1. Overview of Global Space-based Inter-Calibration System (GSICS)
Mitch Goldberg, NOAA/NESDIS
GSICS Executive Panel chair
2nd GSICS User Workshop

2. ECMWF Meeting, Reading UK
June 22, 2001
What is GSICS?
Global Space-based Inter-Calibration System (GSICS)
Goal - Enhance calibration and validation of satellite observations and to intercalibrate critical components global observing system
Part of WMO Space Programme
GSICS Implementation Plan and Program formally endorsed at CGMS 34 (11/06)
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt

3. ECMWF Meeting, Reading UK
June 22, 2001
GSICS Mission
To provide sustained calibration and validation of satellite observations
To intercalibrate critical components of the global observing system to climate quality benchmark observations and/or reference sites
To provide correction coefficients and algorithms to the user community for current and historical data 3
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt

4. ECMWF Meeting, Reading UK
June 22, 2001
Or in technical terms:
Quantify the differences – magnitude and uncertainty
Correct the differences – physical basis and empirical removal
Diagnose the differences – root cause analysis
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt

5. Motivation
Demanding applications require well calibrated and intercalibrated measurements
Climate Data Records
Radiance Assimilation in Numerical Weather Prediction
Data Fusion
Growing Global Observing System (GOS) 5 5

6. Calibration is Critical for Climate Change Detection
Before intercalibraion
After intercalibration
Trend of global oceanic total precipitable water decreases from 0.54 mm/decade to 0.34 mm/decade after intercalibrations! Calibration uncertainties translate to uncertainties in climate change detection 6 6 6

7. Satellite Intercalibration SSM/I Observations (Yang et al. , J. Appl
Satellite Intercalibration SSM/I Observations (Yang et al., J. Appl. Meteor. and Climatology)

8. SSMIS calibration biases cause regional weather patterns
Do we Care about Satellite Biases in NWP? After McNally, Bell, et al. ECMWF, 2005 & 2009
Yes! Because:
1) We wish to understand the origin of the bias and ideally correct instrument / RT / NWP model at source
2) In principle we do not wish to apply a correction to unbiased satellite data if it is the NWP model which is biased. Doing so is likely to:
Re-enforce the model bias and degrade the analysis fit to other observations
Produce a biased analysis (bad for re-analysis / climate applications)
The biases for all satellite channels used in NWP data assimilation are closely monitored at NWP centers. For example, the NCEP system is available at:
SSMIS calibration biases cause regional weather patterns
More accurate satellite observations will facilitate discovery of model errors and their correction. Additional gains in forecast accuracy can be expected. 8 8

9. Difference between JRA25 vs ECMWF and JRA25 vs ERA Interim
(column water vapor above 500 mb)
Larger differences with ERAIM ~13 % vs ~1 % with ECMWF WX analysis
ERAIM - JRA25
ECMWF - JRA25
ECMWF Weather analysis and Reanalysis is different because
the ECMWF bias tuning changed in 2006.

10. Problems associated with Radiance Applications
Multiple error sources:
Observations (biases)
Spectral Response Functions
Radiative transfer
Inversion Technique/Data Assimilation
Model Physics

11. Building Blocks for Satellite Intercalibration
Collocation
Determination and distribution of locations for simultaneous observations by different sensors (space-based and in-situ)
Collocation with benchmark measurements
Data collection
Archive, metadata - easily accessible
Coordinated operational data analyses
Processing centers for assembling collocated data
Expert teams
Assessments
communication including recommendations
Vicarious coefficient updates for “drifting” sensors

12. Other key building blocks for accurate measurements and intercalibration
Extensive pre-launch characterization of all instruments traceable to SI standards
Benchmark instruments in space with appropriate accuracy, spectral coverage and resolution to act as a standard for inter-calibration
Independent observations (calibration/validation sites – ground based, aircraft)

13. GSICS organization Organizations contributing to GSICS:
CMA, CNES, EUMETSAT, ISRO, JAXA, JMA, KMA, NASA, NIST, NOAA, WMO, ESA
Overseen by GSICS Executive Panel
Assisted by Research Working Group
and Data management Working Group
GSICS activities rely on:
GSICS Coordination Centre (GCC)
operated by NOAA/NESDIS
Processing & Research Centres (GPRC)
operated by each satellite operator
Calibration Support Segments (CSS)
including field sites and laboratories
GSICS as an element of the space-based component of the Global Observing System
GSICS activities rely on a GSICS Coordination Centre (GCC) operated by NOAA/NESDIS, several GSICS Processing and Research Centres (GPRC) operated by each satellite operator, and will benefit of Calibration Support Segments including field sites and laboratories.
Activities are overseen by a GSICS Executive Panel assisted by a Research Working Group and a Data management Working Group.
Coordination Center
Calibration Support Segments (reference sites, benchmark measurements, aircraft, model simulations)
Regional Processing Research Centers at Operational Space
Agencies

14. ECMWF Meeting, Reading UK
June 22, 2001
Current focus of GSICS
Interagency collaboration on algorithms (GRWG) and data exchange and formats (GDWG)
Product acceptance and documentation requirements, metadata standards, data formats, website standards
Routine intercalibration (monitor and correct) of all operational GEO Infrared imagers using IASI and AIRS
MODIS and Deep Convective Clouds for visible channels 14 14
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt 14

15. ECMWF Meeting, Reading UK
June 22, 2001
Current focus of GSICS
Intercalibration of LEO instruments
HIRS, SSMI, AMSU, MHS, AVHRR, AIRS, IASI, FY3,
GOME-2, OMI, SBUV
Traceability
Campaigns
Key collocation datasets
Requirements for pre-launch calibration
Root causes and corrections 15 15
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt 15

16. First GSICS Guideline document
Best Practice Guidelines for Pre-Launch Characterization and Calibration of Instruments for Passive Optical Remote Sensing
Report to GSICS Executive Panel
R.U. Datla, J.P. Rice, K. Lykke and B.C. Johnson (NIST)
J.J. Butler and X. Xiong (NASA)
September 2009

17. ECMWF Meeting, Reading UK
First international coordinated GSICS project is the intercalibration
of geostationary infrared channels with IASI and AIRS
June 22, 2001
Ch6
Ch4
Ch3
Ch2
IASI
AIRS
Web Accessible 17 17
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt 17

18. ECMWF Meeting, Reading UK
GSICS Correction Algorithm for Geostationary Infrared Imagers
GSICS will provide correction coefficients for all GEOs
from 2003 (beginning of AIRS record) to present
ECMWF Meeting, Reading UK
June 22, 2001
The first major deliverable to the user community is the GSICS correction algorithm for geostationary satellites.
The user applies the correction to the original data using GSICS provided software and coefficients.
The correction adjusts the GOES data to be consistent with IASI and AIRS.
The figures to the left show the difference between observed and calculated brightness temperatures (from NCEP analysis) before and after correction
The bias is reduced from 3 K to nearly zero.
Before 3K Bias
After: ~ 0K Bias 18
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt

19. ECMWF Meeting, Reading UK
June 22, 2001
CNES SADE Data Base is critical for assessing stability of visible/near infrared reference instruments for intercalibration
Time series of the ratio of the ESA MERIS to
NASA MODIS micron visible channel
reflectance from observations at 19 desert sites
in North Africa and Saudi Arabia.
The results show very good agreement and
stability between the two sensors
19 sites selected over North Africa and Arabia
GSICS – Feb 2008 – Claire Tinel / CNES 19
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt 19

20. Calibration Opportunity Prediction Data Acquisition Scheduler
Integrated Cal/Val System Architecture
Calibration Opportunity Prediction
Data Acquisition Scheduler
Calibration Opportunity Register
(CORE)
Raw Data Acquisition for Calibration Analyses
Stored Raw Data for Calibration
Analyses
SNO/
SCO Rad. Bias and Spectral Analysis
Calibration Parameter Noise/
Stability Monitoring
RTM Model Rad. at Calibration Reference Sites
Inter-sensor Bias and Spectral Analysis
Earth & Lunar Calibration
Geolocation Assessment
(Coastlines, etc.)
Assessment Reports and Calibration Updates

21. Promoting access to instrument performance monitoring is also part of GSICS
GSICS beginning to develop observing system monitoring tools for all GSICS partners. Above is example of NOAA tool for NOAA-19 OV and long-term monitoring
The system has detected instrument anomalies, provided an important tool for diagnoses, data quality assurance, and for short and long term applications
To better support the operations, we have developed comprehensive on-orbit verification for newly launched satellites and an integrated cal/val system.
Courtesy of F.Weng 21

22. Engage User Community (First GSICS Users’ Workshop – September 2009)
Satellite Community – generation of CDRs
New WMO Space Programme SCOPE-CM
ISCCP - Blended geo & leo cloud information
Agency CDR programs - NOAA, EUMETSAT
Satellite Community - NWP direct radiance assimilation
Reanalysis Community
Next ECMWF reanalysis –
GSICS first major deliverable: Correction Coefficients for all GEO IR using IASI/AIRS as reference from 2003 – 2010+
Satellite Acquisition Programs
Prelaunch instrument characterization guidelines
Cal/Val Plans
Request for new products:
GEO solar channels, LEO microwave instruments

23. MSU/AMSU FCDR Coordination funded by NOAA CDR Program and NESDIS STAR
Develop consistent radiance Fundamental Climate Data Record (FCDR) to support consistent modeling reanalysis and consistent satellite retrievals
Develop consistent atmospheric temperature thematic climate data record (TCDR) for climate service support  climate change research, climate change monitoring, validating climate model simulation…
And also, for temperature channels, we don’t have the dynamic range problem. Which means the SNO ranges covers 80% of the global temperature, which is not bad at all compare to the water vapor channels.

24. Before and After Inter-Satellite Calibration AMSU CH6 (Cheng-Zhi Zhou)
NOAA operational calibrated inter-satellite difference time series (Before Inter-satellite calibration); Ocean Mean
NOAA-15 CH6 has strong calibration nonlinearity
Time-dependent calibration coefficients for NOAA-15 are introduced at Level-1c
After SNO recalibration, inter-satellite differences close to zero
Recalibrated trend is expected to be different from NOAA-15
Inter-satellite difference time series after SNO Inter-satellite calibration, AMSU-A CH6
The project is recently funded by the NOAA SDS program. A science team for the CDR development is established.

25. GPS-RO simulated MSU channel 4 vs SNO recalibrated MSU channel 4 products
Scattering plots of 10 x10 degree binned TLS from to
Mean(RSS‐RO_AMSU) =‐0.99
Std (RSS‐RO_AMSU) =1.67
R=0.99
Mean(UAH‐RO_AMSU) =0.02
Std (UAH‐RO_AMSU) =2.06
R=0.99
Mean(NOAA‐RO_AMSU) =-0.49
Std (NOAA‐RO_AMSU) =0.5
R=0.99
From Ben Ho of NCAR, NOAA CDR Workshop, Silver Spring, Maryland
March 22-24

26. Bias Correction of Recalibrated MSU data in NCEP CFSR Reanalysis
Recalibrated MSU level-1c data were assimilated into NCEP CFSR and NASA MERRA reanalysis systems
Bias correction pattern for recalibrated MSU data are much smoother, since instrument errors were removed before assimilation
I do not want to say the sno calibration solves everything, but it indeed improves the radiance in many aspects. Here is the summary of the MSU issues I listed before and how the sno calibration improves them
Before Recalibration After Recalibration
MSU Channel 2 bias correction patterns in NCEP CFSR reanalysis from Recalibrated MSU data after 1987 were assimilated into CFSR (plot from Saha et al. 2010)

27. GPM affiliation with GSICS
ECMWF Meeting, Reading UK
June 22, 2001
GPM affiliation with GSICS
Research working group
- Consensus algorithms
Data working group
- Formats, Servers
GSICS Executive Panel
GRWG
GDWG
GCC
CSS
GPRC
Calibration Support
Segments
Coordination Center
Regional Processing Research
Centers at Satellite Agencies
GPM X-Cal
Working Group
CSS
GPRC
CSS
GPRC
GPM X-cal:
Plan to sustain X-cal in future
Similar Structure to GSICS
Interested in using GSICS product acceptance procedure
Precipitation Processing
System at Goddard
N:\Briefings\International\ECMWF June \ECMWF Jun_22_2001.ppt

28. GSICS Interface to SCOPE-CM
Sustained & Coordinated Processing of Environmental satellite data for Climate Monitoring (SCOPE-CM)
WMO Initiative based on activities of existing initiatives (GOS, GCOS and GSICS)
Provides international cooperation and coordination framework
To generate multi-mission and global satellite climate data records (FCDRs & TCDRs)
Serve users and other organisations (e.g. WMO Regional Climate Centres RCC, National Weather Services)
=> The way toward operational production of high quality ECVs on a global scale
Aside – as an example of how GSICS can fit into the scheme of things.

29. GSICS Outcome
Coordinated international intersatellite calibration program
Exchange of critical datasets for cal/val
Best practices/requirements for monitoring observing system performance
Best practices/requirements for prelaunch characterisation (with CEOS WGCV)
Establish requirements for cal/val (with CEOS WGCV)
Advocate for benchmark systems
Quarterly reports of observing system performance and recommended solutions
Improved sensor characterisation
High quality radiances for NWP & Climate