, Bias -20% (after inlet correction) (Schmid et al.,

2000)

Southern Africa, 2000, Bias -4% (Magi et al., 2003)

U.S. East Coast, 2001, Bias -15% Magi et al. (2004)

Eastern Asia, 2001

C-130: Bias: +6% Redemann et al., 2003

Twin Otter: Bias -13% Schmid et al., 2003

Oklahoma, 2003, Bias -14%

Aerosol IOP

Conclusions• When compared to airborne sunphotometer extinction,

in-situ measurements are biased low (2 - 4 Mm-1 VIS 12-17%, NIR 45%)

• Looking at results from 6 field campaigns, airborne Neph+PSAP measurements of extinction tend to be biased slightly low (<15%, VIS) when compared to airborne sunphotometer extinction.

• Micro-Pulse Lidar measurements biased high – MPLNET 4 Mm-1,13% – MPLARM 7 Mm-1, 24%

• Raman Lidar measurements biased high (29 Mm-1 UV 54%). We expect better agreement from a “healthier” Raman lidar.

Suggestion to AWG

• Improved Raman Lidar needs validation• Newly developed MPLARM extinction

profile algos need validation• Enhanced IAP plane will need validation

With a small amount of flight hours (i.e. AATS-14 on J-31) these objectives could be met in a cost effective way