1 of 14

Evaluation of GEOS-CHEM Aerosol Optical Thicknesses : comparisons to satellite data for 2003

S. Generoso & I. Bey

LMCA, EPFL, Lausanne, Switzerland

Acknowledgements : F.M: Bréon (POLDER data),

Q. Li & R. V. Martin (Data for the Mie calculation)

2 of 14

T2 : projet / questions scientifiq

We plan to use GEOS-CHEM to ...

1. Characterize aerosol distributions (e.g. anthro. vs natural) / focus over the Asian continent and Pacific/Indian Oceans

2. Investigate processes (e.g. emission heights, deposition, transport …)

comparisons with multi observations (satellites, ground-based measurements)

3 of 14

T2 : projet / questions scientifiq

The method chosen to start

Year 2003 : Both POLDER & MODIS data

Large interannual aerosol events in Asia during spring/summer 2003 (to characterize)

Data : POLDER, MODIS, AERONET … (others are welcome)

GEOS-CHEM : version used v07-02-03

Resolution 2°x 2.5°

30 levels

GEOS 4

GEOS-CHEM : version v07-02-03

Modifications for this study :

--> AOT @ 550 & 865 nm on-line

--> biomass burning emissions for 2003 based on ATSR fire counts

4 of 14

ATSR fire count and emission inventories

Adapted from Generoso et al, ACP, 2003

Definition of “homogeneous” regions

An emission constant (g/detected fire) computed for each region

ATSR fire counts (night time)

Available 07/96 => 10/03

annual emission used in GEOS-CHEM

(XXX.bioburn. seasonal.geos.2x25)

Seasonal cycle, Interannual variations and Spatial distribution follow satellite observation

Emitted quantities depend on the inventories used

5 of 14

T2 : projet / questions scientifiq

Differences introduced in the BB emissions used in GEOS-CHEM ...

OC in kg /year (x 1010)

“ATSR” “CURRENT”

2000 1.70 2.11

2002 2.25 2.67

2003 2.00 2.34 (CLIM)

CURRENTLY USED IN GEOS-CHEM v-07-02-03

THIS STUDY

… in terms of emitted quantities :

6 of 14

… in terms of spatial distribution

CLIM

THIS STUDY

OCPO Emissions JUNE 03

kg /month

1.4 x 108 kg

7.3 x 108 kg

7 of 14

The POLDER mission

(Polarization and Directionality of the Earth Reflectance)

POLDER-1 : Nov 96 to Jun 97

POLDER-2 : Apr 03 to Oct 03

(On ADEOS)

Originality : polarization - over ocean and land

~ column aerosol load in the fine mode

Classical method - Over ocean only

AOT ~ column aerosol load

May 2003

Fine mode AOT

Total AOT

(Aerosol Optical Thickness)

8 of 14

Comparisons to POLDER AOT : FINE mode

Sulf + BC + OC + SSac

Overestimate in Northern Hemisphere (mainly sulfates over Europe)

Underestimate in B.B. regions

9 of 14

Comparisons to POLDER AOT: COARSE mode

DUST + SSc

Underestimate in Bay of Bengal

Dust export to Atlantic rather well represented

10 of 14

Conclusions

1. Biomass Burning emissions for 2003 available

May be underestimated

Can be provided when the ATSR data are available (from 96/07 to 03/10)

2. AOT @ 550 & 865 nm available

3. 1st comparisons to satellite data show that there is a lot to investigate

  • Overestimation of fine mode AOT in Northern Hemisphere (mainly sulfates)
  • Underestimation of fine mode AOT in BB regions
  • Underestimation of coarse mode AOT in Bay of Bengal

🡪 Dust export to Atlantic rather well represented

11 of 14

Additional slides

12 of 14

T2 : projet / questions scientifiq

... in terms of seasonal cycles

month

OC in kg C

Clim

2003 “ATSR”

2002 “ATSR”

2002 “Current”

13 of 14

Spatial distribution … modified

initial

ATSR 98

ATSR 00

Burnt Area 00

Better agreement with burnt area observations

14 of 14

ATSR vs other biomass burning sat. products

Sahel

Indonesia

North Australia

South America

ATSR

AVHRR

TRMM

GLOBSCAR

GBA2000

The ATSR night time restrictions show a seasonal cycle that is consistent with the other proxies