Quantitative Interpretation of Satellite and Surface Measurements of Aerosols over North America

Quantitative Interpretation of Satellite and Surface Measurements of Aerosols over

North America

Aaron van Donkelaar

M.Sc. Defense

December, 2005

Aerosols – Why do we care?

• Climate Change– Direct Effect– Indirect Effect

• Health Effects (PM2.5)– Lung cancers– Pulmonary Inflammation

• Visibility

Image from http://cariari.ucr.ac.cr/~faccienc/temas2/planeta.htm

Part I – Remote Sensing of Ground-Level PM2.5

1

3

4 eQrM

115.2,

3

5.2 3

4zrQ

r

rPM ed

zd

Column Mass Loading:

Ground-Level PM2.5:

• ρ – particle mass density• r – effective radius• τ – aerosol optical depth

• Qe – Mie extinction efficiency

• z – Height of regional air mass

subscript d denotes dry conditions

InstrumentationMODIS• Moderate Resolution Imaging

Spectroradiometer• 32 channels (7 used for

Aerosol Retrieval): 0.47, 0.55, 0.67, 0.87, 1.24, 1.64 um

• Approx. daily global coverage• Requires dark surface for AOD

retrieval

MISR• Multiangle Imaging

Spectroradiometer• 4 spectral bands at 9 different

viewing angles• 6-9 days for global coverage• No assumption regarding

surface reflectivity

GEOS-CHEM

• 50 Tracers• 1º x 1º resolution• 30 vertical levels (lowest at ~10, 50, 100, 200, 300 m)• GMAO fields: temperature, winds, cloud properties, heat flux and

precipitation• sulphate, nitrate, mineral dust, fine/coarse seasalt, organic and black

carbon• Aerosol and oxidant simulations

coupled through– formation of sulphate and nitrate

– heterogeneous chemistry

– aerosol effect of photolysis rates

• Seasonal average biomass burning

Remote vs. Ground PM2.5

MODIS MISR

standard 0.68 0.54

constant vertical structure (τz/τ) 0.29 0.29

constant AOD 0.54 0.38

constant aerosol properties (Qe, r, rd, ρd)

0.73 0.52

115.2,

3

5.2 3

4zrQ

r

rPM ed

zd

Scatter Plot Comparison/Table Holding Constants

Temporal Correlation

Glo

bal P

M2.

5

Part II –Organic Aerosol Sources

• Primary Sources:– combustion (biomass/biofuel)

• Secondary Sources:– condensation of gaseous species– not well understood

• GEOS-CHEM OA Simulation– Seasonally varying biomass burning inventories– Inversion removed– SOA based upon Chung and Seinfeld [2002]

• Biogenic emissions from MEGAN inventory

• HxCy + (O3, OH, NO3) → semi-volatile products

IMPROVE Organic Aerosol

IMPROVE – GEOS-CHEM Organic Aerosol

Isoprene conversion fits within model biases

Large effect from non-OA condensation

Conclusions

• Remote PM2.5

– significant correlation (MODIS: R=0.68, MISR:0.54)– dominant factors include AOD and vertical structure

– reveals global regions of high PM2.5

• Sources of Organic Aerosol– isoprene conversion reduces model bias– non-OA condensation unclear