uw cris sdr review december2013 20131213 · 200 220 240 260 280 300 bt (k) 700 800 900 1000 −0.2...

30
CrIS Radiometric Calibra1on: Uncertainty Es1mates and Evalua1ons On behalf of the CrIS SDR Cal/Val team Dave Tobin, Hank Revercomb, Joe Taylor, Bob Knuteson, Dan DeSlover, Lori Borg, Graeme Mar1n Space Science and Engineering Center, University of WisconsinMadison SUOMI NPP SDR Science and Validated Product Maturity Review NOAA Center for Weather and Climate Predic1on, College Park, MD 1820 December 2013

Upload: others

Post on 27-Sep-2020

5 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

CrIS  Radiometric  Calibra1on:    Uncertainty  Es1mates  and  Evalua1ons  

On  behalf  of  the  CrIS  SDR  Cal/Val  team    

Dave  Tobin,  Hank  Revercomb,  Joe  Taylor,  Bob  Knuteson,  Dan  DeSlover,  Lori  Borg,  Graeme  Mar1n  Space  Science  and  Engineering  Center,  University  of  Wisconsin-­‐Madison  

 SUOMI  NPP  SDR  Science  and  Validated  Product  Maturity  Review    

NOAA  Center  for  Weather  and  Climate  Predic1on,  College  Park,  MD  

18-­‐20  December  2013    

 

Page 2: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Outline  •  Radiometric  Uncertainty  Es1mates  

–  Perturba)on  of  Calibra)on  Equa)on  and  Parameter  uncertain)es  –  On-­‐orbit  RU  es)mates  –  Other  terms  

•  Nonlinearity  Refinements  –  Mo)va)on  and  Methodology,  and  changes  to                  NLC  equa)on  and  coefficients  –  Reprocessed  dataset  –  Example  changes  in  calibrated  spectra  

•  Evalua1ons  –  AircraF  underflights  –  CrIS/VIIRS  comparisons  –  CrIS/IASI  comparisons  –  CrIS/AIRS  comparisons  –  Clear  sky  Obs-­‐Calc  

•  Con1nuing  Work  –  Spectral  Ringing  –  Polariza)on  –  SW  band  biases  

•  Summary  and  Conclusions  

2  

Page 3: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Radiometric  Uncertainty  (RU)  Es1mates  •  Perturba)on  of  Calibra)on  Equa)on  and  Parameter  uncertain)es  •  On-­‐orbit  RU  es)mates  •  Other  terms    Ø  Required  in  order  to  understand  the  size  and  dependencies  of  the  

primary  contributors  to  the  CrIS  SDR  uncertain)es,  for  calibra)on  improvements,  weather,  process,  trend,  and  inter-­‐calibra)on  applica)ons.  

3  

Page 4: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Radiometric  Uncertainty  Es1mates  

Simplified  On-­‐Orbit  Radiometric  Calibra1on  Equa1on:  

     Rscene  =  Re  {(C’scene  –  C’SP)  /(C’ICT-­‐C’SP)}  RICT      with:    

Nonlinearity  Correc)on:    C’  =  C  �  (1  +  2  a2  VDC)  ICT  Predicted  Radiance:    RICT  =  εICT  B(TICT)  +  (1-­‐εICT)  [  0.5  B(TICT,  Refl,  Measured)  +  0.5  B(TICT,  Refl,  Modeled)]  

   

   

Following  Tobin  et  al.  (2013),  Suomi-­‐NPP  CrIS  radiometric  calibra>on  uncertainty,  J.  Geophys.  Res.  Atmos.,  118,  10,589–10,600,  doi:10.1002/jgrd.50809.  

Parameter  Uncertain1es:  Parameter   Nominal  Values   3-­‐σ  Uncertainty  

TICT   280K   112.5  mK*  

εICT   0.974-­‐0.996   0.03  

TICT,  Refl,  Measured   280K   1.5  K  

TICT,  Refl,  Modeled   280K   3  K  

a2  LW  band     0.01  –  0.03  V-­‐1   0.00403  V-­‐1  

a2  MW  band   0.001  –  0.12  V-­‐1   0.00128  –  0.00168  V-­‐1  

4  

*Exelis  at-­‐launch  es>mate  

Page 5: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

200

220

240

260

280

300

BT (K

)

700 800 900 1000

−0.2

−0.1

0

0.1

0.2

0.3

wavenumber

3−si

gma

RU

(K)

200 220 240 260 280 3000

0.1

0.2

0.3

BT (K)

3−si

gma

RU

(K)

wavenumber700 800 900 1000

1200 1300 1400 1500 1600 1700wavenumber

200 220 240 260 280 300BT (K)

wavenumber1300 1400 1500 1600 1700

2200 2300 2400 2500wavenumber

200 220 240 260 280 300BT (K)

wavenumber2200 2300 2400 2500

Example 3-sigma RU estimates

For a typical warm, ~clear sky spectrum

200

220

240

260

280

300

BT (K

)

700 800 900 1000

−0.2

−0.1

0

0.1

0.2

0.3

wavenumber

3−si

gma

RU

(K)

200 220 240 260 280 3000

0.1

0.2

0.3

BT (K)

3−si

gma

RU

(K)

wavenumber700 800 900 1000

1200 1300 1400 1500 1600 1700wavenumber

200 220 240 260 280 300BT (K)

wavenumber1300 1400 1500 1600 1700

2200 2300 2400 2500wavenumber

200 220 240 260 280 300BT (K)

wavenumber2200 2300 2400 2500

5  

500 1000 1500 2000 25000

0.05

0.1

0.15

0.2

wavenumber (cm−1)

3m B

T R

U (K

)

TICT¡ICTTICT,Refl,MeasTICT,Refl,ModelTST¡STTST,Refla2Total RU

500 1000 1500 2000 25000

0.05

0.1

0.15

0.2

wavenumber (cm−1)

3m B

T R

U (K

)

TICT¡ICTTICT,Refl,MeasTICT,Refl,ModelTST¡STTST,Refla2Total RU

Page 6: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

BT (K)

3−si

gma

RU

(K)

200 250 3000

0.1

0.2

0.3

0.4

0.5

BT (K)

200 250 300

log(count)0 5 10

BT (K)200 250 300

Example 3-sigma RU estimates

LW MW SW

Log scale RU distributions for one orbit of CrIS Earth view data, including all FOVs and spectral channels within the band:

6  

Ø  Uncertain)es  are  greatly  reduced  due  to  re-­‐analysis  of  the  TVAC  data  and  on-­‐orbit  FOV-­‐2-­‐FOV  analysis.    In  par)cular,  MW  band  uncertain)es  are  greatly  reduced  due  to  the  high  degree  of  linearity  of  MW  reference  FOV9.  

Ø  Overall,  RU  is  <0.3K  (LW),  <0.15K  (MW),  <0.15K  (SW):  Bemer  than  spec  by  approximately  a  factor  of  4.  

Page 7: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Other  Terms  

Ø  Spectral  Ringing  Ø  Polariza)on  Ø  Possible  SW  Nonlinearity  •  Other  smaller/negligible  terms:  

–  Detector  temperature  changes,  Changes  in  DA  Bias  )lt  over  4  minutes,  Changes  in  op)cal  flatness,  OPD  sample  rate  driF  over  4  minutes,  Electronic  gain  driF  over  4  minutes,  Electronic  delay  driF  over  4  minutes,  FOV  to  FOV  crosstalk  in  same  band,  FOV  to  FOV  crosstalk  between  bands,  Stray  light,  Op)cs  temperature  change  during  cal,  Changes  in  channel  spectra  

Smaller contributors not currently accounted for in the calibration algorithm or included in current RU estimates:

7  

Addressed  in  later  slides  

Page 8: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Nonlinearity  Refinements  •  Mo)va)on  and  Methodology,  and  changes  to  NLC  equa)on  

and  coefficients  •  Reprocessed  dataset  •  Example  changes  in  calibrated  spectra  

8  

Page 9: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Nonlinearity  Refinements  (Refinements  developed  since  the  Provisional  Review)  

Mo1va1on  •  To  provide  improved  traceability  of  the  Nonlinearity  algorithm  and  coefficients  to  the  CrIS  TVAC  

External  Calibra)on  Target  residuals  and  analysis.  •  This  is  most  important  for  the  LW  band  where  all  FOVs  display  a  significant  level  of  nonlinearity,  and  

less  so  for  the  MW  band  where  some  FOVs  display  high  levels  of  linearity  and  can  be  used  as  an  in-­‐orbit  reference  to  assess  the  other  MW  FOVs.  

Methodology  •  Involves  determina1on  of  the  a2  nonlinearity  coefficients  (other  NL  related  terms  are  rela1vely  well  

known):  1.  Ini)al  values  determined  from  analysis  of  TVAC  ECT  view  data.  2.  Change  from  pre-­‐launch  to  in-­‐orbit  es)mates  based  on  analysis  of  Diagnos)c  Mode  data  (out-­‐

of-­‐band  harmonics),  leading  to  ini)al  in-­‐orbit  values.  3.  Followed  by  the  selec)on  of  a  reference  FOV  and  adjustments  of  a2  values  for  the  remaining  

FOVs  to  create  op)mal  agreement  with  the  reference  FOV  for  Earth  view  data  

Resul1ng  changes  to  the  NLC  equa1on  and  coefficients  –  New  a2  values  (with  respect  to  current  opera)onal  values,  includes  an  overall  increase  for  all  LW  

FOVs,  along  with  smaller  refinements  to  improve  FOV-­‐2-­‐FOV  agreement  for  LW  and  MW  bands)  –  Equa)on  change  from    C’  =  C  /  (1  -­‐  2a2V)    to    C’  =  C  �  (1  +  2a2V)  –  New  modula)on  efficiency  (emod  )  values  

9  

Page 10: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Reprocessed  Dataset  The  refined  CrIS  SDRs  for  the  full  mission  are  available  at:  

Fp://peate.ssec.wisc.edu/allData/products/results/cris/cspp/SDR_1_4b_ILS_NLC_v33a-­‐04/  

 Differences  with  respect  to  the  opera1onal  IDPS  dataset:  

1.  Includes  Nonlinearity  algorithm  and  coefficient  refinements*  2.  Includes  ILS  algorithm  and  coefficient  refinements*  3.  Includes  consistent  SDR  algorithm  processing  for  the  full  mission  4.  Processing  takes  place  with  ~24  hour  latency  to  avoid  missing  packet  issues  

*  The  same  Nonlinearity  and  ILS  refinements  are  expected  to  be  implemented  in  IDPS  processing  with  MX8.1  and  EPv36  in  February  2014.  

Evalua1ons  of  the  CrIS  RU  presented  here  use  the  reprocessed  dataset.  Some  example  differences  with  respect  to  the  opera1onal  IDPS  processing  are  shown  on  the  following  slides.  

 

10  

Page 11: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Example  changes:    New  versus  Old  Nonlinearity  Correc1on  

600 650 700 750 800 850 900 950 1000 1050 1100200

220

240

260

280

300SCRIS_npp_d20120920_t0007219_e001239: FOR = 15 FOV = 9

CrIS

FM

1 B

.T. (

K)

600 650 700 750 800 850 900 950 1000 1050 1100

−0.2

−0.1

0

0.1UW NLC (mod4) minus CSPP default

Wavenumber (cm−1)

B.T

. Diff

(K)

672-­‐682  cm-­‐1  

11  

Ø  LW  stratospheric  channels  shiF  to  colder  brightness  temperatures  (order  ~0.1K)  everywhere.  

Ø  For  cold  scenes,  LW  window  channels  also  shiF  to  colder  brightness  temperatures.  

Page 12: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Example  changes:  Long  term  FOV-­‐2-­‐FOV  differences  

Apr Jul Oct Jan Apr Jul

−0.2

−0.1

0

0.1

0.2

BT D

iff (K

)

Apr Jul Oct Jan Apr Jul

FOV1FOV2FOV3FOV4FOV5FOV6FOV7FOV8FOV9

LW  (672-­‐682  cm-­‐1)  BT  Differences  with  respect  to  FOV5  

IDPS  processing  

Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug−0.25

−0.2

−0.15

−0.1

−0.05

0

0.05

0.10.15

0.2

BT D

iff (K

)

Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug

FOV1FOV2FOV3FOV4FOV5FOV6FOV7FOV8FOV9

Reprocessed  

Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug−0.25

−0.2

−0.15

−0.1

−0.05

0

0.05

0.10.15

0.2

BT D

iff (K

)

Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug

FOV1FOV2FOV3FOV4FOV5FOV6FOV7FOV8FOV9

12  

Ø  Reprocessing  with  the  NLC  refinements  removes  the  inconsistencies  in  the  IDPS  )me  series  and  also  minimizes  the  magnitude  of  FOV-­‐2-­‐FOV  differences.  

Page 13: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Evalua1ons  of  RU  es1mates      (“Cal/Val”)  

•  AircraF  underflights  •  CrIS/VIIRS  comparisons  •  CrIS/IASI  comparisons  •  CrIS/AIRS  comparisons  •  Clear  sky  Obs-­‐Calc  

Ø  A  range  of  techniques,  with  various  levels  of  uncertainty/sta)s)cs/traceability,  to  assess  the  CrIS  SDRs  and  associated  RU  es)mates.  

13  

Page 14: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

May  15  Underflight  example:  S-­‐HIS  and  CrIS  895–900  cm-­‐1  BTs  overlaid  on  VIIRS  true  color  image  

May  2013  Suomi-­‐NPP  JPSS  Aircrah  Campaign  Scanning-­‐HIS  evalua1ons  of  CrIS  Calibra1on  

14  

BT  [K]  

BT  [K]  

Page 15: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Double  Obs-­‐Calc  Comparison  Methodology  and  Uncertainty  (CrISobs-­‐CrIScalc)-­‐(SHISobs-­‐SHIScalc)  

15  

650 700 750 800 850 900 950 1000 1050 11000

0.1

0.2

0.3

0.4

BT (K

)

wavenumber

Double Obs−Calc Uncertainty

Spatial VariabilitySHIS 3−sigma RUObs−Calc MethodologyTotal (RSS)

Page 16: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

S-­‐HIS  Calibra1on,  Calibra1on  Verifica1on,  and  Traceability  

16  

NIST  TXR  Valida)on  of  S-­‐HIS  Radiances  •  Pre  and  post  deployment  end-­‐to-­‐end  calibra)on  verifica)on    •  Instrument  calibra)on  during  flight  using  on-­‐board  calibra)on  blackbodies  •  Periodic  end-­‐to-­‐end  radiance  evalua)ons  under  flight-­‐like  condi)ons  with  NIST  transfer  sensors  

Post  SNPP  End-­‐to-­‐End  Calibra)on  Verifica)on  

333K  

318K  

298K  

273K  

Channel  1  @  5µm  

Channel  2  @  10µm  

Page 17: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

CrIS/S-­‐HIS  Underflight  Results  Hamming  apodiza)on  

Ø  AircraF  underflights  provide  periodic  end-­‐to-­‐end  verifica)on  of  CrIS  RU  es)mates  with  0.1-­‐0.2K  uncertainty  over  most  of  the  spectrum.  

17  

Mean  DOMC    ±  uncertainty  CrIS  RU  

Page 18: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

CrIS/VIIRS  comparisons  Example  Daily  Comparisons,  M15  band  @  10.8µm,  Descending  

CrIS  convolved  with  VIIRS  SRF   VIIRS  mean  within  CrIS  FOVs  

VIIRS  standard  devia1on  within  CrIS  FOVs   Differences  for  uniform  scenes  

BT  (K)  210      220        230          240        250      260          270        280          290        300  

0                  1                  2                  3              4                  5                6                    7                8                    9              10   -­‐5            -­‐4                -­‐3                -­‐2              -­‐1                0                1                  2                  3                4                  5  

BT  (K)  210      220        230          240        250      260          270        280          290        300  

BT  (K)   VIIRS  -­‐  CrIS  (K)  

Ø  Each  day  includes  ~500,000  coloca)ons  which  pass  a  spa)al  uniformity  test  18  

Page 19: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

−0.1

−0.05

0

0.05

0.1VI

IRS −

CrIS

BT

(K)

CrIS/VIIRS Daily Mean Differences

−0.02

−0.01

0

0.01

0.02

VIIR

S −

CrIS

BT

(K)

Mar

Apr

May

Jun Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun Jul

Aug

Sep

Oct

Nov

Dec

Jan

−0.1

−0.05

0

0.05

0.1

VIIR

S −

CrIS

BT

(K)

CrIS/VIIRS Daily Mean Differences

M13, 4umM15, 10.8umM16, 12um

−0.02

−0.01

0

0.01

0.02

VIIR

S −

CrIS

BT

(K)

Mar

Apr

May

Jun Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun Jul

Aug

Sep

Oct

Nov

+0.074  K  

-­‐0.064  K  

-­‐0.022  K  

Mean  Bias  over  past  21  months:  

with  mean  biases  subtracted  off  

19  

-­‐6.4  ±  1.0  -­‐1.6  ±  1.2  -­‐4.4  ±  1.0  

Trends,  mK/yr:  

Ø  CrIS/VIIRS  daily  mean  differences  are  <  0.1K  and  trends  are  <  10  mK/yr  

Page 20: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

CrIS/VIIRS comparisons with uncertainties

Legend

Mean differences

CrIS 3σ RU

VIIRS 3σ RU

Uncertainty due to Out-of-Band SRF effects

Statistical uncertainty

RSS

20  

Ø  VIIRS  RU  es)mates  provided  by  Jeff  McIn)re  et  al.  

Ø  CrIS/VIIRS  differences  are  bounded  by  combined  RU  for  all  scene  temperatures  

Ø  Larger  VIIRS  RU  for  cold  scenes  at  M15  and  M16  are  due  to  c0  offset  term  and  under  inves)ga)on  by  the  VIIRS  SDR  team.  

VIIR

S −

CrIS

BT

(K)

BT (K)

M13 @ 4µm

200 220 240 260 280 300 320−0.5

0

0.5

VIIR

S −

CrIS

BT

(K)

BT (K)

M15 @ 10.8µm

200 220 240 260 280 300 320−0.5

0

0.5

VIIR

S −

CrIS

BT

(K)

BT (K)

M16 @ 12µm

200 220 240 260 280 300 320−0.5

0

0.5

Page 21: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

SNO  Comparison  Methodology  

21  

−2 0 2101

102

103

104

AIRS − CrIS (K)

The  SNO  comparison  technique  is  aimed  at  minimizing  differences  in  the  spa)al/temporal  colloca)on  process  and  providing  well  understood  uncertain)es  to  iden)fy  persistent  biases  between  two  sensors.  

A  sample  SNO  showing  CrIS  and  AIRS  footprints  within  100  km  of  the  SNO  loca)on.  

Colloca)on  difference  distribu)ons  for  a  large  ensemble  of  SNOs  for  various  

ranges  of  spa)al  variability.  

0

50

100

21:53:59 UTC

AIRSCrIS

700 800 900 1000 11000

2

4

6

wavenumber (cm−1)

0

50

100

Mea

n (m

W/(m

2 sr.

cm−1

))

22:31:24 UTC

700 800 900 1000 11000

2

4

6

wavenumber (cm−1)

STD

EV (m

W/(m

2 sr.

cm−1

))

LW  mean  and  standard  devia)on  spectra  for  two  example  SNOs  collected  on  20120816.  

830-­‐840  cm-­‐1  

Page 22: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

SNO  Datasets  

22  

CrIS/AIRS:  1.2M  “Big  Circle”  SNOs  collected  to  date  (March  2012  to  Nov  2013);    20  minute  window;  -­‐30  to  30  deg  scan  angle,  <=2  deg  scan  angle  diff.  

AIRS  V5  L1B;  CrIS  ADL/CSPP  SDR_1.4b_NLC_ILS  

835  cm-­‐1  CrIS/AIRS  SNO  BT  Diffs  

CrIS/IASI-­‐A:  5270  “Big  Circle”  SNOs  collected  to  date  (March  2012  to  Nov  2013);    20  minute  window;  nadir.    ~20  days  of  coincidences,  ~30  day  gaps,  

~half  at  +72.4  deg,  ~half  at  -­‐72.4  deg.  IASI_xxx_1C_M02;  CrIS  ADL/CSPP  SDR_1.4b_NLC_ILS  

2510  cm-­‐1  CrIS/AIRS  SNO  BTs  

CrIS/IASI  SNO  loca1ons  

Page 23: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

CrIS/IASI  Northern  SNOs    Hamming  apodiza)on  

Ø  Results  shown  for  IDPS  processing  and  with  the  reprocessed  dataset  including  the  NLC  refinements  presented  earlier.  

Ø  Differences  of  ~0.2K  or  less  

Ø  NLC  refinements:  •  Improved  agreement  in  the  

LW  band.  •  Negligible  changes  in  the  

the  MW  band  (as  expected).  

mean  CrIS  mean  IASI  

Mean  Difference  ±  Uncertainty  

Page 24: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

−0.4

−0.2

0

0.2

0.4

BT

Dif

(K)

−0.4

−0.2

0

0.2

0.4

BT

Dif

(K)

200 220 240 260 280 300 320

−0.4

−0.2

0

0.2

0.4

BT (K)

BT

Dif

(K)

200 220 240 260 280 300 320BT (K)

672-­‐682  cm-­‐1   830-­‐840  cm-­‐1  

1382-­‐1408  cm-­‐1   1585-­‐1600  cm-­‐1  

2360-­‐2370  cm-­‐1   2500-­‐2520  cm-­‐1  

24  

Summary  of  SNO  results  for  6  representa1ve  spectral  regions,  

and  VIIRS/CrIS  comparisons:  

VIIRS–CrIS,  M15  VIIRS–CrIS,  M16  VIIRS–CrIS,  M13  

600 700 800 900 1000 1100220

240

260

280

300

BT (K

)

1200 1300 1400 1500 1600 1700 1800220

240

260

280

300

BT (K

)

1800 2000 2200 2400 2600 2800220

240

260

280

300

wavenumber

BT (K

)

VIIRS  M16  

VIIRS  M15  

VIIRS  M13  

672-­‐682  cm-­‐1   830-­‐840  cm-­‐1  

1382-­‐1408  cm-­‐1   1585-­‐1600  cm-­‐1  

2360-­‐2370  cm-­‐1   2500-­‐2520  cm-­‐1  

IASI-­‐CrIS  AIRS-­‐CrIS  

Ø  LW  differences  display  only  small  dependence  on  scene  BT  for  both  IASI  and  AIRS  SNOs.  

Ø  MW  differences  are  rela)vely  independent  of  scene  BT  for  IASI  and  for  AIRS  at  1382-­‐1408  cm-­‐1;  Differences  for  AIRS  at  1585-­‐1600  cm-­‐1  range  from  ~+0.3K  at  200K  to  -­‐0.1K  at  265K.  

Ø  SW  differences  are  rela)vely  flat  above  ~240K;  Below  ~230K  larger  differences  between  all  three  sensors  are  observed.  

Ø  Consistent  with  SNO  results  shown  in  L.  Strow  presenta)on,  and  reported  by  L.  Wang  et  al.  at  NOAA  STAR.  

Page 25: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

25  

800 1000 1200 1400 1600 1800 2000 2200 2400 2600−1.5

−1

−0.5

0

0.5

1

1.5

Wavenumber (cm−1)

Ham

min

g O

bs−

Calc

in K

ECMWF IFS 38r1

ECMWF IFS 38r2

Difference

1000 1500 2000 2500−0.5

0

0.5

1

1.5

2

2.5

Wavenumber (cm−1)

Sta

nd

ard

De

via

tion

of

Ob

s−C

alc

(K

)

ECMWF IFS 38r1

ECMWF IFS 38r2

Difference X10

c/o  Yong  Chen  and  Yong  Han,  NOAA  STAR:  

c/o  Larrabee  Strow,  UMBC:  

Clear  Sky  Obs-­‐Calc  Analyses  

Ø  Behavior  of  mean  biases  and  standard  devia)on  of  obs-­‐calcs  are  consistent  with  forward  model  and  atmospheric  state  uncertain)es  and  imply  very  good  radiometric  performance  for  CrIS.  

                   Before  7/25/13  (ECMWF  IFS  38r1)                      AFer  7/25/13        (ECMWF  IFS  38r2)                      Difference  

                   Before  7/25/13  (ECMWF  IFS  38r1)                      AFer  7/25/13        (ECMWF  IFS  38r2)                      Difference  

Page 26: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Con1nuing  Work:    Calibra1on  and  RU  Refinements  

•  Spectral  Ringing  •  Polariza)on  •  SW  band  biases  

26  

Page 27: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Spectral  Ringing  

•  Considered  to  be  part  of  the  RU  budget,  but  is  covered  separately  in  Dan  Mooney’s  talk  

•  For  the  large  majority  of  spectral  channels,  the  associated  ar)facts  are  very  small,  and  with  apodiza)on  are  negligible  everywhere.  

27  

Page 28: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

0 10 20 30 40 50 60 70 80 90−0.1

−0.05

0

0.05

0.1

CrIS cross track scan index

VIIR

S −

CrIS

(K)

200−210 K210−220 K220−230 K230−240 K240−250 K250−260 K260−270 K270−280 K280−290 K290−300 K300−310 K310−320 K320−330 K

Error  from  Scene  Mirror  Induced  Polariza1on  •  CrIS  uses  a  45°  gold  scene  mirror  that  provides  low  sensi)vity  to  polariza)on;  no  

correc)on  is  included  in  the  SDR  algorithm/processing.  •  However,  it  seems  almost  certain  that  CrIS  should  have  polariza)on  effects  of  ~0.1  K  for  

especially  warm  and  cold  brightness  temperatures  in  some  spectral  regions.  •  A  correc)on  should  be  developed  based  on  CrIS  characteriza)on  tests  yet  to  be  

conducted  (measurements  of  scene  mirror  degree  of  polariza)on,  pr,  and  interferometer  polariza)on  sensi)vity,  pt)  

•  Radiance  error  dependence  ~2prpt  (N  –  BICT)  •  Sugges)ons  of  this  type  of  behavior  in  CrIS/VIIRS  comparisons  vs.  scan  angle:  

28  *  Biases  removed  for  mean  of  INDICES  1-­‐10  &  80-­‐90  

Scan  Angle  and  Scene  BT  dependence  of  VIIRS/CrIS  Comparisons  at  10.8µm*  

Page 29: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

29  

200

250

300

BT (K

)

200

250

300

BT (K

)

2150 2200 2250 2300 2350 2400 2450 2500 25500

0.05

0.1

wavenumber

BT (K

)

SW  Band  Biases  •  FOV-­‐2-­‐FOV  analyses  and  differences  with  

respect  to  other  sensors  suggest  small  ar)facts  in  the  SW  band,  both  in  Mean  biases  and  FOV-­‐2-­‐FOV  differences.        E.g.  Differences  with  respect  to  IASI    

•  Mechanisms  inves)gated  to  date:  –  Spectral  shiF  –  Thermal  SP  view  contamina)on  –  Solar  SP  view  contamina)on  –  Noise  –  Polariza)on  Ø  Low  level  Nonlinearity  

•  Displays  FOV  dependent  behavior  •  Has  plausible  spectral  and  scene  level  

dependencies    

0.20

0.20.40.60.8

FOV 1

−0.20

0.20.40.60.8

FOV 2

−0.20

0.20.40.60.8

FOV 3

0.20

0.20.40.60.8

FOV 4

−0.20

0.20.40.60.8

FOV 5

−0.20

0.20.40.60.8

FOV 6

22002300240025000.2

00.20.40.60.8

FOV 7

2200230024002500−0.2

00.20.40.60.8

FOV 8

2200230024002500−0.2

00.20.40.60.8

FOV 9

CrIS-­‐IA

SI  (K

)  CrIS-­‐IA

SI  (K

)  CrIS-­‐IA

SI  (K

)  

wavenumber   wavenumber   wavenumber  

                 Mean  for  all  FOVs                    Difference  for  this  FOV  

Sample  spectra  (top)  and  Effect  of  small  NLC  (bomom)  

Page 30: uw cris sdr review December2013 20131213 · 200 220 240 260 280 300 BT (K) 700 800 900 1000 −0.2 −0.1 0 0.1 0.2 0.3 wavenumber 3 − sigma RU (K) 200 220 240 260 280 300 0 0.1

Summary  and  Conclusions  

30  

Ø  On-­‐orbit  Radiometric  Uncertainty  (RU)  characterized:  –  Based  on  careful  es)ma)on  of  3-­‐sigma  uncertain)es  of  the  primary  calibra)on  parameters  and  

perturba)on  of  the  radiometric  calibra)on  equa)on  –  Overall  RU  es)mates  are  <  0.3K  (LW),  <  0.15K  (MW),  <  0.15  K  (SW)  

Ø  Nonlinearity  correc)on  algorithm  and  coefficients  refined    –  Provides  improved  traceability  of  the  nonlinearity  coefficients  to  the  TVAC  External  Calibra)on  Target  –  Refinements  reduce  overall  RU  in  LW  band  and  reduce  FOV  dependence  in  LW  and  MW  bands  –  SDRs  reprocessed  using  NLC  (and  ILS)  refinements  and  distributed  

Ø  CrIS  SDRs  and  RU  es)mates  have  undergone  extensive  verifica)on/evalua)on  over  a  range  of  representa)ve  condi)ons  –  Periodic  aircraF  underflights  provide  high  quality,  SI  traceable  verifica)on  of  the  CrIS  RU  –  CrIS/VIIRS  comparisons  imply  excellent  stability  of  both  sensors,  and  scene  BT  dependence  further  

characterized  and  diagnosed  –  SNOs  of  CrIS/AIRS  and  CrIS/IASI  show  excellent  mean  agreement  and  behavior  with  scene  temperature  –  Clear  sky  obs-­‐calcs  imply  very  good  CrIS  radiometric  performance  

Ø  Areas  of  further  refinement  have  been  iden)fied  and  are  under  inves)ga)on  –  Spectral  Ringing  –  Polariza)on  –  Possible  low  level  SW  band  nonlinearity