nucaps carbon trace gas product validation status...2010, 2011) – aircore. balloon-borne . in situ...

NUCAPS Carbon Trace Gas Product Validation Status

Nicholas R. Nalli1,2, C. Tan1,2, J. Warner3, M. Divakarla1,2, M. Wilson1,2, T. Zhu1,2, Z. Wei3, K. Pryor2, T. Wang1,2, C. D. Barnet4, et al.

1IMSG, Inc., Rockville, Maryland, USA2NOAA/NESDIS/STAR, College Park, Maryland, USA3UMD/CICS, College Park, Maryland, USA 4STC Inc., Columbia, Maryland, USA

1

2020 NASA Sounder Science Team Virtual MeetingMay 2020

Acknowledgments

• Sounder EDR Validation Dataset collection– Carbon Trace Gases: Monika Kopacz (NOAA/UCAR), Greg Frost (NOAA/ESRL)– NASA Sounder Science Team: T. Pagano, E. Fetzer (NASA/JPL)– Total Carbon Column Observing Network (TCCON) (Debra Wunch et al.), TCCON Data Archive,

hosted by the Carbon Dioxide Information Analysis Center (CDIAC), tccon.onrl.gov– Atmospheric Tomography (ATom) Mission: Kathryn McCain, Colm Sweeney (NOAA/ESRL),

https://doi.org/10.3334/ORNLDAAC/1581– AirCore: Colm Sweeney, Bianca Baier (NOAA/ESRL)

• The NOAA Joint Polar Satellite System (JPSS-STAR) Office (Lihang Zhou, M. D. Goldberg, et al.) and the NOAA/STAR Satellite Meteorology and Climatology Division (Satya Kalluri).

• Sounder validation effort (past and present): A. Gambacorta, A.K. Sharma, T. Reale, B. Sun, F. Iturbide-Sanchez, E. Maddy, W. W. Wolf (STAR)

May 2020 Nalli et al. – 2020 Spring NSSTM 2

https://doi.org/10.3334/ORNLDAAC/1581

Outline

• JPSS Sounder EDR Cal/Val Recap– JPSS Level 1 Requirements– Validation Hierarchy recap– NUCAPS Algorithm

Overview of Recent Version Upgrades

• Carbon Trace Gas (CO/CH4/CO2) EDR Validation Status– NUCAPS SNPP and NOAA-20– Overview of Trace Gas Validation

Hierarchy– Results versus Independent Truth

Datasets TCCON (ground-based spectrometers) ATom (aircraft in situ)


JPSS SOUNDER EDR CAL/VAL RECAPNUCAPS Validation


JSTAR Cal/Val Program (Zhou, Divakarla, and Liu 2016)

• JSTAR Cal/Val Phases– Pre-Launch– Early Orbit Checkout (EOC)– Intensive Cal/Val (ICV)

Validation of EDRs against multiple correlative datasets

– Long-Term Monitoring (LTM) Routine characterization of all EDR

products and long-term demonstration of performance

• Sounder EDR validation methodology is based upon AIRSand IASI (Nalli et al., 2013, JGR Special Section on SNPP Cal/Val)

• J-1 (NOAA-20) sounder EDR Cal/Val Plan (Dec 2015)– The Cal/Val Plan included for the first

time the validation of carbon trace gas EDRs (CO, CH4 and CO2), but the details had not been completely ironed out at that time.


Validation Methodology Hierarchies

1. Numerical Model (e.g., ECMWF, NCEP/GFS) Global Comparisons– Large, truly global samples acquired from Focus Days– Useful as “transfer standard” (via double-differences), bias tuning and regression– Limitation: Not independent truth data

2. Satellite Sounder EDR (e.g., AIRS, ATOVS, COSMIC) Intercomparisons– Global samples acquired from Focus Days (e.g., AIRS)– Limitation: Similar error characteristics

3. Conventional PTU/O3 Sonde Matchup Assessments– WMO/GTS operational sondes or O3-sonde network (e.g., SHADOZ)– Representation of global zones, long-term monitoring– Large samples after a couple months (e.g., Divakarla et al., 2006; Reale et al. 2012)– Limitations: Skewed distributions; mismatch errors; non-uniform radiosondes,

assimilated into NWP4. Dedicated/Reference PTU/O3 Sonde Matchup Assessments

– Dedicated for the purpose of satellite validation– Reference sondes: CFH, GRUAN corrected RS92/RS41– E.g., ARM sites (e.g., Tobin et al., 2006), AEROSE, CalWater/ACAPEX , BCCSO,

PMRF– Limitation: Small sample sizes, geographic coverage

5. Intensive Field Campaign Dissections– Include dedicated sondes, some not assimilated into NWP models– Include ancillary datasets, ideally funded aircraft campaign(s)– E.g., SNAP, AEROSE, RIVAL, CalWater, JAIVEX, AWEX-G, EAQUATE

1. Numerical Model Global Comparisons– Examples: ECMWF CAMS– Large, truly global samples acquired from Focus Days– Limitation: Not independent truth data

2. Satellite Sounder EDR Intercomparisons– Examples: AIRS, TROPOMI, OCO-2, MLS– Global samples acquired from Focus Days (e.g., AIRS)– Limitation: Similar error characteristics

3. Surface-Based Network Matchup Assessments– Total Carbon Column Observing Network (TCCON) spectrometers (Wunch

et al. 2010, 2011)– AirCore balloon-borne in situ profile observations (Membrive et al. 2017)– Provide routine independent measurements representing global zones

akin to RAOBs– Limitations: Small sample sizes, uncertainties in unit conversions, different

sensitivities to atmospheric layers

4. Intensive Field Campaign In Situ Data Assessments– Include ancillary datasets, ideally funded aircraft campaign(s)– ATom, WE-CAN, ACT-America, FIREX


T/H2O/O3 Profiles(e.g., Nalli et al., JGR Special Section, 2013)

Carbon Trace Gases

JPSS Specification Performance RequirementsCrIS Trace Gas EDR Uncertainty (O3, CO, CO2, CH4)


Source:(L1RD, 2014, pp. 45-49)

Ozo

ne P

rofil

eCa

rbon

Gas

es

NOAA Unique Combined Atmospheric Processing System (NUCAPS) Algorithm

• Operational algorithm– NOAA Enterprise Algorithm for CrIS/IASI/AIRS (AST v5.9;

after Susskind, Barnet & Blaisdell, 2003; Susskind et al., 2011)

– Global non-precipitating conditions– Atmospheric Vertical Temperature and Moisture Profiles

(AVTP, AVMP)– Trace gases: O3, CO, CO2, CH4

• Users– Weather Forecast Offices (AWIPS)

Nowcasting / severe weather Alaska (cold core)

– NOAA/CPC (OLR)– NOAA/ARL (IR ozone, trace gases)– NOAA TOAST product (IR ozone EDR)– Basic and applied science research (e.g., Pagano et al.,

2014) Stability Indices (e.g., Bloch et al. 2019; Iturbide-Sanchez et al. 2018)

May 2020 Nalli et al. – 2020 Spring NSSTM 8https://www.star.nesdis.noaa.gov/jpss/EDRs/products_Soundings_2018.phphttp://www.ospo.noaa.gov/Products/atmosphere/soundings/nucaps/index.html

AVTP (500 hPa) AVMP (500 hPa)

Ozone (50 hPa) Carbon Monoxide (500 hPa)

Carbon Dioxide (500 hPa)Methane (300 hPa)

https://www.star.nesdis.noaa.gov/jpss/EDRs/products_Soundings_2018.phphttp://www.ospo.noaa.gov/Products/atmosphere/soundings/nucaps/index.html

SNPP/NOAA-20 CARBON TRACE GASNUCAPS Validation


Overview of Carbon Trace Gas Validation

• Carbon trace gas EDR validation versus JPSS program established uncertainty specifications were a relatively new sounder validation requirement that began during the transition period to the FSR CrIS NUCAPS

• Validation strategy leverages global truth datasets, including– Satellite EDRs from Global Focus Days (Cal/Val Method #2)

Valuable for inter-satellite stability Aqua AIRS; TROPOMI; potential future work: OCO-2, MLS (not shown here)

– Total Carbon Column Observing Network (TCCON) (Wunch et al. 2011) (Cal/Val Method #3) Global network of ground-based FTS that accurately measure total column abundances of CO2,

CO, CH4, N2O trace gases Provides “spot checks” for verifying NUCAPS and AIRS

– Atmospheric Tomography (ATom) campaigns (Cal/Val Method #4)– AirCore (Cal/Val Method #3; not shown here)

• Collocation Methodology– Include all FOR within threshold radius

(e.g., 125 km) and time window (e.g., ±2 hours)

• Trace Gas Quality assurance (QA)– NUCAPS IR+MW quality flag– NUCAPS trace gas QA flags

CO and CH4 trace gas flags developed and tested CO2 trace gas flags still undergoing development


TCCON Stations and Averaging Kernels (AKs)

Collocated Focus Day StationsNOAA-20 Trace Gas QA, ±2 hr, 125 km

Column AKs

• Global network of ground-based FTS that accurately measure total column abundances of CO2, CO, CH4, N2O trace gases

• Serves as SNPP ↔ NOAA-20 transfer standard, complementing the ATom analysis

Total Carbon Column Observing NetworkTCCON (Wunch et al. 2011)


NUCAPS v2.7.2 Total Column Trace Gases vs TCCON: Statistical Analysis

NOAA-20 SNPPBox-Whisker Statistics by TCCON Station

(Sorted in order of latitude)

NUCAPS − TCCON


NOAA-20 NUCAPS v2.7.2 Total Column Trace Gases vs TCCON

Outlier cases expected to be removed when new QA flags become available

Histograms ScatterplotsStatistical Analysis Focus Days: 20180401, 20180615, 20180820, 20181015, 20181215, 20190215

NUCAPS − TCCON


SNPP NUCAPS v2.7.2 Total Column Trace Gases vs TCCON

Histograms ScatterplotsStatistical Analysis Focus Days: 20180401, 20180615, 20180820, 20181015, 20181215, 20190215

NUCAPS − TCCON


NUCAPS v2.7.2 Total Column Trace Gases vs TCCON: Statistical Analysis

Focus Day Time Series Spanning Annual Cycle20180401, 20180615, 20180820, 20181015, 20181215, 20190215NOAA-20 SNPP

NUCAPS − TCCONMay 2020 Nalli et al. – 2020 Spring NSSTM 15

• ATom deployed extensive gas and aerosol payloads on the NASA DC-8 aircraft for global-scale sampling of the atmosphere, profiling continuously from 0.2–12 km altitude

• Flights occurred spanning 4 seasons originating from the Armstrong Flight Research Center, Palmdale, California

• We use the NOAA Picarro ATom-1, -2 and -4datasets (2016-2018)

– We have fully implemented the ATom VALAR dataset for NOAA-20, based upon 2018 ATom-4

– Additionally, we have enhanced the SNPP ATom VALAR dataset Updated the ATom-1,-2 truth profiles with the latest release

versions Included the ATom-4 along with the ATom-1,-2 data, creating a

larger collocation ensemble

Atmospheric Tomography (ATom) Mission (Wofsy et al. 2018)

NOAA Picarro PIs: Kathryn McKain and Colm Sweeney (CIRES, U. of Colorado, NOAA/ESRL)

Source: https://espo.nasa.gov/atom/


https://espo.nasa.gov/atom/

• SNPP and NOAA-20 retrievals are comparable

• T/H2O/O3 favorably compare with ATom truth, are comparable with OPS and have not suffered any degradation

• SNPP OPS and v2.7.2 CO retrievals are comparable to Oct 2019 results and have not suffered degradation

• NUCAPS CO, CH4 and CO2 trace gas retrievals improve the a priori in the layers of sensitivity

• CH4 retrievals are slightly improved over OPS (offline v2.5.2.2)

• CO2 retrievals and a priori are significantly improved over OPS

• CO and CO2 retrievals are “near-perfect” after applying AKs to the ATom data

• With the exception of CH4, which does not meet the overly-stringent precision requirement of 1%, NUCAPS trace gasretrievals using averaging kernels (AKs) meet JPSS L1 Requirements

NUCAPS v2.7.2 ATom Trace Gas Statistical Summary


ATom-NUCAPS v2.7.2 Extended Sample Collocations (ATom-1,-2,-4)

SNPP Extracted ATom Profile Collocations

NOAA-20 Extracted ATom ProfileCollocations


NUCAPS v2.7.2 Retrieval vs ATom w AKs: 100 Layer Profile Statistical Analysis

SNPP NOAA-20Temperature / H2O / O3

NUCAPS − ATom


NUCAPS v2.7.2 A Priori vs ATom: 100 Layer Profile Statistical Analysis

SNPP NOAA-20CO / CH4 / CO2

NUCAPS − ATom


NUCAPS v2.7.2 Retrieval vs ATom: 100 Layer Profile Statistical Analysis


NUCAPS − ATom


NUCAPS OPS (v2.5.2.2) Retrieval vs ATom: 100 Layer Profile Statistical Analysis


NUCAPS − ATom


NUCAPS v2.7.2 Retrieval vs ATom: 100 Layer Profile Statistical Analysis


NUCAPS − ATom


NUCAPS v2.7.2 Retrieval vs ATom w AKs: 100 Layer Profile Statistical Analysis


NUCAPS − ATom


NUCAPS OPS A Priori & Retrieval vs ATom: Total Column Statistical Analysis


NUCAPS − ATom


NUCAPS v2.7.2 A Priori & Retrieval vs ATom: Total Column Statistical Analysis


NUCAPS − ATom


Summary and Future Work

• Maturity Review for SNPP/NOAA-20 NUCAPS EDRs was held on 23 April 2020– CO and O3 Validated Maturity (Oct 2019)– CH4 Validated Maturity– CO2 Provisional Maturity– Recent NUCAPS upgrades have focused

on upgrades/optimizations of the CH4 and CO2 trace gas EDRs cf. Warner et al. presentation this session

– Ongoing Validation versus ATom, TCCONand satellite truth datasets are showing the products are now generally meet requirements (with the exception of the CH4 1% precision requirement)

– https://www.star.nesdis.noaa.gov/jpss/documents/AMM/Recommendations/Review_Team_Recommendations_for_20200423_Maturity_Review.pdf

• Future Work– Ongoing NUCAPS development, Cal/Val and

Long-Term Monitoring Continue algorithm optimizations (e.g., MW-only) Further upgrades/optimizations for CH4 and CO2

products (Warner et al.) Continued validation versus other datasets (e.g.,

AirCore) Publication of results in MDPI Remote Sensing Special

Issue

– Other Related Work Surface emissivity upgrades/updates

o IR sea surface emissivity (IRSSE) model upgrades (cf. Nalli et al. talk Tuesday next week)

o CAMEL database a priori Continued support of EDR applications (COVID-19,

AWIPS, AR/SAL, atmospheric chemistry users)


https://www.star.nesdis.noaa.gov/jpss/documents/AMM/Recommendations/Review_Team_Recommendations_for_20200423_Maturity_Review.pdf

This Special Issue will focus on the calibration / validation (cal/val) of advanced passive sensors (IR and/or MW) essential for Earth (atmospheric/oceanic) observation onboard both operational and experimental environmental satellites.

We invite papers in the areas of sensor (SDR) calibration, algorithm/retrieval (EDR) validation (including intensive campaigns), and sensitivity/impact on derived product (e.g., EDR) applications.

Special Issue

Satellite Data Application, Validation and Calibration for Atmospheric ObservationSpecial Issue Guest Editors:Dr. Nicholas R. NalliIMSG, Inc. at NOAA/NESDIS/STARMs. Lori A. BorgUniversity of Wisconsin-Madison CIMSSDr. Quanhua LiuNOAA/NESDIS/STAR

Submission Deadline: 31 August 2020.

IMPACT FACTOR

4.118


THANK YOU! QUESTIONS?NUCAPS Validation


BACKUP SLIDESNUCAPS Validation


NUCAPS Carbon Trace Gas Product Validation StatusAcknowledgmentsOutlineJPSS SOUNDER EDR CAL/VAL RECAPJSTAR Cal/Val Program (Zhou, Divakarla, and Liu 2016)Validation Methodology HierarchiesJPSS Specification Performance Requirements�CrIS Trace Gas EDR Uncertainty (O3, CO, CO2, CH4)NOAA Unique Combined Atmospheric Processing System (NUCAPS) AlgorithmSNPP/NOAA-20 Carbon Trace GasOverview of Carbon Trace Gas ValidationTCCON Stations and Averaging Kernels (AKs)NUCAPS v2.7.2 Total Column Trace Gases vs TCCON: Statistical AnalysisNOAA-20 NUCAPS v2.7.2 Total Column Trace Gases vs TCCONSNPP NUCAPS v2.7.2 Total Column Trace Gases vs TCCONNUCAPS v2.7.2 Total Column Trace Gases vs TCCON: Statistical Analysis Atmospheric Tomography (ATom) Mission (Wofsy et al. 2018)NUCAPS v2.7.2 ATom Trace Gas Statistical SummaryATom-NUCAPS v2.7.2 Extended Sample Collocations (ATom-1,-2,-4)NUCAPS v2.7.2 Retrieval vs ATom w AKs: 100 Layer Profile Statistical AnalysisNUCAPS v2.7.2 A Priori vs ATom: 100 Layer Profile Statistical AnalysisNUCAPS v2.7.2 Retrieval vs ATom: 100 Layer Profile Statistical AnalysisNUCAPS OPS (v2.5.2.2) Retrieval vs ATom: 100 Layer Profile Statistical AnalysisNUCAPS v2.7.2 Retrieval vs ATom: 100 Layer Profile Statistical AnalysisNUCAPS v2.7.2 Retrieval vs ATom w AKs: 100 Layer Profile Statistical AnalysisNUCAPS OPS A Priori & Retrieval vs ATom: Total Column Statistical AnalysisNUCAPS v2.7.2 A Priori & Retrieval vs ATom: Total Column Statistical AnalysisSummary and Future WorkSlide Number 28Thank you! Questions?Backup slides

nucaps carbon trace gas product validation status...2010, 2011) – aircore. balloon-borne . in situ...

Documents