evaluation of terra modis aerosol and water vapor ......• used modis, sgp data from march 2000...

Evaluation of Terra MODIS Aerosol and Water VaporMeasurements Using ARM SGP Data

Richard Ferrare, Lorraine Heilman Brasseur,Dave Turner, Dave Whiteman, Lorraine Remer, Bo-Cai Gao

MODIS Atmosphere Group MeetingDecember 17, 2001

Outline

• MODIS aerosol comparisons

• MODIS water vapor comparisons

• Aerosol and water vapor profile diurnal variability

• Average aerosol and water vapor profiles during Terra overpass

• DOE ARM Aerosol IOP experiment

• Summary

DOE ARM SGP Measurements Used for MODIS Validation

• Aerosol optical thickness (AOT)• Cimel Sun photometer (Cimel)• Multi-Filter Rotating Shadowband Radiometer (MFRSR)

• Precipitable Water Vapor (PWV)• Microwave Radiometer (MWR)• CART Raman Lidar (CARL)• Cimel Sun photometer (Cimel)

• Aerosol and Water Vapor Profiles• CART Raman Lidar (CARL)

• Used MODIS, SGP data from March 2000 through September 2001

Example MODIS and SGP Aerosol Measurements

MODIS AOT (470 nm) CARL aerosol extinction profiles

CARL and Cimel AOT

0.01

0.1

1

20012000OctJulAprJanOctJulAprJan

AO

T

AOT (470 nm) MODIS Cimel MFRSR

0.01

0.1

1


AO

T

AOT 670 nm MODIS Cimel MFRSR

Aerosol optical thickness over SGP during MODIS measurements

• SGP AOT measured by Cimel, MFRSR• AOT generally low (~between 0.02-0.3)

470 nm 670 nm

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.70.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7 Slope=0.74Intercept=0.11N=29 R=0.58rms diff=0.09 (68%)bias diff=0.05 (38%)

- - - - MODIS AOT uncertainty

MO

DIS

AO

T (

470

nm

)

SGP AOT

Cimel (470 nm) MFRSR (500 nm) linear fit 95% pred. level

0.0 0.1 0.2 0.3 0.40.0

0.1

0.2

0.3

0.4

N=29 R=0.18rms diff=0.067 (67%)bias diff=-0.01 (-10%)

- - - MODIS AOT uncertainty

MO

DIS

AO

T (

660

nm

)

SGP AOT

Cimel (670 nm) MFRSR (674 nm)

MODIS AOT vs. SGP AOT

• (470 nm) • MODIS biased high at low AOT • differences generally within MODIS uncertainty

• (660 nm)• large scatter, no linear trend for low AOT• differences generally within MODIS uncertainty

ARM SGP (March 2000-Sept 2001) clear-sky subset

Cimel AOT (340 nm)

CARL A

OT (355 nm

)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

cloudf1>1 and car470a2>0 and car470a2<2 and cim470ao>0 and cim470ao<2

Slope=0.94Intercept=0.01R=0.80 N=66rms diff=0.07 (33%)


Cimel AOT (470)

CART Ram

an Lidar AO

T (470)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

cloudf1>1 and car470a2>0 and car470a2<2 and cim470ao>0 and cim470ao<2


CART Raman lidar AOT (355 nm), Cimel AOT (340 nm) agree within about5% in mean

Using Angstrom exponent (340-500 nm) from CimelAOT to extrapolate CARL AOT from 355 nm to 470 nmgives generally good results

CART Raman lidar AOT vs. Cimel AOT

ARM SGP (March 2000-Sept 2001) MODIS clear-sky subset

Cimel Angstrom Exponent (440 - 670 nm)

MO

DIS A

ngstrom Exponent (470 - 660 nm

)

0

1

2

3

4

5

0 1 2 3 4 5

cloudf1>1 and car470a2>0 and car470a2<2 and cim470ao>0 and cim470ao<2 and mod470ao>0 and mod470ao<2


Cimel Angstrom Exponent (440-670 nm)

Cimel A

ngstrom Exponent (3

40-5

00 nm

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0



Cimel AOT (470 nm)

CART Ram

an Lidar AO

T (470 nm

)

0.0

0.1

0.2

0.3

0.4

0.5

0.0 0.1 0.2 0.3 0.4 0.5

Used Cimel (340-500) Angstrom ExpSlope=0.94 Intercept=0.01R=0.80 N=66rms diff=0.07 (33%)

Used MODIS (470-660) Angstrom Exp.Slope=0.32 Intercept=0.06R=0.37 N=26rms diff=0.07 (33%)

Using MODIS(470-660) Ang. ExpoUsing Cimel (340-500) Ang. Expo

Can not use CARL AOT (355 nm) alone toevaluate MODIS AOT (470 nm) because wecan not use MODIS Angstrom exponent δ(470-660 nm) to extrapolate CARL AOTfrom 355 nm to 470 nm

Cimel δ (350-500 nm) ≠ Cimel δ (470-660 nm) MODIS δ (470-660 nm) ≠ Cimel δ (470-660 nm)

CARL AOT at 470 nm

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

MODIS lower

MODIS higher


AO

T D

iffe

ren

ce

MODIS (470 nm) - Cimel (470 nm) MODIS (470 nm) - MFRSR (500 nm) MODIS (660 nm) - Cimel (670 nm) MODIS (660 nm) - MFRSR (674 nm) -300

-200

-100

0

100

200

300

400

500

MODIS lower

MODIS higher


AO

T D

iffe

ren

ce (

%)

MODIS (470) - Cimel (470) MODIS (470) - MFRSR (500) MODIS (660) - Cimel (670) MODIS (660) - MFRSR (674)

(MODIS - SGP) AOT difference vs. date

No obvious trend with time at either 470 or 660 nm

Absolute differences Relative differences

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

MODIS lower

MODIS higher

AO

T d

iffe

ren

ce

SGP AOT

MODIS (470) - Cimel (470) MODIS (470) - MFRSR (500) MODIS (660) - Cimel (670) MODIS (660) - MFRSR (674)

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35-200

-100

0

100

200

300

400

500

MODIS lower

MODIS higher

AO

T d

iffe

ren

ce (

%)

SGP AOT

MODIS 470 - Cimel 470 MODIS 470 - MFRSR 500 MODIS 660 - Cimel 670 MODIS 660 - MFRSR 674

(MODIS - SGP) AOT difference vs. AOT


• Relative differences decrease with increasing AOT• (470 nm) differences < 0-20% (MODIS higher) for AOT > 0.1

MODIS near IR PWVMODIS IR PWV

CARL water vaporprofiles

CARL and MWR PWV

0

1

2

3

4

5

6

20012000OctJulAprOctJulApr JanJan

PW

V (

cm)

MODIS near IR MODIS IR MWR

Precipitable water vapor over SGP during MODIS measurements

PWV varies between 5 – 50 mm

-80

-60

-40

-20

0

20

40

60

80

MODIS near IR drier

MODIS near IR wetter

OctJulAprJanOctJulAprJan20012000

PW

V D

iffe

ren

ce (

%)

Date

MODIS Near IR - SGP MWR MODIS Near IR - SGP Cimel

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

MODIS near IR drier



PW

V d

iffe

ren

ce (

cm)

MODIS near IR - MWR MODIS near IR - Cimel

(MODIS near IR - SGP) PWV difference vs. date

• Nov. 1 2000• modified water vapor transmittance lookup table to improve “continuum”absorption• changed to side b electronics improved 1.24 µm radiometric calibration

• changes reduced MODIS near IR overestimates of PWV • Jun. 1 2000 regenerated lookup tables using new HITRAN2000 database

• have not yet analyzed sufficient data to evaluate change


MWR PWV (cm)

MO

DIS near IR PW

V (cm

)

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

After Nov. 1, 2000Slope = 1.15 Intercept = -0.08 cmN = 55 R = 0.99rms diff= 0.25 cm (16%)bias diff (modis nir-mwr)=0.156 cm (10%)

Before Nov. 1, 2000Slope = 1.63 Intercept = -0.623 cmN = 25 R = 0.93rms diff= 1.1 cm (54%)bias diff (modis nir-mwr)=0.67 cm (33%)

MODIS near IR PWV comparison vs. Microwave radiometer (MWR) After Nov. 1, 2000

• bias difference decreases from 0.67 cm (33%) to 0.16 cm (10%)

• rms difference decreases from 1.1 cm (54%) to 0.25 cm (16%)


MWR PWV (cm)

PWV

Diff (%

) (MO

DIS near IR - M

WR)

-80

-60

-40

-20

0

20

40

60

80

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Before Nov. 1, 2000After Nov. 1, 2000


MODIS near IR drier

MODIS near IR PWV comparison vs. Microwave radiometer (MWR) After Nov. 1, 2000

• bias difference decreases from 0.67 cm (33%) to 0.16 cm (10%)



MWR PWV (cm)

Cimel PW

V (cm

)

0

1

2

3

4

5

0 1 2 3 4 5

Slope = 1.07Intercept = -0.029 cmN = 44R = 0.996rms diff= 0.156 cm (9%)bias diff (cimel-mwr)=0.089 cm (5%)

MWR vs. Cimel PWV comparison

If you use Cimel PWV to validate MODIS…

PWV derived using Cimel processing using LOWTRAN 7 database is about 5-8% higherthan ARM MWR

[If Cimel PWV processing usedGiver corrected HITRANdatabase (which increased 940 nmline strengths by ~14%), thenCimel PWV decreases by about14%.]

See Schmid et al. Comparison ofcolumnar water-vapormeasurements from solartransmittance methods. AppliedOptics, Vol. 40, No. 12, 1886-1896(2001).


Cimel PWV (cm)

MO

DIS near IR (PW

V)

0

1

2

3

4

5

6

7

8

0 1 2 3 4 5 6 7 8

After Nov. 1, 2000Slope = 1.05 Intercept = -0.02 cmN = 32 R = 0.997rms diff= 0.11 cm (7%)bias diff (modis nir-mwr)=0.051 cm (3%)

Before Nov. 1, 2000Slope = 1.69 Intercept = -0.794 cmN = 12 R = 0.95rms diff= 1.31 cm (56%)bias diff (modis nir-mwr)=0.82 cm (35%)

After Nov. 1, 2000• bias difference decreases from 0.82 cm (35%) to 0.05 cm (3%)


MODIS near IR PWV comparison vs. Cimel

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0After Nov. 1, 2000

mean diff = 0.11 cmrms diff = 0.18 cm

MODIS near IR drier


PW

V d

iffe

ren

ce (

cm)

SGP PWV (cm)


0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5-30

-20

-10

0

10

20

30


MODIS near IR drier

mean diff = 6.3%rms diff = 8.2%

After Nov. 1, 2000

PW

V d

iffer

ence

(%

)

SGP MWR (cm)


(MODIS near IR - SGP) PWV difference vs. PWV

• MODIS near IR PWV about 5-15% higher than SGP MWR and Cimel PWV• No systematic variations with PWV

ARM SGP (March 2000-Sept 2001) MODIS night clear-sky subset

MWR PWV (cm)

CARL PW

V (cm

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Slope = 1.002Intercept = -0.002 cmN = 42R = 0.999rms diff= 0.045 cm (3%)bias diff (carl-mwr)=0.0012 cm (0.1%)

CART Raman Lidar water vapor measurements• Nighttime profiles extend through troposphere (0.06 – 10 km) permitting PWV retrievals• Daytime profiles limited by background skylight (0.06 - ~ 3.5 km) – no PWV retrieved• CARL calibrated so that CARL PWV matches MWR PWV• CARL PWV has excellent long-term stability when compared with MWR


MWR PWV (cm)

MO

DIS IR PW

V (cm

)

0

1

2

3

4

5

6

0 1 2 3 4 5 6

irpwv>0 and irpwv<7 and mwrpwv>0 and mwrpwv<7 and radius<30 and cldflag1>1 and mwrliq<0.01

Slope = 0.73 Intercept = 0.66 cmN = 150 R = 0.77rms diff= 0.684 cm (38%)bias diff (modis ir-mwr)=0.18 cm (21%)

MODIS IR PWV vs. MWR PWV

• (daytime+nightime) offset (intercept) = 6.8 mm • bias difference ~ 2 mm (~20%) (MODIS IR wetter), rms difference 7 mm (~40%)


CART Raman Lidar PWV (cm)

MO

DIS IR PW

V (cm

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

irpwv>0 and irpwv<7 and carlpwv>0 and carlpwv<7 and radius<30 and cldflag1>1 and mwrliq<0.01 and daynight=0

Slope = 0.56 Intercept = 0.72 cmN = 41 R = 0.78rms diff= 0.524 cm (34%)bias diff (modis ir-carl)=0.28 cm (18%)

• (nightime) offset (intercept) = 7 mm • bias difference ~ 3 mm (~20%) (MODIS IR wetter), rms difference 5 mm (~30%)

MODIS IR PWV vs. CARL PWV


Cimel PWV (cm)

MO

DIS IR PW

V (cm

)

0

1

2

3

4

5

6

0 1 2 3 4 5 6

irpwv>0 and irpwv<7 and cimpwv>0 and cimpwv<7 and radius<30 and cldflag1>1 and mwrliq<0.01 and daynight=1

Slope = 0.92 Intercept = 0.53 cmN = 44 R = 0.875rms diff= 0.694 cm (39%)bias diff (modis ir-cimel)=0.385 cm (21%)

MODIS IR PWV vs. Cimel PWV

• (daytime) offset (intercept) = 5 mm • bias difference ~ 4 mm (~20%) (MODIS IR wetter), rms difference 7 mm (~40%)

-100

-50

0

50

100

150

200

MODIS IR drier

MODIS IR wetter


PW

V d

iffe

ren

ce (

%)

Date

MODIS IR - MWR MODIS IR - CARL MODIS IR - Cimel

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

MODIS IR drier

MODIS IR wetter


PW

V d

iffe

ren

ce (

cm)

Date


(MODIS IR - SGP) PWV difference vs. Date

• relative error increases in winter due to low PWV


0 1 2 3 4 5-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

MODIS IR drier

MODIS IR wetter

mean diff = 0.19 cmrms diff = 0.63 cm

PW

V d

iffe

ren

ce (

cm)

(MO

DIS

IR -

SG

P)

SGP PWV (cm)


0 1 2 3 4 5-100

-50

0

50

100

150

200

mean diff = 28%rms diff = 52%

MODIS IR drier

MODIS IR wetter

PW

V D

iffe

ren

ce (

%)

(MO

DIS

IR -

SG

P)

SGP PWV (cm)


(MODIS IR - SGP) PWV difference vs. PWV

• large relative error at low PWV due to MODIS IR PWV offset (floor around 5-7 mm)• mean difference ~ 2 mm (~25%) (MODIS wetter), rms difference 6 mm (~50%)

0 1 2 3 4 50

1

2

3

4

5

NightSlope=0.55Intercept=0.75 cmR=0.77 N=113rms diff=0.45 cm

MO

DIS

IR P

WV

(cm

)

SGP PWV (cm)

MWR CARL linear fit 95% prediction level

0 1 2 3 4 50

1

2

3

4

5

DaySlope=0.99Intercept=0.46 cmR=0.87 N=78rms diff=0.54 cm

MO

DIS

IR P

WV

(cm

)

SGP PWV (cm)

MWR Cimel linear fit 95% prediction level

(MODIS IR - SGP) PWV daytime vs. nighttime performance

MODIS IR PWV has better agreement with SGP PWV for daytime measurements (smaller offset, increase in slope closer to unity, higher linear correlation)

Night Day


MWR PWV (cm)

MO

DIS IR PW

V (cm

)

0

1

2

3

4

5

0 1 2 3 4 5

NightDay

DaySlope = 0.94 Intercept = 0.51 cmN = 78 R = 0.87rms diff= 0.68 cm (41%)bias diff (modis ir-mwr)=0.65 cm (39%

NightSlope = 0.54 Intercept = 0.76 cmN = 72 R = 0.76rms diff= 0.68 cm (36%)bias diff (modis ir-mwr)=0.24 cm (12%

(MODIS IR - SGP) PWV daytime vs. nighttime performance

MODIS IR PWV has better agreement with SGP MWR PWV for daytime measurements (smaller offset, increase in slope closer to unity, higher linear correlation)

(This time used single instrument, MWR, to look at diurnal changes)

Winter Spring

Summer

Fall

Aer

osol

Ext

inct

ion

(km

-1)

Average Diurnal Variation of Aerosol Extinction Profiles

• CARL aerosol extinction profiles averaged over 837 days between March 98 - October 01• Higher extinction concentrated over smaller vertical extent at night

Sunrise

Sunrise

Sunrise

Sunrise

Sunset Sunset

Sunset Sunset

Ter

ra T

erra

Aqu

a

Aqu

a

Winter

SummerA

eros

ol E

xtin

ctio

n (k

m-1

)

Average Diurnal Variation of Aerosol Extinction Profiles and AOT• Large changes in vertical profile; smaller changes in AOT (st. dev ~ 10%)

Sunrise

Sunrise

Sunset

Sunset

Ter

ra

0 2 4 6 8 10 12 14 16 18 20 22 240.0

0.1

0.2

0.3

0.4

0.5

0.6

Aq

ua

Aq

ua

Aq

ua

Te

rra

Fall (day) Fall (night) Winter (day) Winter (night)

Aero

so

l O

pti

cal T

hic

kn

ess (

355 n

m)

Time (UT)

Summer (day) Summer (night) Spring (day) Spring (night)

Aqu

a AOT

Wat

er V

apor

Mix

ing

Rat

io (

g/kg

)

Winter Spring

Summer Fall

Sunrise

Sunrise

Sunrise

Sunrise

Sunset Sunset

Sunset Sunset

Ter

ra T

erra

Aqu

a

Aqu

a

Average Diurnal Variation of Water Vapor Profiles

• generally smaller diurnal changes than aerosol extinction near the surface• larger diurnal changes in spring and summer near top of mixed layer

0 2 4 6 8 10 12 14 16 18 20 22 240

10

20

30

40

50

60

Aq

ua

Terr

a

Fall (day) Fall (night) Winter (day) Winter (night)

Inte

gra

ted

Wa

ter

Va

po

r (a

rb.

un

its

)

Time (UT)

Summer (day) Summer (night) Spring (day) Spring (night)

Wat

er V

apor

Mix

ing

Rat

io (

g/kg

)

Winter

Summer

Sunrise

Sunrise

Sunset

Sunset

Ter

ra

Average Diurnal Variation of Water Vapor Profiles

• smaller diurnal changes in profiles and integrated water vapor (st. dev ~3-5%)

Aqu

a

Integrated water vapor

Rel

ativ

e H

umid

ity

(%)

Winter Spring

Summer

Fall

Sunrise

Sunrise

Sunrise

Sunrise

Sunset Sunset

Sunset Sunset

Ter

ra T

erra

Aqu

a

Aqu

a

Average Diurnal Variation of Relative Humidity Profiles• RH computed using CARL water vapor, AERI+model temperatures• Increase in aerosol extinction near surface at night correlated to RH• Terra, Aqua measurements occur when RH (and aerosol size, composition?) vary with z

0

1

2

3

4

5

6

7

30 40 50 60 70

Relative Humidity (%)

Alti

tude

(km

) Summer Spring Fall Winter

0

1

2

3

4

5

6

7

0 2 4 6 8 10 12 14

Water Vapor Mixing Ratio (g/kg

Alt

itu

de

(km

)

Summer Spring Fall Winter

0

1

2

3

4

5

6

7

0.00 0.05 0.10 0.15 0.20 0.25

Aerosol Extinction (355 nm) (km-1)

Alt

itu

de

(km

)

Summer Spring Fall Winter

Average Profiles at time of Terra Overpass

• Aerosol extinction:• profile shape varies with season• scale height varies with season and AOT

• Water Vapor• profile shape and scale height are constant

See Turner, Ferrare, Brasseur, GRL, 28, 4441-4444, 2001.

DOE ARM Proposed Aerosol Experiment• Aerosol IOP (Intensive Operations Period) (~May, 2003 at SGP)

Objectives:•Diffuse Flux closure

Use new and additional measurements of aerosol absorption and extinction toaccurately constrain aerosol absorption to resolve differences between measuredand modeled diffuse radiation

• CCNInvestigate relationship between CCN number concentrations at the surface andcloud base, and determine whether profiles of aerosol extinction and RH can beused to determine cloud nucleating properties just below cloud base

• AOT closureCharacterize routine (Raman, MPL) lidar and aircraft in situ profilingmeasurements of aerosol scattering and extinction and how aerosol humidificationfactor varies with altitude

Measurements:• Use one (possibly two) instrumented aircraft• Additional surface aerosol and radiation measurements• Possible coordination with DOE Tropospheric Aerosol Program (TAP) (i.e. chemistry)

Potential MODIS Terra, Aqua validation/science opportunity

Summary

• Aerosol Optical Thickness (AOT)- low range of AOT (0-0.3) hampers full evaluation and gives large rms differences- MODIS AOT (470 nm) higher by 30-40% for all AOT, 10-20% for AOT>0.1- MODIS AOT (660 nm) not well correlated to SGP AOT for low AOT- comparisons show results generally fall within MODIS AOT uncertainties

• Precipitable Water Vapor (PWV)- before Nov. 1, 2000 MODIS near-IR biased high by about 35-40%- after Nov. 1, 2000 MODIS near-IR biased high by ~10%- relatively small (10-20%) rms differences- MODIS IR has apparent offset (PWV floor around 5-7 mm)- MODIS IR biased high 2-4 mm(~10-20%) due to offset, rms diff ~ 6 mm (50%)- MODIS IR daytime retrievals in better agreement with SGP than nighttime

• Vertical Variability of Aerosols- Raman lidar profiles show diurnal variability in aerosol, water vapor profiles

- diurnal variability of AOT ~ 10% (st. dev), PWV ~ 3-5% (st. dev)- average aerosol extinction profiles vary with season and AOT- average water vapor profiles have constant shape and scale with PWV

• Proposed DOE ARM Aerosol IOP (SGP, ~ May 2003)- Diffuse radiation (aerosol absorption) and AOT closure, CCN- Potential for joint Terra/Aqua validation/science

Recent Publications

Turner, D.D., W.F. Feltz, and R.A. Ferrare, Continuous water vapor profiles from operational ground-based active and passiveremote sensors, Bull. Amer. Meteor. Soc., 81, 1301-1317, 2000.

Kato, S., M.H. Bergin, T.P. Ackerman, T.P. Charlock, E.E. Clothiaux, R.A. Ferrare, R.N. Halthore, N. Laulainen, G.G. Mace, J.Michalsky, and D.D. Turner, A comparison of the aerosol optical thickness derived from ground-based and airbornemeasurements, J. Geophys. Res., 105, No. D11, 14701-14717, 2000.

Peppler, R.A., C.P. Bahrmann, J.C. Barnard, J.R. Campbell, M.-D. Cheng, R.A. Ferrare, R.N. Halthore, L.A. Heilman, D.L.Hlavka, N.S. Laulainen, C.-J., Lin, J.A. Ogren, M.R. Poellot, L.A. Remer, K. Sassen, J.D. Spinhirne, M.E. Splitt, D.D. Turner,ARM Southern Great Plains Site Observations of the Smoke Pall Associated with the 1998 Central American Fires, Bull. Amer.Meteor. Soc., 81, 2563-2592, 2000.

Whiteman, D.N., G. Schwemmer, D. O'C. Starr, K.D. Evans, B. Demoz, T. Berkoff, S.H. Melfi, M. Cadirola, and G. Jedlovec,"The use of Raman Lidar in Cloud Studies", in Advances in Laser Remote Sensing, Selected Papers Presented at the 20thInternational Laser Radar Conference (ILRC), Vichy, France, 10-14 July 2000, A. Dabas, C. Loth, and J. Pelon, eds., Ecolepolytechnique, France, pp. 271-274, 2001.

Turner, D.D., R.A. Ferrare, L.A. Heilman, T. T. Tooman, A Two Year Climatology of Water Vapor and Aerosols in the LowerTroposphere Measured by a Raman Lidar, in Advances in Laser Remote Sensing, Selected Papers Presented at the 20thInternational Laser Radar Conference (ILRC), Vichy, France, 10-14 July 2000, A. Dabas, C. Loth, and J. Pelon, eds., Ecolepolytechnique, France, pp. 309-312, 2001.

Whiteman, D. N., K. D. Evans, B. Demoz, D. O'C. Starr, D. Tobin, W. Feltz, G. J. Jedlovec, S. I. Gutman, G. K. Schwemmer, M.Cadirola, S. H. Melfi, F. J. Schmidlin, 2001: Raman lidar measurements of water vapor and cirrus clouds during the passage ofhurricane Bonnie, J. of Geophys. Res., 106, No. D6, 5211-5225.

Ferrare, R.A., D.D. Turner, L.A. Heilman, O. Dubovik, and W. Feltz, Raman Lidar Measurements of the Aerosol Extinction-to-Backscatter Ratio Over the Southern Great Plains, J. Geophys. Res., 106, 20333-20347, 2001.

Turner, D.D., R.A. Ferrare, and L.A. Brasseur, Average aerosol extinction and water vapor profiles over the Southern GreatPlains, Geophys. Res. Letters, 28, 4441-4444, 2001.

Turner, D.D., R.A. Ferrare, L.A. Heilman, W.F. Feltz, and T. Tooman, Automated Retrievals of Water Vapor and AerosolProfiles over Oklahoma from an Operational Raman Lidar, J. Atmos. Oceanic Tech., in press, July, 2001.


Sensor zenith Angle (deg)

PWV

difference (%) (M

OD

IS IR - MW

R)

-100

-50

0

50

100

150

200

0 10 20 30 40 50 60 70

NightDay

(MODIS IR - SGP) PWV difference vs. sensor zenith angle

• Nearly Continuous Operation• Nd:YAG (355 nm) (day/night)

– 12 W• 61 cm telescope• Wavelengths

– Rayleigh/Aerosol (355 nm)– Depolarization (355 nm)– Raman water vapor (408 nm)– Raman nitrogen (387 nm)

• 39 meter range resolution• low, high sensitivity channels• measures

• water vapor and aerosolprofiles

• precipitable water vapor andaerosol optical thickness

• aerosol and clouddepolarization

Additional information: http://www.arm.gov/docs/instruments/static/rl.html

Southern Great Plains (SGP) CART Raman Lidar (CARL)

Raman lidar observations of the aerosol vertical variability

17.0 17.5 18.0 18.5 19.0 19.5 20.0

7

6

5

4

3

2

1

Date (May, 1998)

Aerosol Extinction (km-1)

0

1

2

3

4

5

6

7

0.0 0.1 0.2 0.3 0.4 0.5

Aerosol Extinction (km-1)

Alt

itu

de

(km

)

30 40 50 60 70 80 90 100 110 120

May 18, 1998 19:50-20:20 UTExtinction/Backscatter Ratio

Extinction profile derived from CARL backscatter profileand mean S

a = 59 sr

CARL Extinction Profile

Aerosol Extinction/Backscatter Ratio Sa (sr)

• measures vertical variability in aerosol extinction/backscatter (Sa)• variability due to changes in size, composition of aerosols

Atmospheric st. deviation of Sa (sr)

Pe

rce

nta

ge

of O

bs

erv

atio

ns

Nu

mb

er o

f Ob

se

rva

tion

s

48.4%

31.0%

14.1%

5.0%

1.5%

0

52

104

156

208

260

312

364

416

468

520

572

624

676

728

780

0%

10%

20%

30%

40%

50%

<= 5 5-10 10-15 15-20 > 20

• Average values show increase at top of BL, then a slight decrease with altitude• Large (>20% or >10 sr) variations in Sa vertical profile occur about 20% of time

Raman lidar measurements of aerosol extinction/backscatter ratio Sa

0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.020

30

4050

60

7080

90

100110

120Slope = 38Intercept = 16.5R = 0.72N = 19

CA

RL

Sa (355 n

m)

Angstrom Exponent (443/872 nm)

Comparison of Raman lidar Sa with Cimel aerosol measurements

• Sa increases with accumulation mode aerosol• limited to cases with small Sa vertical variability and large AOT (>0.4)

20

30

40

50

60

70

80

90

100

110

120

0.2 0.3 0.4 0.5 0.6 0.7

Slope=108.6Intercept=16.2R=0.90N=19

Volume Ratio = Volume (fine mode)/Volume (total)

Volume Ratio

CA

RL

Sa (sr)

How does Sa profile correlate with IAP measurements of • Angstrom exponent profile• single scattering albedo profile

DOE ARM In-Situ Aerosol Profiling

Sample Inlet Port view of rack

Objective: Obtain a statistically-significant data set of verticaldistribution of aerosol properties

Measurements: aerosol scatteringand absorption, plus chemicalcomposition, above a similarlyinstrumented surface site

2-3 profiles/week for 1 year

0

1000

2000

3000

4000

0 10 20 30 40scattering (1/Mm)

alt

itu

de

(m

ms

l)

0.6 0.7 0.8 0.9 1.0

single-scattering albedo

SGPMay 4,2000

John Ogren, Betsy AndrewsNOAA/CMDL

Summary of IAP Flight Hours

96

75

93

# complete profiles

-over CART site

-with SGP data

2.1 hoursAverage flight duration

215 hoursTotal flight time

104 flightsTotal flights

98 flight days/

280 day period

Total days

March 25 – December31, 2000

Flight period

John Ogren

Betsy Andrews

NOAA/CMDL

Aerosol Extinction (km-1 )

Alt

itude (

km

) <= .4

(.4,.75]

(.75,1.1]

(1.1,1.35]

(1.35,1.75]

(1.75,2.5]

(2.5,3.1]

> 3.1

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16

95%5%

75%25%Median

Angstrom Exponent (450/550 nm)

Alt

itude (

km

) <= .4

(.4,.75]

(.75,1.1]

(1.1,1.35]

(1.35,1.75]

(1.75,2.5]

(2.5,3.1]

> 3.1

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5

95%5%

75%25%

Median

SGP IAP measurements

wo (550 nm) (ambient)

Alt

itude (

km

) <= .4

(.4,.75]

(.75,1.1]

(1.1,1.35]

(1.35,1.75]

(1.75,2.5]

(2.5,3.1]

> 3.1

0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05

95%5%75%25%Median

wo (550 nm) (dry)

Alt

itude (

km

) <= .4

(.4,.75]

(.75,1.1]

(1.1,1.35]

(1.35,1.75]

(1.75,2.5]

(2.5,3.1]

> 3.1

0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05

95%5%75%25%Median

SGP IAP measurements

St. dev of Angstrom Exponent (45/550 nm)

Percent of observations

19.2%

16.7%

17.9%

11.5%

7.7%

12.8%

3.8% 3.8%

1.3%

5.1%

0%

5%

10%

15%

20%

<= .1(.1,.15]

(.15,.2](.2,.25]

(.25,.3](.3,.4]

(.4,.5](.5,.6]

(.6,.7]> .7

(~45% with st. dev > 0.2

St. deviation of Angstrom Exponent (450/550 nm)

Percent of observations

33.3%

38.5%

12.8%

3.8%6.4%

3.8%1.3%

0%

5%

10%

15%

20%

25%

30%

35%

40%

<= .1(.1,.15]

(.15,.2](.2,.25]

(.25,.3](.3,.4]

(.4,.5](.5,.6]

(.6,.7]> .7

(weighted by aerosol extinction at 550 n m

(~15% with st. dev > 0.2

St. dev of wo (550 nm)

No of observations

2.6%

24.4%

17.9%

9.0%7.7%

38.5%

0

3

6

9

12

15

18

21

24

27

30

33

<= .01 (.01,.02] (.02,.03] (.03,.04] (.04,.05] > .05

St. dev of wo (550 nm)

No of observations

9.0%

32.1%

25.6%

11.5%

6.4%

15.4%

0

2

4

6

8

10

12

14

16

18

20

22

24

26

<= .01 (.01,.02] (.02,.03] (.03,.04] (.04,.05] > .05

(weighted by aerosol extinction (550 nm )

evaluation of terra modis aerosol and water vapor ......• used modis, sgp data from march 2000...

Documents