Inversion of combined OMIMODIS data sets for

1
Inversion of combined OMI-MODIS data sets
for detailed aerosol characterization
Annual report 2007
Lorraine Remer NASA/GSFC With S.K.
Satheesh/IIS Omar Torres/Hampton Univ. J.
Vanderlei Martins/UMBC
2
Year 1 Objectives and Initial Challenge

• Objective 1. Co-locate and intercalibrate MODIS
  and OMI at common
• wavelengths

Initial Challenge - Because OMI does not
fly at precisely MODIS nadir point, the two
instruments view each target with different
geometries. Combining the radiances into one
super instrument will require parallel Look Up
Tables with simultaneous different geometry, and
will not be easy.
Objective 2. Evaluate whether MODIS L2 retrievals
can provide information to OMI and improve OMI
retrievals. Using L2 instead of radiances
directly avoids geometry issues. Objective 3.
Develop method to observe enhanced absorption in
the UV to further remote sensing and
identification of organic aerosols.
3
Comparison of OMI and MODIS at the same
wavelength (466 nm)
Omar Torres Changwoo Ahn
OMI Scan Angle
The differences in viewing geometry of the same
target create differences in OMI-MODIS
reflectance measurements at 0.466 µm of nearly
20 in some geometries. These differences make a
joint retrieval using L1 radiances from both
instruments extremely challenging.
4
However, we can avoid geometry by using L2
products. We can extrapolate MODIS AOD from the
visible to the UV and match measured AERONET
values.
Maldives
Ascension Island
MODIS AOD 340 extrap
AERONET AOD 340 nm
AERONET AOD 340 nm
Non-linear extrapolation to the UV
S.K. Satheesh
5
When MODIS supplies the AOD to the OMI retrieval,
the OMI retrieval Is free to obtain aerosol
height and single scattering albedo. Currently
we do not have validation for the single
scattering albedo, but we do have validation for
height.
Arabian Sea
Adding MODIS info to OMI retrieval greatly
improves OMI height product, compared to
OMI initial assumption based on climatology
Layer height from OMI (km)
S.K. Satheesh
Aerosol layer height from airborne lidar (km)
6
Because we see strong improvement in the
height retrieval, we also expect a more
accurate retrieval of single scattering albedo in
the UV, but this remains unvalidated.
Ground truth measurements of absorption in the
UV are rare.

• We are developing methods, applied to
  ground-based
• observations to make these UV absorption
  measurements.
• Modified shadow-band radiometer deployed in Santa
  Cruz,
• Bolivia shows promising results (Krotkov and
  Torres).
• 2. Continue filter collection and analysis
  (Martins).

(1) and (2) are funded under linked but separate
proposals
7
We have developed a method to identify absorbing
orgaincs in the AERONET sun observations, similar
to what has been shown in the filter data below.
Our AERONET observations of this enhanced UV
absorption are the FIRST observations of the
absorption in total column, ambient conditions.
8
Objectives for Year 2

• Complete and submit two papers on (a)
  improvements to
• the OMI retrieval using MODIS AOD and (b)
  observations
• of enhanced UV absorption using AERONET
• 2. Incorporate MODIS retrieved information on
  aerosol model
• to the OMI inversion, for even more accurate
  OMI results.
• 3. Investigate the possibility of making a two
  channel UV
• retrieval with a MODIS-OMI combination.
  Spectral absorption
• information in the UV will indicate enhanced
  organic
• absorption from space.
• 4. Expand, enhance and intercompare our abilities
  to measure
• absorption in the UV from the ground to be used
  as validation
• for satellite retrievals.