IASI FM2 on METOP A
Performances after 1.5 year in orbit
International TOVS Study Conference 16
7–13, May 2008
Angra dos Reis, Brazil
D. Blumstein1, E.Pequignot1, B.Tournier2, R.Fjortoft1, L.Buffet1, C.Larigauderie1,
T.Phulpin1, I.Gaudel1 and the IASI TEC Team
(1) Centre National d'Etudes Spatiales (CNES), Toulouse, France
(2) NOVELTIS, Toulouse, France
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein CNES DCT/PO/EV
Outline
■ First decontamination
■ Stability of the instrument
■ Spectra rejections
Š Day 2 improvements of processing
■ Intercalibration IASI / AIRS
see IASI performances assessment at the 1st IASI Conference
• http://smsc.cnes.fr/IASI
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
2
Evolution of contamination effect (ice) 2007-2008
IASI FM2 Instrument Noise (pixel 1)
0.7
NeDT at 280 K (K)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
650
850
1050
1250
1450
1650
1850
Wave Number (cm-1)
Spec
ITSC-16, Angra dos Reis, 7-13 May 2008
Spec
Dec Spec
2006
Dec Spec
2006
Dec
Apr 2007
2006
Apr 2007
Sept
Apr 2007
2007
Sept 2007
Feb 2008
Dec
2006
D.Blumstein DCT/PO/EV
3
After first IASI decontamination
IASI FM2 Instrument Noise (pixel 1)
0.7
NeDT at 280 K (K)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
650
850
1050
1250
1450
1650
1850
Wave Number (cm-1)
Dec 2006
Spec
end of March 2008
■ First IASI decontamination end of March 2008 (1.5 year after launch)
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
4
Evolution of the measured transmission loss
Evolution of maximum loss of transmision (at 850 cm-1)
Smoothed and debiased ratios
1.03
1
1.01
0.97
Ratio of calibration coefficients
0.95
0.93
0.91
0.89
0.87
0.85
0.83
0.9
0.85
20/03/2008
20/02/2008
20/01/2008
20/12/2007
20/11/2007
20/10/2007
20/09/2007
20/08/2007
20/07/2007
0.8
20/06/2007
850 cm-1
20/05/2007
2645
20/04/2007
2145
20/03/2007
1645
Wave number (cm-1)
20/02/2007
1145
20/01/2007
0.81
645
0.95
20/12/2006
Ratio of calibration coefficients
0.99
Date
■ First IASI decontamination
Š end of March 2008 (1.5 year after launch)
■ Successul
Š recovery of the initial radiometric noise IASI
• Initially measured beginning of December 2006
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
■ Contamination rate
Š Now 1/4 of the initial rate
Š No need for new decontamination before 2 years
5
Fixed Cube Corner Offset
Fixed Cube Corner Offset
■ Accurate
determination
-11
Š Std < 0.2 µm
-13
-12
■ Stability over
16 months
Offset (µm)
-14
Š Small drift
Š 1 µm
■ Period of
analysis
Š 4th Dec 2006
Š 31th Mar 2008
ITSC-16, Angra dos Reis, 7-13 May 2008
-15
-16
-17
-18
-19
-20
-21
500
1500
2500
3500
4500
5500
6500
7500
Orbit Number
Y0 CD0
D.Blumstein DCT/PO/EV
Z0 CD0
Y0 CD1
Z0 CD1
6
Interferometric Axis Position
■ Stability over 12 months
Š Negligible drift
Š Results provided at Anglet
conference confirmed
• Obtained at that time from 5.5
months of data
■ Period of analysis
Š 16 th Apr 2007
Š 1 th Mar 2008
11 July 2007
Parameter
updating
■ Small evolution 11 July 2007
Š Cause by parameter updating
(spectral database)
Š Amplitude 30 µrad
Š Equivalent to Δν/ν = 5 10-7
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
7
Long term evolution of other parameters
Black Body
5K
0.1 K
Optical Bench
Temperatures
Š 1 point per orbit
Š average over 1 orbit
■ Period of analysis
Š 16 th Apr 2007
Š 1 th Apr 2008
1 AVHRR
Pixel (~1km)
■ Long term evolution over 1 year
Geometry IASI / AVHRR
■ Reminder : orbital stability
verified during Cal/Val
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
8
Regulation margin of the detectors temperature
CBS power regulation (mW)
observation of 2 orbits every 2 months
10
9
8
power (mW)
7
6
5
4
3
2
1
0
30/11/2006
29/01/2007
30/03/2007
29/05/2007
28/07/2007
26/09/2007
25/11/2007
24/01/2008
24/03/2008
■ Stable on the long term
Š Small seasonal effect, small orbital variation
■ Very good indicator that the temperature of the detectors will be kept low
Š Strong impact on the radiometric noise for long wavelength
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
9
Fraction of Spectra rejected by on-board processing
End of Cal/Val
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
10
Fraction of Spectra rejected by on-ground processing
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
11
Spectra affected by Spikes
■ 98.5 % of earth views (groups of 4
soundings) not affected by spikes
Histogramme of the number of spikes per box 50 x 50 km2
affected by at least 1 spike
90%
82.17%
80%
■ Among the 1.5 % of earth views affected by
spikes
Š
Š
Š
Š
E.g. over the South Atlantic Anomaly (SAA)
82.2 % have more than 3 spectra available
97.9 % have more than 2 spectra available
99.8 % have more than 1 spectrum
70%
60%
50%
40%
30%
20%
15.75%
10%
1.92%
0.16%
3
4
0%
1
2
Histogram of the number of spikes per scan
■ If the 4 IASI pixels of each Earth View are
not assimilated
Š Dynamic selection of the selected sounding
increase availability of the measurements
1.4%
1.282%
1.2%
1.0%
0.8%
0.6%
■ On the long term
0.4%
0.246%
Š Proposal for Day 2 evolution of IASI
processing
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
0.2%
0.030%
0.003%
3
4
0.0%
1
2
12
Spectra not computed because of NZPD error
■ Small fraction of spectra not available
because not computed by on-board
processing
Š Between 0.15 % (Pixel 1&4) and 0.3 % (Pixel 2)
Š Stable over 9 months
■ Geographic repartition
Š 1 or 2 occurrences max per month per bin of
0.5 x 0.5 deg2
Feb 2008
Dec 2007
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
13
Spectra not computed because of NZPD error
■ Brightness Temperatures from the IIS imager
Š Black curve : Histogram of BT in the vicinity of rejected spectra
• 1/4 of the IIS image
Š Red curve : Histogram of BT in the IASI footprint for rejected spectra
■ Conclusion
Š Affected pixels : 0.3 %
Š Histogram of rejected pixels
FWHM = 25 K
Š Close shape of the 2 histograms
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
14
Proposed evolutions of IASI processing for the day 2
■ Add more detailed information for the cause of rejected spectra
Š Spike reason,
Š NZPD reason,
Š OverFlows,
Š Other
■ In case of a spike occurrence, provide B1 and B2 spectra when available
Š With proper flagging
■ Add AVHRR L1B cloud mask information in the L1C product
■ Add IIS image Brightness Temperature average and variance in L1C prod
Š For easy spatially uniform scenes detection
■ Add minor modifications for improving (or easing) the monitoring of IASI
performance by the TEC
Š No impact on the L1C products
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
15
Intercalibration IASI-AIRS (GSICS)
■ Method : for intercomparison at high spectral resolution
Š Precompute TF of each AIRS Spectral Response Function ÎAk, k=1..2378
Š IASI calibrated spectra (L1B or L1C TBC) Îinterferogram I0
Š For each k, Sk=TF-1(I0/Af*Ak) , … AIRS like spectrum, interpolated at nuk
• ÎSAIRS_like (k)
• Af is IASI apodisation function
(G if L1C spectrum, self-apodisation if L1B spectrum)
■ Validation
0.3 K
Š Over 2000 spectra from the TIGR dataset
Red: stddev, Blue : Bias
■ On-going activities
1. Intercomparison with GCC/GSICS
results
2. Increase the number of
comparison opportunities
(generalization of the SNO
concept)
-0.3 K
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
16
Conclusions
■ After more than 17 months in orbit
Š IASI is performing very well
• all mission requirements are met
• both instrument and processing
■ All performances very stable in the long term
Š Radiometry, spectral, geometry
■ During the routine phase, IASI Technical Expertise Center
(IASI TEC) located in CNES/Toulouse takes care of
Š In-depth Performance monitoring
Š Processing parameters updating
■ In parallel with the operational monitoring performed by the EUMETSAT
EPS/CGS teams
Š Near Real Time
Š Radiances monitoring (wrt Radiative Transfer)
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
17
Spares
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
18
IASI outages : SEU/SET anomalies
Date
Mission outage Origine
Subsystem
Détails
15/05/2007
~ 1day 5h
SEU
LAS
Target LAS T° corruption, due to an SEU on CCE or LAS
13/06/2007
~ 1day 7h
SEU
DPS
DMC checksum error
20/07/2007
~ 1day 9h
SEU
DPS
DPC pixel 3A checksum error
01/11/2007
~ 10h
SEU
DPS
DPC pixel 4A checksum error
08/11/2007
~ 11h
SEU
DPS
DMC checksum error
18/11/2007
~ 4days 9h
SET
DPS
Communication error IMS-DPS with OOLs on DPS voltages due to
an SET on DMC converter
04/02/2008
7h 50min
SEU
CCM
Overflow on ALU computation. Error disapear after reset which
confirms SEU, probably on CCE RAM
09/02/2008
3h 45min
SEU
DPS
DPC pixel 3B checksum error
SEU
OBDH
OBDH corruption zone OBDH (EDAC counter anomalie) without
mission outage
5 occurrences
*
■ Proton or Heavy ion events caused IASI to go into safe mode (1 or 2)
*
long outage due to detectors temperature stabilization after safe mode 2
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
19
2007-05-15-13:18
SEU laser
2007-11-08-11:03
SEU DMC memory
2008-02-09-10:37
SEU DPC 3B
2007-06-13-10:23
SEU DMC memory
2007-11-01-13:35
SEU DPC 4A
2007-11-18-01:31
SET DMC CV
2007-07-20-04:31
SEU DPC 3A
2008-02-04-12:46
SEU CCM ALU
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
20
IASI availability in Operational modes
% in each mode between 07/05/07 and 03/31/08
Operational
Modes
90 %
Non Operational
Modes
10 %
Normal OP
89%
External Calibration
1%
decontamination, IASI anomalies
and platform anomalies
■ EUMETSAT / CNES / ALCATEL Working group has proposed
recommendations for diminishing impact of SEU anomalies (on-board)
■ Before these recommendations are implemented
Š Strong involvement of the EUMETSAT and CNES operational teams to
reduce the duration of unavailability periods
Š E.g. IASI decontamination implemented consecutively to a plate-form
anomaly to save 2 days in non operational modes
ITSC-16, Angra dos Reis, 7-13 May 2008
D.Blumstein DCT/PO/EV
21