looking forward: using modis for aerosol science over the ......ο fmf base the uv is sensitive to...

Looking Forward:Looking Forward:

Using MODIS for aerosolUsing MODIS for aerosolscience over the next 3 yearsscience over the next 3 years

Lorraine A. Remer, J. Lorraine A. Remer, J. Vanderlei Vanderlei Martins,Martins,Robert Levy,Robert Levy, Shana Shana Mattoo, Charles Mattoo, Charles IchokuIchoku,,Richard Richard KleidmanKleidman,, Hongbin Hongbin Yu, Yu, Ilan Ilan KorenKoren

Where do we go from here?Where do we go from here?--The MODIS aerosol algorithmThe MODIS aerosol algorithm-- Aerosol-cloud interaction Aerosol-cloud interaction-- Absorption and composition Absorption and composition-- Air quality and long range transport of aerosols Air quality and long range transport of aerosols-- LidarLidar

The MODIS Algorithm: AccomplishmentsThe MODIS Algorithm: Accomplishments

Recent improvements to the over ocean retrievalRecent improvements to the over ocean retrieval

The MODIS Algorithm:The MODIS Algorithm:AccomplishmentsAccomplishments

Recent improvements to the over land algorithm.Recent improvements to the over land algorithm.

Levy et al. (2006)Levy et al. (2006)

The MODIS Algorithm: For the futureThe MODIS Algorithm: For the futureDevelop an urban product: surfaceDevelop an urban product: surface

Robert LevyRobert Levy’’s work (ATBD)s work (ATBD)Urban: y = 0 +Urban: y = 0 +0.77x0.77x nonUrbannonUrban: y=0.007 + : y=0.007 + 0.51x0.51x

The MODIS Algorithm: For the futureThe MODIS Algorithm: For the future

-0.1

-0.05

0

0.05

0.1

0.15

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

3 km AOT

10 km AOT

diffe

renc

e in

AO

T

distance in degrees latitude

Sao PauloDifference between MODIS and sunphotometer

Develop anDevelop anurban product:urban product:

Fine resolutionFine resolution

Work from Work from Shana Shana MattooMattooData from Andrea Data from Andrea CastanhoCastanho


0.1

1

10

100

1000

0 40 80 120 160

MODIS empiricalVolten sahara Volten quartz Dubovik spheroid spherical r_eff =1.5 µmspherical r_eff = 0.10 µmMODIS observations

phas

e fu

nctio

n (6

60 n

m)

scattering angle

-0,005

0

0,005

0,01

0,015

0,02

0,025

0,03

80 100 120 140 160 180

(a-1)West Mediter. Sea

R=1.12Spherical model

Aer

osol

rad

ianc

es La λ, L

p,a λ

Scattering angle Θ

λ=670nmλ=865nm

-0,005

0

0,005

0,01

0,015

0,02

0,025

0,03

80 100 120 140 160 180

(a-2)West Mediter. Sea

R=1.12Non-spherical model

Aer

osol

rad

ianc

es La λ

, Lp,

a λScattering angle Θ

λ=670nmλ=865nm

POLDERPOLDER

Use POLDER with MODIS to solveUse POLDER with MODIS to solvethe nonsphericity issues once andthe nonsphericity issues once andfor all.for all. Herman et al. (2005)Herman et al. (2005)


refle

ctan

cere

flect

ance

Distance from clouds (km)Distance from clouds (km)

Work by Work by Ilan Ilan KorenKoren

MODIS image MODIS image Distance from cloudsDistance from clouds


1.1. Develop an urban product: surface reflectance Develop an urban product: surface reflectance assumptions and finer resolution. assumptions and finer resolution.2.2. Combine MODIS and POLDER to solve Combine MODIS and POLDER to solve nonspherical problems.nonspherical problems.3.3. Elevated reflectance in the vicinity of clouds:Elevated reflectance in the vicinity of clouds: real or artifact? Deal with it. real or artifact? Deal with it.

Cloud-Aerosol InteractionCloud-Aerosol Interaction

In this elevation of reflectance near clouds, is any of itIn this elevation of reflectance near clouds, is any of itreal?real?

0 10 20 30 40 50

0

5

10

15

20

25

30

35

40

45

col

row

0.000 0.047 0.095AER_REF066&CloudMask

We donWe don’’t think it is classicalt think it is classicalcloud contamination.cloud contamination.

Work by A. MarshakWork by A. Marshak


Wen et al. (2006; in preparation)Wen et al. (2006; in preparation)

What about enhancement of radiation from scattering lightWhat about enhancement of radiation from scattering lightfrom cloud sides into the satellite view?from cloud sides into the satellite view?

That is a possible artifact.That is a possible artifact.

Cloud-Aerosol InteractionCloud-Aerosol InteractionHowever, it could also be realHowever, it could also be real-- Hygroscopic growth to larger, more optically effectiveHygroscopic growth to larger, more optically effective

sizessizes-- Cloud outflow and decayCloud outflow and decay

0

1000

2000

3000

4000

5000

6000

0.1 1

uncorrectedcorrected

dN/d

log(

d) (

#/cm

3 )

diameter ( µm)

no-bag

0.1 1diameter ( µm)

pcasp

Remer et al. (1997)Remer et al. (1997)

From work byFrom work bySantiago Santiago GassGassóó


The moreThe moreabsorbing the aerosolabsorbing the aerosolthe more likelythe more likelyit will act to increaseit will act to increasecloud covercloud cover

Kaufman and KorenKaufman and KorenScience (2006)Science (2006)


We think that the aerosol absorption is the We think that the aerosol absorption is the keykey to tounderstanding the aerosol role in modifying cloud cover.understanding the aerosol role in modifying cloud cover.

We will approach this problem with a measurement-basedWe will approach this problem with a measurement-basedstrategy using a variety of instruments, but supported bystrategy using a variety of instruments, but supported bymodeling.modeling.

The end, as always, is a better estimate of the aerosolsThe end, as always, is a better estimate of the aerosols’’roles in influencing climate change and the hydrologicalroles in influencing climate change and the hydrologicalcycle.cycle.

Aerosol absorption and compositionAerosol absorption and composition

Work by J. Work by J. Vanderlei Vanderlei MartinsMartins

fine

fine

coar

seco

arse

From From SedeSede BokerBoker, Israel, Israel

Both size and absorption characteristics matterBoth size and absorption characteristics matterin determining composition.in determining composition.


More work byMore work byJ. J. Vanderlei Vanderlei MartinsMartins

Note that in the UV, there is a deviation from the Note that in the UV, there is a deviation from the λλ-1-1 curve curvethat characterizes that characterizes ““black carbonblack carbon””

Kirchstetter Kirchstetter et al. (2004) shows similar plots. Theet al. (2004) shows similar plots. Thedeviation is due to different combustion efficiencies.deviation is due to different combustion efficiencies.


wavelengthwavelength

Δρ

Δρ

ωωοο

fmffmf

basebase

The UV is sensitive to absorption. The The UV is sensitive to absorption. The midIRmidIR to size. to size.

Work byWork byViktor Viktor ZubkoZubko

We will combine We will combine OMIOMI and and MODISMODIS to cover this range. to cover this range.


Critical Reflectance

-0.1-0.08-0.06-0.04-0.02

00.020.040.060.080.1

0 0.1 0.2 0.3 0.4 0.5

Surface Reflectance

TO

A-R

su

p f

or

ae

ros

ols

Tau=0.4 Tau=0.6 Tau=0.8 Tau=1.0

ωo = 0.8

ωo = 0.95

ρρSUPSUP

ρρTO

ATO

A --

ρρ SUPSU

P

Relationship between Rcrit and ω oSeveral wavelengths (0.35- 2.5µm)

Smoke Aerosol Model

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 0.2 0.4 0.6 0.8 1ωo

Rcr

it

Work byWork byJ. J. Vanderlei Vanderlei MartinsMartins

There is a one-to-oneThere is a one-to-oneRelationship betweenRelationship betweenRcritRcrit and and ωωoo..


You need a varietyYou need a varietyof surface reflectancesof surface reflectancesso you need toso you need towork over land.work over land.

You need two images,You need two images,one with aerosol,one with aerosol,one without, havingone without, havingthe same geometry.the same geometry.


Critical ReflectanceCritical Reflectance

Path RadiancePath Radiance

0.900.90

0.700.70

0.800.800.850.85

0.750.75

0.950.95ωωoo

1.01.0

0.20.2

0.00.0

0.60.6

0.40.4

ττ

CriticalCriticalReflectance andReflectance and

ωωoo

Path RadiancePath Radianceand and ττ

2.02.0

MartinsMartins

Aerosol absorption: Laboratory workAerosol absorption: Laboratory workAerosol Mass Absorption Efficiency for Dust Samples

from Bodele/BODEX

0

0.05

0.1

0.15

0.2

0.25

0.3

300 600 900 1200 1500 1800 2100 2400

Wavelength (nm)

Ab

s. E

ff.

(m2/

g)

0

0.05

0.1

0.15

(1-ω

o)

fro

m A

ero

net

AeronetAVG and Stdev Dust 2AVG and Stdev Dust 1

Air Quality and long range transportAir Quality and long range transport

-18-18

-4-4

95953535

2323

1010

55

99

May -May -SeptemberSeptember

Kaufman et al. (2005)Kaufman et al. (2005)

Long rangeLong rangetransport oftransport ofdust.dust.

Air Quality and long range transportAir Quality and long range transport

It doesnIt doesn’’t have to be dust and it doesnt have to be dust and it doesn’’t have to be thet have to be theAtlanticAtlantic

Hongbin Hongbin Yu has been working in this area.Yu has been working in this area.

Other aspects of air quality:Other aspects of air quality:1.1. Producing an urban aerosol product from MODISProducing an urban aerosol product from MODIS2.2. Using results of INTEX-B to refine assumptions inUsing results of INTEX-B to refine assumptions in

retrievals.retrievals.

Fire radiative energyFire radiative energy

Looking Forward:Looking Forward:

1.1. We still have things to do with the MODIS algorithmWe still have things to do with the MODIS algorithm (urban product, nonsphericity, cloud artifacts) (urban product, nonsphericity, cloud artifacts)2.2. Some cloud Some cloud ““artifactsartifacts”” may be real and that would may be real and that would be exciting. be exciting.3.3. Aerosol absorption may be key to understandingAerosol absorption may be key to understanding aerosol influences on cloud cover. aerosol influences on cloud cover.4. We have 3 plans for increasing our knowledge on 4. We have 3 plans for increasing our knowledge on absorption (MODIS-OMI; critical reflectance;absorption (MODIS-OMI; critical reflectance; laboratory analysis). laboratory analysis).5.5. Fire emissionsFire emissions6. MODIS and6. MODIS and CALIPSO joint inversions.CALIPSO joint inversions.

The MODIS aerosol group (more or less) Sept 2006The MODIS aerosol group (more or less) Sept 2006

Announcing the arrivalsAnnouncing the arrivals of:of:Claire Yu, Jordan Levy, Claire Yu, Jordan Levy, Arava Arava KorenKoren

looking forward: using modis for aerosol science over the ......ο fmf base the uv is sensitive to...

Documents