Carmine Serio, Guido Masiello
DIFA, University of Basilicata, Italy
ITSC-18 Toulouse March 2012
Stephen Tjemkes,
Poster in this conference
ITSC-18 Toulouse March 2012
IASI stand-alone cloud detection scheme
The baseline
Test
Method
Threshold
hs-test
Look for similarity in shape of a
given observed spectrum with a
reference spectrum for clear sky
Critically depending on surface
emissivity
Surface temperature test
In super window channels the
corresponding BT is almost
proportional
to
surface
temperature (need a reference
surface temperature)
Depends on surface emissivity
The super-window tests @ 1168
and 830 cm-1
Look at the slope of the window
The slope can depend on surface
emissivty
The CO2 split-window test @ 791
cm-1
Sensitive to the pressure surface
Mostly depending
pressure
The LW-SW regression test
Look at the coherency between
short and long wave regions of the
spectrum sensitive to temperature
(CO2) channels
Depends on the surface properties
ITSC-18 Toulouse March 2012
on
surface
Our cloud detection approach and hs-methodology for
IASI has been largely published in the peer reviewed
literature
 Carmine Serio, Alberta Marcella Lubrano, Filomena Romano, and Haruisha Shimoda, "Cloud
Detection Over Sea Surface by use of Autocorrelation Functions of Upwelling Infrared Spectra in the
800–900-cm−1
Window
Region,"
Appl.
Opt.
39,
3565-3572
(2000)
 Guido Masiello, Marco Matricardi, Rolando Rizzi, and Carmine Serio, "Homomorphism between
Cloudy and Clear Spectral Radiance in the 800-900-cm-1 Atmospheric Window Region," Appl. Opt.
41,
965-973
(2002)

Masiello, G., Serio, C., Shimoda, H. Qualifying IMG tropical spectra for clear sky, Journal of
Quantitative Spectroscopy and Radiative Transfer, 77 (2), p.131-148, Mar 2003
doi:10.1016/S0022-4073(02)00083-3
 Guido Masiello, Carmine Serio, and Vincenzo Cuomo, "Exploiting Quartz Spectral Signature for the
Detection of Cloud-Affected Satellite Infrared Observations over African Desert Areas," Appl. Opt.
43, 2305-2315 (2004)
 Grieco, G., Masiello, G., Matricardi, M., Serio, C., Summa, D. and Cuomo, V. (2007), Demonstration
and validation of the φ-IASI inversion scheme with NAST-I data. Quarterly Journal of the Royal
Meteorological Society, 133: 217–232. doi: 10.1002/qj.162
ITSC-18 Toulouse March 2012
Surface emission can be separated from atmospheric emission
with Correlation Interferometry, that is by properly exploiting the
concept of partial interferogram
 Kyle,
T.G. Temperature soundings with partially scanned
interferograms, Appl. Opt., 1977, 16/2, 586 326-332.
 Smith, W.L.; Howell, H.B.; and Woolf, H.M. The use of interferometric
radiance measurements for sounding the atmosphere, J. Atmos.
Science 1979, 36, 566-575
 G. Grieco, G. Masiello, C. Serio, R. L. Jones, and M. I. Mead, "Infrared
Atmospheric Sounding Interferometer correlation interferometry for
the retrieval of atmospheric gases: the case of H2O and CO2," Appl.
Opt. 50, 4516-4528 (2011)
ITSC-18 Toulouse March 2012
Exemplifying the concept of emissivity as a low frequency component
through spectra with different emissivity, but same atmospheric state
vector. The emissivity signal is very strong in window regions
Spectrum
IASI Interferogram at full
range
0.14
400
0.1
-1
250
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
0.12
200
-2
-2
-1
Interferogram,(W m sr )
300
-1 -1
350
Interferogram,(W m sr (cm ) )
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
150
100
50
0
0.08
0.06
0.04
0.02
-50
-100
10
-3
-2
-1
10
10
optical path difference, (cm)
10
0
ITSC-18 Toulouse March 2012
0
800
1000 1200 1400 1600 1800 2000 2200 2400 2600
wave number, cm-1)
IASI INTERFEROGRAM CUT
BELOW 0.023 CM
SPECTRUM
400
0.14
0.1
-1
250
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
0.12
200
-2
-2
-1
Interferogram,(W m sr )
300
-1 -1
350
Interferogram,(W m sr (cm ) )
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
150
100
50
0
0.08
0.06
0.04
0.02
-50
-100
0
0.005
0.01
0.015
optical path difference, (cm)
0.02
0
800
1000 1200 1400 1600 1800 2000 2200 2400 2600
wave number, cm-1)
ITSC-18 Toulouse March 2012
IASI INTERFEROGRAM
BETWEEN 1.5 – 1.9 CM
DIFFERENCE SPECTRUM
0.1
8
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
-4
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
6
-1 -1
2
-2
0
4
-1
Interferogram,(W m sr (cm ) )
-2
-1
Interferogram,(W m sr )
0.05
x 10
-0.05
-0.1
0
-2
-4
-6
-0.15
1.5
1.6
1.8
1.7
optical path difference, (cm)
1.9
2
-8
500
1000
1500
2000
-1
wave number, cm )
ITSC-18 Toulouse March 2012
2500
3000
LOW FREQUENCY ANALYSIS
(THE NOISE HAS BEEN PROPERLY SCALED)
2
x 10
HIGH FREQUENCY ANALYSIS
(THE NOISE HAS BEEN PROPERLY SCALED)
-3
1.5
-1 -1
Difference,,(W m sr (cm ) )
-1 -1
-2
-4
-6
-2
-8
-10
500
0.5
-1
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
± 1σ
-1
-2
-4
1
0
Difference,(W m sr (cm ) )
x 10
0
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
± 1σ
-0.5
-1
1000
1500
2000
2500
3000
-1.5
500
-1
1000
1500
2000
-1
wave number, cm )
wave number, cm )
ITSC-18 Toulouse March 2012
2500
3000
Use the difference spectrum!
Spectrum
Difference spectrum
300
8
290
-4
masuda sea surface
deathvalley
Black Body
aridisol
entisol
mollisol
-1 -1
2
-2
260
4
-1
270
Interferogram,(W m sr (cm ) )
6
280
BT(K)
x 10
250
Masuda sea surface
Death Valley
Black Body
Aridisol
Entisol
mollisol
240
230
220
210
200
500
1000
1500
2000
2500
-1
wave number (cm
3000
0
-2
-4
-6
-8
500
1000
1500
2000
wave number, cm-1)
ITSC-18 Toulouse March 2012
2500
3000
50
60
40
40
30
Latitude (degrees)
Latitude (degrees)
80
20
0
-20
20
10
0
-10
-40
-20
-60
-30
-40
-80
-150
-100
50
0
-50
Longitude (degrees)
100
150
-50
-20
ITSC-18 Toulouse March 2012
0
40
20
Longitude (degrees)
60
80
Seviri cloud mask co-located with
IASI
Colocating SEVIRI to IASI
ITSC-18 Toulouse March 2012
Details of the co-location
80
60
40
Latitude (degrees)
Results: orbit 1
Sea Surface: Aden Gulf, Red sea and Indian ocean
Land surface: Turkey and Arabian peninsula, Eastern Africa
20
0
-20
-40
-60
-80
-150
-100
50
0
-50
Longitude (degrees)
100
Sea surface: coincidence table
Land surface: coincidence table
Seviri
Cloudmask
Seviri
Cloudmask
IASI d-spectrum
Cloud mask
150
IASI d-spectrum
Cloud mask
Clear
Cloudy
Total
Clear
Cloudy
Total
Clear
1803
(83%)
377
(17%)
2180
(100%)
Clear
7639
(89%)
949
(11%)
8588
(100%)
Cloudy
189
(18%)
861
(82%)
1050
(100%)
Cloudy
237
(16%)
1250
(84%)
1487
(100%)
Total coincidence: 82.5%
Total coincidence: 88.2%
ITSC-18 Toulouse March 2012
Blue is clear sky
Red is cloudy
ITSC-18 Toulouse March 2012
Blue is clear sky
Red is cloudy
ITSC-18 Toulouse March 2012
80
60
Results: orbit 2;
Sea surface: Mediterranean area, Atlantic Ocean (Namibia)
Land surface: Eastern Europe, Sahara desert, rain forest,
Kalahari desert
Latitude (degrees)
40
20
0
-20
-40
-60
-80
-150
Sea surface: coincidence table
Seviri
Cloudmask
IASI d-spectrum
Cloud mask
Clear
Cloudy
Total
Clear
1758
(89%)
209
(11%)
1967
(100%)
Cloudy
136
(5%)
2584
(96%)
2720
(100%)
Total coincidence: 92.6%
-100
-50
0
50
Longitude (degrees)
100
Land surface: coincidence table
Seviri
Cloudmask
IASI d-spectrum
Cloud mask
Clear
Cloudy
Total
Clear
6748
(97%)
226
(3%)
6974
(100%)
Cloudy
373
(12%)
2794
(88%)
3167
(100%)
Total coincidence: 94.1%
ITSC-18 Toulouse March 2012
150
80
60
40
Latitude (degrees)
Results: orbit 3;
Sea surface: Mediterranean area, then Atlantic Ocean
Land surface: Western Europe, Western Sahara desert, rain forest
20
0
-20
-40
-60
-80
-150
Sea surface: coincidence table
Seviri
Cloudmask
IASI d-spectrum
Cloud mask
Clear
Cloudy
Total
Clear
540
(82%)
121
(18%)
661
(100%)
Cloudy
50
(2.7%)
1764
(97.3%)
1814
(100%)
Total coincidence: 93.1%
-100
-50
0
50
Longitude (degrees)
100
Land surface: coincidence table
Seviri
Cloudmask
IASI d-spectrum
Cloud mask
Clear
Cloudy
Total
Clear
4202
(98%)
86
(2%)
4288
(100%)
Cloudy
432
(16%)
2298
(84%)
2730
(100%)
Total coincidence: 92.6%
ITSC-18 Toulouse March 2012
150
80
60
Latitude (degrees)
40
20
0
-20
-40
-60
-80
-150
-100
-50
0
50
Longitude (degrees)
100
150
Over 37626 IASI spectra covering
1.
2.
Land surface (desert soil, rain forest, organic soil, canopy,
mountains)
Sea surface
ITSC-18 Toulouse March 2012





Combining
Correlation
interferometry
with
hsmethodology we have devised a IASI stand-alone cloud
detection scheme, which can work on land and sea
surface.
The cloud detection looks at the clear-sky atmospheric
fingerprint and does not need any information about the
optical properties of the surface (emissivity)
A method has been developed, which uses the Chevalier
data base for the training. The surface emissivity has been
modeled with the Masuda sea surface emissivity.
No real IASI observations have been added to the training
data base.
The comparison with SEVIRI imagery and cloud mask
shows a coincidence which is higher than 90%.
ITSC-18 Toulouse March 2012