MODIS Geolocation Status · accuracy goal (50 m) •Large errors occur ~1.5 hr after maneuvers...
Transcript of MODIS Geolocation Status · accuracy goal (50 m) •Large errors occur ~1.5 hr after maneuvers...
(Terra, Aqua) MODIS Geolocation Status
MODIS-VIIRS Science Team Meeting Calibration Workshop
April 15, 2013 5/1/2014 5/18/20156/6/2016
Guoqing (Gary) Lin,
Robert Wolfe, James Kuyper, Zhangshi Yin, and Bin Tan
Geometric Calibration Group
NASA GSFC Code 619
2
Geolocation C5/C6 results
• Excellent results –Root Mean
Square Error (RMSE) in nadir
equivalent units is better than
accuracy goal (50 m)
• Large errors occur ~1.5 hr after
maneuvers (about 12 per year)
- accuracy in following orbit
suspect
• Good results – RMSE is better
than goal (50 m) in track direction
but slightly over goal in scan
direction (but much better than
specification – 150 m)
• Definitive ephemeris is used for
best results – causes up to 27 hr
processing delay
Terra Aqua
C5 C6 C5 C6
Along-track RMSE (m) 44 43 47 46
Along-scan RMSE (m) 44 44 53 53
Years of Data 15.8 16.0 13.6 13.8
Ground Control Point Match-ups/day
Days missing (no residuals)
260 218
134 62
224 189
99 10
Terra Aqua
Results up to April 30, 2016
Note: These results are for MODIS Band 1, which is used in the control point matchup. Other bands must be offset by the band-offsets published by the MODIS calibration team. Lin et al., 6 June 2016
Terra trend and update details
Lin et al., 6 June 2016
Lin et al., 6 June 2016 4
Terra C6 long-term trend (uncorrected)
RMSE with no correction: Track: 49 m (+6 m vs C6) Scan: 48 m (+4 m vs C6)
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
Lin et al., 6 June 2016
Lin et al., 6 June 2016 5
Actual Terra C5 residuals
C5 RMSE Track: 44 m Scan: 44 m, nadir equivalent
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
Lin et al., 6 June 2016 6
Actual Terra C6 residuals
C6 RMSE Track: 43 m Scan: 44 m, nadir equivalent
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
Terra Safe Hold
• Pointing/geolocation look ok after re-activation Lin et al., 6 June 2016 7
-75
-50
-25
0
25
50
75
16.00 16.08 16.17 16.25 16.33
Tra
ck r
es (m
) .
Years since Jan. 1, 2000
-75
-50
-25
0
25
50
75
16.0 16.1 16.2 16.3 16.3
Scan r
es(m
) .
Years since Jan. 1, 2000
Aqua trend and update details
Lin et al., 6 June 2016
Lin et al., 6 June 2016 9
Aqua C6 Long-term Trend (unorrected)
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
Lin et al., 6 June 2016
RMSE with no correction: Track: 68 m (+22 m vs C6) Scan: 54 m (+1 m vs C6)
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
Lin et al., 6 June 2016 10
Actual Aqua C5 residuals
C5 RMSE Track: 47 m, Scan: 53 m, nadir equivalent
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
Lin et al., 6 June 2016 11
Actual Aqua C6 residuals
C6 RMSE Track: 46 m, Scan: 53 m, nadir equivalent
next slide
-75
-50
-25
0
25
50
75
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
-60
-40
-20
0
20
40
60
11 11.5 12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17
Sca
n (
ad
j.)
res.
(m)
Years since Jan. 1, 2000
-60
-40
-20
0
20
40
60
11 11.5 12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17
Tra
ck (a
dj.)
res.
(m)
Years since Jan. 1, 2000
Daily 16-day GlobalEffects of AMSR-E (in)activities
Lin et al., 6 June 2016 12 • LUT update soon (shown here are plots from C6)
AMSR-E, 0->2 rpm
2013-12-5, 16:50 UTC
AMSR-E, 2->0 rpm
2015-12-4, 5:34 UTC
AMSR-E, 40->0 rpm
2011-10-4, 7:27 UTC
-20
0
20
40
60
80
4019 4384 4750 5115 5480 5845 6211
Cor
rect
ion
(m)
Day (c. 2000)
c6 actual pitch (track)
c6 actual roll (scan)
c6 actual yaw
LUT 6.0.25
2016-02-02 00z
Future work
Lin et al., 6 June 2016 13
1) Routine monitor and LUTs update as needed
2)Update to correct AMSR-E effects in Aqua geolocation from late 2011
3)C7 update?
4)Update of Landsat ground control point chip library
Concluding Remarks
Residuals Terra C5 Aqua C5 Terra C6 Aqua C6 VIIRS C1.1
Track mean -1 m 7 m 2 m 2 m 8 m
Scan mean 5 m -3 m -1 m 0 m 5 m
Track RMSE 44 m 47 m 43 m 46 m 72 m
Scan RMSE 44 m 53 m 44 m 53 m 61 m
Data-days 5777 (15.8 yrs) 4951 (13.6 yrs) 5849 (16.0 yrs) 5040 (13.8 yrs) 1566 (4.3 yrs)
Missing days 134 99 62 10 1
Daily matched
GCPs w/ B1/I1
260 224 218 189 130
• Nadir equivalent accuracy (RMSE – Root Mean Square Error)
– Mostly within 20% band B1 HSI (250 m) = 50 m @ nadir ;
– Within 10 % for HKM bands and 5% for KM bands
• Band-to-band mis-registration adds bias to RMSE to other bands:
• Other features for MODIS geolocation
– Aqua uses definitive ephemeris data 27 hour latency (Terra uses TDRSS-based on-board ephemeris)
– C6 uses 0.5 km resolution DEM for terrain corrections (C5 uses older 1 km DEM, same as the VIIRS), plus other improvements over C5
Lin et al., 6 June 2016
22 RMSE
14
Backup slides: sun angle dependent Residuals
Lin et al., 6 June 2016 15
y = 0.0025x2 - 0.5159x + 20.521R² = 0.6431
y = -0.0074x2 + 1.2704x - 44.415R² = 0.8514, C4 used for C5
-100
-80
-60
-40
-20
0
20
40
60
80
100
-20 0 20 40 60 80 100 120 140 160 180
Sh
ift (
me
ters
)
Sun angle (degrees)
Track
y = 0.0017x2 - 0.3666x + 21.435R² = 0.5203
y = -0.0068x2 + 1.3044x - 49.118R² = 0.8886, C4 used for C5
-100
-80
-60
-40
-20
0
20
40
60
80
100
-20 0 20 40 60 80 100 120 140 160 180
Sh
ift (
me
ters
)
Sun angle (degrees)
Scan
y = -9E-05x2 - 0.0406x + 5.9642R² = 0.27
y = -0.0025x2 + 0.5511x - 22.682R² = 0.7863, C5 used for C6
-100
-80
-60
-40
-20
0
20
40
60
80
100
-20 0 20 40 60 80 100 120 140 160 180
Sh
ift (
me
ters
)
Sun angle (degrees)
Track
Lin et al., 6 June 2016 16
Terra Sun angle Correction C6 results C5 results
C6
y = 0.0002x2 - 0.0579x + 3.1975R² = 0.0876
y = -0.0031x2 + 0.7185x - 34.362R² = 0.8647, C5 used for C6
-100
-80
-60
-40
-20
0
20
40
60
80
100
-20 0 20 40 60 80 100 120 140 160 180
Sh
ift (
me
ters
)
Sun angle (degrees)
Scan
Small overcorrection in C5 – was corrected in C6
y = 0.0011x2 - 0.1535x + 4.7188R² = 0.4607
y = 0.0026x2 - 0.3546x + 4.3033R² = 0.7477, C5 used for C6
-100
-80
-60
-40
-20
0
20
40
60
80
100
0 20 40 60 80 100 120 140 160 180
Sh
ift (m
ete
rs)
Sun angle (degrees)
Track
y = -0.0022x2 + 0.3632x - 11.523R² = 0.5996
y = -0.0043x2 + 0.6999x - 19.695R² = 0.7678, C5 used for C6
-100
-80
-60
-40
-20
0
20
40
60
80
100
0 20 40 60 80 100 120 140 160 180
Sh
ift (m
ete
rs)
Sun angle (degrees)
Scan
Lin et al., 6 June 2016 17
Aqua Sun angle Correction C5 results C6 results
y = 0.0025x2 - 0.3173x + 9.4398R² = 0.7822
-100
-80
-60
-40
-20
0
20
40
60
80
100
0 20 40 60 80 100 120 140 160 180
Sh
ift (
me
ters
)
Sun angle (degrees)
Track
y = -0.0045x2 + 0.757x - 26.315R² = 0.7628
-100
-80
-60
-40
-20
0
20
40
60
80
100
0 20 40 60 80 100 120 140 160 180
Sh
ift (m
ete
rs)
Sun angle (degrees)
Scan
no correction in C5
(SNPP+J1) VIIRS Geolocation Status
NASA VIIRS Characterization Support Team (VCST)
Geometric Calibration Group
Guoqing (Gary) Lin, SSAI/GSFC code 619
Robert E. Wolfe, NASA/GSFC Code 619
Zhangshi Yin, Ping Zhang, Bin Tan, SSAI/GSFC Code 619
John Dellomo, GST/GSFC Code 619
James C. Tilton, NASA/GSFC Code 606
NASA MODIS/VIIRS STM Calibration Workshop
Silver Spring, Maryland
18 May 2015 6 June 2016
Acknowledgements
• Thanks Carol Davidson & her Land SIPS (formerly Land
PEATE) Team for processing control point residuals and
possible two line element (TLE) use from both IDPS and
LSIPS forward-&re-processed VIIRS geolocation
products, and testing Geo LUTs updates
• Thanks Geo JAMs -- Robert Williamson (now moved on)
& Rosalie Marley -- for helping us resolving DRs in the
AMP (formally DPA) at the GRAVITE
• Thanks Fred Patt of NASA Ocean Group for helping us
resolving issues related to ephemeris and attitude data
Lin et al., 6 June 2016 VCST/GEO 2
Outline
• Geolocation Performance and Trends
• I- & M-bands geolocation
• DNB geolocation, up to ~ 9/1/2013
• Remaining issues, concerns and challenges
• Some expectations of J1 VIIRS
• Conclusions
Lin et al., 6 June 2016 VCST/GEO 3
Overall Geolocation Performance Residuals IDPS Land SIPS
C1.1
Aqua
MODIS C6
Terra
MODIS C6
Track mean -1 m 8 m 2 m 2 m
Scan mean -1 m 5 m 0 m -1 m
Track RMSE 75 m 72 m 46 m 43 m
Scan RMSE 61 m 61 m 53 m 44 m
Data-days 1127 (3.1 yrs) 1566 (4.3 yrs) 5040 (13.8 yrs) 5849 (16.0 yrs)
Missing days 21 1 10 62
Daily matched
GCPs w/ I1/B1
130 130 189 218
• Nadir equivalent accuracy (RMSE – Root Mean Square Error) . (MODIS for reference)
– Meet Spec: 133 m (1s); within 20% I1 HSI (375 m) = 75 m @ nadir for VIIRS
– Band-to-band mis-registration adds bias to RMSE to other bands:
– Land PEATE/SIPS time period: 19 Jan 2012 to 3 May 2016
• MODIS – VIIRS differences
Aqua use definitive ephemeris data 27 hour latency
SNPP attitude data is not as good, see Slides10 & 11
DEM resolutions: older 1 km for VIIRS vs newer 0.5 km for MODIS C6
22 s RMSE
Lin et al., 6 June 2016 VCST/GEO 4
-100-75-50-25
0255075
100
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
Tra
ck (a
dj.)
res.
(m)
.
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
-100
-75
-50
-25
0
25
50
75
100
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
Sca
n (
ad
j.)
res.
(m)
.
Years since Jan. 1, 2000
Lin et al., 6 June 2016
NASA LSIPS C1.1 geolocation errors
Started 1/19/2012
Cryo door open
VCST/GEO 5
C1.1 RMSE Track: 72 m Scan: 61 m, nadir equivalent
y = 4.5999x - 57.255R² = 0.3709
-40-30-20-10
010203040
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
Tra
ck (a
dj.)
res.
(m)
.
Years since Jan. 1, 2000
16-day Global16-day Global Linear (16-day Global)
y = 1.8391x - 21.165R² = 0.1346
-40
-30
-20
-10
0
10
20
30
40
12 12.5 13 13.5 14 14.5 15 15.5 16 16.5
Sca
n (
ad
j.)
res.
(m)
.
Years since Jan. 1, 2000
Possible Long-Term Trending and Correction
Lin et al., 6 June 2016 Land SIPS Re-processed, will be corrected
Linear trend and annual cycle
Linear trend and annual cycle
VCST/GEO 6
y = 0.0018x2 - 0.4397x + 31.08
R² = 0.3885
-100
-75
-50
-25
0
25
50
75
100
0 20 40 60 80 100 120 140 160 180
Tra
ck r
esid
uals
(m
)
Sun angle (deg) y = 0.001x2 - 0.1782x + 13.271
R² = 0.1088
-100
-75
-50
-25
0
25
50
75
100
0 20 40 60 80 100 120 140 160 180
Scan r
esid
uals
(m
)
Sun angle (deg)
VIIRS Sun Angle Residuals
Lin et al., 6 June 2016
Track
Scan
Land SIPS Re-processed, will be corrected VCST/GEO 7
DNB geolocation error trending
based on coastal area GCP matching
As of Nov 4, 2013, the DNB geolocation accuracy is
Scan: 8 ± 33 μrad Track: -35 ± 68 μrad Scan: 7 ± 28 m Track: -29 ± 57 m over coastal areas
(nadir equivalent with mean altitude of 838.8 km)
Courtesy
of NGAS
2nd update
2/15/2013 1st update
3/30/2012
4/25/2013 fine tuned
4/18/2013 3rd & last update
8/22/2013
Lin et al., 6 June 2016
89
89
Star tracker
induced shift
Star tracker
shift corrected
VCST/GEO 8
DNB TC
geolocation (appending
fields (lat, lon, height,
QF)_TC to the
ellipsoid DNB
geolocation product)
was TTOed on
5/22/2014, 14:30 GMT
(data observation time)
in IDPS.
DNB errors track
with I1 band errors
no more
monitoring
~ 9/1/2013
Lin et al., 6 June 2016
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
17
:07
:00
17
:09
:00
17
:11
:00
17
:13
:00
17
:15
:00
17
:17
:00
17
:19
:00
17
:21
:00
17
:23
:00
17
:25
:00
17
:27
:00
17
:29
:00
17
:31
:00
17
:33
:00
17
:35
:00
Att
itud
e, L
atit
ude
(Deg
)
Time (UTC, 2015-09-23)
SNPP Inclination Adjust Maneuver
Roll
Pitch
Yaw
-18-16-14-12-10
-8-6-4-202468
1012141618
23:36:00 23:42:00 23:48:00 23:54:00 0:00:00
Att
itu
de
(deg
)
Time (UTC, 2015-11-05)
SNPP Collision Avoidance Maneuver
Roll
Pitch
Yaw
“False” attitude
reported during delta-V
maneuvers
A
B C E
D .
.
. . .
A
B C E
D .
.
. . .
• On-orbit control
reference frame is
changed ~ 17 deg to
orient the SC for
thruster burns between
points A & C.
delta-V
delta-V
VCST/GEO 9
• Geolocation error ~ 250
km between A & C
• We plan to mask the
VIIRS geolocation with
FILLs
0
17
34
51
68
85
102
119
136
153
170
187
204
0
108
216
324
365 730 1095 1460
99
.73
% o
uta
ge
(ti
me
s/o
rbit
)
Att
itu
de
err
or
(3*
Std
ev,
arc
sec)
Time (days since launch)
SNPP attitude (Spec 3s<108") non-compliance
Max 3*Stdev in RPY
Max 3*Stdev outage in RPY
Max 99.73% outage in RPY
SNPP attitude issues
Lin et al., 6 June 2016
60
80
100
120
140
160
180
200
-600
-400
-200
0
200
400
600
800
16:54:30 16:55:00 16:55:30 16:56:00 16:56:30 16:57:00
Lat/
Lon
(deg
)
Att
itu
de
(arc
sec)
Time (UTC, 2013-03-25)
RollPitchYawLonLat
-150
-120
-90
-60
-30
0
30
60
90
120
150
180
210
-1500
-1200
-900
-600
-300
0
300
600
900
1200
1500
1800
2100
9:13:00 9:13:30 9:14:00 9:14:30 9:15:00 9:15:30 9:16:00
Lat/
Lon
(deg
)
Att
itu
de
(arc
sec)
Time (UTC, 2015-03-04)
Roll Pitck
Yaw Lon
Lat
• SNPP attitude under-performed, in
99.73% interpretation of 3s, as early as
2013-03-25
• The MOT is working with the Flight
Project and the SC vendor to tune the star
trackers
• We plan to flag/mask geolocation for 100
< |R, P, Y| < 1800 arcsec (> 1800 arcsec
already has geolocation masked)
Thanks Fred Patt and NASA FDF for support VCST/GEO 10
~7 km error
60 scans
Attitude error 2016-05-02 06:48:50 – 06:50:40z
Lin et al., 6 June 2016
Difference in image masking
W. Australian coastline, 2016-04-16 False coastline, 2016-05-02 VCST/GEO 11
RGB (I4, I3, I2), 2016-05-02
-800
-600
-400
-200
0
200
400
600
800
128 130 132 134 136 138 140 142 144 146 148 150 152 154 156 158
attit
ude
(arc
sec)
Scan#
roll pitch yaw
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
M1 M2 M4 M3 I1 I2 M7 M5 M6 M13 M12 I4 I3 M10 M11 M8 M9 I5 M16 M15 M14
Ban
d O
ffse
t (M
-HS
I)
Band Name
Scan Agg1x1
Scan Agg2x1
Scan Agg3x1
Track
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
M1 M2 M4 M3 I1 I2 M7 M5 M6 M13 M12 I4 I3 M10 M11 M8 M9 I5 M16 M15 M14
Ban
d O
ffse
t (M
-HS
I)
Band Name
Scan Agg1x1Scan Agg2x1Scan Agg1x1Track
J1 VIIRS M/I band offsets wrt I1
• J1 scan rate is nominal @3.517 rad/sec (SNPP 3.531)
• Overall BBR band pair performances are compatible between J1 and SNPP
• J1 Bands on SWMWIR and LWIR FPAs shifted from bands on VisNIR FPA, ~ 7% for
M-bands and ~ 14% for I-bands. Mapping uncertainties are affected
Lin et al., 6 June 2016
S/MWIR FPA LWIR FPA VisNIR FPA
22 s RMSE
VCST/GEO 12
J1
SNPP
Hot
Cold
0
0.5
1
1.5
2
2.5
3
3.5
4
1000 1200 1400 1600 1800 2000 2200
Sca
n H
SI (
km)
DNB distance from nadir (km)
Beginning of scan
End of scan
0
0.5
1
1.5
2
2.5
3
3.5
4
0 200 400 600 800 1000 1200 1400 1600
Hor
izon
tal S
ampl
ing
Inte
rval
(km
)
Distance from Nadir (km)
Along-Scan Direction HSIM-band - No Agg
M-band
I-band
DNB
J1 DNB Geometric Performance DNB LSFs are mostly square
Baselined pixel size is ~ 750 m
“Option21” has pixel size up to 1.6 km
within 56.5o
Geolocation is extrapolated post-nadir for
scan angle > ~56.5o (pixel size up to 3.9
km @60.5o)
Lin et al., 6 June 2016
“Option21” – default, in km
extrapolated
Geo
Baseline -- SNPP
J1 DNB
default
below
scan angle > ~56.5o
Nominal maximum
scan angle (~56.5o)
4546474849505152535455565758596061
0
0.5
1
1.5
2
2.5
3
3.5
4
1400 1500 1600 1700 1800 1900 2000 2100 2200
Scan
An
gle
(deg
)
Scan
Grd
Sam
p D
ist (
km)
DNB Sample#, "Option21"
Pre_nadir
Post-nadir
Scan_Angle
“Option21” – default, in Samp#
scan angle
> ~56.5
BL:2032
extrapolated
Geo
VCST/GEO 13
BL:1480
Conclusions
• SNPP VIIRS geolocation performance is good
– The attitude system underperforms, geolocation error of 7 km occurred
occasionally. Attitude/star trackers need maintenance.
• Geolocation mean errors for I- & M-bands are near 0 and
uncertainties are ~ 70 m at nadir, statistically
– Caveat: DNB geolocation errors are not monitored after ~9/1/2013
• J1 geolocation expectations – Bands on VisNIR FPA should be good; Bands on cold FPAs will be ~ 50 m off in the
track direction
– DNB geolocation pixels will be larger beyond S#1480, 1100 km off nadir
– DNB geolocation accuracy will be TBD
Lin et al., 6 June 2016 VCST/GEO 14
Questions?
؟ Do random variables exist?
Lin et al., 6 June 2016 VCST/GEO 15
؟ Where can we find biases & remove them?
؟ Is ergodicity a fair assumption?
؟ What do we mean by “mean values?”
؟ in what time/spatial scales?
؟ in any functional forms?
Backup Slides
Lin et al., 6 June 2016 VCST/GEO 16
On-orbit Geolocation LUT Updates
Lin et al., 6 June 2016
Update Date Description Comments
a 1/19/2012 Cryo-radiator door open All VIIRS band available,
LPEATE re-process start date
1 2/23/2012 Initial mounting coef. update Removed bias ~ 1.3 km
2 3/30/2012 Initial DNB FPA center update Removed bias ~ 1 km
b 11/22/2012 Scan control electronics (SCE) was switched
from B-side to A-Side Caused bias ~ 300 m
3 12/11/2012 Correction after SCE was switched from B-
Side to A-side Removed bias ~ 300 m
4 2/15/2013 Second, fine DNB FPA center update Removed DNB bias ~ 300 m
5 4/18/2013 Second, scan angle dependent, fine Geo LUT
update
Fine tuned and removed scan
dependent biases
c 4/25/2013 Star tracker maintenance/re-alignment Caused bias ~ 25 m
6 8/22/2013 Correction to the star tracker re-alignment Removed bias ~ 25 m
DNB only All bands impacted Key: External event
VCST/GEO 17
-100-75-50-25
0255075
100
12 12.25 12.5 12.75 13 13.25 13.5 13.75 14 14.25 14.5 14.75 15 15.25
Tra
ck (a
dj.)
res.
(m)
.
Years since Jan. 1, 2000
Daily 16-day Global 16-day Southern Hemisphere 16-day Northern Hemisphere
-100
-75
-50
-25
0
25
50
75
100
12 12.25 12.5 12.75 13 13.25 13.5 13.75 14 14.25 14.5 14.75 15 15.25
Sca
n (
ad
j.)
res.
(m)
.
Years since Jan. 1, 2000
Lin et al., 6 June 2016
IDPS (NRT) geolocation errors
Started 2/23/2012
1.3 km error before
300 m error in scan direction
2012 11/22 - 12/11
25 m error Star Tracker adjustment
2013 4/25 - 8/22
VCST/GEO 18
R² = 0.3933
-100
-80
-60
-40
-20
0
20
40
60
80
100
-60 -40 -20 0 20 40 60
Tra
ck r
esid
uals
(m
)
Scan angle (deg)
R² = 0.0007
-100
-80
-60
-40
-20
0
20
40
60
80
100
-60 -40 -20 0 20 40 60
Scan resid
uals
(m
)
Scan angle (deg)
VIIRS Scan Angle Residuals
Lin et al., 6 June 2016
Track
Scan
Land SIPS Re-processed VCST/GEO 19
-200
-150
-100
-50
0
50
100
150
200
50 60 70 80 90
Tra
ck r
esid
ua
ls (
m)
Day of Year (2012)
-200
-150
-100
-50
0
50
100
150
200
50 60 70 80 90
Sca
n r
esid
ua
ls (
m)
Day of Year (2012)
-2000
-1500
-1000
-500
0
500
30 50 70 90
Tra
ck r
esid
ua
ls (
m)
Day of Year (2012)
-500
0
500
1000
1500
2000
30 50 70 90
Sca
n r
esid
ua
ls (
m)
Day of Year (2012)
Lin et al., 6 June 2016
Initial on-orbit geolocation LUTs Update
375 m
Nadir equivalent units;
Bias
(m)
RMSE
(m)
Track -21 80
Scan -8 64
Error after
LUT update
(2/23/2012, doy 54)
27 days with average of
142 matchups/day
(minus 12 outliers/day)
Feb
. 2
3,
2012
Biases removed: Track -755 m, Scan 1118 m
Track
Scan
VCST/GEO 20
Scan Control Electronics (SCE) Side Switch,
Geolocation Error and Correction
Lin et al., 6 June 2016 (Nadir equivalent units)
-600
-500
-400
-300
-200
-100
0
100
200
12.86 12.87 12.88 12.89 12.90 12.91 12.92 12.93 12.94 12.95 12.96 12.97
Tra
ck r
esid
ua
ls (
m)
Years since Jan. 1, 2000
-600
-500
-400
-300
-200
-100
0
100
200
12.86 12.87 12.88 12.89 12.90 12.91 12.92 12.93 12.94 12.95 12.96 12.97
Sca
n r
esid
ua
ls (
m)
Years since Jan. 1, 2000
After A-side SCE
coefficient update
in IDPS (Dec. 11,
2012)
After switch to A-side SCE (Nov. 22,
2012)
Scan direction geolocation
shifted ~325 m for 19 data-days
Operated on B-side
SCE since launch
Track
Scan
VCST/GEO 21
Star Tracker Re-alignment and Correction
Lin et al., 6 June 2016
-75
-50
-25
0
25
50
75
13.2 13.3 13.4 13.5 13.6 13.7 13.8
Tra
ck (
ad
j.)
res.
(m)
.
Years since Jan. 1, 2000
Daily 16-day Global
-75
-50
-25
0
25
50
75
13.2 13.3 13.4 13.5 13.6 13.7 13.8
Scan (
adj.)
res. (m
) .
Years since Jan. 1, 2000
6
8/22/2013
5
4/18/2013 4/25/2013
4/18/2013:
Geo LUTs fine tuned
4/25/2013:
Star tracker re-
alignment
8/22/2013
Error ~ 25 m found
and corrected
Track
Scan
VCST/GEO 22
RTA/HAM sync losses
• Sync losses happen more frequently lately
• Sync losses cause data loss (events detected and flagged in SDR/Geo
SW)
• 63 times so far, ~ 100 sec each (60 scans), equivalent to 1.0 orbits total Lin et al., 6 June 2016
0
10
20
30
40
50
60
70
0 365 731 1,096 1,461
Occ
urr
ence
Time (Days since Launch 2011-10-28, A-Side started 2012-11-22)
SNPP VIIRS Sync Loss History (up to the one on 16 May 2016)
A-Side
B-Side
VCST/GEO 23
An outlier period
Lin et al., 6 June 2016
-1000
-500
0
500
1000
1500
0 3 6 9 12 15 18 21 24
VIIR
S G
eo
loca
tio
n E
rro
r (m
)
Time (UTC Hours, 19 Aug 2015)
Track_shiftScan_shift
1 second time error
onboard of SNPP
14:23Z 21:12Z
“SCC Clock slope was performed on August 19, 2015 at 14:23Z”
“SCC Clock was demonstrating a +999.989 ms drift relative to the
GPS time, “
LPEATE/LSIPS archives exclude the datasets now. Could
“manually” correct time error later.
Error in time code
VCST/GEO 24
SNPP, & J1+, “824 km orbit”
Lin et al., 6 June 2016 VCST/GEO 25
SNPP Altitude: =838.8 km, Max = 856 km, Min = 828 km
Wolfe et al., May 13, 2014
838.4
838.6
838.8
839
839.2
0 365 730 1095 1460
Orb
ital
Mea
n A
ltit
ud
e (k
m)
Time (Days since Launch 28 October 2011)
827.5
828
828.5
829
829.5
855
855.5
856
856.5
857
0 365 730 1095 1460
Time (Days since Launch 28 October 2011)
Alt_max (km) Alt_min (km)
VCST/GEO 25
13:23
13:24
13:25
13:26
13:27
13:28
13:29
13:30
13:31
13:32
13:33
13:34
0 365 730 1,095 1,460
LT
AN
Time (Days since Launch on Oct 28, 2011)
SNPP Equatorial Crossing Time
-20
-16
-12
-8
-4
0
4
8
12
16
20
101.49
101.5
101.51
0 365 730 1,095 1,460
BW
D D
Lo
n (
km
)
Orb
ital
P
eri
od
(m
in)
Time (Days since Launch on Oct 28, 2011)
Orbital Period 16-day Repeat Deviation
23 DMUs (DAMs, CAMs, RMMs)
5 IAMs
5 IAMs