Jennifer Wei1,2, Antonia Gambacorta1,2,

1
AIRS/IASI Trace Gas Products (Level 2)

• Jennifer Wei1,2, Antonia Gambacorta1,2,
• Eric Maddy1,2, Xiaozhen Xiong1,2, Fengyin Sun1,2,
  
• Xingpin Liu1,2, Murty Divakarla2,3..
• Chris Barnet2, Mitch Goldberg2
• START08/pre-HIPPO Workshop
• Jan. 09, 2008

1Perot Systems Government Services 2NOAA/NESDIS/ST
AR 3IMSG
2
Trace Gas Product Potential from Operational
Thermal Sounders
Haskins, R.D. and L.D. Kaplan 1993
3
Example of Ozone from AIRS

• What has been learned so far..
• High degree of consistency with dynamical
  variability of UTLS
• Realistically map chemical transitions between
  stratosphere and troposphere
• Show reasonable agreement with aircraft data over
  a large dynamical range of ozone
• Comparisons with ozonesonde show good agreement
  between 400-50 mb range

Bian J., A. Gettelman, H. Chen, L. L. Pan (2007),
Validation of satellite ozone profile retrievals
using Beijing ozonesonde data, J. Geophys. Res.,
112, D06305, doi10.1029/2006JD007502. Monahan
K. P., L. L. Pan, A. J. McDonald, G. E. Bodeker,
J. Wei, S. E. George, C. D. Barnet, E. Maddy
(2007), Validation of AIRS v4 ozone profiles in
the UTLS using ozonesondes from Lauder, NZ and
Boulder, USA, J. Geophys. Res., 112, D17304,
doi10.1029/2006JD008181. Divakarla et al.
(2008), JGR-A, submitted.
4
X
5
Ozone A priori for Version 6 retrieval
Consideration of a tropopause referenced
climatology
Altitude
Tropopause referenced Relative Alt.
Pan et al, 2004
6
Example of Carbon Monoxide from AIRS

• What has been learned so far.
• CO can be used to estimate horizontal and
  vertical transport during combustion events.
• CO can be used to help us distinguish combustion
  sources (fossil or biomass) from other
  sources/sinks in our methane and carbon dioxide
  products.
• CO and Ozone may help us improve atmospheric
  vertical transport models in the mid-troposphere.

Warner, J., M. M. Comer, C. D. Barnet, W. W.
McMillan, W. Wolf, E. Maddy, and G. Sachse
(2007), A comparison of satellite tropospheric
carbon monoxide measurements from AIRS and MOPITT
during INTEX-A, J. Geophys. Res., 112, D12S17,
doi10.1029/2006JD007925.
7
AIRS CO and Trajectories
500 mb 700 mb 850 mb
CO from southern Alaska Fires was transported to
Europe at high altitudes (5 km)
CO from northern Alaskan fires was transported to
the lower atmosphere in SE of US
8
Example of Methane from AIRS (Xiaozhen.Xiong_at_noaa.
gov)

• What has been learned so far
• The accuracy is about 0.5-1.5 depending on
  different altitudes, and sensitive region is at
  200-300mb in the tropics and 300-500 mb in the
  high northern hemisphere (HNH).
• Observed significant summer enhancement of CH4 in
  HNH (possibly due to wetland emissions/thawing
  permafrost, paper submitted to GRL by Xiong et
  al.)
• Observed significant plume of CH4 over the
  Tibetan Plateau (collaborate with S. Houweling,
  paper in preparation)
• Use AIRS CH4 in conjunction with model
  simulations to better quantify the source region

9
Comparison of CH4 product ESRL/GMD Continuous
Ground Site
Barrow Alaska
3deg. x 3deg. gridded retrieval averaged over
60-70 lat, -165 to -90 long.
10
AIRS CH4 comparison to ESRL Aircraft
Xiong, X., C. Barnet, C. Sweeney, E. S. Maddy, X.
Liu, L. Zhou, and M. D. Goldberg (2008),
Characterization and Validation of Methane
Products from the Atmospheric Infrared Sounder
(AIRS), J. Geophys. Res., doi10.1029/2007JG000500
, in press.
11
CH4 plume over Tibetan Plateau
Paper is in preparation
12
Xiong et al., Satellite Observed Increase of
Tropospheric Summer Methane Concentration Is it
due to Wetland Emission over the High Northern
Hemisphere?, GRL, 2008 (submitted)
AIRS CH4 at 300 mb
Courtesy of L. Pan
13
Example of Carbon Dioxide from AIRS (Eric.Maddy_at_no
aa.gov)

• What has been learned so far.
• Maximum measurement sensitivity from AIRS in the
  middle to upper troposphere broadly weighted
  column measurement.
• Retrievals require significant spatial and
  temporal averaging (5 day / 400 km) to improve
  S/N.
• Total uncertainty in middle-to-upper troposphere
  
• 1 ppmv in tropics vs. high altitude aircraft
  (JAL Matsueda)
• 2 ppmv in middle/high latitudes vs. ESRL/GMD
  aircraft.

CO2 Retrievals from the Atmospheric Infrared
Sounder Methodology and Validation, Maddy, E. S.
and Barnet, C. D. and Goldberg, M. D. and
Sweeney, C. and Liu, X., Accepted to JGR-A
14
Validation AIRS CO2 (6km 8km) vs ESRL/GMD
Aircraft (2.5km 8km)

• Right Taylor diagram Taylor, JGR, 2001 for
  2005 aircraft matchups illustrates retrieval
  skill (end of arrow) relative to a priori
  (beginning of arrow).

• Left AIRS retrieval and ESRL aircraft timeseries
  at Poker Flat, Alaska shows good agreement in
  placement of seasonal cycle and year-to-year
  variability.

15
ESRL/GMD Aircraft vs. AIRS Retrieval CO2
16
NOAA AIRS CO2 Product is Still in Development

• Product is CO2(p) profile with associated
  averaging kernels.
• Measuring a product to 0.5 is inherently
  difficult
• Cloud clearing error (also error estimates)
  strongly impacts the CO2 product.
• Errors in moisture of 10 is equivalent to 0.7
  ppmv errors in CO2.
• Errors in surface pressure of 5 mb induce 1.8
  ppmv errors in CO2.
• AMSU side-lobe errors corrupt the ability to use
  the 57 GHZ O2 band as a T(p) reference point.
• Reduction of Core product retrieval errors is
  critical for CO2.
• Currently, we can characterize seasonal and
  latitudinal mid-tropospheric variability to test
  product reasonableness.
• The real questions is whether thermal sounders
  can contribute to the source/sink questions.
• Requires accurate vertical horizontal transport
  models
• Having simultaneous O3, CO, CH4, and CO2 products
  is a unique contribution that thermal sounders
  can make to improve the understanding of
  transport.

17
IASI AIRS Global Measurements 4 Times/Day
METOP
AQUA

• Initial Joint Polar System an agreement between
  NOAA EUMETSAT to exchange data and products.
• NASA/Aqua in 130 pm orbit (May 2002)
• EUMETSAT/IASI in 930 am orbit (October 2006)

18
The NOAA Unique Level 2
Processing System

• The NOAA level 2 processing is a unique system to
  compute atmospheric core and trace gas products.
• The whole architecture is a file-driven system
  compatible with multiple instruments.
• This system has been developed during the Aqua
  mission, using AIRS/AMSU/MODIS Instruments.
• Although the system was built for AIRS, it was
  designed to be expandable for both IASI and CrIS.
• This system has been thoroughly validated using
  several in-situ measurement campaigns (e.g.,
  ESRL/GMD Aircraft, JAL, INTEX, etc.)

• This system is a reliable, well tested and fast
  package that we are migrating into operations for
  IASI.

19
IASI AIRS Carbon Monoxide (
October 22nd 2007)
1
2
3
4
20
Spectral Coverage Comparison AIRS, IASI, CrIS
CO2
O3
CH4
CO2
CO
21
Preliminary selection of IASI channels for
physical retrieval.(NOTE All channels except
non-LTE are used in regression)
Ignored non- LTE
22
Instrument Noise, NE?T at 250 K
CO2
CO2
CH4
CO
23
RMS Simulation Inter-Comparison
24
Preliminary validation resultsTemperature,
water vapor, ozone(focus day October 19th, 2007)
25
Selected IASI Nighttime Ascending Granules
26
Near AIRS Nighttime Descending Granules
(4 hour difference)
27
Standard deviation of retrievals-ECMWFIASI
(blue), IASI CLEAR (red) AIRS (cyan), AIRS CLEAR
(green)
Ozone, O3(p)
Temperature, T(p)
Water, q(p)
Standard Deviation w.r.t. ECMWF
Dashed lines are NOAA Cloudy Regression and Solid
lines are Physical Retrieval Using Physical QA
28
Bias of retrievals ECMWFIASI (blue), IASI
CLEAR (red) AIRS (cyan), AIRS CLEAR (green)
BIAS w.r.t. ECMWF
Dashed lines are NOAA Cloudy Regression and Solid
lines are Physical Retrieval Using Physical QA
29
Towards Operational Status (April 08)

• Have not installed the regression derived from
  cloud cleared radiances. Will be installed in
  Jan. 2008.
• Have not computed tuning for AMSU MHS (used
  Aqua AMSU tuning). Will be installed in Jan.
  2008.
• Have not installed the local angle correction
  (needed for cloud clearing). Will be installed
  Jan. 2008.
• No attempt has been made to perform sub-pixel ILS
  correction.
• There is an advantage to cloud clearing in that
  FOVs are averaged with clearest having highest
  weight.
• This will be studied and installed in version 2.
• Only quick optimization has been done.
• Need to derive optimal functions regularization
  parameters, Jan/Feb. 2008.
• Preliminary list of channels looks good, minor
  changes to channel list.
• Pre-launch Radiative Transfer Algorithm (RTA).
  Post-launch RTA from UMBC, expected any time soon
• Empirical bias corrections, empirical noise term
  (to compensate for sub-pixel ILS), etc. are still
  very crude.

30
Our interest in participating to the START
08/pre-HIPPO campaign

• Compare with in situ co-located trace gas
  measurements to validate and assess the
  performance of IASI L2 products
• Exchange data and products
• 
  chris.barnet_at_noaa.gov
• 
  antonia.gambacorta_at_noaa.gov