Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
Challenges and progresses in the intercalibrationof microwave humidity sounders
Isaac Moradi1∗, James Beuchamp1, and Ralph Ferraro2
1 University of Maryland, College Park, MD2 STAR, SCSB, NOAA, College Park, MD
∗ Now at GMAO, GSFC, NASA, Greenbelt, MD
NASA Sounder Science Team MeetingGreenbelt Marriott HotelGreenbelt, MD, 20770
September 16, 2016
AMSU-B/MHS Climate Data Records
AMSU-B Channels
Isaac Moradi Satellite Climate Data Records September 16, 2016
AMSU-B/MHS Climate Data Records
AMSU-B/MHS Level 1b to FCDR
Isaac Moradi Satellite Climate Data Records September 16, 2016
L1b to FCDR Corrections
RFI correction for AMSU-B at-tributed to a mismatch of mate-rials (significant for NOAA-15 andNOAA-17)
Antenna Pattern Correction forboth AMSU-B and MHS
Geolocation correction for allAMSU-A, AMSU-B, and MHSinstruments
Calibration drift (corrected throughintercalibration and RT simula-tions)
Scan Asymetry (ongoing)
NOAA−15 AMSU−A Channel 1, 1−1−2004 to 1−31−2004
120° E
135° E 150
° E
165° E
45° S
30° S
15° S
−50
0
50
NOAA−15 AMSU−A Channel 1, 1−1−2004 to 1−31−2004
10 ° W
0° 10
° E 20
° E
30° E
40° E
30 ° N
40 ° N
50 ° N
−50
0
50
AMSU-B/MHS Climate Data Records
AMSU-B Antenna Pattern Correction
Isaac Moradi Satellite Climate Data Records September 16, 2016
HEWISON AND SAUNDERS: MEASUREMENTS OF THE AMSU-B ANTENNA PATTERN 411
g 1.5
c 0
� u
I!! :::, 1.0
l ! m c c
.! !f 0.5
�,-.----E> PFM Channel 16
l!,..6,----..AA PFM Channel 17
l]i-----,EJ PFM Channel 19
2 3 4
Space View
Fig. 6. Predicted antenna temperature corrections for the 4 space views for channel 16 ¢, channel 17 6, and channel 19 D of the AMSU-B PPM. The correction should be added to the measured brightness temperatures.
16, 17, and 19 of the PPM assuming a mean Earth radiance equivalent .to 230, 270, and 255 K, respectively (derived by averaging corresponding measured SSM/T-2 brightness temperatures of the Earth), and the platform with 100% reflectance at these frequencies. For channel 16, space view 3 clearly has the lowest radiance with least influence from the Earth and platform and is predicted to measure the equivalent of 3.6 K which is 0.9 K greater than the nominal microwave brightnyss temperature of space (2.73 K). For channels 17 and 19 space views 2 or 3 are predicted to be 0.4 and 0.7 K warmer than cold space. This difference between the predicted radiance and the space radiance for each channel gives a correction factor which must be included in the calculation of the space view radiance for the radiometric calibration to avoid a consistent bias in the radiometric gain. Once the instrument is in orbit the relative radiances in the four space views can be examined to verify these predictions and also to provide a better estimate of the mean Earth brightness temperature. The stability of the space radiance around an orbit and for different seasons will also need to be monitored to determine if this correction factor can be assumed to be a constant.
2) Effect on Earth Views: Similarly the Earth view radiances can be corrected for the effect of cold space and platform radiation in the antenna side lobes. This may become significant especially for fields of view on the edges of the swath where cold space radiance will reduce the measured
g c 0
0.8
� 0.6
u
I!!
!.
� m c c
.! c cc
0.4
0.2
15
+----+ PFM Channel 16
30 45
Earth Pixel
60 75 90
Fig. 7. Predicted antenna temperature corrections for a mean Earth scene brightness temperature of 230 K for channel 16 of the AMSU-B PPM. The correction should be added to the measured brightness temperatures.
Earth view radiances. The predictions for an assumed Earth brightness temperature of 230 K using the measured antenna pattern are shown as the points in Fig. 7 for channel 16 of the PPM. Pixels 1 -25 are affected both by cold space and the warmer platform as shown in Fig. 5 and so the correction is reduced depending on the Earth radiance. On the space view side of the swath (pixels 70-90) only the cold space radiance influences these pixels and so the correction is larger than on the other side of the swath. The largest biases are for pixel 90 where for channel 16 the predicted reduction in the measured brightness temperature due to cold space in the side lobes is 0.6 K for a mean Earth radiance equivalent to 2.30 K. The effect will be larger for warmer Earth brightness temperatures.
VIII. CONCLUSION
The antenna patterns of the AMSU-B flight models have been measured on a Compact Antenna Test Range and all flight models met or were close to the required specifications. The main exceptions were channel 16 beam efficiencies being slightly less than required and channel 17 exceeding the 2% cross-polarization specification at one extreme of the scan. However, the extensive knowledge of AMSU-B's antenna performance based on the measurements reported here will allow accurate corrections to be applied to the data. Users
Hewison and Saunders, TGRS 1996.
AMSU-B/MHS Climate Data Records
Radiometric Correction
Isaac Moradi Satellite Climate Data Records September 16, 2016
Intercalibration of satellite data
Simultanous Nadir Observa-tions (SNO)
Averages over Tropical Oceans
Polar Averages (only nighttime)
Issues?
Slight Frequency Differences
Polariztion DifferencesIsaac Moradi, GMAO/GSFC/NASA - 10/16/2014
Diurnal Cycle of RH
Isaac Moradi Satellite Climate Data Records September 16, 2016
Diurnal amplitude of RH
Moradi et al., EGU ACP, 26, 2016.
Diurnal peak time of RH
NOAA-15 vs. NOAA-17 over ocean
10505
1050N-70N
10505
1030N-50N
10505
1030S-30N
10505
1030S-50S
2002 2003 2004 2005 2006 2007 2008 2009 201010
505
1050S-70S
Chan 1Chan 2
Chan 3Chan 4
Chan 5
NOAA-15 vs. NOAA-17 over land
10505
10 50N-70N
10505
10 30N-50N
10505
10 30S-30N
10505
10 30S-50S
2002 2003 2004 2005 2006 2007 2008 2009 201010
505
10 50S-70S
Chan 1Chan 2
Chan 3Chan 4
Chan 5
Land/Ocean double differences (N15 - N17)
2002 2003 2004 2005 2006 2007 2008 2009 2010date
10
5
0
5
10
∆ T
[K
]
Chan 1
Chan 2
Chan 3
Chan 4
Chan 5
Impact of Clouds (N15 - N17)
2002 2003 2004 2005 2006 2007 2008 2009 2010date
4
2
0
2
4
∆ T
[K
]
Chan 1
Chan 2
Chan 3
Chan 4
Chan 5
Polarization Difference (N17 vs. N18)
42024 FOV01
42024 FOV30
42024 FOV45
42024 FOV60
2002 2003 2004 2005 2006 2007 2008 2009 201042024 FOV90
Chan 1Chan 2
Chan 3Chan 4
Chan 5
L1b Tb Averages Over Tropical Ocean
240
250Chan 1
275
280
285 Chan 2
245
250
255 Chan 3
260
265
270Chan 4
2002 2003 2004 2005 2006 2007 2008 2009 2010265
270
275
280 Chan 5
N15N16
N17N18
N19MOA
Correction Schemes
225 230 235 240 245 250 255 260 265 270NOAA-17 Tb [K]
225
230
235
240
245
250
255
260
265
270N
OA
A-1
5 T
b [
K]
70N-90N
50N-70N
30N-50N
30S-30N
30S-50S
50S-70S
70S-90S
210 220 230 240 250 260 270 280 290NOAA-17 Tb [K]
210
220
230
240
250
260
270
280
290
NO
AA
-15 T
b [
K]
70N-90N
50N-70N
30N-50N
30S-30N
30S-50S
50S-70S
70S-90S
225 230 235 240 245 250 255NOAA-17 Tb [K]
225
230
235
240
245
250
255
NO
AA
-15 T
b [
K]
70N-90N
50N-70N
30N-50N
30S-30N
30S-50S
50S-70S
70S-90S
220 230 240 250 260 270NOAA-17 Tb [K]
220
230
240
250
260
270
NO
AA
-15 T
b [
K]
70N-90N
50N-70N
30N-50N
30S-30N
30S-50S
50S-70S
70S-90S
AMSU-B Chan. 1 AMSU-B Chan. 3
AMSU-B Chan. 2 AMSU-B Chan. 4
Level 1b Averages Over Tropical Oceans
42024 Chan 1
42024 Chan 2
42024 Chan 3
42024 Chan 4
2002 2003 2004 2005 2006 2007 2008 2009 201042024 Chan 5
N15N16
N17N18
N19MOA
FCDR Averages Over Tropical Oceans
42024 Chan 1
42024 Chan 2
42024 Chan 3
42024 Chan 4
2002 2003 2004 2005 2006 2007 2008 2009 201042024 Chan 5
N15N16
N17N18
N19MOA
AMSU-B/MHS Climate Data Records
Conclusions
The NWP calibrated AMSU-B measurements can be significantlyimproved by recalibrarion using RFI and APC information
The geolocation error in AMSU-B and MHS instruments is notsignificant compared to the geolocation error in AMSU-A
Tropical ocean and night-time polar regions can be used asradiometrically stable targets for intercalibration of micrwovehumidity sounders
Several channels of AMSU-B on NOAA-15 and NOAA-16satellites show a large calibration drift
Frequency and polarization differences do not allow tointer-calibrate AMSU-B and MHS
Isaac Moradi Satellite Climate Data Records September 16, 2016
Thank you for yourattention!