GOES-R AWG 2nd Validation Workshop 9-10 January 2014, College Park, MD
GOES-R and JPSS SST Monitoring System
Sasha Ignatov, Prasanjit Dash, Xingming Liang,Feng Xu, Yury Kihai, Xinjia Zhou
With support fromJohn Sapper, John Stroup, Boris Petrenko, Marouan Boauli
9 January 2014 GOES-R SST Validation 1
SQUAM - SST Quality Monitor www.star.nesdis.noaa.gov/sod/sst/squam/ Monitor SST Products (L2, L3, L4) for Self- and Cross-
Consistency; Validate against in situ SSTs (iQuam)
iQuam - In situ Quality Monitor www.star.nesdis.noaa.gov/sod/sst/iquam/ QC in situ SSTs, Monitor on Web, Distribute to users Input to SQUAM
MICROS - Monitoring IR Clear-sky Radiances over Oceans for SST www.star.nesdis.noaa.gov/sod/sst/micros/ Monitor Clear-sky ocean radiances for Self- and Cross-
Consistency; Validate against CRTM simulations
NOAA JPSS and GOES-R Cal/Val Tools
9 January 2014 2GOES-R SST Validation
SST Community Resources, Support JPSS and GOES-R
3
Principles of NOAA SST Monitoring
9 January 2014 GOES-R SST Validation
Monitor automatically, in Near-Real Time, Globally, Online- Satellite SSTs against reference SST- Satellite sensor radiances associated with SST- Reference SSTs (in situ, L4)
Monitor GOES-R and JPSS SST products in context of other community SST products
Monitor deviations from a reference state (expected to be centered at zero, and have a narrow Gaussian distribution)
Supplement heritage VAL against in situ SST (sparse, geographically biased, suboptimal quality, not available in NRT) with global consistency checks against another reference, L4 analysis fields (global coverage, large statistics, more uniform data quality, available in NRT)
SST Monitoringin SST Quality Monitor (SQUAM)
www.star.nesdis.noaa.gov/sod/sst/squam/
9 January 2014 4GOES-R SST Validation
Das, et al: SST Quality Monitor. JTECH, 2010.
DAY: VIIRS L2 minus OSTIA L428 December 2013
• Deviation from Reference SST is flat & close to 0• Residual Cloud/Aerosol leakages seen as cold spots• Warm spots show areas with diurnal warming during daytime
9 January 2014 GOES-R SST Validation 5
DAY: VIIRS L2 minus in situ SST28 December 2013
• Much fewer match-ups with in situ data than with L4 • Validation statistics may not be globally representative• Daily validation statistics may not be accurate and stable
9 January 2014 GOES-R SST Validation 6
DAY: VIIRS L2 minus OSTIA L428 December 2013
• Shape close to Gaussian but skewed positively due to diurnal warming• Domain & Performance Stats close to expected
9 January 2014 GOES-R SST Validation 7
DAY: VIIRS L2 minus in situ SST28 December 2013
• Shape close to Gaussian • Domain & Performance Stats better than spec
9 January 2014 GOES-R SST Validation 8
DAY 28 December 2013 – Summary
NOBS Min/ Max Mean/ STD Med/ RSD Skew/ Kurt
117,600,000 -8.7/ +8.6 +0.29/0.60 +0.22/0.45 +1.5/ +7.9
ΔT = “VIIRS minus OSTIA” SST
9 January 2014
NOBS Min/ Max Mean/ STD Med/ RSD Skew/ Kurt
1,909 -1.5/ +2.5 +0.06/0.39 +0.09/0.32 -0.1/ +2.8
ΔT = “VIIRS minus in situ” SST
9GOES-R SST Validation
DAY STD DEV wrt. In situ SST
Outliers come from satellite and in situ data
IDPS shows larger STDalthough improved recently
9 January 2014 GOES-R SST Validation 10
• Daily val against in situ SST is noisy due to small match-up sample & lack of global coverage• IDPS SST improved recently but has been out of family for the full 2 years of monitoring
DAY STD DEV wrt. Reynolds L4
9 January 2014 GOES-R SST Validation 11
IDPS shows much larger STD
• IDPS SST has been out of family but improved in mid 2012. Still significantly deviates from family, but less than earlier in the mission
DAY STD DEV wrt. OSTIA L4
9 January 2014 GOES-R SST Validation 12
IDPS shows much larger STD
• Comparison with OSTIA SST suggest the same qualitative observations as against Reynolds and in situ, although quantitatively, comparison statistics may differ
VIIRS, MODIS, and AVHRRRadiance Monitoring in MICROS
www.star.nesdis.noaa.gov/sod/sst/micros/
9 January 2014 13
Liang, Ignatov: Monitoring IR Clear-Sky Radiances over Oceans for SST. JTECH, 2011.
GOES-R SST Validation
Model minus Observation (“M-O”) Biases- M (Model) = Community Radiative Transfer Model (CRTM) simulated
TOA Brightness Temperatures (w/ Reynolds SST, GFS profiles as input)- O (Observation) = Clear-Sky sensor (AVHRR, MODIS, VIIRS) BTs
Double Differences (“DD”) for Cross-Platform Consistency
- “M” used as a “Transfer Standard”- DDs cancel out/minimize effect of systematic errors & instabilities in BTs
arising from e.g.- Errors/Instabilities in Reynolds SST & GFS- Missing aerosol- Possible systemic biases in CRTM- Updates to ACSPO algorithm
M-O Biases and Double Differences (“DD”)
)]([)]([ OMREFOMSATREFSAT
9 January 2014 14GOES-R SST Validation
• All AVHRRs (except N16 and Metop-B), MODISs, and VIIRS are now consistent to within ±0.1K• VIIRS Cal change 7 Mar 2012 reset BT@M15 by +0.14K – now better in family• Before 13 Sep 2012: Terra & Aqua/MODIS were biased by -0.6K (suboptimal CRTM coeffs in v2.02)• Both are back in family now, after CRTM V2.1 implemented on 13 Sep 2012• Metop–B out of family by ~+0.3 K (likely due to suboptimal CRTM coefficients in CRTM V2.1)
N16: unstable and out of family
CRTM V2.1 implemented
Metop-B: out of family
Double Differences in IR11 (VIIRS M15)
9 January 2014 15
VIIRS recalibration
GOES-R SST Validation
• All AVHRRs, MODISs and NPP/VIIRS SSTs are consistent to within ±0.1K• As a result of VIIRS Cal Change on 7 Mar 2012, VIIRS SST went out of family • Was brought back in family when new SST coefficients implemented 3 May 2012• CRTM update resulted regression SSTs more noise, and the new coefficients have been implemented
since Dec. 2012. More data is needed to understand their performance
N16: unstable and out of family
VIIRS recalibration CRTM V2.1 implemented
Double Differences in SST
9 January 2014 16
New Reg. Coeff. used
GOES-R SST Validation
In situ SST QC & Monitoring in iQuam
www.star.nesdis.noaa.gov/sod/sst/iquam/
9 January 2014 17
Xu, Ignatov: In situ SST Quality Monitor. JTECH, 2014.
GOES-R SST Validation
9 January 2014 GOES-R SST Validation 18
In situ Quality Monitor (iQuam) performs the following functions
QC: Accurate/flexible QC of in situ SSTs, consistent with wider Meteorological and Oceanographic communities
Monitoring: Report online statistical summaries of in situ minus reference L4 SST (stratified by ships, drifters, tropical & coastal moored, ARGO floats; and individual platforms)
Data Serving: Serve QCed in situ SST data online for SST community
Objectives of in situ SST Quality Monitor
9 January 2014 GOES-R SST Validation 19
Quality Control – Consistent with UK MO
Category Check Type of error handled Physical basis
Preprocessing Duplicate Removal
Duplicates arise from multiple transmission or data set merging
Identical space/time/ID
Plausibility Plausibility checks
Unreasonable field values Range of single fields &Relationships among them
Internal consistency
Tracking Points falling out of track Travel speed exceeds limit
Spike check Discontinuities in SST time series along track
SST gradient exceeds limit
External consistency
Reference Check
Measurements deviating far away from reference
Bayesian approach (*) (Ref. SST: Daily OI SST v2)
Mutual consistency
Cross-platform Check
Mutual verification with nearby measurements (“buddies check”)
Bayesian approach (*) based on space/time correlation of SST field (Correlation model: 2-scale SOAR, Martin et al., 2002)
(*) Lorenc and Hammon, 1988; Ingleby and Haddleston, 2007
9 January 2014 GOES-R SST Validation 20
http://www.star.nesdis.noaa.gov/sod/sst/iquam/
9 January 2014 GOES-R SST Validation 21
Monthly Statistical Summaries
Outliers detected by each QC check
Moments of ΔTS=Tin situ - TReynolds
Histograms of ΔTS
9 January 2014 GOES-R SST Validation 22
Time Series of Monthly Statistics (1981-pr)
No of Platforms No of Observations
Bias in ΔTS
SD of ΔTS
ships
drifters
moorings
9 January 2014 GOES-R SST Validation 23
Monitoring individual platforms
List of platforms & individual statistics
Error Rate History
Time Series of ΔTS
Monthly Trajectory
9 January 2014 GOES-R SST Validation 24
Data for Download
Last monthly file updated in NRT every 6hrs. Initial QC performed on the fly. Final QC requires ~7 days.
QC’ed data in HDF format available for download (1981-pr)
9 January 2014 25GOES-R SST Validation
Status of NOAA SST Monitoring
Well established for polar products AVHHR (NOAA, NAVO, EUMETSAT) MODIS-Terra, MODIS-Aqua (NOAA) VIIRS (NOAA)
Progress made towards Monitoring Geo products Some elements of SQUAM and MICROS web design in place Prototyped using MSG SEVIRI (NOAA) Not linked to SQUAM and MICROS main pages yet
Future work
9 January 2014 GOES-R SST Validation 26
Polar Products Bring into SQUAM remaining community SST products -
MOD28/MYD28 and (A)ATSR Update and unify functionality
Geo products Complete development of Geo functionality in SQUAM and MICROS Set up match-up processing with iQuam in situ data Link to main SQUAM and MICROS pages
Get ready for GOES-R launch!