The AIRS V5 Ozone Validation The AIRS V5 Ozone Validation An Outline - - PowerPoint PPT Presentation

1

• An Outline of the Paper Submitted

An Outline of the Paper Submitted To JGR (Presently under review).

To JGR (Presently under review).

(Let me know if you wish a copy of the paper) (Let me know if you wish a copy of the paper)

Evaluation of Atmospheric Infrared Sounder (AIRS) Ozone Profiles and Total Evaluation of Atmospheric Infrared Sounder (AIRS) Ozone Profiles and Total Ozone Retrievals with Matched Ozonesonde Measurements, ECMWF Ozone Data, Ozone Retrievals with Matched Ozonesonde Measurements, ECMWF Ozone Data, and Ozone Monitoring Instrument (OMI) Retrievals* and Ozone Monitoring Instrument (OMI) Retrievals*

• On-Going Activities and Upcoming Plans

On-Going Activities and Upcoming Plans

Murty Murty Divakarla, Christopher Barnet, Mitchell Goldberg, Eric Maddy, XingpinLiu, Divakarla, Christopher Barnet, Mitchell Goldberg, Eric Maddy, XingpinLiu, Walter Wolf, Lawrence Flynn, Gordon Walter Wolf, Lawrence Flynn, Gordon Labow Labow@

@, Xiaozhen Xiong, Jennifer Wei, and

, Xiaozhen Xiong, Jennifer Wei, and Lihang Zhou Lihang Zhou NOAA/NESDIS, Camp Springs, MD 20746; @NASA/GSFC NOAA/NESDIS, Camp Springs, MD 20746; @NASA/GSFC Bill Irion and Mike Bill Irion and Mike Newchurch Newchurch NASA/JPL, Univ. of Alabama NASA/JPL, Univ. of Alabama

Thanks To: Thanks To: Shuntai Shuntai Zhou, NOAA/NCEP, Craig Long, NOAA/NCEP, Eric Beach NOAA, Zhou, NOAA/NCEP, Craig Long, NOAA/NCEP, Eric Beach NOAA, Bojan Bojan Bojkov Bojkov NASA/GSFC NASA/GSFC

The AIRS V5 Ozone Validation The AIRS V5 Ozone Validation

2

What We Did What We Did

• Validation of AIRS Retrieved Ozone profile, and Total

Validation of AIRS Retrieved Ozone profile, and Total Ozone with WOUDC O Ozone with WOUDC O3

3SNDS and Total Ozone

SNDS and Total Ozone Measurements. Measurements.

» » Both V4 and V5 Versions (Retrieval Emulations at Both V4 and V5 Versions (Retrieval Emulations at NOAA/NESDIS) NOAA/NESDIS) » » AIRS Ozone Profile Retrievals with O AIRS Ozone Profile Retrievals with O3

3SNDS

SNDS » » AIRS Total Ozone with BD Measurements AIRS Total Ozone with BD Measurements – – Simultaneous Aqua-AIRS, Aura-OMI and WOUDC Total Simultaneous Aqua-AIRS, Aura-OMI and WOUDC Total Ozone Measurements Ozone Measurements – – Individual Individual STNs STNs, NH, SH, Tropics, Global , NH, SH, Tropics, Global

• Analysis of AIRS Global Ozone Grids

Analysis of AIRS Global Ozone Grids

» » Using Using Gridded Gridded Monthly Averages (2004 and 2005) of AIRS, Monthly Averages (2004 and 2005) of AIRS, OMI, SBUV and Global Forecast System (NCEP-GFS) OMI, SBUV and Global Forecast System (NCEP-GFS) – – Seasonal Trends, Patterns Seasonal Trends, Patterns » » How Point Measurements/Validation How Point Measurements/Validation Corroborate to Global Corroborate to Global Perspective Seen with the Global Grids Perspective Seen with the Global Grids » » Annual Cycles Annual Cycles

3

Data Credits and Acknowledgments Data Credits and Acknowledgments

» » Ozone Ozone sondes sondes and Brewer Dobson Measurements and Brewer Dobson Measurements – – WOUDC Ozone Data WOUDC Ozone Data ( (www.woudc.org www.woudc.org) ) » » Gridded Gridded Monthly Averages for AIRS, SBU, OMI and GFS Monthly Averages for AIRS, SBU, OMI and GFS for the Years 2004 and 2005 for the Years 2004 and 2005 – – AIRS V4 and V5 Retrievals Emulated at NOAA/NESDIS AIRS V4 and V5 Retrievals Emulated at NOAA/NESDIS

  Global Grid Resolution: 3

Global Grid Resolution: 3o

• x 3

x 3o

• 

 Generated by NOAA/NESDIS/IOSSPDT

Generated by NOAA/NESDIS/IOSSPDT – – V4 and V5 V4 and V5 Emulation Emulation – – SBUV/2 on NOAA-16 SBUV/2 on NOAA-16

  Global Grid Resolution: 2.5

Global Grid Resolution: 2.5o

• x 2.5

x 2.5o

• 

 Monthly Data Provided by

Monthly Data Provided by Shuntai Zhou Shuntai Zhou, NCEP , NCEP

  Download :

Download : ftp://ftp.cpc.ncep.noaa.gov/SMOBA ftp://ftp.cpc.ncep.noaa.gov/SMOBA – – Ozone Monitoring Instrument (OMI) on Aura Ozone Monitoring Instrument (OMI) on Aura

  Global Grid Resolution: 1.25

Global Grid Resolution: 1.25o

• x 1.0

x 1.0o

• 

 Downloaded from:

Downloaded from:

  ftp://toms.gsfc.nasa.gov/pub/omi/data/monthly_averages/ozo

ftp://toms.gsfc.nasa.gov/pub/omi/data/monthly_averages/ozo ne ne – – Global Forecast System (GFS) Global Forecast System (GFS)

  Monthly Data provided by

Monthly Data provided by Craig Long Craig Long

  ftp://ftp.cpc.ncep.noaa.gov/long/gfs_ozone/month/

ftp://ftp.cpc.ncep.noaa.gov/long/gfs_ozone/month/

4

The AIRS Ozone Validation Ladder The AIRS Ozone Validation Ladder

2 1 3 4 5 6

5

What We Did and What is Achieved (1) What We Did and What is Achieved (1)

• Performed validations for V4 and V5

Performed validations for V4 and V5

• AIRS Retrieved Ozone Profiles with WOUDC O

AIRS Retrieved Ozone Profiles with WOUDC O3

3SNDS

SNDS

• Results:

Results:

• The V5 algorithm significantly improves the retrieval

The V5 algorithm significantly improves the retrieval bias and RMS differences for the lower troposphere bias and RMS differences for the lower troposphere and especially over the tropical regions. and especially over the tropical regions.

• Retrieval Improvement for the Lowest Layer (1000-

Retrieval Improvement for the Lowest Layer (1000- 260 260 mb mb) is Mainly due to the Climatology First ) is Mainly due to the Climatology First Guess Guess

• The Retrieval Statistics with Global O

The Retrieval Statistics with Global O3

3SNDS

SNDS

• Bias ~ 5%

Bias ~ 5%

• RMS Difference ~ 20%

RMS Difference ~ 20%

6

V5 Physical Retrievals - Different Stations V5 Physical Retrievals - Different Stations

ALL (N=353), ALL (N=353), STN 089 (N=53, NH), STN 089 (N=53, NH), STN 101(N=50,SH), STN 101(N=50,SH), Tropics (N=54, ±12HR) Tropics (N=54, ±12HR)

10 100 1000 20 40 60 80 100 120 140 160 Ozone RMS Difference (%) Pressure (hPa) V5.0.7_ALL_3H V4.0_ALL_3H V5.0.7_STN089 V4.0_STN089 V5.0.7_STN101 V4.0_STN101 V5.0.7_TRP V4.0_TRP 10 100 1000

• 60
• 40
• 20

20 40 60 80 100 120 140 Ozone Bias (%) Pressure (hPa) V5.0.7_ALL_3H V4.0_ALL_3H V5.0.7_STN089 V4.0_STN089 V5.0.7_STN101 V4.0_STN101 V4.0_TRP V5.0.7_TRP

RMS Difference RMS Difference Bias Bias

1000 mb 10 mb 100 mb

0 20 40 60 80 100 120 140

• 60 -40 -20 0 20 40 60 80 100

V4 Dotted lines, V5 Solid Lines V4 Dotted lines, V5 Solid Lines ALL_3H ALL_3H STN089 STN089 STN101 STN101 TRP_12H TRP_12H ALL_12H ALL_12H

V5 Climatology First Guess V5 Climatology First Guess Helps Physical Helps Physical Retrival Retrival for for the Lowest Layer 1 (1000- the Lowest Layer 1 (1000- 260 260 mb mb) )

7

What We Did and What is Achieved (2) What We Did and What is Achieved (2)

• Performed Validations for V4 and V5 Total Ozone

Performed Validations for V4 and V5 Total Ozone

• With WOUDC Brewer/Dobson Station Measurements.

With WOUDC Brewer/Dobson Station Measurements.

• Results:

Results:

• Total ozone from both the V4 and V5 versions agrees well

Total ozone from both the V4 and V5 versions agrees well

• Bias ~ 4% and an RMS difference ~ 8%.

Bias ~ 4% and an RMS difference ~ 8%.

 

V5 ozone retrievals are better than V4 V5 ozone retrievals are better than V4 » » Over Desert Regions (Due to Improvements in Over Desert Regions (Due to Improvements in Emissivity Emissivity Regression implemented in V5) Regression implemented in V5) » » In Depicting Ozone Hole Events, Trends, Seasonality In Depicting Ozone Hole Events, Trends, Seasonality Patterns etc. Patterns etc.

 

However, V5 Algorithm However, V5 Algorithm 1. 1. Slightly underestimates total ozone in the tropics Slightly underestimates total ozone in the tropics 2. 2. Slightly overestimates total ozone in the midlatitude and Slightly overestimates total ozone in the midlatitude and high latitudes regions. high latitudes regions. 3. 3. Emissivity Emissivity retrievals may require further refinements. retrievals may require further refinements.

8

AIRSPR_Bias(%)

• 10
• 5

5 10 POLAR Mid-LAT TROPICS GLOBAL Region Bias(%) AIRSPR_Bias(%)

V5 - Total Ozone Statistics with BD Measurements V5 - Total Ozone Statistics with BD Measurements Polar, Mid-Lat, Tropics, and Global Polar, Mid-Lat, Tropics, and Global

V5 Physical Retrieval V5 Physical Retrieval Bias and RMS Bias and RMS Difference Difference

Slightly underestimates Slightly underestimates total ozone in the total ozone in the tropics tropics Slightly overestimates Slightly overestimates total ozone in the total ozone in the midlatitude midlatitude and and polar regions. polar regions.

V5 Bias V5 Bias V5 RMS V5 RMS

AIRSPR_RMS(%)

5 10 15 20 25 POLAR Mid-LAT TROPICS GLOBAL Region RMS (%) AIRSPR_RMS(%)

Polar Midlat Tropics Global

9

What We Did and What is Achieved (3) What We Did and What is Achieved (3)

• Using Simultaneous AIRS & OMI Observations

Using Simultaneous AIRS & OMI Observations and BD Total Ozone Measurements and BD Total Ozone Measurements

» » Relative Performance Assessment of AIRS and OMI Relative Performance Assessment of AIRS and OMI Total Ozone Retrievals. Total Ozone Retrievals.

• Results:

Results:

» » The OMI Retrieval Bias The OMI Retrieval Bias – – ~ 2-3% for most of the stations ~ 2-3% for most of the stations » » The AIRS Retrieval bias is ~ 6% for high latitude STNS The AIRS Retrieval bias is ~ 6% for high latitude STNS » » The AIRS retrieval shows The AIRS retrieval shows – – Slight underestimation in the tropics Slight underestimation in the tropics – – and a slight overestimation in the high latitudes. and a slight overestimation in the high latitudes. » » OMI Data Are Extremely Good and Could be Used as a OMI Data Are Extremely Good and Could be Used as a transfer standard for the truth in Analyzing Global transfer standard for the truth in Analyzing Global Grids. Grids.

10

100 150 200 250 300 350 400 450 500 550 600 100 150 200 250 300 350 400 450 500 550 600 Total Ozone from Brewer/Dobson Station Measurements (DU) Total Ozone from V5_PR and OMI (DU) V5_PR OMI N=694 Correlation (R2) BD Stations vs. V5_PR 0.73 BD Stations vs. OMI 0.97

Simultaneous Measurements of AIRS, OMI Simultaneous Measurements of AIRS, OMI and WOUDC BD Measurements. and WOUDC BD Measurements.

AIRS_PR AIRS_PR Aura-OMI Aura-OMI Brewer/Dobson Total O Brewer/Dobson Total O3

3 Measurements (DU)

Measurements (DU)

■ V5 ○ OMI

OMI: 15 km resolution, Overpass retrievals collocated within 25 km (CT Pixel) AIRS : 50 km resolution, collocated within 100 km distance to BD Measurements Intent: Just to get a feel for the OMI, and see whether OMI global grids could be used as a transfer standard for the truth in analyzing AIRS global grids

11

• 30
• 20
• 10

10 20 30 111_P_90S 101_P_69S 029_M_55S 159_M_32S 265_M_25S 191_T_14S 214_T_01N 002_T_23N 245_M_24N 190_M_26N 010_M_29N 152_M_30N 376_M_31N 007_M_32N 287_M_33N 014_M_36N 346_M_38N 082_M_39N 261_M_41N 012_M_43N 226_M_44N 019_M_47N 100_M_47N 099_M_48N 331_M_49N 053_M_51N 318_M_52N 316_M_52N 174_M_52N 021_M_54N 116_M_56N 077_M_59N 043_P_60N 105_P_65N 262_P_67N 315_P_80N

Dobson Measurement Station Bias, STDEV (%)

OMI_Bias(%) AIRSPR_Bias(%)

Total Ozone Total Ozone – – Percent Bias Percent Bias V5 RET V5 RET vs. vs. BD STNS BD STNS; ; OMI OMI vs. vs. BD STNs. BD STNs.

Dobson Station ID Sorted Dobson Station ID Sorted 90S-90N 90S-90N Saharan Desert South Pole

Spurious Station Data

90S 90S 90N 90N

12

What We Did and What is Achieved (4) What We Did and What is Achieved (4)

• Analyzed two years (2004-2005) of AIRS

Analyzed two years (2004-2005) of AIRS V4 and V5 Ozone profile Retrievals for V4 and V5 Ozone profile Retrievals for Characteristic Features/Trends Characteristic Features/Trends

• Results:

Results: The AIRS V5 ozone profiles The AIRS V5 ozone profiles

» » Show the ozone hole events clearly Show the ozone hole events clearly » » Depicts Brewer/Dobson circulation patterns Depicts Brewer/Dobson circulation patterns as expected. as expected.

13

AIRS V5 Clearly Depicts Ozone Hole AIRS V5 Clearly Depicts Ozone Hole

V4 09/2004 V4 09/2004 V5 09/2004 V5 09/2004 V4 09/2005 V4 09/2005 V5 09/2005 V5 09/2005

• 90S Latitude 90N
• 90S Latitude 90N

1000 mb 100 mb 10 mb 1000 mb 100 mb 10 mb

14

90S Latitude 90N

150 200 250 300 350 400 450 500

• 90
• 80
• 70
• 60
• 50
• 40
• 30
• 20
• 10

10 20 30 40 50 60 70 80 90 Latitude Total Ozone from V4, V5, SBUV - September 2004 Sept_AIRS_V4 Sept_AIRS_V5

V4, V4, V5 V5 Total Ozone - September 2004-2005 Total Ozone - September 2004-2005

150 200 250 300 350 400 450 500

• 90
• 80
• 70
• 60
• 50
• 40
• 30
• 20
• 10

10 20 30 40 50 60 70 80 90 Latitude Total Ozone from V4, V5, SBUV - September 2005 Sept_AIRS_V4_2005 Sept_AIRS_V5_2005

2.5 2.5o

• Zonal Averages

Zonal Averages

V4 has difficulty Depicting Ozone Hole V5 Depicts Ozone Hole Clearly

• Sep. 2004
• Sep. 2004
• Sep. 2005
• Sep. 2005

90S Latitude 90N

15

BD Circulation Features As Seen with AIRS V5 BD Circulation Features As Seen with AIRS V5 Monthly Averages of O3 Profile Monthly Averages of O3 Profile Latitude vs. Pressure Latitude vs. Pressure

V5 12/2004 V5 12/2004 V5 03/2005 V5 03/2005 V5 10/2004 V5 10/2004

100 mb 10 mb 100 mb 10 mb 1000 mb 1000 mb

V5 06/2005 V5 06/2005 V5 08/2005 V5 08/2005 V5 10/2005 V5 10/2005

90S Latitude 90N 90S Latitude 90N 90S Latitude 90N

Pressure (mb) Pressure (mb)

We can generate these maps for each day – next slide is a movie loop with 15-day maps. UTLS/STE

16

AIRS V5 BD Circulation MOVIE LOOP AIRS V5 BD Circulation MOVIE LOOP

17

AIRS V5 BD Circulation MOVIE LOOP AIRS V5 BD Circulation MOVIE LOOP

18

What We Did and What is Achieved (5) What We Did and What is Achieved (5)

• Analyzed Global Monthly Maps of AIRS total

Analyzed Global Monthly Maps of AIRS total

• zone with the OMI and SBUV maps:
• zone with the OMI and SBUV maps:
• To evaluate the ability of the AIRS retrievals in

To evaluate the ability of the AIRS retrievals in reproducing the characteristic trends and seasonal reproducing the characteristic trends and seasonal cycles as depicted by the OMI and SBUV instruments cycles as depicted by the OMI and SBUV instruments. .

• Results:

Results:

» » Analysis of two years (2004-2005) of V5 total ozone Analysis of two years (2004-2005) of V5 total ozone retrievals reveal: retrievals reveal: – – The AIRS V5 retrieval shows trends and patterns in The AIRS V5 retrieval shows trends and patterns in concurrence with the OMI (for 2005) and SBUV concurrence with the OMI (for 2005) and SBUV depictions. depictions. – – The AIRS retrieval shows a tendency The AIRS retrieval shows a tendency

 

Slight underestimation in the tropics Slight underestimation in the tropics

 

Slight overestimation in the high latitudes. Slight overestimation in the high latitudes.

19

Total Ozone Total Ozone – – February 2005 February 2005 AIRS V4,V5-PR, OMI, SBUV, and GFS AIRS V4,V5-PR, OMI, SBUV, and GFS

V5-FG V5-FG Climatology Climatology V5-PR V5-PR V4-FG V4-FG (Regression) (Regression) V4-PR V4-PR OMI OMI GFS GFS SBUV SBUV

ECMWF Training for ECMWF Training for T(p T(p), ), q(p q(p) regression is good, (because ) regression is good, (because ECMWF uses RAOB ECMWF uses RAOB T(p T(p), ), q(p q(p) in its analysis) but what about ) in its analysis) but what about O3SNDs (?). O3SNDs (?). NCEP-GFS Assimilates SBUV Ozone profiles (better choice ? Or O3SNDS ?)

20

Total Ozone Total Ozone – – September 2005 September 2005 AIRS V4,V5-PR, OMI, SBUV, and GFS AIRS V4,V5-PR, OMI, SBUV, and GFS

V5-PR V5-PR V4-PR V4-PR OMI OMI GFS GFS SBUV SBUV V5-FG V5-FG Climatology Climatology V4-FG V4-FG (Regression) (Regression)

Training data for FG Reg. should adequately represent Ozone hole events and other strategic locations

21

Biomass Burning Biomass Burning Related Ozone Changes Seen by AIRS? Related Ozone Changes Seen by AIRS?

February, 2004 February, 2004 SEVERI Derived Map of Active SEVERI Derived Map of Active Fire Detections (AGU, Gareth Fire Detections (AGU, Gareth Roberts) Roberts) AIRS February, 2004 AIRS February, 2004 AIRS March, 2004 AIRS March, 2004

AIRS !!

22

Biomass Burning Biomass Burning Related Ozone Changes Seen by AIRS? Related Ozone Changes Seen by AIRS?

AIRS CO AIRS CO AIRS Total Ozone AIRS Total Ozone November 2004 November 2004 OMI Total Ozone OMI Total Ozone MODIS Fire Counts MODIS Fire Counts November 2004 , November 2004 , Kevin Bowman, JPL Kevin Bowman, JPL OMI Total Ozone OMI Total Ozone November 2004 November 2004

AIRS !!!

23

Total Ozone Difference Maps (%) Total Ozone Difference Maps (%) (AIRS_V5 (AIRS_V5 – – OMI)/OMI; (SBUV-OMI)/OMI; (GFS-OMI)/OMI OMI)/OMI; (SBUV-OMI)/OMI; (GFS-OMI)/OMI AIRS-OMI 11/2005 AIRS-OMI 11/2005 GFS-OMI 11/2005 GFS-OMI 11/2005 AIRS-OMI 10/2005 AIRS-OMI 10/2005 GFS-OMI 10/2005 GFS-OMI 10/2005 SBUV-OMI 10/2005 SBUV-OMI 10/2005 SBUV-OMI 11/2005 SBUV-OMI 11/2005 Characteristics Invisible (or Slightly Visible) to Other Instruments are Characteristics Invisible (or Slightly Visible) to Other Instruments are Visible to AIRS (Hope they are not Artifacts) Visible to AIRS (Hope they are not Artifacts)

24

Total Ozone Difference Map (AIRS_V5 Total Ozone Difference Map (AIRS_V5 – – OMI)/OMI) OMI)/OMI)

AIRS-OMI 10/2005 AIRS-OMI 10/2005

CO-Ozone Correlations CO-Ozone Correlations (by Jennifer (by Jennifer Wei Wei) ) CO-Ozone and Biomass CO-Ozone and Biomass Burning Burning

25

150 200 250 300 350 400 450 500

• 90
• 80
• 70
• 60
• 50
• 40
• 30
• 20
• 10

10 20 30 40 50 60 70 80 90 Latitude Total Ozone from V4, V5, SBUV - February 2005 Feb_SBUV_2005 Feb_GFS_2005 Feb_AIRS_V5_2005 Feb_OMI_2005

Total Ozone February 2005 Total Ozone February 2005 AIRS-V5 (PR), AIRS-V5 (PR), OMI OMI, , SBUV SBUV, and GFS , and GFS

February February 2.5 2.5o

• Bins

Bins

AIRS V5 Higher than OMI in HighLAT Lower than OMI in the Tropics SBUV and OMI Match Well

We analyzed 2 years of zonal averages to make conclusion about OMI and AIRS Let me know if you wish to see off-line.

26

Annual Cycle 2005 AIRS-V5 Annual Cycle 2005 AIRS-V5

Months 1-12

27

AIRS V4, V5 (PR and FG) OMI, SBUV 2005 AIRS V4, V5 (PR and FG) OMI, SBUV 2005 Annual Cycle Annual Cycle

AIRS-V5 RET AIRS-V5 RET SBUV 2005 SBUV 2005 AIRS-V5 FG AIRS-V5 FG OMI 2005 OMI 2005 AIRS-V4 FG AIRS-V4 FG AIRS-V4 RET AIRS-V4 RET

SBUV Appended with GFS Beyond 80 Degrees.

AIRS Can See Day/Night

Sun-Lit Portions Only

28

Total Ozone Annual Cycle for 2 Years (2004-2005) Total Ozone Annual Cycle for 2 Years (2004-2005)

AIRS V4,V5-PR, OMI, SBUV, and GFS AIRS V4,V5-PR, OMI, SBUV, and GFS

V5-PR V5-PR V4-PR V4-PR OMI OMI GFS GFS SBUV SBUV Months 1-24 Months 1-24 Months 1-24 Months 1-24 Months 1-24 Months 1-24 Months 1-24 Months 1-24 Annual Cycle for 2 Years (2004-2005) Annual Cycle for 2 Years (2004-2005) with 2.5 Degree Zonal Averages with 2.5 Degree Zonal Averages V4-FG V4-FG (Regression) (Regression) V5-FG V5-FG Climatology Climatology

29

Q:

Q: Whether the

Whether the AIRS Instrument Measures Ozone AIRS Instrument Measures Ozone Profiles and Total Ozone Reasonably ? Profiles and Total Ozone Reasonably ? YES. YES. V5 Algorithm V5 Algorithm

 

Retrieval is Solely Due to Channel Radiances Retrieval is Solely Due to Channel Radiances

 

Ability to Capture Trends and Seasonal Patterns Ability to Capture Trends and Seasonal Patterns

 

Reasons for Deficiencies are Understandable and Reasons for Deficiencies are Understandable and Remedies can be implemented Remedies can be implemented. .

– – Improvements Improvements

 

Algorithm Optimization Algorithm Optimization

 

Playing with Damping Factor and Channel Selection Playing with Damping Factor and Channel Selection

 

Further Further Emissivity Emissivity Upgrades are Welcome Upgrades are Welcome

 

Consistent Positive Bias for High Latitude Regions Consistent Positive Bias for High Latitude Regions – – Possibly Due to Static Radiance Bias Corrections Possibly Due to Static Radiance Bias Corrections (Tuning Coefficients) (Tuning Coefficients)

V4 Algorithm V4 Algorithm

 

Good for its time - Good for its time - T(p),q(p T(p),q(p) REG with ECMWF is ) REG with ECMWF is good, but not for O3(p) good, but not for O3(p)

 

First Guess Regression Improvements - Proper First Guess Regression Improvements - Proper Training (O3SNDs ?) (If we Wish to Continue) Training (O3SNDs ?) (If we Wish to Continue)

30

Outlook into Near-Future Outlook into Near-Future

• Development of a IASI /

Development of a IASI / T(p T(p), ), q(p q(p), O ), O3

3(p) Validation

(p) Validation System (IASI System (IASI vs. vs. RAOBs RAOBs; AIRS ; AIRS vs. vs. RAOBs RAOBs) )

– – AIRS First Guess with RAOB/O AIRS First Guess with RAOB/O3

3SND Regression for

SND Regression for T(p T(p), ), q(p q(p) and O ) and O3

3(p) (Attempted Earlier with Success)

(p) (Attempted Earlier with Success) – – Metrics for Validation Metrics for Validation

• Synergetic Use of A-Train Products (Aqua, Aura,

Synergetic Use of A-Train Products (Aqua, Aura, NOAA-18 (With reference to Ozone) NOAA-18 (With reference to Ozone)

» » Validation and Inter-comparison with a Same Set of Qualified Validation and Inter-comparison with a Same Set of Qualified Measurements (O3SNDs) for Relative Performance Assessment Measurements (O3SNDs) for Relative Performance Assessment » » Generation of New Products Generation of New Products – – Visualize Integrated Observing System Products Visualize Integrated Observing System Products

 

Generate Merged Products - AIRS Retrievals, Aura OMI; Generate Merged Products - AIRS Retrievals, Aura OMI; NOAA-SBUV/2 NOAA-SBUV/2

31

Thank You All for your Patience Thank You All for your Patience The NOAA AIRS Team at The The NOAA AIRS Team at The

Integrated Observing System Science Integrated Observing System Science And Product Development Team (IOSSPDT) And Product Development Team (IOSSPDT)

NOAA/NESDIS Camp Spring, MD, USA NOAA/NESDIS Camp Spring, MD, USA

The contents are solely the opinions of the authors and do not The contents are solely the opinions of the authors and do not constitute a statement of policy, decision, or position on constitute a statement of policy, decision, or position on behalf of the NOAA,NASA, or the U.S. Government. behalf of the NOAA,NASA, or the U.S. Government.

Backup Slides Backup Slides

32

AIRS Ozone Retrieval ? AIRS Ozone Retrieval ? Reliable Reliable Improvements/Remedies Possible ? Improvements/Remedies Possible ? YES YES

33

O3WGT Version 5.0.7 (set 48_49.xls) O3WGT Version 5.0.7 (set 48_49.xls)

RMS Difference RMS Difference Bias Bias V5.0.7 O3WGT=1.0 is Chosen V5.0.7 O3WGT=1.0 is Chosen

1000 mb 10 mb 100 mb

0 20 40 60 80 100 120 140 160

• 60 -40 -20 0 20 40 60 80 100 120 140

10 100 1000 20 40 Ozone RMS Difference (%) Pressure (hPa) PR_V5P0P7_D0.02 PR_V5.0.7_D0.40 PR_V5.0.7_D0.75 PR_V5.0.7_D1.00 PR_V5P0P7_D1.25 PR_V5.0.7_D1.50 PR_V5.0.7_D1.75 PR_V5.0.7_D2.00 PR_V5.0.7_NEWCLM 10 100 1000

• 20

20 Ozone Bias (%) Pressure (hPa) PR_V5P0P7_D0.02 PR_V5.0.7_D0.40 PR_V5.0.7_D0.75 PR_V5.0.7_D1.00 PR_V5P0P7_D1.25 PR_V5.0.7_D1.50 PR_V5.0.7_D1.75 PR_V5.0.7_D2.00 PR_V5.0.7_NEWCLM

34

AIRS V5 AIRS V5 and and V4 Retrieved Layer Ozone V4 Retrieved Layer Ozone Amounts Amounts vs.

• vs. O

O3

3SND Layer Ozone Amounts

SND Layer Ozone Amounts

20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 O3SND Layer Ozone Amount (DU) V4_PR Layer Ozone Amount (DU)

V4_PR_L07 V4_PR_L06 V4_PR_L05 V4_PR_L04 V4_PR_L03 V4_PR_L02 V4_PR_L01

(a) 20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 O3SND Layer Ozone Amount (DU) V5_PR Layer Ozone Amount (DU)

V5_PR_L07 V5_PR_L06 V5_PR_L05 V5_PR_L04 V5_PR_L03 V5_PR_L02 V5_PR_L01

(b)

V4-PR V4-PR V5-PR V5-PR

Layer 1

V5 Climatology First Guess V5 Climatology First Guess Helps Physical Helps Physical Retrival Retrival for the for the Lowest Layer 1 (1000-260 Lowest Layer 1 (1000-260 mb mb) )

PR Layer Ozone Amounts PR Layer Ozone Amounts O3SND Layer Ozone Amounts O3SND Layer Ozone Amounts

Difficulty reaching Difficulty reaching extreme values extreme values Slight Slight Overestimation Overestimation

Layer 7 4 8 mb Layer 6 8 16 mb Layer5 16 32 mb Layer 4 32 66 mb Layer 3 66 126 mb Layer 2 126 260 mb Layer 1 260 1100 mb

35

O3WGT for V5.0.7 O3WGT for V5.0.7

O3WGT Factors 0.02 0.40 0.75 1.0 1.25 1.50 1.75 2.0 NONE

• bson M

0.02 0.40 0.75 1.00 1.25 1.5 1.75 2.00

36

V5 Physical Retrievals - Different Stations V5 Physical Retrievals - Different Stations

ALL (N=353), ALL (N=353), STN 089 (N=53, NH), STN 089 (N=53, NH), STN 101(N=50,SH), STN 101(N=50,SH), Tropics (N=54, ±12HR) Tropics (N=54, ±12HR)

10 100 1000 20 40 60 80 100 120 140 160 Ozone RMS Difference (%) Pressure (hPa) V5.0.7_ALL_3H V4.0_ALL_3H V5.0.7_STN089 V4.0_STN089 V5.0.7_STN101 V4.0_STN101 V5.0.7_TRP V4.0_TRP 10 100 1000

• 60
• 40
• 20

20 40 60 80 100 120 140 Ozone Bias (%) Pressure (hPa) V5.0.7_ALL_3H V4.0_ALL_3H V5.0.7_STN089 V4.0_STN089 V5.0.7_STN101 V4.0_STN101 V4.0_TRP V5.0.7_TRP

RMS Difference RMS Difference Bias Bias

1000 mb 10 mb 100 mb

0 20 40 60 80 100 120 140

• 60 -40 -20 0 20 40 60 80 100

V4 Dotted lines, V5 Solid Lines V4 Dotted lines, V5 Solid Lines ALL_3H ALL_3H STN089 STN089 STN101 STN101 TRP_12H TRP_12H ALL_12H ALL_12H

V5 Climatology First Guess V5 Climatology First Guess Helps Physical Helps Physical Retrival Retrival for for the Lowest Layer 1 (1000- the Lowest Layer 1 (1000- 260 260 mb mb) )