AIRS-TES T and H2O Comparison

1
Improved Tropospheric Ozone Profiles using OMI
and TES Radiances John Worden1, Xiong Liu2,
Kevin Bowman1, Kelly Chance2, Reinhard Beer1,
Annmarie Eldering1, Michael Gunson1, and Helen
Worden1 1 Jet Propulsion Laboratory, California
Institute of Technology 2 Harvard Smithsonian
2
Scientific and Technical Objectives

• Global boundary layer ozone measurements are
  critical for characterizing the venting of ozone
  from polluting regions and the subsidence of
  pollution into populated areas.
• Currently TES has limited boundary layer ozone
  measurement capability (hot polluted conditions)
  and OMI ozone estimates cannot distinguish
  boundary layer ozone from free troposphere.
• But what if we combine the two radiance
  measurements?

3
Case Study
Simulated True Ozone profiles from GEOS-CHEM
for the Caribbean, Texas, and Midwest Use TES
observed temperature and water over the
Caribbean, Texas, and the Midwest during November
2005 (challenging thermal conditions!) Combine
TES and OMI Jacobians, a typical signal-to-noise
of TES and OMI radiances, along with TES a
priori constraints (and constraint matrices) to
linearly calculate ozone profile retrievals of
each scene No clouds, but aerosol loading for
ocean and land are included in UV radiance and
Jacobian calculation
4
Linear or Synthetic Profile Retrieval
Xa a priori (same for all profiles, clean
atmosphere)
A Averaging kernel
Key to optimally combining TES and OMI
Kweighting function
Sm Measurement Error Covariance
5
(No Transcript)
6
Improved resolution in boundary layer
All OMI Averaging Kernels show sensitivity to
boundary layer
Vertically resolved estimates of free troposphere
ozone
7
(No Transcript)
8
Summary
Combining TES IR and OMI UV measurements results
in significant vertical resolution improvement in
boundary layer ozone retrievals and in the free
troposphere Global measurements of boundary
layer ozone and other pollutants have been
identified as a key requirement for future air
quality satellite missions (Air Quality Workshop
February 2006) We can use existing data sets and
algorithms to assess requirements for such
missions Paper on this topic recently submitted
to GRL.