Ocean Vector Wind as Essential Climate Variable [email protected]

Ocean Vector Wind as

Essential Climate Variable

[email protected]

mailto:[email protected]

IOVWST, May 2011

Context• WCRP, World Climate Research Program• WOAP, WCRP Observation Assimilation Panel• GCOS, Global Climate

Observing System• Essential Climate Variable

– Ocean Vector Wind• Fundamental Climate

Data Records FCDR inventory

www.wmo.int/pages/prog/sat/Databases.html#UserRequirements

www.wmo.int/pages/prog/gcos/index.php?name=ClimateObservationNeeds

IOVWST, May 2011

Satelite Ocean Winds• NRT ASCAT, QSCAT, OSCAT: NOAA, ISRO, OSI SAF• L2/L3/L4 archive, all scatterometers, PO.DAAC, Ifremer, RSS, FSU,

EUMETSAT, …• WindSat (Bettenhausen, NRL) NRT / archive• Wind speeds from radiometers (RSS, HOAPS) archive (SMMR,

SSMI(S), AMSR, MIS, .. )• Altimeter wind speeds (GeoSat, Topex, GFO, ERS, Jason, ENVISAT,

CryoSat, HY2 (2011), SARAL (2012), Jason3 (2013), S3 (2013))• SAR (off-line) (ERS, RadarSat ($), ENVISAT, S1 (2012), .. )

Variables• Wind vector or speed, stress vector, curl, divergence, see also white

paper OceanObs’09 paper by Bourassa et al.

IOVWST, May 2011

Sustainability• Radiometer missions, wind speed swath

– SSM/I, WindSat (vector > 8 m/s), MIS• Altimeter missions, wind speed track • Historic scatterometer missions, wind vector swath

– SeaSat, 3 months in 1978– ERS1&ERS2 AMI 1991-2000 (ERS2 is regional since)– NSCAT and SeaWinds1, each 9 months– QuikScat, 1999-2009– ASCAT, 2007-– OceanSat-2, 2009-

Continuous wind vector coverage since 1991

IOVWST, May 2011 sw 24feb11Approved

10 11 12 13 14 15 16 17 18 19 20 21 22

Ku-band

Combined C- and Ku-band

0908

C-band

LaunchDate

10/06

6/99

Operating

QuikSCAT USA

GLOBAL SCATTEROMETER MISSIONS (CEOS VC)

FY-3E with 2FS China

Operational Series with 2FS India

EPS SG Europe

ExtendedExtended

Global availability uncertain

?

Availability ?

IOVWST, May 2011

GCOS needs1. Full description of all steps taken in the generation of FCDRs and ECV products, including

algorithms used, specific FCDRs used, and characteristics and outcomes of validation activities 2. Application of appropriate calibration/validation activities 3. Statement of expected accuracy, stability and resolution (time, space) of the product, including,

where possible, a comparison with the GCOS requirements 4. Assessment of long-term stability and homogeneity of the product 5. Information on the scientific review process related to FCDR/product construction (including

algorithm selection), FCDR/product quality and applications 6. Global coverage of FCDRs and products where possible 7. Version management of FCDRs and products, particularly in connection with improved algorithms

and reprocessing 8. Arrangements for access to the FCDRs, products and all documentation 9. Timeliness of data release to the user community to enable monitoring activities 10. Facility for user feedback11. Application of a quantitative maturity index if possible 12. Publication of a summary (a webpage or a peer-reviewed article) documenting point-by-point the

extent to which this guideline has been followed

GCOS-143 (WMO/TD No. 1530)

These are NRT ops. production needs too (planned up to L2/L3)

IOVWST, May 2011

Assessment of ocean ECVs• Geophysical parameter and related ECV• Existing and/or potential users• History and outlook; sustainability• Availability and DOI registration• Maturity (Bates & Barkstrom maturity index or others)• Point-by-point description of how the effort adheres to the GCOS

guidelines• Strengths and weaknesses or limitations• Uncertainty estimates, possibly as a function of time• Dataset details, such as time period, spatial resolution, data

formats From FCDR inventory; I need your help !

IOVWST, May 2011

Wind stress ECV• Radiometers/scatterometers measure ocean roughness• Ocean roughness consists in small (cm) waves generated by air impact

and subsequent wave breaking processes; depends on water mass density sea= 1024±4 kg m-3 and e.m. sea properties (assumed constant)

• Air-sea momentum exchange is described by = air u* u* , the stress vector; depends on air mass density air , friction velocity vector u*

• Surface layer winds (e.g., u10) depend on u* , atmospheric stability, surface roughness and the presence of ocean currents

• Equivalent neutral winds, u10N , depend only on u* , surface roughness and the presence of ocean currents and is currently used for backscatter geophysical model functions (GMFs)

√ air . u10N is suggested to be a better input for backscatter GMFs (under evaluation by IOVWST)

IOVWST, May 2011

Users FCDR/ECV stress

• Oceanography, eddy scale winds (MyOcean) • Re-analyses (data assimilation uses wind)• IOVWST; process studies (air-sea momentum

exchange, cyclones, extreme winds, convection, tropical circulation, ...)

• Climate, fluxes (incl. carbon)• Design, policy-makers, wind energy, adaptation, ..

IOVWST, May 2011

Strengths / Limitations Scatterometer / Passive

• Excellent precision, mature algorithms, complete coverage• Small scales (25 km), order better than NWP• Since 1991 vector winds

• Intercalibration, accuracy assessment speed scale• Calibration above 30 m/s (truth ?)• Rain (bias) for Ku band and passive systems• Temporal coverage does not match scales (yet)• Low spatial resolution (physical processes)• Ambiguous direction retrieval

IOVWST, May 2011

Strengths/Limitations Altimeter• Continuity/sustainability• Potentially low uncertainty

• Limited speed range (25 m/s)• Instrument anomalies, calibration, ..• Very low coverage (track)• Sea state

IOVWST, May 2011WOAP workshop, April 2011

L1 Calibration• Transponder procedure in development for ASCAT• Rain forest (stable points)• Sea ice / snow /desert (stable points) Geographically limited, while some errors may be orbit

phase dependent

Need to combine all methods of calibration, including ocean calibration

Calibration procedures and GMFs need to be shared between producers to achieve intercalibrated NRCS


ASCAT stability - Ocean calibration• Trends of 0.1 m/s just

visible (10 year req.)• Sampling error to be

accounted for (buoy)


Precision, accuracy: triple collocation

Spatial representation error from spectrum difference integrated from scales from 25 km to 800 km

u v

Bias ASCAT (m/s)Bias ECMWF (m/s)

0.150.28

-0.020.08

Trend ASCATTrend ECMWF

1.011.03

1.011.04

ASCAT (m/s) ECMWF (m/s)

0.691.50

0.811.52

Representation error *) (m/s) 0.79 1.00

Representation error is part of ECMWF error

OSI SAF NRT req. 2 m/s, WMO in speed/dir.

See also Vogelzang et al., JGR, 2011


Spatial resolution

• Spectral analysis of collocated fields

• Comparison to in situ spectra in time / space to check energy cascade behaviour

• Verification of variances and resolution by averaging products (e.g., QSCAT 100km vs 25km, ASCAT 25km vs 12.5km)


Define Uncertainty, Stability, Resolution• Users have little clue how different products compare and

whether they use the product most fit for their purpose Standardization of methods (software?) to assess uncertainty,

resolution and stability to be discussed in the IOVWST• NWP ocean calibration, triple collocation, CDF matching• The resulting speed scale standard would be applicable to

scatterometers, radiometers, altimeters and SAR• Accuracy of speed scale TBD (speed dependent)

Producers to share match-up data bases / independent cal/val Publish / post results for users (in central place(s) ?)

IOVWST, May 2011

Maturity Index

IOVWST, May 2011

Summary Several producers provide OVW FCDRs, which are defensible by their

own verification metric These products cannot be easily understood nor combined by the user

community• Mature (5) stable products exist over long times, but not reprocessed

according to GCOS guidelines; some uncoordinated RP plans exist• Matchup data bases exist too, but by producer• Moored buoys are the main reference, but lacking in open ocean• Quality metrics and assessment standards (software) exist too by

producer, but resolution, wind scale, wind quality to be coordinated/agreed

• An IOVWST has been set up last year, which could address ECV coordinated needs when mandated as such

• CEOS Virtual Constellation coordinates satellites/products

IOVWST, May 2011

Suggested actions• Obtain data set details from producers and make ECV inventory (me)• Reprocessing of all satellite winds following GCOS guidelines • Share matchup data bases • Extend moored buoy network in open ocean• Coordinate quality metrics and assessment standards (software) on

resolution, wind scale, wind quality • IOVWST to be mandated to address wind ECV coordinated needs (by

satellite agencies) (incl. altimeters)• CEOS VC to promote satellite coordination and intercalibration• Maintain L1 reprocessing facilities (e.g., ESA ERS)• Complete efforts in ASCAT and OSCAT calibration• Perform ASCAT-ERS and QSCAT-OSCAT NRCS intercalibration• Finally, develop a reference wind scale (intercalibration) for all satellite

winds, scatterometer, radiometer, altimeter, SAR

IOVWST, May 2011

Discussion !

IOVWST, May 2011ISRO 15 Dec 200821

WMO


Triple collocation for winds wbuoy = t + δbuoy w = wind component

wscat = ascatt + bscat + δscat

wback = abackt + bback + δback

• Calculate first and second (mixed) moments• Test assumptions on errors and derive spatial representation error• Eliminate <t> and <t 2> (get rid of the truth)• Solve for calibration coefficients and error variances <δ 2>• Apply CDF matching for higher order calibration (beyond linear)

For derivation:• Ad Stoffelen, 1998, Toward the true near-surface wind speed: error modeling and

calibration using triple collocation. J. Geophys. Res. 103C4, 7755-7766


Ocean momentum The atmospheric stress forces wave motion by momentum transfer

Wave momentum depends on water mass density

Varies by .5% (< 0.05 m/s)

® Wikipedia

IOVWST, May 2011IOVWST meeting, 18-20 May 2010. (c) ECMWF, Hans Hersbach

Air-density effect on satellite ocean windMicrowave roughness relates to stress

Lower air density (Tropics) relates to higher winds 10% change pole vs Tropics, gives 5%, or ~0.4 m/s

IOVWST, May 2011

Air-sea interaction ECMWF weak

ECMWF model ECMWF model

Chelton et al., Science

IOVWST, May 2011

© ECMWF, Hans Hersbach; A. Brown et al., 2005

Lack of ageostrophic flow

IOVWST, May 2011

6-hourly EC wind change (4 cycles)

Forcing is dominated by transient or temporal effects - temporal wind variance larger than in small scales

IOVWST, May 2011

ECMWF increments

ECMWF analysis increments modest wrt spatial deficit (1.2 m2s-2) Most mesoscale scatterometer information remains unexploited Develop MyOcean L3/L4 products

IOVWST, May 2011

GCOS guidelines1) Full description of all steps taken in the generation of FCDRs and ECV products,

including algorithms used, specific FCDRs used, and characteristics and outcomes of validation activities

2) Application of appropriate calibration/validation activities3) Statement of expected accuracy, stability and resolution (time, space) of the product,

including, where possible, a comparison with the GCOS requirements4) Assessment of long-term stability and homogeneity of the product5) Information on the scientific review process related to FCDR/product construction

(including algorithm selection), FCDR/product quality and applications6) Global coverage of FCDRs and products where possible7) Version management of FCDRs and products, particularly in connection with

improved algorithms and reprocessing8) Arrangements for access to the FCDRs, products and all documentation9) Timeliness of data release to the user community to enable monitoring activities10) Facility for user feedback11) Application of a quantitative maturity index if possible12) Publication of a summary (a webpage or a peer-reviewed article) documenting point-

by-point the extent to which this guideline has been followed

Documents

Ocean Vector Wind as Essential Climate Variable [email protected]