satellite data assimilation in regional models: promises
TRANSCRIPT
![Page 1: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/1.jpg)
1
Satellite Data Assimilation in Regional Models: Promises and Challenges
Department of Earth, Ocean and Atmospheric Sciences
Florida State University, USA
Xiaolei Zou
September 4, 2013
A Presentation to the HFIP Telecon
![Page 2: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/2.jpg)
• Important Building Blocks for Satellite DA
Outline
• GOES Imager Radiance Assimilation in GSI/ARW
• POSE MHS Radiance Assimilation in GSI/ARW
• SNPP ATMS Radiance Assimilation in GSI/HWRF
• Summary & Conclusions
![Page 3: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/3.jpg)
J(x0 ) =12(x0 − x
b )T B−1(x0 − xb ) + 1
2(H (x0 ) − y
obs )T (O + F)−1(H (x0 ) − yobs )
F
xb yobsx0 H
B O
Choice of analysis variable
What data to assimilation?
Which forward model to use?
What background to start with?
How to estimate elements in B?
Where to find their values?
How to quantify it?
⇔
Model State Observed Space
J(xa ) = minx0J(x0 )
Important Building Blocks for Satellite DA
![Page 4: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/4.jpg)
4
An Evaluation of Added Benefits of GOES Imager Radiances to Other Satellite Data Assimilation
Part I
![Page 5: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/5.jpg)
5
GOES Imager Radiance Assimilation in GSI/ARW
1) Comparison of Single Type Satellite Data Assimilation
AMSU-A
AIRS
2) GOES Imager Added to Different Types of Satellite Data
HIRS/3
MHS
HIRS/4
AIRS
GSN
GOES Imager
3) GOES Imager Added to All Satellite Data Assimilation
4) Impact of Quality Control on MHS Data Assimilation
![Page 6: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/6.jpg)
6
GOES West (11) and GOES East (12)
![Page 7: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/7.jpg)
Weighting Function
Ch3 (6.8µm) Ch2 (3.9µm)
Ch5 (12µm)
Ch4 (10.7µm)
Weighting Function
Pres
sure
(hPa
) GOES WEST
GOES EAST
Pres
sure
(hPa
) Ch3 (6.48µm)
Ch2 (3.9µm) Ch6 (13.3µm)
Ch4 (10.7µm)
Ch2: near-infrared (low cloud, fog, fire) Ch3: infrared (upper-level water vapor) Ch4: infrared (surface and cloud-top T) Ch5: infrared (low-level water vapor) Ch6: infrared (cloud detection)
![Page 8: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/8.jpg)
Mean Jacobian of Brightness Temperature
dTb/dT dTb/dq
Pres
sure
(hPa
)
Pres
sure
(hPa
)
Sensitivity to T Sensitivity to q
-Ch3 -Ch4 -Ch5 -Ch6
![Page 9: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/9.jpg)
Resolution: 10 km, 27 layers Domain size: 250x200x27 Microphysics: WRF single-moment 3-class scheme PBL: Yonsei University scheme Cumulus: Kain-Fritsch scheme Radiation: Dudhia scheme
Advanced Research WRF (ARW) Model Domain
![Page 10: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/10.jpg)
GSI DA Cycling (1200 UTC May 22 to 0000 UTC May 23, 2008)
36-h ARW forecasts
6-h DA window 6-h DA window 6-h DA window
6-h ARW forecasts
AMSU-A, AIRS, HIRS/3, HIRS/4, MHS, GSN,
Assimilation of Different Combinations of Observations
NCEP GSI 3D-Var Data Assimilation System
GOES imager, Conventional data
![Page 11: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/11.jpg)
Satellite Instruments Satellite Instruments Satellite Instruments
NOAA-14 [HIRS/2](1)
MetOp-A
HIRS/4
GOES-11
(SNDR)
[MSU] AMSU-A Imager
NOAA-15 AMSU-A MHS SNDRD1
AMSU-B [IASI] SNDRD2
NOAA-16
(HIRS/3) (2)
Aqua
AIRS SNDRD3
(AMSU-A) (AMSU-A) SNDRD4
AMSU-B (AMSRE)
GOES-12
(SNDR)
(AVHRR3) F13 (SSMI) Imager
NOAA-17
HIRS/3 F14 (SSMI) SNDRD1
(AMSU-A) F15 (SSMI) SNDRD2
AMSU-B F16 (SSMIS) SNDRD3
(AVHRR3) SNDRD4
NOAA-18
(HIRS/4)
GOES-13
(SNDR)
AMSU-A (Imager)
MHS (SNDRD1)
(AVHRR3) (SNDRD2) (1)Data not available for this case. (2)Instruments removed from operational data assimilation.
(SNDRD3)
(SNDRD4)
![Page 12: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/12.jpg)
Total Data Count
2008
1200 UTC May 22
1800 UTC May 22
0000 UTC May 23
0600 UTC May 23
1200 UTC May 23
1800 UTC May 23
0000 UTC May 24
![Page 13: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/13.jpg)
HIRS/3
MHS 1200 UTC 1800 UTC 0000 UTC
1200 UTC 1800 UTC 0000 UTC
UTC Dependence of POES Data Count
HIRS/4
AIRS
1200 UTC 1800 UTC 0000 UTC
MetOp-A (09:30am)
1200 UTC 1800 UTC 0000 UTC
NOAA-17 (10:00am)
Aqua (01:36pm)
NOAA-18 (01:45pm)
MetOp-A (09:30am)
![Page 14: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/14.jpg)
1200 UTC 1800 UTC 0000 UTC
G11-Imager G12-Imager
GOES Imager Data Count
![Page 15: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/15.jpg)
O-B for AMSU-A (NOAA-18) Ch6
Ch7
Ch8
O-B (K)
1800 UTC May 22, 2008
Only three AMSU-A channels are assimilated in GSI/ARW due to a too low model top (~50 hPa)
for satellite data assimilation!
![Page 16: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/16.jpg)
Observed BT of GOES-11 Ch5 on May 23, 2008
0300-0306 UTC 0600-0606 UTC 0900-0906 UTC
1200-1206 UTC 1500-1506 UTC 1800-1806 UTC
![Page 17: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/17.jpg)
12-h Forecast Differences at 850 hPa
Mixing Ratio GOES_Img - CONV
1200 UTC May 23 2008
Temperature GOES_Img - CONV
![Page 18: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/18.jpg)
Threat Scores of 3-h Accumulative Rainfall at 1mm thresholds Averaged over 24 Hours
![Page 19: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/19.jpg)
Threat Scores of 3-h Accumulative Rainfall Averaged
over 24 Hours
Assimilation of a Single Type of Satellite Observations
1 mm
5 mm
10 mm
15 mm
![Page 20: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/20.jpg)
Threat Scores of 3-h Accumulative Rainfall at 5mm thresholds Averaged over 24 Hours
5 mm
CONV GSN HIRS/3 HIRS/4 AIRS MHS AMSU-A ALL
Left bar: without GOES Imager data
Right bar: with GOES Imager data
![Page 21: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/21.jpg)
Threat Scores of 3-h Accumulative
Rainfall Averaged over 24 Hours
1 mm
5 mm
10 mm
15 mm
Left bar: without GOES imager Right bar: with GOES Imager
CONV GSN
HIRS/3 HIRS/4 AIRS
MHS AMSU-A
ALL
GOES Imager improves the assimilation of a single
type of satellite data.
![Page 22: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/22.jpg)
Verification with GOES Sounder
SNDRD1 SNDRD2
SNDRD3 SNDRD4
Time (UTC) Time (UTC)
Δσ = σCONV + AMSU - A+GOES_Img − σCONV +AMSU −A
Δσ < 0 ↔ Improvement
![Page 23: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/23.jpg)
Verification with Independent GOES Sounder Data
Ch12
Ch11
Ch10
Ch8
Pres
sure
(hPa
)
Channel Wavelength,um (wavenumber, cm-1)
Purpose
10 7.43(1345) Low-level moisture
11 7.02(1425) Midlevel moisture
12 6.51(1535) Upper-level moisture
Δσ = σCONV + AMSU - A+GOES_Img − σCONV +AMSU −A
DA cycle Forecast period
![Page 24: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/24.jpg)
12-h Forecast Verification with AIRS
NWP Channel Number Data Count
Stan
dard
Dev
iatio
n (K
) D
ata count
-
-
σCONV +AMSU −A
σCONV + AMSU - A+GOES_Img
![Page 25: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/25.jpg)
Threat Scores of 3-h Accumulative Rainfall
10 mm
MHS HIRS/3/4
AIRS AMSU-A
AMSU-A
Without GOES Imager data
With GOES Imager data
GSN MHS
HIRS/3/4 AIRS
AMSU-A
HIRS/3 HIRS/4 AIRS
AMSU-A
HIRS/4 AIRS
AMSU-A
AIRS AMSU-A
CONV
![Page 26: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/26.jpg)
Threat Scores of 3-h Accumulative Rainfall
10 mm
CONV GOES_Img MHS
HIRS/3 HIRS/4 AIRS
AMSU-A CONV
HIRS/3 HIRS/4 AIRS
AMSU-A CONV
HIRS/4 AIRS
AMSU-A CONV
AIRS AMSU-A
CONV
AMSU-A CONV
More data, better forecasts.
![Page 27: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/27.jpg)
Threat Scores of 3-h Accumulative Rainfall
10 mm
CONV AMSU-A CONV
AIRS AMSU-A
CONV
HIRS/4 AIRS
AMSU-A CONV
HIRS/3 HIRS/4 AIRS
AMSU-A CONV
MHS HIRS/3 HIRS/4 AIRS
AMSU-A CONV
GSN MHS
HIRS/3 HIRS/4 AIRS
AMSU-A CONV
GOES_Img GSN MHS
HIRS/3 HIRS/4 AIRS
AMSU-A CONV
10 mm
More data, not necessarily better forecasts.
![Page 28: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/28.jpg)
O-B (MHS Channel 3 at 1800 UTC 05/22/08) MHS MHS after thinning
GOES-11 GOES-12
![Page 29: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/29.jpg)
Threat Scores (May 23, 2008)
Threshold (mm) Threshold (mm)
0600 UTC 0900 UTC
1200 UTC 1500 UTC
MHS MHS collocated with GOES
![Page 30: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/30.jpg)
30
Improved QPFs by MHS Radiance Data Assimilation with a Newly Added Cloud Detection Algorithm
Part II
![Page 31: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/31.jpg)
31
MHS Data Quality Control (QC)
A new MHS QC for cloud detection
MHS QC in GSI and GSI QC results
Impact of the modification of MHS QC to QPFs
Pres
sure
(hPa
)
Weighting Function
Ch3
Ch4
Ch5
![Page 32: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/32.jpg)
LWP Index Used for MHS QC in GSI
Over Ocean:
LWPindexocean =
0.13× Tb,1o −Tb,1
m( )− 33.58×Tb,2
o −Tb,2m( )
300−Tb,2o
⎧⎨⎪
⎩⎪
⎫⎬⎪
⎭⎪, if Tb,2
o ≤ 300
9, otherwise
⎧
⎨⎪⎪
⎩⎪⎪
LWPindex
land = 0.85× Tb,1o −Tb,1
m( )− Tb,2o −Tb,2
m( )Over Land:
Tb, 1o −Tb, 1
m
Tb, 2o −Tb, 2
m } O-B differences of MHS channels 1-2
![Page 33: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/33.jpg)
Three Steps for MHS Data Rejection in GSI
Step I:
TPWindex >1
Step II:
or:
O − B > 3 ei × 1−TPWindex
2( )× fH ×τ itop( )
O − B > 6K
Step III: All five channels if data of any other channel was removed by the first two QC steps
fH=2000/H, H is terrain height>2km is ransmittance at model top τ i
top
ei is accuracy of obs.
TPWindex ≡ Tb,1
o −Tb,1m( )− 7.5× LWPindex
⎡⎣
⎤⎦ 10.0{ }2
+ LWPindex2 >1
![Page 34: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/34.jpg)
(O-B
) mhs
,ch1
(K)
GSI LWPindex (mm)
Diagnosis of MHS QC Results
Open circle: data removed Dot: Data assimilated
GSI TPWindex
TPWindex = 1
1800 UTC May 22, 2008
![Page 35: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/35.jpg)
(O-B)Ch2 (K)
TPW
inde
x
(O-B)Ch1 (K) TP
Win
dex
Diagnosis of MHS QC Results (cont.)
1800 UTC May 22, 2008
Data eliminated by • Step I defined by TPWindex • Step II defined by O-B • Step III defined by inter-channels relationship
![Page 36: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/36.jpg)
Diagnosis of MHS GSI QC
Data that pass GSI QC
MHS O-B
GOES 10.7µm
RH at 300-hPa
Modeled RH at 300 hPa
![Page 37: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/37.jpg)
Infrared O-B More Sensitive to Cloud Than Microwave
(O-B
) MH
S C
h3 (K
)
(O-B)GOES ch4 (K)
![Page 38: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/38.jpg)
“(O-B)GOES” Regressed by MHS Channels 1, 2 and 5
Over Ocean:
Over Land:
Observations of MHS channels 1-2, 5 are used in the regression.
(O − B)GOES ,ch4
regression = −0.536×Tb,MHSch1
obs +1.132×Tb,MHSch2
obs + 0.537 ×Tb,MHSch5
obs − 321.318
(O − B)GOES ,ch4
regression = 0.009×Tb,MHSch1
obs + 0.085×Tb,MHSch2
obs + 0.877 ×Tb,MHSch5
obs − 274.255
• Channel 5 is most sensitive to scattering from thin clouds
• Channels 1-2 are affected by the radiation from both the Earth’s surface and emission and scattering from ice phase clouds
![Page 39: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/39.jpg)
Thresholds for Cloud Detection
(O-B
) GO
ES C
h4 (K
)
Land
Data Number
(O − B)GOES ,land
regression ≤ −4K
(O − B)GOES ,ocean
regression ≤ −2K
(O-B
) GO
ES C
h4 (K
)
Data Number
MHS data from NOAA-18 on May 17, 18, 19, 20, 21, 2008
![Page 40: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/40.jpg)
(O-B
) MH
S C
h2 (K
)
(O-B)GOES Ch4 (K)
(O-B
) MH
S C
h3 (K
)
Points that pass GSI QC but removed by the new scheme
Points that pass both GSI QC and the new QC QC
(O-B)GOES Ch4 (K) (O-B)GOES Ch4 (K)
O-B Scatter Plots for MHS Channels 1-5 versus GOES-12 Ch4 (O
-B) M
HS
Ch1
(K)
(O-B
) MH
S C
h4 (K
)
(O-B
) MH
S C
h5 (K
)
![Page 41: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/41.jpg)
(unit: mm)
00-03UTC
03-06UTC
06-09UTC
09-12UTC
24-h QPFs of 3-h Accumulative Rainfall EXP1: The original GSI QC for MHS data is used.
00-03UTC
03-06UTC
06-09UTC
09-12UTC
May 23, 2008
![Page 42: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/42.jpg)
24-h Forecasts of 3-h Accumulative Rainfall
00-03
03-06
06-09
09-12
12-15
15-18
18-21
21-24
Obs. Modified QC Obs. Modified QC GOES 10.7µm
(unit: mm) (unit: K)
GOES 10.7µm
![Page 43: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/43.jpg)
Threat scores (TS) of 3-hour Accumulative Rainfall
TS
Time (UTC)
TS
Time (UTC)
TS
Time (UTC)
TS
Time (UTC)
EXP1:CONV+AMSU-A+MHS EXP2:Same as EXP1 except for modified MHS QC
1 mm 5 mm
10 mm 15 mm
![Page 44: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/44.jpg)
44
Impacts of ATMS Data Assimilation on Hurricane Track and Intensity Forecasts
Part III
![Page 45: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/45.jpg)
45
Suomi NPP Satellite Instruments
VIIRS!
CrIS!
OMPS!
CERES!
ATMS!
VIIRS --- Visible/Infrared Imager/Radiometer Suite CrIS --- Cross-track Infrared Sounder ATMS --- Advanced Technology Microwave Sounder
OMPS --- Ozone Mapping and Profiler Suite CERES --- Cloud and Earth Radiant Energy System
![Page 46: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/46.jpg)
Channel Characteristics of ATMS and AMSU Channel Frequency (GHz) NEΔT (K) Beam width (o) Peak WF (hPa)
ATMS AMSU ATMS AMSU ATMS AMSU ATMS AMSU ATMS AMSU
1 23.8 0.50 0.30 5.2 3.3 Surface 2 31.4 31.399 0.60 0.30 5.2 3.3 Surface 3 50.3 50.299 0.70 0.40 2.2 3.3 Surface
4 51.76 0.50 2.2 Surface 5 4 52.8 0.50 0.25 2.2 3.3 1000 6 5 53.596±0.115 0.50 0.25 2.2 3.3 700 7 6 54.4 0.50 0.25 2.2 3.3 400 8 7 54.94 0.50 0.25 2.2 3.3 270 9 8 55.5 0.50 0.25 2.2 3.3 180 10 9 57.29 0.75 0.25 2.2 3.3 90 11 10 57.29± 0.217 1.00 0.40 2.2 3.3 50 12 11 57.29± 0.322± 0.048 1.00 0.40 2.2 3.3 25
13 12 57.29± 0.322 ± 0.022 1.25 0.60 2.2 3.3 12
14 13 57.29± 0.322 ± 0.010 2.20 0.80 2.2 3.3 5
15 14 57.29± 0.322± 0.0045 3.60 1.20 2.2 3.3 2
16 15 88.2 89.0 0.30 0.50 2.2 3.3 Surface 17 16 165.5 89.0 0.60 0.84 1.1 1.1 1000 Surface 18 17 183.31±7.0 157.0 0.80 0.84 1.1 1.1 800 Surface
19 18 183.31±4.5 183.31±1.0 0.80 0.60 1.1 1.1 700 400
20 19 183.31±3.0 0.80 0.70 1.1 1.1 600
21 20 183.31±1.8 183.31±7.0 0.80 1.06 1.1 1.1 500 800
22 183.31±1.0 0.90 1.1 400
![Page 47: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/47.jpg)
47
ATMS Weighting Functions and HWRF Model Levels Pr
essu
re (h
Pa)
Weighting Function
Pres
sure
(hPa
)
Weighting Function
Δp (hPa) Δp (hPa)
50 hPa
0.5 hPa
![Page 48: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/48.jpg)
48
ATMS FOV
ATMS FOV
A consistent FOV distribution between temperature and humidity channels on ATMS makes the cloud detection easy to implement.
Channels 1-2 (larger FOV)
Channels 17-22 (smaller FOV)
Channels 3-16 (larger FOV)
Channels 17-22 (smaller FOV)
Same FOVs for All ATMS Channels
![Page 49: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/49.jpg)
49
MHS FOV
Scan
line
AMSU-A FOV
An inconsistent FOV distribution between AMSU-A and MHS channels makes the cloud detection for MHS data difficult.
Inconsistent FOVs between AMSU-A and MHS
![Page 50: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/50.jpg)
50
The outer domain, ghost domain, middle nest and inner nest of HWRF
Surface pressure at 0000 UTC May 27, 2012 for Hurricane Beryl.
![Page 51: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/51.jpg)
51
Four 2012 Atlantic Hurricanes which Made Landfall
Debby Beryl
Sandy
Isaac
![Page 52: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/52.jpg)
52
O-B Values for Those Data Points that Pass QC
Tb,4AVHRR O-B
0600 UTC October 26, 2012
![Page 53: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/53.jpg)
53
Data Points Removed by QC
the 5th QC the 6th QC the 7th QC the 8th QC the 9th QC
criteria
0600 UTC October 26, 2012
![Page 54: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/54.jpg)
54
Convergence of ATMS Data Assimilation (Isaac)
(FOV)
(FOV)
O-B
(K)
O-A
(K)
Data count
Data count
ATMS channel 6
![Page 55: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/55.jpg)
55
Convergence of ATMS Data Assimilation (Isaac)
(FOV)
(FOV)
O-B
(K)
O-A
(K)
Data count
Data count
ATMS channel 9
![Page 56: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/56.jpg)
56
Standard Deviation before and after Data Assimilation For Hurricane Isaac
Stan
dard
Dev
iatio
n (K
)
ATMS Channel Number
![Page 57: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/57.jpg)
57
Daily Mean Forecast Errors
For Isaac
Forecast Time (h) Forecast Time (h) Forecast Time (h)
08/22
08/23
08/24
08/25
08/26
08/27
Track (nm) Vmax (kts) ps (hPa)
![Page 58: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/58.jpg)
58
Impacts of Satellite Data Assimilation on vmax Forecast Errors for Hurricane Isaac
v max
(kts
)
v max
(kts
)
v max
(kts
)
v max
(kts
)
Date (August) Date (August)
Date (August) Date (August)
22 23 24 25 26 27 28 29
![Page 59: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/59.jpg)
59
Impacts of ATMS Data Assimilation on the Track Forecast of Hurricane Sandy
23 24 25 26 27 28 29 (October)
without ATMS with ATMS
![Page 60: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/60.jpg)
60
72-h Forecasts of PV and Wind Vector at 200 hPa
Without ATMS
With ATMS
Sandy
![Page 61: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/61.jpg)
61
84-h Forecasts of PV and Wind Vector at 200 hPa
Without ATMS
With ATMS
Sandy
![Page 62: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/62.jpg)
62
84-h Forecasts of Cloud Liquid Water Valid at 0000 UTC 30 October 2012
Without ATMS With ATMS
kg m-2
GOES-13 Tb,4 for Verification ATMS Tb,18 at 1727 UTC 10/29/12
![Page 63: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/63.jpg)
63
Mean Forecast Errors for 2012 Atlantic Landfalling
Time (hour) Time (hour)
v max
err
or (k
ts)
p c (h
Pa)
Trac
k er
ror (
nm)
v max
err
or (k
ts)
p c (h
Pa)
Trac
k er
ror (
nm)
CONV CONV+ATMS SAT SAT+ATMS
![Page 64: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/64.jpg)
64
Summary (Part I)
• AMSU-A and GOES imager data contribute most significantly to improved QPFs near Gulf of Mexico
• Assimilation of GOES imager radiances contributes positively to any single type of satellite data
• Assimilation of all types of satellite data in the GSI system did not produce a better forecast than any experiment assimilated a single type of satellite data
• An improved cloud detection for MHS observations results in a significant positive impact to coastal QPFs
![Page 65: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/65.jpg)
65
• Some cloudy radiances remain near cloud edges
after the MHS QC in GSI
Summary (Part II)
• The cloud detection algorithm effectively removes
those cloudy radiances remaining near cloud edges
after the MHS QC in GSI
• The MHS data assimilation with the revised QC is
shown to significantly improve coastal QPFs
![Page 66: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/66.jpg)
66
Summary (Part III)
• ATMS data assimilation in GSI/HWRF results in a consistent positive impact on the track and intensity forecasts of the four landfall hurricanes in 2012
• A consistent FOV distribution between temperature and humidity channels on ATMS makes the cloud detection easy to implement
• Hurricane Sandy’s forecasts are significantly improved after ATMS data assimilation when verified with independent GOES and POES observations
![Page 67: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/67.jpg)
67
Zou, X., F. Weng, B. Zhang, L. Lin, Z. Qin and V. Tallapragada, 2013: Impact of ATMS radiance data assimilation on hurricane track and intensity forecasts using HWRF. J. Geophy. Res. JPSS Special Issue, (revised)
Weng, F., X. Zou, X. Wang, S. Yang, and M. D. Goldberg, 2012: Introduction to Suomi NPP ATMS for NWP and tropical cyclone applications. J. Geophy. Res., 117, D19112, 14pp, doi:10.1029/2012JD018144.
Zou, X., Z. Qin, and F. Weng, 2012: Improved coastal precipitation forecasts with direct assimilation of GOES 11/12 imager radiances, Mon. Wea. Rev., 139, 3711-3729.
Qin, Z., X., Zou, and F. Weng, 2013: Evaluating added benefits of assimilating GOES imager radiance data in GSI for coastal QPFs. Mon. Wea. Rev., 141(1), 75-92.
Zou, X., Z. Qin, and F. Weng, 2013: Improved quantitative precipitation forecasts by MHS radiance data assimilation with a newly added cloud detection algorithm, Mon. Wea. Rev., 141, 3203-3221.
More details can be found in
![Page 68: Satellite Data Assimilation in Regional Models: Promises](https://reader035.vdocuments.mx/reader035/viewer/2022071904/62d4f5659691cf3c5a64ce39/html5/thumbnails/68.jpg)
Acknowledgement
This work was supported by NOAA GOES-R risk reduction program and JPSS Proving Ground Program.