Requirement: Provide information to air quality decision makers and

1
Near-real-time Tropospheric NO2 Retrievals for
Air Quality Applications
T. Beck1 (Government Principal Investigator) and
S. Kondragunta1 1NOAA NESDIS STAR
Requirement Provide information to air quality
decision makers and improve NOAAs national air
quality forecast capability. Science How can we
use data from multiple sensors( GOME-2 and OMI)
to understand the diurnal variation of NOX(
nitrogen dioxide and nitric oxide)
Emissions? Benefit Will lead to improved
surface ozone forecasts. Will improve
identification of NOX sources and improved NOX
budget.

• Validation by comparison to OMI. 15 months of
  GOME-2 and OMI slant column NO2 data (2008-2009)
  were compared using SNO analysis. Number of
  matchups for this analysis were 77.
• SNO matchup criteria
• 2 minutes overpass
• Solar zenith angles less than 80o
• View zenith angles less than 40o (nadir)
• OMI row anomaly flag used
• Results
• Mean bias is 0.23 ( 2)
• Correlation coefficient is 0.85

Modeled and Measured Diurnal Variation of NO2 in
the U.S. Ozone Non-Attainment Areas
OMI and GOME-2 measurements are plotted as filled
circles. Continuous lines represent the CMAQ
forecast at both overpass times. OMI is in an
afternoon orbit and GOME-2 is in a morning orbit.
Plans are underway to work with NOAA/ARL and
NOAA/NWS to use the satellite retrievals in
constraining the NOx emissions in the CMAQ model.
The Global Ozone Monitoring Experiment-2 (GOME-2)
is one of the new-generation European instruments
carried on MetOp and will continue the long-term
monitoring of atmospheric NO2 started by GOME on
ERS-2. MetOp was launched on October 19, 2006.
The GOME-2 instrument is a nadir-scanning
UV/visible spectrometer. It includes four main
optical channels which focus the spectrum onto
linear silicon photodiode arrays of 1024 pixels
each, and two Polarization Measurements Devices
(PMDs). The four main channels provide continuous
spectral coverage of the wavelengths between 240
and 790 nm with a spectral resolution between
0.26nm and 0.51nm.
Operational Near Real Time NO2 Monitoring at NOAA
http//www.osdpd.noaa.gov/ml/air/home.html .
The tropospheric NO2 will become operational in
March 2010. The pre-operational images are
available at the above website
Solar Measurements
Algorithm The retrieval uses a two step
procedure to estimate tropospheric NO2. The 1st
step uses the measured spectroscopy in the blue
light region of the visible( 425nm to 455nm).
Using a non-linear least square solver the amount
of NO2 within the viewing field is estimated(
slant column). To be useful for applications,
the vertical column is needed to estimate the
vertical column, radiative transfer must be used
to relate the two quantities. We use the LIDORT(
LInearized DisORT) model.
The CMAQ model summertime Average at the GOME-2
overpass time. The model underestimates rural
NO2. The major sources are comparable.
GOME-2 measured weekday averaged NO2. The
overpass time is 1000AM. The measurements
generally agree however the rural background
values are higher.
The GOME-2 NO2 retrievals are supporting the EPA
AIRNow program. The Near-Real Time measurements
are expected to improve surface ozone forecasts.
http//www.airnow.gov
The profile shape of the Strato-spheric NO2. It
has a well defined Maximum.
In the troposphere Polluted NO2 profiles are
supplied by GEOS-CHEM modeling studies.
A wave 2 fit estimates the global background NO2.
METOP/GOME-2 ?
Once the total column NO2 is found the Reference
Sector Method is used to remove stratospheric
contribution. The assumption is that over much
of the oceans there is none or very little NO2
pollution present( NO2 has a short lifetime). An
average background NO2 amount is found using only
the unpolluted measurements. The unpolluted
field represents the stratospheric NO2. The
initial assumption for the AMF calculation is
there is no tropospheric NO2. When the total
column NO2 exceeds twice the standard deviation
of the zonal mean the measurement is assumed to
be polluted, a new AMF calculation is done.
Based on GEOS-CHEM model runs a polluted NO2
profile is interpolated. The polluted profile is
used to generate the tropospheric AMF.

• Science Challenges
• Lack of in-situ observations.
• Assimilation of data into air quality models.
• Data continuity (NPP/NPOESS OMPS instrument
  cannot make NO2 retrievals).
• No profile information.
• Retrieval efficiency and communication of that
  information to users.
• Next Steps Maintain science support to
  operational product processing
• including algorithm improvements.
• Transition Path Instrument calibration and
  modeling studies will flow
• into improved operational air quality forecasts.

Weekend and Weekday averages for OMI and GOME-2.
The drop in weekend NO2 is captured by both
instruments.
AURA/OMI ?