Organic Carbon during AMAZE08: Preliminary model analysis

1
Organic Carbon during AMAZE-08Preliminary model
analysis
Colette L. Heald (heald_at_atmos.colostate.edu)
With observations from Qi Chen Scot Martin
(Harvard University), Delphine Farmer Jose
Jimenez (CU Boulder), Thomas Karl Alex Guenther
(NCAR), Paulo Artaxo (University Sao Paulo) And
discussion with Andi Andreae, Tony Prenni
AMAZE Science Team Meeting, Boulder July 24, 2008
2
GEOS-Chem AEROSOL SIMULATION
2ºx2.5º global simulation with climatological
emissions
FEB Surface Concentrations
Site isolated from anthropogenic sources, monthly
mean aerosol concentrations dominated by OC and
dustUnfortunately with consistent cloud cover,
no satellite AOD available to validate
distributions
3
ORGANIC CARBON GEOS-Chem
Simulated surface OC at site 0.5-1 µg/m3,
dominated by POA and monoterpene SOA. NOTE
there is no PBAP source here.
4
COMPARISON WITH AMS OBSERVATIONS
Field site in Amazonia, Feb-Mar
Early Feb observe significantly more organic
aerosol than simulated (rain ends this period)
Observations 1-4 µg/m3
Consistent air flow throughout campaign
Point 1 For majority of campaign GEOS-Chem
simulates similar amount of OC (1 µg/m3) Point
2 Observed concentrations are relatively low
(somewhat explains 1) Point 3 Theres an
interesting discrepancy early in the
campaign.source?? Point 4 Consistent airflow
throughout campaign simplifies analysis no
change in source influence
5
ARE FIRES A LIKELY SOURCE?
Fires in the region during early Feb. These are
not reflected in model emission inventories.
Feb 1-9
Feb 21-29
MODIS fire counts http//maps.geog.umd.edu/firms/
maps.asp
Elevated OC is coincident with elevated
acetonitrile and potassium. MODIS detects
actives fires in the immediate region.
6
WHAT ABOUT FIRES IN AFRICA?
Surface OC concentrations simulated by
GEOS-Chem (with climatological fire source)
Fires from early Feb transported across Atlantic?
OC from African fires could be transportedor
potassium from African dust?
7
CONCLUSIONS(SO FAR)

• No significant unexplained sub-micron OC
  observed at field site
• Elevated OC in early Feb likely fire related
• hope that a GEOS-Chem simulation with 2008 fire
  emissions can help determine if South American or
  African fire source is likely
• Note that the AMS m/z 60 / Org ratio does not
  indicate a strong fire influence

Interestingly, This leaves little room for
underestimated BSOA formation or sub-micron PBAP
How does this mesh with our thinking on PBAP?