Cryosat-2 Product Handbook - Earth Online

Cryosat-2ProductHandbookBaselineD1.0

C2-LI-ACS-ESL-5319

CryoSat-2ProductHandbook C2-LI-ACS-ESL-5319

P a g e 2 BaselineD1.0draft

TableofContents1 INTRODUCTION ................................................................................................................................................ 6

1.1 PURPOSE AND SCOPE OF THIS USER GUIDE .................................................................................................................. 6 1.2 THE CRYOSAT-2 MISSION AND ITS CHARACTERISTICS ..................................................................................................... 6 1.3 SIMPLIFIED PRINCIPLES OF MODES OF OPERATION ......................................................................................................... 8

1.3.1 Low Resolution Mode .................................................................................................................................... 8 1.3.2 Synthetic Aperture Radar Mode .................................................................................................................... 8 1.3.3 SAR Interferometry Mode ........................................................................................................................... 10

1.4 SIRAL CHARACTERISTICS ....................................................................................................................................... 12 1.5 DATA ACCESS AND DATA PROCESSING TOOLS ............................................................................................................ 13

2 DATA PARAMETERS AND CORRECTIONS ......................................................................................................... 14

2.1 SPACECRAFT ORBIT AND ORIENTATION ..................................................................................................................... 14 2.1.1 Drifting orbit and long repeat cycle ............................................................................................................ 14 2.1.2 Orbit processing .......................................................................................................................................... 15 2.1.3 Reference Frame Processing ....................................................................................................................... 15

2.2 SPACECRAFT TIME AND LOCATION ........................................................................................................................... 18 2.2.1 Reference Ellipsoid ...................................................................................................................................... 18 2.2.2 Some Basic Altimetry Terms ........................................................................................................................ 18 2.2.1 Geoid and Mean Sea Surface ...................................................................................................................... 20 2.2.2 Sea Surface Height Anomaly ....................................................................................................................... 21 2.2.3 Freeboard .................................................................................................................................................... 21 2.2.4 Timestamps ................................................................................................................................................. 21 2.2.5 Latitude, Longitude and Altitude ................................................................................................................. 21 2.2.6 DORIS Ultra Stable Oscillator Drift .............................................................................................................. 21 2.2.7 Satellite Velocity Vector .............................................................................................................................. 22 2.2.8 Reference Orbit ........................................................................................................................................... 22 2.2.9 Ground Tracks ............................................................................................................................................. 22 Illuminated Areas ............................................................................................................................................ 24 2.2.10 ....................................................................................................................................................................... 24

2.3 INSTRUMENT CORRECTIONS AND NOISE .................................................................................................................... 27 2.3.1 Automatic Gain Control .............................................................................................................................. 28 2.3.2 Signal Phase Corrections ............................................................................................................................. 28 2.3.3 Phase Slope Correction ................................................................................................................................ 28 2.3.4 Hamming Weighting Function for SAR Azimuth FFT ................................................................................... 29 2.3.5 Noise ........................................................................................................................................................... 34 2.3.6 Echo Saturation ........................................................................................................................................... 34 2.3.7 Doppler Correction ...................................................................................................................................... 34

2.4 RETRACKING AND PARAMETERS DERIVED FROM ECHO SHAPE ....................................................................................... 34 2.4.1 Range Window and Window Delay ............................................................................................................. 34 2.4.2 Reception Period ......................................................................................................................................... 34 2.4.3 Range Window Sampling ............................................................................................................................ 34 2.4.4 Echo Positioning and Scaling ....................................................................................................................... 37 2.4.5 Retracking ................................................................................................................................................... 37



2.4.6 Backscattering ............................................................................................................................................ 39 2.4.7 Peakiness ..................................................................................................................................................... 39 2.4.8 Effects of Terrain ......................................................................................................................................... 40 2.4.1 Interferometric Mode .................................................................................................................................. 40 2.4.2 Significant Wave Height .............................................................................................................................. 41 2.4.3 Sea State Bias .............................................................................................................................................. 41 2.4.4 Wind Speed ................................................................................................................................................. 41

2.5 RANGE CORRECTIONS FOR ATMOSPHERIC EFFECTS ..................................................................................................... 41 2.5.1 Dry and Wet Tropospheric Corrections ....................................................................................................... 41 2.5.2 Inverse Barometric Correction .................................................................................................................... 42 2.5.3 Dynamic Atmospheric Correction ................................................................................................................ 42 2.5.4 Ionospheric Correction ................................................................................................................................ 42 2.5.5 Derivation of Corrections ............................................................................................................................ 42 2.5.6 Application of Corrections to Range ............................................................................................................ 43

2.6 RANGE CORRECTIONS FOR TIDAL EFFECTS ................................................................................................................. 43 2.6.1 Ocean Tide .................................................................................................................................................. 43 2.6.2 Ocean Loading Tide ..................................................................................................................................... 43 2.6.3 Solid Earth Tide ........................................................................................................................................... 43 2.6.4 Geocentric Polar Tide .................................................................................................................................. 43 2.6.5 Application of Corrections to Range ............................................................................................................ 44

2.7 DERIVING RANGE ................................................................................................................................................. 44 2.7.1 Geophysical Corrections by Surface ............................................................................................................ 44 2.7.2 Surface Type ................................................................................................................................................ 45 2.7.3 Level-2 Application ...................................................................................................................................... 45

2.8 MODELS USED ..................................................................................................................................................... 46 2.8.1 Static Auxiliary Data Files ............................................................................................................................ 46 2.8.2 Dynamic Auxiliary Data Files ....................................................................................................................... 48

3 CRYOSAT-2 DATA SUPPLY ............................................................................................................................... 50

3.1 CRYOSAT-2 GROUND SYSTEM ................................................................................................................................. 50 3.2 CRYOSAT-2 USER PRODUCTS .................................................................................................................................. 50

3.2.1 Level-2 Geophysical Data Record ................................................................................................................ 50 3.2.2 Freeboard Processing .................................................................................................................................. 50 3.2.3 Slope Correction .......................................................................................................................................... 51 3.2.4 Level-1b Data .............................................................................................................................................. 51 3.2.5 Near Real Time Data ................................................................................................................................... 52 3.2.6 Fast Delivery Marine Mode Data ................................................................................................................ 52

3.3 FILE STRUCTURE ................................................................................................................................................... 52 3.4 FILE NAMING CONVENTION .................................................................................................................................... 53 3.5 OTHER CONVENTIONS ........................................................................................................................................... 54

3.5.1 Latitude and Longitude Limits ..................................................................................................................... 54 3.5.2 Range Correction Sign Convention .............................................................................................................. 55 3.5.3 Timestamp Format ...................................................................................................................................... 55 3.5.4 Time in L2 GDR Records .............................................................................................................................. 55



4 REFERENCES AND ACRONYMS ........................................................................................................................ 57

4.1 REFERENCE DOCUMENTS ....................................................................................................................................... 57 4.2 ACRONYMS AND ABBREVIATIONS ............................................................................................................................ 58

TableofFigures

Figure1:Examplemodemask,uploadedtoCryoSat-2inApril2011......................................................................8Figure2:Geometryofmeasurementforasingleburst..................................................................................................9Figure3:Geometryofmulti-looking....................................................................................................................................10Figure4:SARinmodeacross-trackangledetermination...........................................................................................11Figure5:ExampleoftracksofCryoSat-2'sorbitoverathree-dayperiod..........................................................14Figure6:SatelliteReferenceFrame.....................................................................................................................................16Figure7:Basicaltimetrytermsandapplicablecorrectionsoveropenocean...................................................20Figure8:Variationoffootprintwithlatitude...................................................................................................................25Figure9:ComparisonofAIRwithandwithoutHammingwindow.......................................................................29Figure10:ComparisonofAIRwithandwithoutHammingwindow(fullrange)............................................30Figure11:Hammingweightingfunction............................................................................................................................31Figure12:AzimuthFFTofasingleburst:thenadirclutteristhe'line'inthemiddle...................................32Figure13:AzimuthFFTofasingleburstwithhamming:thenadirclutterisreduced.................................32Figure14:Stack60withouthamming:heretheparabolaisthenadirclutter..................................................33Figure15:Stack60withhamming:thenadirclutterhasdisappeared................................................................33Figure16:Idealisedpulse-limitedwaveforminwindow...........................................................................................38Figure17:Peakinessbylatitude............................................................................................................................................39Figure18:Proportionofspecularpointsbylatitude....................................................................................................40

TableofTables

Table1:Satellitemaincharacteristics....................................................................................................................................6 Table2:L2retrackers.................................................................................................................................................................37 Table3:Staticauxiliaryfiles....................................................................................................................................................46 Table4:RelevantFileTypes....................................................................................................................................................54



DocumentChangelogRevision Date DetailsofchangesD1.0 3-APR-2018 BroughtuptodateforBaselineD.



1 Introduction

1.1 PurposeandScopeofthisProductHandbookThisdocumentisintendedtogiveCryosat-2datausersaconciseoverviewoftheCryosat-2level-2(L2)iceandoceanproducts.ThisrevisionofthedocumenthasbeenreleasedtoaccompanythedeliveryofBaseline-DCryosatproducts.More informationaboutCryosatproductbaselinesmaybe foundherehttps://earth.esa.int/web/guest/missions/cryosat/ipf-baseline.

TheBaseline-Dand laterproducts are in a self-describingnetCDF format, and canbe read into anyprogramthatsupportsreadingnetCDFdata.Thisdocumentisintendedtohelpinunderstandingandusingthecontentsoftheproducts.Foracompletelistofthecontentsandstructureoftheproducts,seetheproductformatdocuments:

• CRYOSATGroundSegmentInstrumentProcessingFacilityL1BProductsSpecificationFormatCS-RS-ACS-GS-5106

• CRYOSATGroundSegmentInstrumentProcessingFacilityL2ProductsSpecificationFormatCS-RS-ACS-GS-5123

A global attribute in the netCDF product named reference_document gives the specific productdocumentIDandrevisionthatappliestothatproduct.

1.2 TheCryosat-2MissionanditscharacteristicsCryosat-2isafour-yearradaraltimetrymission,launchedon8April2010,tomonitorvariationsinthethicknessoftheEarth’smarineicecoverandcontinentalicesheets.Itsprimaryobjectiveistomeasurethe extent of thinning Arctic ice due to climate change. However, beyond the core of the missionobjective,CryoSat-2alsorepresentsavaluablesourceofdatafortheoceanographiccommunity.

TheCryoSat-2satellitereplacestheoriginalCryoSat,whichwaslostduetoalaunchfailureinOctober2005.Itsmaincharacteristicsaresummarizedinthetablebelow:

Table1:Satellitemaincharacteristics

Launch Missionduration Orbit Payload Mass Dimensions

08April2010

Extensionuntil2017approved

- LowEarthOrbit- NonSun-synchronous- Altitude:730km- Inclination:92degree- Repeatcycle:369days

(5344orbits)with30-daysub-cycle

- SARInterferometricRadarALtimeter(SIRAL)

- DORISreceiver- Laserretro-

reflector- 3Startrackers

720kg

4.60m

x

2.4m

x

2.2m



Fromanaltitudeofjustover700kmandreachinglatitudesof88°,Cryosat-2monitorsprecisechangesinthethicknessofthepolaricesheetsandfloatingseaice.Theobservationsmadeoverthefour-yearlifetime of the mission have provided conclusive evidence of rates at which ice cover may bediminishing.

SIRAL is the primary instrument on-board CryoSat-2 and has extended capabilities to meet themeasurementrequirementsforice-sheetelevationandsea-icefreeboard.

Derivedfromthewell-knownPoseidonoceanaltimeterontheJasonsatellite,SIRALisaverycompactassembly, weighing just 90 kilograms. It combines three measurement modes to determine thetopographyofoceans,landandseaicemasses,aswellasicefloesandsignificantelevationtransitions,especiallybetweenlandandicefields:

Ø Overtheoceansandicesheetinteriors,Cryosat-2generallyoperateslikeatraditionalradaraltimeterinLowResolutionMode(socalledLRM).

Ø Overseaiceandfewoceanographicareas,coherentlytransmittedechoesarecombinedviaSyntheticApertureRadarMode(socalledSARprocessing,toreducetheilluminatedsurfacearea.Thismodeismainlyusedtocarryouthigh-resolutionmeasurementsoffloatingseaiceandlandicesheets,enablingtheindirectmeasurementofthesheets’thickness.

Ø CryoSat's most advancedmode is used around the ice sheetmargins and overmountainglaciers.Here,thealtimeterperformsSyntheticApertureRadarprocessingandusesasecondantennaasaninterferometertodeterminetheacross-trackangletotheearliestradarreturns.Thismode(so-calledSINorSARinprovidestheexactsurfacelocationbeingmeasuredwhenthesurfaceisslopingandcanbeusedtostudymorecontrastedterrains, liketheveryactiveareaslocatedatthejunctionoftheicesheetandtheAntarcticcontinentorGreenland.

Themodeofoperationisselectedfromamaskofgeographicalzones.Themaskisupdatedeverytwoweekstoallowforchangesinseaiceextent.Thegeographicalmaskisconvertedintoaweeklytimelineofinstrumentcommandsbythemissionplanningfacility,takingintoaccounttheCryosat-2referenceorbit. An example of the mode mask, downloaded from the ESA website at the URLhttps://earth.esa.int/web/guest/-/geographical-mode-mask-7107isshowninFigure1.



Figure1:Examplemodemask

1.3 Simplifiedprinciplesofmodesofoperation

1.3.1 LowResolutionModeConventionalradaraltimetersoperateinpulse-limitedmodeforwhichtheareaofthesurfaceseenbytheinstrumentislimitedbythelengthoftheradarpulsetransmittedbythealtimeter.InCryoSat-2thismodeisoftenreferredtoasLRMasthedatarateismuchlowerthanfortheothermeasurementmodes.Inpulse-limitedmoderadarpulsesaresentatintervalsofabout500µs(~2000Hz)ensuringthatthereturningechoesareuncorrelated.Manyuncorrelatedechoescanbeaveragedtoreducespeckle.Thismodeofmeasurementissuitableovertherelativelysmoothicesheetinteriorsoroverocean(redonFigure1,andalsowherenoothermodeisshown).

1.3.2 SyntheticApertureRadarModeTheSARmode(green inFigure1) ismainlyusedoverrelatively flatareasofsea ice.TheCryoSat-2altimeter sends aburst of pulseswith an interval of only50µs (~20,000Hz)partitioning thepulse-limitedfootprint.Giventhattheobservedsurfaceisinmovementrelativetothesatellite,theechoesreceived by the radar include amix of Doppler frequencies corresponding to all surface directionscoveredbytheradarantennabeam,fromthefronttotheback,seeFigure2.Inthisfigure,only8beamsareshownforclarity(insteadofthe64whicharegenerated)andthedivergenceanglebetweenthemisexaggerated.



Figure2:Geometryofmeasurementforasingleburst.

Asthereturningechoesarecorrelated,andbytreatingthewholeburstatonce,thedataprocessorcanseparate the echo into64 stripsperpendicular to the trackby exploiting the slight frequency shiftscausedbytheDopplereffect.

Eachstripisabout250mwideandtheintervalbetweenburstsisarrangedsothatthesatellitemovesforwardby80meachtime.Thus,eachgroundlocationislookedmultipletimes(seeFigure3).Inthisfigure,thebluerectangleisthemulti-lookedfootprint.

Thestripslaiddownbysuccessiveburstscanthereforebesuperimposedoneachothertocreateastackofechoeswhichareaveragedtoreducenoise.


P a g e 1 0 BaselineD1.0draft

Figure3:Geometryofmulti-looking

1.3.3 SARInterferometryModeSARinterferometry,orSINorSARIn,mode(purpleinFigure1)isusedacrossicesheetmarginswheretheicesurfacemaybesloping.SIRALhastwoantennae,mountedabout1mapart,whichreceivetheechoalmostsimultaneously.Iftheechocomesfromanywherebutdirectlybelowthesatellite,thenthereisadifferenceinthepathlengthsofthesignaltravellingtobothantennas.Bycomparingthedifferingsignalphases,theangleofarrivalandhencetheechoorigincanbederived.



If one SIRAL receive chain should fail then the instrument can operate in SINmodewith only onechannel.ThisisreferredtoasSINdegradedmode,orSID.Insteadofinterferometrymethod,theslopemodelintendedforLRMmodeisusedtotrytoimprovetheresultbyestimatingtheacross-trackangle.Alongsidetheseoperatingmodestherearespecificinstrumentcalibrationmodes,thepulseresponsecalibration,thereceivechaincalibrationandtheinterferometricexternalcalibration.Informationfromtheseisfedintothehigher-leveldataproductsandcontainedinthelevel-1bdataproducts.

Furthergeneralinformationisgiveninreferences1,2andespecially3.AnoverviewofSIRALisgiveninreference15.

Figure4:SARinmodeacross-trackangledetermination



1.4 SIRALCharacteristicsRadiofrequency 13.575GHz(singlefrequencyKu-bandradar)

Pulsebandwidth 320MHz(40MHzfortrackingonlyinSIN)

PulseRepetitionFrequency(PRF) 1.97kHzinLRM18.181kHzinSARandinSIN;coherentpulsetransmissionforDopplerprocessing

BurstmodePRF Pulses/burst1970HzinLRM,85.7HzinSARand21.4HzinSIN

N/AinLRM,64inSARandinSIN

Compressedpulselength 3.125ns

Pulseduration 44.8µs

Timing RegularPRFinLRM,burstmodeinSARandSIN

Samplesinecho 128inLRMandSAR,512inSIN

RFpeakpower 25W

Antennasize 2reflectors1.2mx1.1m,side-by-side

Antennabeamwidth(3dB) 1.06º(along-track)x1.1992º(across-track)

Antennafootprint 15km

Rangebinsample 0.2342mforSARandSIN,0.4684mforLRM

Datarate 60kbit/sforLRM,12Mbit/sinSAR,2x12Mbit/sinSIN

Instrumentmass(withantennas) 90kgredundant

Instrumentpower 149W

Trackingcycle 47.17ms(notamultipleofPRF)

Burstrepetition 11.8ms(notamultipleofPRF)

Antennabaselinelength 1167.6mm

Thealong-tracksamplingresolutioninSARandSINmodesisvariablearoundtheorbit.Itis,forexample,278mbasedonanaltitudeof750km, flatEarthandhalfpowerof theazimuthbeam.This is then



affectedbytheazimuthhammingweightingfunctionwhichdegradestheresolution.Thesamplingoftheprocessorisdifferenttotheresolution.WiththeeffectofacurvedEarthinsteadofaflatone,thentheDopplercellwidthonthegroundisabout290malongtrack;however,itisafunctionofaltitude,soitvaries.IncludingalsothebeamwideningeffectofthehammingwindowappliedbeforetheazimuthFFTintheprocessor,thentheeffective"resolution"ofthebeamsiswidenedbyafactorofroughly1.3.Sothe"resolution"(asopposedtosampling)inthealong-trackdirectionismorelike380meters,butthisisplusorminusseveralpercentasthealtitudevaries.

Thewidthintheacross-trackdirectionoftheareaoverwhichechoesarecollectedbytheantenna(theantenna-limited illuminated area) is approximately equal to 15 km since it can be computed as2*h*tan(theta/2)beinghthealtitudeandtheta=1.2°theantennabeamwidthacross-track.Thewidthintheacross-trackdirectionof thearea illuminatedbythe leadingedgeof thepulseuntil thetimethetrailingedgefirstintersectsthesurface(thepulse-limitedilluminatedareaaroundthepointofclosestapproach),incaseofalmostflatsurfaceresponse,isequalto2*sqrt(c*h/B),approximately1.6kmforCryoSat,beingBthepulsebandwidth.

Formoredetails,theinterestedreadermayrefertotheESAwebpageat

http://www.eoportal.org/directory/pres_CryoSat2EarthExplorerOpportunityMission2.html

1.5 DataAccessandDataProcessingToolsAccess to Cryosat-2 data is controlled by a registration system. To register, or to download data ifalreadyregistered,visitESA‘EarthnetOnline’Cryosat-2portalat:

https://earth.esa.int/web/guest/missions/esa-operational-eo-missions/cryosat

AveryusefulwebsiteisESAradaraltimetrytutorialsite,whichincludeslinkstotoolsavailablefordataanalysisaswellasageneraldiscussionofradaraltimetryprinciples:

http://earth.eo.esa.int/brat/

BeginningwithBaselineD,Cryosat-2L1BandL2productsareproducedinstandardnetCDFformats.For older baselines, a set of API routines to read Cryosat-2 data products is available through ESA‘Earthnet Online’ web site under the link under ‘Software Routines’. The API libraries have beensuppliedbyUCLandprovideingestionroutineswritteninANSICandIDLthatcanbeusedtoreadtheL1BandL2Cryosat-2products.Thedownloadpackageincludesareferencemanualandexamples.

All Cryosat-2 science products, available on the Cryosat-2 dissemination server (ftp://science-pds.cryosat.esa.int)areprocessedwiththeDORISpreciseorbitsandaregeneratedwithadelayofabout30daysafterSIRALacquisition.



2 DataparametersandcorrectionsPleasenotethattherelevantfieldsinthedatarecordsarereferencedbytheirvariablenamestakenfromtheBaselineDnetCDFproductsandarepresentedinmonospacedfont.

2.1 SpacecraftOrbitandOrientation

2.1.1 DriftingorbitandlongrepeatcycleThechoiceoftheCryoSat-2orbitistheresultoftradeoffbetweenthedesiredhighdensityofcrossoverpointinordertomeetthemissionrequirementsandtheneedtosufficientlycoverthesouthGreenland.For this, theCryoSat-2'sorbithasameanaltitudeof717kmandahigh inclinationof92°,allowingmeasurementsathighlatitudes(upto88°).Thisorbit isnot-sun-synchronousanddriftsthroughallanglestotheSuninapproximately16months.Furtherinformationisgiveninreference4.

The repeat cycle for Cryosat-2 orbit should be 369 days, corresponding to 5344 revolutions. It ishoweverimportanttonotethattheCryosatorbitdoesnotexactlyrepeataftereachcycle,asitisusually the case for previous ocean-oriented altimetric mission. CryoSat's ascending nodes arerepeating fromcycle to cyclewithin few tenmeters inorder tohave equidistant ascending equatorcrossingsinthereferencegroundtrack.Thedescendingnodesarehowevernolongerequidistantduetoaresidualrotationoftheeccentricityvector.EvenifthisissuehasbeenaddressedaftertheLaunchandEarlyOrbitPhase(LEOP),fluctuationsuptonearly4kmcanstillbeobservedonthedescendingnodefromcycletocycle.

Figure5:ExampleoftracksofCryoSat-2'sorbitoverathree-dayperiod



2.1.2 OrbitprocessingInordertomeasureaccurately,thepositionofthesatellite,CryoSat-2carriessomespecificequipment:

• DORIS (Doppler Orbit and Radiopositioning Integration by Satellite) is a trackingsystem carried by CryoSat-2 to detect and measure the Doppler shift on signalsbroadcastfromanetworkofover50radiobeaconsaroundtheworld.Thesesignalsareusedfororbitdetermination,downtomillimetrelevel,andareessentialforaccuratelymeasuringtheheightoftheicesurface.DORISstandsfor.Thesystemhasbeenusedonpreviousaltimetrymissions. http://www.aviso.oceanobs.com/en/doris/index.html

• Asmalllaserretro-reflectorreflectslightbackinexactlythedirectionitcamefrom.AglobalnetworkoflasertrackingstationsfireshortlaserpulsesatCryoSat-2andtimethe interval before the pulse is reflected back, providing independent referencemeasurementsofCryoSat-2position.The InternationalLaserRangingService(ILRS)provides tracking from its global network of laser ranging stations to support themission.

Themainorbit information fordataprocessing is taken from theDORISPreciseOrbit (POE)or theDORISNavigatorLevel-0Data.

SystematicprocessingreliesonthePOE,whichbecomesavailabletothePayloadDataSystemaround30daysafteracquisition.NearRealTimeprocessingpreferablyusestheDORISNavigatorlevel-0datageneratedon-boardthesatelliteanddownlinkedtogroundduringstationvisibility.

Orbitformatsaredescribedinreferences5and6.

Precise orbital details used during processing of a given dataset are found in the instrumentcharacterizationdatafile,andthenameofthefileusedduringprocessingislistedinthenetCDFglobalattributesoftheproductfile,seereferences12and13fordetails.

2.1.3 ReferenceFrameProcessingKnowledgeofthepreciseorientationofthebaselineandthetworeceivingantennasisessentialforthesuccessofthemission.CryoSat-2measuresthisbaselineorientationusingthepositionofthestarsinthesky.Threestar-trackersaremountedonthesupportstructurefortheantennas.Eachcontainsacamera,whichtakeuptofivepicturespersecond.Theimagesareanalysedbyabuilt-incomputerandcomparedtoacatalogueofstarpositions.

TheCryoSat-2spacecraftfliesinanose-downattitudeinclinedat6degreestothepositiveXaxis.Thenadirdirectionisinclined6degreesfromthenegativeZaxis.Theoriginofthesatellitereferenceframeisatthecentreofthesatellitemountingplaneonthelaunchvehicle.Theradarantennasaremountedsuchthattherealbeamdirectionisalongthenadirdirection.



Fulldetailsaregiveninreference7.

This illustrationcomes fromreference7andshowsCryoSat-2,without its thermalcontrolmaterial.NotethatthenegativeZaxisisshowninordertodisplaytheoffsettothenadirdirection.

Level-2andlevel-1bproductsuseaCryosat-2processingreferenceframe(notthesatellitereferenceframe) to describe the interferometric antenna baselineinter_base_vec_20_ku and real antennaboresight(realbeam)directionvectorsbeam_dir_vec_20_ku.Thisreferenceframehasitsoriginatthesatellitecentreofmass(CoM).CryoSat-2'sCoMiscolocatedwithitscentreofgravity(CoG)andhasbeendesignedsothattheCoMcorrespondstothecentreofthesingle,sphericalfueltank.Thefuelisgaseous nitrogen, which uniformly fills the tank. CryoSat-2 has nomoving parts. Consequently thecentreofmasswillnotmoveduringthesatellitelifetime.Overalldesignconcernsaregiveninreference4.

TheXaxisisdirectedfromtheframeoriginandisparallelwiththenormalincidencetothereferenceellipsoid. The Y axis is in the plane defined by the X axis and the spacecraft velocity vector V. It isdeterminedby

Y=(V–X(X.V))/|V–X(X.V)| (Equation2.1)

TheremainingaxisisdeterminedbyZ=YxX (Equation2.2)

Allreferenceframesusedduringprocessingaredescribedindetailinreference20.

Inotherwords, theantennabaselineandrealbeamvectorsaredefined in theCryosat-2processingreferenceframe(CoG,X,Y,Z)orthonormalbutleft-handed,whereXistheunitvectorpointingtonadir

Figure6:SatelliteReferenceFrame



inthe"along-track'plandefinedbytheCOGandthevelocityvector,andYistheunitvectoralmostinthesamedirectionasthevelocityvector.

Theinterferometricbaselinevectorandtherealbeamvectorcomponentsarenormalizedtomakeaunitvector and scaledbyonemillion to get integer components inmicrometres.These components aremeasuredby thestar trackersanglesonboard. In theperfect situation, thebaselinevector is [0;0;-1000000]andthebeamvectoris[1000000;0;0].

Sinceonevectorelementisnominallyoneunitandtheanglesareverysmall,theroll,pitch,andyawanglesoftheantennabenchintheCryosat-2processingreferenceframecanbeapproximatedwith:

roll =+(baseline_vector_x) inmicrorad,whererollisrelativetoaxisYdefinedabove.

yaw =-(baseline_vector_y) inmicrorad,whereyawisrelativetoaxisXdefinedabove.

pitch =-(beam_vector_y) inmicrorad,wherepitchisrelativetoaxisZdefinedabove.

These angles are presented in the variables off_nadir_roll_angle_str_20_ku,off_nadir_pitch_angle_str_20_ku, off_nadir_yaw_angle_str_20_ku, at a resolution of 0.1microdegreesandwithallapplicablecorrectionsandbiasesapplied.

Positive/negativerollcorrespondstorightantennadown/uprespectively.

Positive/negativeyawcorrespondstospacecraftnoseontheright/leftrespectively.

Positive/negativepitchcorrespondstospacecraftnosedown/uprespectively.

InordertolookatEarth,thesatelliteturnsonerevolutionininertialspaceforeachrevolutionthatitmakesaroundtheEarth.SincearevolutionaroundtheEarthtakesroughly6000seconds,therotationrateisabout360degrees/6000seconds,orabout0.06degreespersecond.ItismainlythepitchwhichmustturnininertialspacetofollowtheEarth.Therefore,ifthereisatimingerrorinadatationoracalculation,onecanexpectpitch,inparticular,tobeaffectedbyanamountoforder0.06degrees/sectimeswhateveristhedeltatime.

The SIN phase difference waveforms inL1B data ph_diff_waveform_20_ku do includea phasedifferencebiasinstr_ext_ph_cor_20_ku,whichaccountsfortherollfixedbiascomponentonly,butdonotincludetherollvariablecomponent,whichcanbeextractedintheproductitselfaccordingtotheaboveformulaandshouldbeused,asshownbelow,tocorrecttheinferredangle.

The variable roll off_nadir_roll_angle_str_20_ku can be applied to the inferred angle as avariablecorrection.TheinferredanglealphaisobtainedfromtheretrackedSINL1Bphasedifferencewaveform ph_diff_waveform_20_ku with: alpha [rad] = lambda * retrack_PD / (2*pi*baseline),where retrack_PD is the retracked phase difference in radians, baseline = 1.1676m and lambda =0.022084m

Thenthevariablerollcorrectioncanbeappliedtotheinferredanglewith:alpha-roll.



Notethatthephasedifferencevaluesph_diff_waveform_20_kuareprovidedbetween–piand+pimicroradians.Apositive/negativephasedifference(andacross-trackinferredangle)correspondstoasignalcomingfromtheright/lefthandsideoftheobserversittingontheantennabenchwithitsdorsalspinealignedwiththenadirdirectionandlookingforwardinthealong-trackflightdirection.

Themispointingangleistheanglebetweentheantennapointing,i.e.thedirectionoftheactualbeam,andthenadirdirection.Nominallythisshouldbezerodegree.

2.2 SpacecraftTimeandLocation

2.2.1 ReferenceEllipsoidSatellitealtimeterheightmeasurementsareusuallyreferencedtoanellipsoid,asimplemathematicalfigurethatdescribesaroughapproximationofthegeoid.ThegeoidistheshapethatwouldhavetheseasurfaceunderthesingleinfluenceofEarths’sgravitationandwithoutanysignatureassociatedtotheoceancirculationandtideeffects.ForCryoSat,thereferenceellipsoidisWGS84,definedbytheNationalGeospatialIntelligenceAgencyoftheUSA.Thiscanbeexploredonlineat

https://www1.nga.mil/ProductsServices/GeodesyandGeophysics/Worldgeodeticsystem/Pages/default.aspx

2.2.2 SomeBasicAltimetryTermsBasictermsusedintheCryosat-2missionare(seeFig2.3):

Range:theone-waydistancefromthesatellitetothesensedsurfacebelow.

Altitude:thedistanceofthesatellitecentreofmassabovethereferenceellipsoid.

Corrections:ValuestobesubtractedtotheRangeinordertoavoidundesirableeffectsduetothedelayof the echopropagation through the atmosphere layers, the temporally subsampledhigh frequencygeophysicalsignalsandtheinstrumentalnoise.

Height:DifferencebetweentheAltitudeandthecorrectedRange,i.e.theinstantaneouselevationofthesurfaceobservedabovethereferenceellipsoid.

MeanSeaSurface:TemporalaverageoftheseaHeightoveragivenperiodwhichcontainsboththesignatureofmarinegeoidandoftheseaelevationduetothemeanoceaniccirculation.

SeaHeightAnomaly:DifferencebetweentheHeightandtheMeanSeaSurfacecorrespondingtothevariablecomponentoftheinstantaneouselevation.

MeanDynamicTopography:DifferencebetweentheMeanSeaSurfaceandtheGeoid,correspondingthesurfacesignatureofthesinglemeanoceaniccirculation.



AbsoluteDynamictopography:MeanDynamicTopographyplusSeaHeightAnomaly(orequivalentlyHeight minus Geoid) containing the surface signature of both the mean circulation and transientoceanographicprocesses.

Notethatmostradaraltimetersobservethesurfaceatnadir,butCryosat-2canobserveoff-nadirwhenitoperatesinSARinmode.Iftheobservationisoff-nadirthenthelatitudeandlongitudeoftheactualobservedpointobservedisprovidedinthedataproducts.

At level 1b, only the engineering corrections (including the USO), the Doppler corrections and thecentre-of-massoffset(explainedinsections2.2.6,2.3.7and2.4.1)areappliedtotherangemeasurement(givenasatimedelaywindow_del_20_ku).Thealtitudeisgivenasalt_20_ku.

Atlevel2,theatmosphericandtidalcorrections(explainedinsections2.5and2.6)arealsoappliedtotherangebeforeitisusedtocomputeheight.TheheightspresentedintheL2productsarecontainedinvariablesthathavethestring‘height’inthename,e.g.height_1_20_ku.



Figure7:Basicaltimetrytermsandapplicablecorrectionsoveropenocean

2.2.1 GeoidandMeanSeaSurfaceInthelevel-2dataproduct,valuesofthegeoid(overlandandlandice)oralternativelyofthemeanseasurface(MSSoveropenoceanorclosedsea)arespatiallyinterpolatedatthegroundtracklocations.Themodels used are EGM96 (geoid) geoid_20_ku and UCL13 (mean sea surface MSS)mean_sea_surf_sea_ice_20_ku.

EGM96canbeexploredmorefullyathttp://cddis.nasa.gov/926/egm96/egm96.html.

UCL13 is a hybrid globalmodel compiled fromCryoSat sea-surface heightmeasurements above 60degreesnorthandtheCLS2011MSS(http://www.aviso.oceanobs.com/index.php?id=1615)below50degrees North. Between those two latitudes, a linear blend is performed between themodels. ThepurposeofthemodelistoprovidebetteraccuracyintheArcticocean.Modeldetailsarealsogiveninsection2.8.1andreference8.



2.2.2 SeaSurfaceHeightAnomalySeasurfaceheightanomaly(SSHA)isthedifferencebetweenthelong-termaverageseasurfaceheight,i.e.theMSS,andtheoceanheightactuallyobservedbythesatellite.Thismustbedeterminedinordertoaccuratelycalculatetheseaicefreeboard.Atlevel2,theSSHAisfoundbycomparingtheoceanheightwiththeUCL13MSStogettheoceanheightanomalyssha_20_ku,andthenusinglinearinterpolationtogettheSSHAssha_interp_20_ku.TheSSHAqualityflaggivestheRMSoftheresidualsofthelinearfit ssha_interp_rms_20_ku. Technically another MSS may be chosen if desired, as choice isconfigurable.Modeldetailsarealsogiveninsection2.8.1andreference8.

2.2.3 FreeboardSeaicefreeboardfreeboard_20_kuistheheightbywhichanicefloeextendsabovetheseasurface.Level 2 data includes climatological snow depth snow_depth_20_ku and density valuessnow_density_20_kuforeachrecord.Thesnowdepthvaluesareextractedfromastaticclimatology,UCL04, and can be used to adjust the freeboard estimate to account for snow load. Note that thefreeboardcanpossiblybenegativeifthereisheavysnowloadonthinice.Thereisaseparateclimatologyfileforeachmonthoftheyear,whichprovidessnowdepthvaluesfortheArcticregion.NoclimatologyfortheAntarcticregionisavailable.Snowdensityisprovidedasasingleglobalconstantaverage.Thecurrentvalueis400kg/m3.Modeldetailsarealsogiveninsection2.8.1andreference8.

2.2.4 TimestampsIn LRM, the timestamptime_20_ku for eachmeasurement dataset record is the time atwhich themiddleoftheaveragedgroupofpulsesimpactstheground.InSARandSINmodes,itisthetimeatwhichthesatelliteoverfliesthesurfacesamplelocation.

2.2.5 Latitude,LongitudeandAltitudeThe latitude lat_20_ku, longitude lon_20_ku and altitude alt_20_ku measurements are thoseinterpolatedfromtheorbitattheexacttimerecordedinthetimestamp.Allofthemaremeasuredfromthereferenceellipsoidatnadirtothesatellitecentreofgravity.

2.2.6 DORISUltraStableOscillatorDriftTheDORISUSOdriftuso_cor_20_kuisrequiredtoaccuratelyconverttheSIRALaltimetertimeandrangecountsintoUTCtimetagandrangedelaymeasurements.SIRALderivesallitsradarfrequenciesfromtheDORISUltraStableOscillatorwhosefrequency,intermsofdeviationsfromnominalvalue(onemeasurementevery24hours),ismadeavailableinaDORISLevel1bproduct,whichisanincrementalfile.Thisisdescribedinreference6.

TheUSOcorrectionisappliedtothewindowdelayatL1b.ThecorrectionappliedisgivenintermsofrangeatL2asuso_cor_applied_20_ku.



2.2.7 SatelliteVelocityVectorThesatellitevelocityvectorsat_vel_vec_20_ku isgiven in the InternationalTerrestrialReferenceFrame(ITRF),whichisdefinedandmaintainedbytheInternationalEarthRotationService(IERS).Seetheironlinepresenceat

http://www.iers.org/IERS/EN/Science/science.html

2.2.8 ReferenceOrbit

TwophaseshavebeenincludedintheCryosat-2ReferenceOrbitFilebecauseduringtheperiodbetweenlaunchandJune2010theCryosat-2cyclewasnotconsistentduetocontinuousorbitalmanoeuvreswhichneededtobeperformedinordertoobtainastableorbit.ThesemanoeuvreswereplannedaspartoftheCommissioningPhaseactivities.FromJuly2010onwards,astableorbitwasmaintainedandconsequentlytherearePhase1intheReferenceOrbitFile,whichincludestheperiodbetweenlaunchinApril2010toJune2010,andPhase2whichincludestheperiodfromJuly2010onwards.

Initially,theReferenceOrbitFileuseduntilOctober2011wasincorrectsuchthatitonlycontainedPhase1andasaresultalldataproductswithvalidityupuntilOctober2011currentlycontainthewrongcyclenumberandlengthwhichdoesnotcorrespondwiththeactualdataandisthecauseofthediscrepancyobservedbytheusers.InearlyOctober2011,theOrbitScenarioFilewasupdatedwiththetwophasesdescribedabove.OnecannoticethattheorbitparametershavechangedinOctober2011.ThisiscorrectedinallprocessingbaselinesfromBonwards.

PleasenotethatdatabetweenlaunchandJuly2010willnotbereprocessedasithasbeendecidedthatatthisearlystageofthemissionthedatawereoflittlescientificuseduetothemanoeuvers.

2.2.9 GroundTracksThescopeofthissectionistoexplaintheretrievalandanalysisofCryosat-2orbitsandgroundtracksfiles.Themaininterestofthisanalysisistoinvestigatethedifferencesinthegroundtrackswhenusingdifferentpredictionsoftheorbit.InformationanddescriptionaboutavailablefilescanbefoundontheESACryosat-2webpages.

Thefollowinginformationisextractedfromthiswebsite:

The Reference Orbit file contains information on the orbits over a full repetition cycle (369 days),generatedatthestartofeveryorbitalrepetitioncycle.

TheReferenceGroundTrackfileisgeneratedfromtheReferenceOrbitfile.

ThePredictedOrbitfileisahighaccuracyorbitprediction,generatedclosertothetimeoftheoverpass.Itisgeneratedeveryday,andcontainsoneOrbitStateVector(OSV)perorbit,locatedat(orverynearto)theAscendingNode.Thecoverageisfor30daysstartingatthegenerationtime.

Asaresult,theorbitsforagivendaycanbeevaluatedfromtheReferenceOrbitfileorfrom30differentPredictedOrbitfileswhichallcontainthatday,butwithincreasinglyaccuratepredictions.



The following paragraphs show the differences among these predictions, and furthermore set adocumentedexampleofhowtheinformationcanbeextractedandanalysed.

Inparticular,theReferenceGroundTracksareanalysed,theReferenceOrbitandPredictedOrbitfilesarecomparedandthedifferencesamong30consequentgroundtracks,retrievedfromthepredictedorbits,areassessed.

AllthementionedfilescanberetrievedfromtheFTPserver:

FTPserver:ftp://calval-pds.cryosat.esa.int Username:ground Password:tracks

UponconnectiontotheFTPserver,thefollowingdatastructurecanbeobserved:

• 3ReferenceOrbitfiles,oneperyear(2010,2011,2012,etc.)• PredictedOrbitfiles,about30filespermonth• GroundTracks,availableonamonthlyoryearlybasisandwiththreedifferentsamplingsteps

(2,10or60seconds)• Softwarefolder,whichcontainsthenecessarytoolstogenerategroundtracksfromtheorbit

filesandotherfunctionalities.

TheGroundTracksfileisasimple.txtfile.ForJanuary2012,thisfilecontainsallthereferencegroundtracksfrom01January2012to31January2012sampledevery10seconds.

InordertocompareReferenceOrbitsandPredictedOrbitsfor2012,thefollowingfilescanberetrievedfromtheFTPserver:

TheReferenceOrbitsspanfrom17June2011to19June2012.

ThePredictedOrbitsspanfrom02January2012to01February2012

ThetimesamplingoftheReferenceOrbitsfileis100minutes,thereforeonlyoneOSVisgivenforeachorbit (100minutes is approximately the orbit period) for the local ascending node. The orbit filescontentisfullydescribedintheCryosat-2groundsystemMissionFilesFormatSpecifications(pages9-27).There,thetime,numberoforbitandstatevectorarestored.TheX,YandZcoordinatescanbeusedtocalculatetheOrbitHeightH=(X*X+Y*Y+Z*Z)1/2-Requator

The differences on X, Y and Z coordinates, comparing the reference and the predicted orbits, inpercentage,arealways<5%.

Alltheinformationtocalculatethegroundtracksfromtheorbitfilesisinthedocument:



“CRYOSAT-2: TRANSPONDER PASS TOOLS - FILETRANSFER DOCUMENT” from EOCFI-FTD-004_1_5.zip

Startingwith a ReferenceOrbit or PredictedOrbitfile,theexecutablefunctionreturnsthegroundtrackpoints(atatemporalsamplingdefinedinTimeStep)which are contained in the area defined in ZoneDatabase,foragivensetoforbitsorforagiventimeinterval.

Thegroundtracksamplesretrievedfromtheoldestpredictedorbit(onemonthbefore)andthegroundtracksamplesretrievedfromthenewestpredictedorbit (one day before) can be a bit different. Thisdifferencemainlydependson the timeordistancefromthelastorbitmaneuver.Thespatialdifferencebetween the tracks cannot be calculated as thesimple distance between two samples of thedifferentgroundtracks,becausethereisatimedelaywhichturnsintoaspatialshift.However,theacross-track distance can be easily calculated with somegeometry. Nominally, the across-track distanceneverexceeds1kmandrarelyexceeds500minpolarareas.

2.2.10 IlluminatedAreas

2.2.10.1 Beam-limitedfootprint

Thebeam-limitedilluminatedareaisdefinedasthewholeareaontheEarthsurfaceonwhichabeamechoiscollected.Itdependsmainlyontheantennailluminationpattern,asaconsequenceitdoesnotvarywiththeacquisitionmode.Itcanbedescribedbyitswidthsinthealong-trackdirectionandintheacross-trackdirection,thatdependontheantennapattern.Being𝝑theantennabeamwidthat-3dB,𝜽𝑩theangleofthecentralbeamdirectionwithrespecttothenadirandhthealtitudeofthesatellite,thewidthofthebeam-limitedfootprintresultsin𝑫 = 𝒉 ∙ tan(𝜽𝑩 + 𝝑 𝟐⁄ ) − 𝒉 ∙ tan(𝜽𝑩 − 𝝑 𝟐⁄ )wheretheflatEarthapproximationisused.

Consideringanaltitudeof730km,𝝑=1.06degreesand𝜽𝑩=0degreesalongtrack,while𝝑=1.1992degrees and 𝜽𝑩 = 0 degrees across track, the beam-limited illuminated area for SIRAL can bedecomposedin:

h

J

Bq

D



• widthofthebeam-limitedareaalongtrackapproximatelyequalto13.5km

• widthofthebeam-limitedareaacrosstrackapproximatelyequalto15.3km

Both the along and across beam-limited areas are not constant but are dependent on the orbitcharacteristicsandthereforetheychangewithlatitude.Thewidthofthebeam-limitedareaalongtrackvariesbetween13.2and14.0kmwhilethebeam-limitedareaacrosstrackvariesbetween14.9and15.8km.

Figure8:Variationoffootprintwithlatitude

2.2.10.2 Pulse-limitedfootprint

Thepulse-limited illuminatedarea isdefinedas the illuminatedareaongroundaroundthepointofclosestapproachanditcorrespondstotheareailluminatedbytheleadingedgeofthepulseuntilthetimethetrailingedgefirstintersectsthesurface.

For LRM, the pulse-limited footprint depends only on the compressed pulse duration, since LRMacquisitionhasonlyoneindependentvariable(thetimedelay).

TheDopplerbeamformationinSARandSINmodesallowsdiscriminatingthedirectionofarrivaloftheechoesinthealong-trackdirection,sothatapulse-Dopplerfootprintcanbedefined.ThisisduetothefactthatinSARandSINmodestherearetwoindependentvariables,thealong-trackpositionandtheacross-trackposition,relatedtotimedelay.

ThewidthoftheilluminatedareaforLRMandSAR/SINmodesisdefinedbelowusingtheflatEarthapproximationandassumingaquasi-flatgroundsurface.



2.2.10.3 Pulse-limitedfootprintforLRM

Inconventionalaltimeteracquisition,suchasLRM,theareaaroundtheclosestapproachpointisfullyilluminatedwhentherearofthepulsereachestheEarthsurface.Thepulse-limitedilluminatedareacanthereforebeapproximatedasacircularareawithradiusequalto:

where denotesthespeedoflight,h=730kmisthealtitudeofthesatelliteandthecompressedpulselengthis with beingthepulsebandwidth.

ForSIRAL,thepulse-limitedilluminatedareaisabout2.15km2,correspondingtoawidthoftheareaapproximatelyequal to1.65km. Ithas thesamewidth inboth thealong-trackand theacross-trackdirectionssinceinLRMthereisonlyoneindependentvariable,thatisthetimedelay.

Bchchr ×=××= t

cB/1=t B

t=Tt=T+tt=T+2t

r

Pulse-limited footprint

Time-delay rings

t=Tt=T+tt=T+2t

Doppler beams

Along-track

Acro

ss-t

rack

Pulse-Doppler limited footprint

Time-delay rings

r

xD



2.2.10.4 Pulse-Doppler-limitedfootprintforSARandSINmodes

InSAR/SINacquisitionmodes,theprocessorallowstosharpentheareaofthepulseinthealong-trackdirection.Asstatedbefore,theDopplerbeamformationinSAR/SINmodeallowstodiscriminatethedirectionofarrivaloftheechoesinthealong-trackdirectioninadditiontothemeasureofthetimedelay.Thiswaytheilluminatedareawidthisdefinedindependentlyinbothindependentdirections,thealong-trackandtheacross-trackdirections.

Intheacross-trackdirection,theilluminatedareawidthforSIRALisdefinedasthepulse-limitedwidthinLRM.However,inthealong-trackdirection,theilluminatedareawidthforSIRALisdefinedasthesharpenedbeam-limitedarea.Thepulse-Doppler-limitedareaforSAR/SINcanbeapproximatedbyarectangleareagivenbythepulse-limitedareawidthacrosstrackandbythesharpenedbeam-limitedareawidthalongtrack.

SincethebandofDopplerfrequenciesthatisunambiguouslysampledbythePRFgoesfrom–PRF/2to+PRF/2,and64differentsharpenedbeamsareequallyspacedintheDopplerdomain,thewidthofthesharpenedbeam-limitedarearesultsin

where isthealtitudeofthesatellite, isthewavelength, isthevelocityofthespacecraftand isthePulseRepetitionFrequency.

ForCryoSat,thepulse-limitedareawidthintheacrosstrackdirectionisapproximatelyequalto1.65kmwhilethesharpenedbeam-limitedareawidthinthealong-trackdirectionisapproximatelyequalto305m,whichinturncorrespondstoanalong-trackresolutionapproximatelyequalto401m,usingflatEarthapproximation.Hence,thepulse-Doppler-limitedareaforSAR/SINisabout0.5km2.

2.3 InstrumentCorrectionsandNoiseErrorcomponentscanbecharacterisedaccordingtotheirtimedependenceasfollows:

Abiaserrorisaresidualfixedoffseterror,whichisstablethroughoutthemissionbydefinition.Biasesareassumedtohaveauniformdistribution.

Adrifterrorisavariationduetoageing,orothereffects,whichappearsasaslowchangewithtime,havingnoperiodiccharacter,withthepossibilityofdiscretesteps.

Aharmonicerrorvariesperiodically,wheretheperiod is typicallyof theorderof theorbitalorhalforbitalperiodbutmaybemuchsmallerormuchlonger.Theerrorhasameanofzero.

Arandomerrorvariesinanunpredictablemanner,relativelyquicklyinrelationtoanorbitalperiod,inwhichthereisnocorrelationbetweensuccessiverealisations.TheseerrorscanbeassumedashavingaGaussiandistributionwithzeromeanandgivenstandarddeviation.

642PRFv

hx l=D

h l v PRF



Instrumentcorrectionsarederivedfromcharacterisationandcalibrationdata.Thealtimeterelectronicsystemintroducesfixedbiasesthatcanbedeterminedduringgroundcalibration.Perturbationsduetoe.g.differentialheatingaroundtheorbit,causecyclicandlong-termdrift,andduringthemissionareestimated using data from periodic on-board calibrations. The internal path delay correction is acombinationofbothafixedpartandfrequentcalibrations.TheinternalpathdelayiscomputedusingaspecificcalibrationcalledCAL1,whichisperformedabout10timesperweek,andperformedforeachSIRALmode.Physically thepathdelay is thesame forLRMandSARmodes.HoweverLRMdataareprocessedonboardwhereasmallshiftisaddedequalto1/64rangebin≈7mm.SotheLRMpathdelayisdifferentfromtheSARone,itisactuallysmallerby1/64rangegate:thisdoescompensatetheshiftintroducedbytheLRMon-boardprocessor.ThevaluesareprovidedintheL1Bproducts.However,theL1Bproductcontainsinadditionanothercomponentofthepathdelayrelativetothewaveguides,whichisaconstantvalue,measuredbeforelaunchandcommontoallmodes.

Bothantennaeneedseparatecorrectionforinstrumenteffectsonrangeinstr_cor_range_rx_20_kuand instr_cor_range_tx_rx_20_ku and on gain instr_cor_gain_rx_20_ku andinstr_cor_gain_tx_rx_20_ku.

2.3.1 AutomaticGainControlCryoSat-2usesautomaticgaincontrol(AGC),whereinformationabouttheprevioussignallevelisusedtoadjustthegaininanticipationofthenext.Theaimistokeepthesignallevelasconstantaspossible.Theactualsettingoftheon-boardreceiverattenuatorattheobservationtimeiswhatisrecordedinthedataagc_ch1_20_kuandagc_ch2_20_ku.TheAGCisconstantwithintheradarcycle.

2.3.2 SignalPhaseCorrectionsSignalphasecorrectionsaresplitintointernalandexternaltypes,internalfortheelectronicsystemandexternal for antenna subsystem andwaveguide effects. Signal phase is used in SINmode only. Thecorrections are derived from in-flight calibration data. The internal phase correctioninstr_int_ph_cor_20_kuistheinternalphasecorrectioncomputedfromtheCAL-4packetsduringtheazimuthimpulseresponseamplitude(SINonly).ItissetfromthelatestavailableCAL-4packet.Theexternalphasecorrectioninstr_ext_ph_cor_20_kuisanexternalphasecorrectiontakenfromtheIPF database file (SIN mode only) to be added to the internal phase correction term. This is thetemperature-averaged component of external inter-channel phase difference derived from phasedifferencesensitiveantennasubsystem,waveguidesandinstrumentwaveguideswitches.Itdoesnotcontaininternalinstrumenttemperaturedependenteffectsofcalibrationcouplerandduplexer,whicharedealtwithbytheCAL-4signalanditssubsequentprocessingandisreportedintheinternalphasecorrectionparameter.

2.3.3 PhaseSlopeCorrectionAcrossthewholebandwidththereisaconstantchangeinphasedifference.Someofthiswasmeasuredbeforelaunch;therestiscalibratedat1Hz.Bothmeasurementsarecombinedtogivethephaseslopecorrectionph_slope_cor_20_ku.Sincethehighestdatarateis20Hz,theclosestcorrectionvalueintimetothatofthesampleistheoneused.



2.3.4 HammingWeightingFunctionforSARAzimuthFFTAhammingwindowoflengthN=”numberofechoesinburst”(always64forSARandSIN)isgeneratedonceatthebeginningoftheSARprocessing,andappliedinazimuthdirection,toallsamplesofallechoesofeveryburstattheverybeginningofthebeam-formingstep.

Thewindowingprocessisperformedwiththefollowingparametersfromtheprocessorconfigurationfile:

• Apply_Azimuth_Hamming:whichdefineswhetherornotapplythewindow.• Azimuth_Hamming_c1andAzimuth_Hamming_c2:whichareusedtochangetheshape

ofthewindowo H(x)=c1+c2(cos((πx/N)–(π/2)))2

• IntheIPF1processorreleaseVK1.0(February2012)theaboveparametersaresetto:o Apply_Azimuth_Hamming:ono Azimuth_Hamming_c1:0.08o Azimuth_Hamming_c2:0.92

forbothSARandSINspecializedprocessors.Whenthehammingwindowisapplied,thepowerisnotcompensatedbyany factor.Seebelowtheazimuth impulseresponseofaCAL1SARburst,withandwithouthammingwindow(zero-padding=64).

Figure9:ComparisonofAIRwithandwithoutHammingwindow



Figure10:ComparisonofAIRwithandwithoutHammingwindow(fullrange)



Figure11:Hammingweightingfunction

Themainbeamwideningfactorisabout1.5sotheazimuthresolutionisdecreased.Thepeakpowerdecreaseisabout5.4dBbutthisshouldnothaveanyimpactontheL2data.Theweightingfunctionaimistoreducethesidelobes.Thisalwayshasthesideeffectsofincreasingthemainbeamwidth(whichmeansa lossinresolution),anddecreasingthemainbeampowerlevel.Thereareseveralweightingfunctions.Hamminghasthecharacteristicofmakingthefirstsidelobesverylow,reducedbyabout-42dB.Hammingiscurrentlyusedintheprocessor,inazimuthprocessingbeforebeamforming(64beamsarecreatedfromeachburstof64waveforms),butthiscouldbechangedinthefuture,ifamoresuitableoneisrecommended.Itisneededforspecularsurfaceswherethenadirechoisverystrong,andbeamspointingoffnadircontainastrongenergyfromnadir,whichistheclutterinthiscase.Thiscanbeseenin the following four figures forsea ice,showinghowthenadirclutter, removedbywindowing,canimpacttheL1Bwaveformshapeandmainlytheleadingedge.



Figure12:AzimuthFFTofasingleburst:thenadirclutteristhe'line'inthemiddle

Figure13:AzimuthFFTofasingleburstwithhamming:thenadirclutterisreduced



Figure14:Stack60withouthamming:heretheparabolaisthenadirclutter

Figure15:Stack60withhamming:thenadirclutterhasdisappeared

NotethattheL1Bwaveformisobtainedaveragingallthesebeamsinthestack.



2.3.5 NoiseNoiseontheechosignalisalsomeasuredduringin-flightcalibrationnoise_power_20_ku.Atlevel1b,anestimateof thenoise isgivenas thenoisepowermeasurement,but it isnotsubtracted fromthewaveforms.At level2,noise is re-estimatedaspartof the retrackingcalculations. InSINmode, thisestimateissubtractedfromthebinspowerbeforetheretrackingisdone.

2.3.6 EchoSaturationEchosaturationoccurswhenthepowerinthepulsevariesmorethanthedynamicrangeofthereceiver,causingdistortionsintheechowaveform.Thismostlyoccursinhigh-amplitudeechoesfromflatbrightsurfacessuchasleadsinseaiceormeltpondsonicesheets.ThisconditionisflaggedintheMCDwordflag_mcd_20_kuusingbitecho_saturated

2.3.7 DopplerCorrectionWhenthereisacomponentofsatelliteorbitalvelocityvectordirectedalongthelineofsightfromthesatellitetotheechoingpointonthesurface,theradarechoundergoesacorrespondingDopplershift.Thisiscompensatedbyarangecorrectiondop_cor_20_ku.Allrangecorrectionsaredefinedsuchthattheywereaddedtotherangemeasurementtogivethetruerange.

2.4 RetrackingandParametersDerivedFromEchoShape

2.4.1 RangeWindowandWindowDelayThedistancebetweenthesatelliteandtheEarthsurfacevariesarounditsorbitbyseveralkilometres.However, in the vertical dimension the altimeter can only receive echoeswithin a 'rangewindow'specifiedbythewidthofthefrequencyspectrum.Inordertoreceiveechoesallaroundtheorbit,thetimeofthereceptionperiodmustbecontinuouslyadjustedtotrytokeepaspecificpointontheleadingedgeoftheechowaveformatareferencepositionwithinthewindow.TheCryosat-2referencepositionisconfigurable,butitisfornowcentrallylocated,atbin64oftherangewindowinLRMmode,bin128in SAR mode, and at bin 512 in SIN mode. Bins are counted from bin 0. The window delaywindow_del_20_kureferstothe2-waytimebetweenthepulseemissionandthereferencepointatthecentreoftherangewindow.

2.4.2 ReceptionPeriodInformationaboutthereceptionperiodtimeisrepresentedintwoways.Eitheraninitialheightisgivenh0_applied_20_ku, followed by the rate of change in this height per tracking cyclecor2_applied_20_ku, or a coarse height h0_lai_word_20_ku and fine height adjustmenth0_fai_word_20_ku aregiven,whichmustbeadded together.The trackingcycle isa fundamentalintervalcommontoallmodes,of47.2ms.

2.4.3 RangeWindowSamplingInLRM,therangewindowexploredisabout60m,coveredby128datasamples.InSARmodeitisabout60m,againcoveredby256datasamples.InSINmode,duetotheslopevariationinicesheetmargins,



therangewindowisincreasedtoabout240m,andiscoveredby1024samples.Duringprocessingtheexact,calibrated,lengthofthedatabinsisretrievedfromtheinstrumentcharacterisationdatafile,thefilenameofwhichisfoundinthedatasetdescriptionheaderwithinthedatafile,asdescribedinsection6.5.

TherearetwotypesofwaveformsprovidedintheLRM,SARandSINlevel-1bproducts.

1.WaveformsderivedfromclassicalLRMorthosegeneratedfromSAR/SINazimuthprocessingthatareprovidedinthewaveformgroupofthelevel-1bproducts(L1BProductSpecTable2.3.4-1fields71,76and82).

2. Pulse-width limited data extracted from all modes averaged over 1 second and provided in theaveragewaveformgroupofthelevel-1bproduct(L1BProductSpecTable2.3.4-1fields56and66).

Itisafactthatradaraltimeterpowerechowaveformsarealiased(see“Radaraltimetergatetracking:theoryandextension”,J.R.Jensen,IEEETransactionsonGeoscienceandRemoteSensing,Vol37,No2,March1999,sectionIIIB.)

DuringCryoSat-2commissioningphase,itbecameclearthatSARprocesseddataoverspecularsurfaceswasaliasedandthiswasparticularlyevidentwhenthesatelliterateofchangeofaltitudewaslarge.TheresultrenderedSAR(andSIN)level-1bdatahighlydegradedoverspecularsurfaces.

Ingeneral, thealiasingeffectappearsminoroverdiffuse surfaces suchasoceanbut isneverthelesspresent.Inordertoavoidaliasing,rawcomplexSARandSINechoesarenowoversampled(since01February2012,productsversionB)intheFBRtolevel-1bprocessorandasaresultanyuserofthedataneedstotakethisintoaccountwhenusingwaveformsanddeterminingrange.However,althoughthesamplinghaschanged,theresolutionhasnotsinceitisfixedbytheinstrumentimpulseresponse.

LRMsamplingremainsthesameasinearlierversionsoftheCryosat-2processorsassamplingisfixedon-boardandoversamplingof theon-boardcomplexdata isnotpossible.However, theLRMdata isaffectedbyaliasingasinotheraltimetermissions.

Whatfollowsisadescriptionofhowtocomputerangetoagivenrangebinforallcases.

ForALLLRM20Hz,ALL1Hz,andforALLBaseline-Adata,andforanydatagatheredbefore01-FEB-2012,use

Giventhewaveformarray, (pwr_waveform_avg_01_kuorpwr_waveform_20_kuwhere isthewaveformsamplenumberwithavalue (anyvaluebetween0and )and isthemodedependentnumberofsamplesinanywaveform:

€

Φ n( )

€

n

€

∀n∈[0,Ns −1]

€

Ns −1

€

Ns

Ns =

128, LRM128, SAR512, SIN

!

"#

$#



Therange towaveformsample isgivenby meters

where isthewindowdelayconvertedtosecondswindow_del_avg_01_kuorwindow_del_20_ku

referencedtothecentralrangebin ,thefirstrangebinbeing ,

is the speed of light in vacuum, c=299792458 m/s, is the measured chirp bandwidth,B=320000000Hz.

Hence,therangetothefirstwaveformsample, , meters.

Eachwaveformsamplecoversarangeof oradelayof

ForSAR&SIN20HzoversampledechoesinproductsversionB,gatheredafter01-FEB-2012

Given the oversampled power echowaveform array, pwr_waveform_20_ku where is thewaveformsamplenumberwithavalue (anyvaluebetween0and )and isthemodedependentnumberofsamplesinanywaveform:

Therange towaveformsample isgivenby [m]

where is thewindowdelayconverted tosecondswindow_del_20_ku referenced to thecentral

rangebin ,thefirstrangebinbeing ,

is the speed of light in vacuum, c=299792458 m/s, is the measured chirp bandwidth,B=320000000Hz.

Hence,forexample,therangetothefirstwaveformsample, , meters.

Eachwaveformsamplecoversarangeof oradelayof

ForSAR&SIN20HzoversampledechoesinproductsversionCandlater,gatheredafter01-FEB-2012

Asabove,butforSARmode𝑁3 = 256,andforSARinmode𝑁3 = 1024.

€

R n( )

€

n

€

R n( ) =Twc2

−Nsc4B

+nc2B

=c2B

TwB −Ns

2+ n

#

$ %

&

' (

€

Tw

€

n = Ns2

€

n = 0

€

c

€

B

€

n = 0

€

R(0) =Twc2

−Nsc4B

€

Δr =c

2B= 46.84 cm Δt = 1

B= 3.125 ns

€

Φ n( )

€

n

€

∀n∈[0,Ns −1]

€

Ns −1

€

Ns

Ns =128, SAR512, SIN

!"#

$#

€

R n( )

€

n

€

R n( ) =Twc2

−Nsc8B

+nc4B

=c8B

4TwB − Ns + 2n( )

€

Tw

€

n = Ns2

€

n = 0

€

c

€

B

€

n = 0

€

R(0) =Twc2

−Nsc8B

Δros =c

4B= 23.42 cm Δtos =

12B

=1.5625 ns



2.4.4 EchoPositioningandScalingInthedataproductsthepositionoftheechowithintherangewindowvarieswithSIRALmode.TheinstrumentdynamicallysetstherangewindowandinLRMthewindowpositionisnotchangedbythedataprocessing.InSARandSINmodesechoesareaveragedduringdataprocessingonthegroundandthepositionofthewindowisthenselectedtobestaccommodatetheresultantwaveform.

Inordertoretainasmuchinformationaspossible,theindividualsamplesofeachechowaveformarescaledtofitwithintherange0to65535.Theycanbeconvertedtopowerinwattsusingequation4.2-1with the parameters echo_scale_pwr_20_ku and echo_scale_factor_20_ku (orecho_scale_pwr_avg_01_kuandecho_scale_factor_avg_01_ku)

2.4.5 RetrackingAtlevel2,aprocedurecommonlyreferredtoasretrackingisperformed.Aspecificpointontheecho'sleadingedge, knownas the retrackingpoint, is used tomark thepointofmeasurementof range tosurface.Theretrackingpointisdefinedrelativetotheshapeofthewholeechoandfoundusingamodel-fittingmethod.Theoffsetoftheretrackingpointfromareferencepoint,whichforCryosat-2isinthecentreoftherangewindow,isthencalculated.Thisistheretrackingcorrection.

Thealgorithmsusedforretrackingvarywithmode,seeTable2below.ForLRMandFDM(anysurfacetype),theamplitudeandepochreturnedbytheOCOGretrackperformeduponthewaveform,areusedastheinitialguessesforthefitsperformedbytheCFIandland-iceretrackers.Theprimaryresultsarereturnedinthevariablesheight_1_20_ku,height_1_20_ku,andheight_1_20_ku.

Table2:L2retrackers

RetrackerID ModeLRM ModeSAR ModeSARin1 OceanCFImodelfit Diffuseecho–CPOMthresholdoffirstpeak Wingham/Wallismodelfit

Specularecho–Gilesmodelfit2 UCLland-ice notused notused3 OCOG notused notused

Furtherdetailsaregiveninreference18,andbackgroundinformationinreferences4and19,however(inbrief)theCLSoceanretrackerandUCLland-iceretrackersaremodel-fitstotheLRMwaveform.TheoceanretrackerfitsaHaynemodelwaveformthatisnotadaptedforCryoSat2andtheUCLretrackerfitsaBrownmodeladaptedforCryoSat2.OCOGiswelldescribedintheliterature.

ForSARmode,theGilesretrackerisamodelfitforspecularleadwaveforms,andathresholdretracktothe first peak of a smoothed waveform for diffuse echoes from sea-ice. In SARin mode, theWingham/Wallisretrackerfitsatheoreticalmodeltotheportionofthepowerwaveformthatexhibitsthemaximumcoherence.Thephasedifferenceisestimatedattheretrackingpointtogivetheacross-track angle. The SARmode retracking is also performed on the SARinwaveforms,with the resultsappearinginheight_sea_ice_floe_20_ku(diffuse)andheight_sea_ice_lead_20_ku(specular)



Figure16:Idealisedpulse-limitedwaveforminwindow

In the L2I, the variables retracker_1_cor_20_ku, retracker_1_cor_20_ku, andretracker_1_cor_20_kuholdthecorrectiontorangeappliedduetotheretrackingforeachofthe(upto)threeretrackers.

ItisnotpossibletolocatetheretrackingpointwiththeL2productsonly.OnewouldneedtouseL1BandL2intandemandinordertoachievethis:

1)ConvertL2heighttorangeusingthealtitudeat20HzfromtheL1B

2)Removethegeophysicalcorrections

3)RemovetheL1Brange(thewindowdelayconvertedtorange)

Overocean,whatisleftfollowingthestepsaboveistheactionoftheretracker.Thisisslightlymorecomplicatedovericeduetotheactionoftheslope.

The L2I also contain the range with only retracking and instrumental corrections applied (nogeophysicalcorrections)inthevariablesrange_1_20_ku,range_1_20_ku,andrange_1_20_ku.Thisallowstheusertoapplytheirowngeophysicalcorrections.



2.4.6 BackscatteringTheradarbackscatteringcoefficient(sigma-0)providesinformationabouttheobservedsurface.Itisafunction of the radar frequency, polarisation and incidence angle and the target surface roughness,geometric shape and dielectric properties. A discussion of all these factors is given online athttp://earth.esa.int/applications/data_util/SARDOCS/spaceborne/Radar_Courses/Radar_Course_III/parameters_affecting.htm

Atlevel2,thebackscattercoefficientisfullycorrected,includinginstrumentgaincorrectionsandbias.Ameasurementismadeforeachretrackersig0_1_20_ku,sig0_2_20_ku,andsig0_3_20_ku.

2.4.7 PeakinessPeakinessisameasureofhowsharplypeakedanechoispeakiness_20_ku.Itisessentiallytheratioofthehighestbinvaluetothemeanvalueofallbinsabovetheretrackingpoint.Thehighertheratio,themorepeakedtheecho.Highpeakinessindicatesaveryspecularreflection,suchasthatfromleadsinseaice.

Furtherinformationisgiveninreference18.

Figure17:Peakinessbylatitude

Aboveshowsthetotalnumberofpointsavailableperdegreelatitude,between140Wand180WforMarch 2011 alongwith the number of pointswith ‘Pulse Peakiness’ (PP) above certain thresholds.HigherPPindicatesmore‘specular’waveformsi.e.moreliketooriginatefromleads.ThetotalnumberofCryosat-2measurementsbetween75Nand88N(between140Wand180W)is515000.

0100002000030000400005000060000700008000090000

75.5 76.5 77.5 78.5 79.5 80.5 81.5 82.5 83.5 84.5 85.5 86.5 87.5Center latitude (deg N)

Number of points per degree latitude between 140W and 180W, March 2011

Total points

Points PP>0.15

Points PP>0.20

Points PP>0.25



BelowshowsthenumberofpointswithPPgreaterthancertainthresholdsasaproportionofthetotalnumber of points recorded, per degree latitude, forMarch 2011. The number of points varieswithlatitudeaccordingtotheproportionofseaicecover,aswellas,presumably,climatologicalconditions.

Figure18:Proportionofspecularpointsbylatitude

2.4.8 EffectsofTerrainSincethetypeofsurfaceaffectstheradarechoshape,adescriptionofthesurfaceatnadir,derivedfromthe TerrainBase digital terrain model from the NOAA, is also given surf_type_01, andsurf_type_20_ku.Seealsosection2.7.2andreference8.

2.4.1 InterferometricModeThe purpose of the SIN mode is to allow the across-track angle offset of the echoing point to bedetermineddirectly(using interferometricphase information). IntheL2SINdata, the20Hz latitudelat_poca_20_kuandlongitudelon_poca_20_kuvaluesprovidetheestimatedpositionofthepointofclosestapproach(ascomputedfromthephasedifference)andnotthenadir.ThenadirlocationisonlypresentintheL2dataat1Hzlat_01andlon 01.Each20Hzsampleproducesonlyonelocationontheground.

Itisimportanttonotethatphase-wrappingcanoccurwhentheacrosstrackoffsetisgreatenoughandthiscanhavetheeffectoftheechoappearingtocomefromtheothersideofthegroundtrack.Normally,this is flagged flag_sarin_ambiguity_warning_20_ku by use of an ambiguity DEMdem_height_20_kuthatshowswherethiseffectislikelytohaveoccurred,howevertheseDEMsareonlyavailableforGreenlandandAntarctica.Iftheuserobservesasequenceofrecordswithincreasing

0%

5%

10%

15%

20%

25%

75.5

76.5

77.5

78.5

79.5

80.5

81.5

82.5

83.5

84.5

85.5

86.5

87.5

Center latitude (deg N)

Proportion of 'specular' points per degree latitude between 140W and 180W, March 2011

Points PP>0.15

Points PP>0.20

Points PP>0.25



across-track offset and then an abrupt switch to the other side of the track, it is likely that phase-wrappinghasoccurred.

2.4.2 SignificantWaveHeightSignificantwaveheight(SWH)swh_ocean_20_ku istheaveragewaveheight,troughtocrest,oftheone-thirdlargestwavesinaparticulargeographiclocation.SWHcanbederivedfrominspectionoftheecho waveform. The reflection from the wave crest is returned earlier than that from the trough,'stretching' the leadingedgeof thewaveform.Thestretching increaseswith increasingwaveheight.Althoughnotphysicalbutonlynumerical,negativevaluesofSWHarepreserved.

2.4.3 SeaStateBiasSeastatebiassea_state_bias_20_kuisacombinationofeffectsdependentonthewaveheightatthesurface. The electromagnetic (EM) bias effect arises because wave troughs are better reflectors ofenergy thanwavecrests,due to theirdifferentdistributionsofspecular-reflecting facets.This is thecorrectiongivenunderthetitleofseastatebiasinthelevel-2data.SeastatebiasisderivedfromamodelprovidedbyCLStoESA.Forageneraltreatmentofseastatebiasseereference14.

2.4.4 WindSpeedThe Abdalla2007 model is used to calculate wind speed wind_speed_alt_20_ku from the radarbackscatteringcoefficient.Seereference8.

2.5 RangeCorrectionsforAtmosphericEffectsAnaltimeterpulseisslowedslightlyasitpassesthroughtheEarth'stroposphereduetotherefractiveindexof the atmosphere.The conversionof timedelay to rangeusing the speedof light in vacuumthereforehastobecorrectedforthissmalladditionaldelay.Thespeedoflightusedintheprocessoris299792458.0metrespersecond.

2.5.1 DryandWetTroposphericCorrectionsThe dry tropospheric correction mod_dry_tropo_cor_01 compensates for the effect of non-polargasessuchasoxygenandnitrogen.Ithasatypicalrangefrom1.7to2.5m.

Thewettroposphericcorrectionmod_wet_tropo_cor_01compensatesfortheeffectofpolargases,mainlywatervapour.Ithasatypicalrangefrom0to50cm.

The dry and wet corrections rely on knowledge of the atmospheric pressure and temperatureobservationsmadeon theEarth surface.Theseare suppliedbyMeteoFrancevia theCNESSSALTOsystem,asdescribedinsection2.8andreference8.

TheWetTroposphericCorrectionisretrieveddirectlyfromECMWFanalysedgrids.Thesecorrectionfiles are then simply formatted to theCryosat-2PDS file standardbeforebeingdirectlyused in theprocessor.TheDryTroposphericCorrectiondoesnotcomefromanauxiliaryfilebutiscomputedbytheprocessorusingECMWFsurfacepressurefiles.Overwater(openwater,semi-enclosedseasorenclosed



seasandlakes),thesurfacepressureisequaltothemeanpressureminustheclimatology(computedwiththepressuregrids).PhasesandamplitudesofS1andS2tidewavesprovidedasStaticAuxiliaryDatafilesarealsotakenintoaccount.Overland,theconsideredsurfacepressureisthemeanpressuresotheDryTroposphericCorrectionoverlandiscomputedwiththissolemeanpressure.

2.5.2 InverseBarometricCorrectionTheinversebarometriccorrectioninv_bar_cor_01compensatesforvariationsinseasurfaceheightduetoatmosphericpressurevariations,whichisknownasatmosphericloading.Ithasatypicalrangefrom-15to+15cmandiscalculatedfromdataprovidedbyMeteoFranceviatheCNESSSALTOsystem,asdescribedinsection2.8andreference8.ThiscorrectionisonlyusedinSARmodeandwherethesurfacetypeis“openocean”.

2.5.3 DynamicAtmosphericCorrectionThe dynamic atmospheric correction hf_fluct_total_cor_01 compensates for variations in seasurfaceheightduetoatmosphericpressureandwinds.Ithasatypicalrangefrom-15to+15cmandistakenfromtheMOG2DmodeldataprovidedbyMeteoFranceviatheCNESSSALTOsystem,asdescribedinsection2.8andreference8.ThiscorrectionisusedbySIN,SID,LRMandFDMwhenthesurfacetypeis“openocean”.

2.5.4 IonosphericCorrectionThe ionospheric correction, also called the ionospheric bias correction, compensates for the freeelectrons intheEarth's ionosphereslowingtheradarpulse.Solarcontrolof the ionosphere leadstogeographicandtemporalvariationsinthefreeelectroncontent,whichcanbemodelled,ormeasured,forexampleusingtheGPSsatellitenetwork.Thecorrectionhasatypicalrangefrom0.06to0.12m.Thelevel-1b data product contains values derived from the Global Ionospheric Map (GIM)iono_cor_gim_01,whichusesGPSmeasurements,andfromtheBentmodeliono_cor_01.TheGIMisthenominalchoicewiththeBentmodelavailableasaalternativesolutionifGIMdataisnotavailable.TheGIMisdescribedonlineat:

http://iono.jpl.nasa.gov/gim.html

The Bentmodel is based on knowledge of a solar activity index, such as sunspot numbers, and isdescribedonlineat:

http://modelweb.gsfc.nasa.gov/ionos/bent.html

Thesolaractivityindexused(theR12index)isissuedmonthlyandisdefinedinreference5.

2.5.5 DerivationofCorrectionsMostoftheabovecorrectionsareprovidedonameteogrid,agroupof6dynamicfilescontaininggridsofsurfacepressure,oceanmean-sea-levelpressure,wetanddrytroposphericcorrectionsandU(east-west)andV(south-north)windcomponents.Inadditionastaticfileforthegriddefinitionisrequired.ThisistheGaussianAltimetricCorrectionGrid.Ameteogridcoversatimespanof6hours,andtwo



meteogrids,onebeforeandoneaftertherequesteddateforthecorrection,areneeded.Thegridsareinterpolatedintimeandspacetoderiveacorrectionvalueforeachmeasurement.

Thesefilesarefurtherdefinedinreference8.

2.5.6 ApplicationofCorrectionstoRangeTheabovecorrectionsareaddedtotherangemeasurementtogivethetruerange.

Furtherbackgroundinformationisgiveninreferences9and10.

2.6 RangeCorrectionsforTidalEffectsAlltidalcorrectionsadjusttherangetoappearasifitcamefromthemeanlandicesurface,ortide-freeseasurface.ThefieldTotalGeocentricOceanTide(field#55)intheFDMproductsisnotthesameastheOceanTidefieldinthenominalL1BandL2products.WiththeFDMproducts,thesameconventionasmanyotheraltimetryproducts, suchas forEnviSat, Sentinel-3and Jason-2, isused,where the totalgeocentric ocean tide provided in the products includes the loading tide and the long period tide.HoweverinthenominalL1Bproducts,the“oceantide”fielddoesnotincludetheloadingtideorthelongperiodtide.FortheL2product,theusersdonotneedtoapplyanytidesastheyareappliedduringthecomputation of height. However, if they areworkingwith the L1B product and deriving their ownsurfaceheightthentheyshouldapplythesetoftidesandmeteorologicalcorrectionsintheL1Bthatissuitablefortheirspecificusage.

2.6.1 OceanTideTheoceantidecorrectionocean_tide_01removestheeffectsof localtides i.e. thosecausedbytheMoon,andthelong-periodequilibriumoceantidecorrectionocean_tide_eq_01removestidaleffectsduetotheSun.Typicallythiscorrectionrangesfrom-50to+50cm.TheFES2004oceantidemodelisusedforthiscorrection.Thisisfurtherdescribedinsection2.8andreference8.

2.6.2 OceanLoadingTideTheoceanloadingtidecorrectionload_tide_01removesthedeformationoftheEarth'scrustduetotheweightoftheoverlyingoceantides.Typicallythiscorrectionrangesfrom-2to+2cm.TheFES2004loadingtidemodelisusedforthiscorrection.Thisisfurtherdescribedinsection2.8andreference8.

2.6.3 SolidEarthTideThesolidEarthtidecorrectionsolid_earth_tide_01removesthedeformationoftheEarthduetotidalforcesfromtheSunandMoonactingontheEarth'sbody.Typically,thiscorrectionrangesfrom-30to+30cm.TheCartwrightmodelisusedforthiscorrection.Itisfurtherdescribedinsection2.8andreference8.

2.6.4 GeocentricPolarTideThegeocentricpolar tide correctionpole_tide_01 removesa long-perioddistortionof theEarth’scrust caused by variations in centrifugal force as the Earth’s rotational axis moves its geographic



location.Typically,thiscorrectionrangesfrom-2to+2cm.Itisderivedfromthepolelocationfiles.Itisfurtherdescribedinsection2.8andreference8.

2.6.5 ApplicationofCorrectionstoRangeTheabovecorrectionsareaddedtotherangemeasurementtogivethetruerange.

Forfurtherinformationseereferences9and11.

2.7 DerivingRangeThewindowdelayismeasuredfromthesignalemissiontimetothecentreoftherangewindow.Atlevel1btheengineeringcalibrationsarealreadyappliedtothewindowdelay,astheDopplercorrectionandtheCOMoffset. Thus, the rangemeasurementhas its origin atCryoSat-2COM (asdoes the altitudemeasurement),andendsat thecentreof therangewindow,point(b) in fig2.5.5-1.Thegeophysical(atmospheric and tidal corrections) are not applied at level 1b. Range can be calculated using theparametersinthedataproduct,takingintoaccountthesurfacetype.

Thefollowingequationisapplicable:

range=(0.5*c*windowdelay)+retrackingcor.+surface-dependentgeophysicalcor. (Eqn2.7-1)

wherecisthespeedoflightinvacuum.

AtL2,20measurementshavebeencorrectedusingthe1Hzcorrections.Nofurtherinterpolationisdoneandthesame1Hzcorrectionisusedforall20measurementsasthecorrectionsareslowlyvarying.The20Hzand1Hzmeasurementscanbeassociatedviatheindexvariablesind_first_meas_20hz_01andind_meas_1hz_20_ku.

2.7.1 GeophysicalCorrectionsbySurfaceSurface height w.r.t. reference ellipsoid = altitude - (retracked range + corrections), where thegeophysicalcorrectionsnecessaryovereachsurfaceareasfollows:

• SeaIce:o Tides:

§ Oceantide§ Longperiodequilibriumoceantide§ Oceanloadingtide§ SolidEarthtide§ Geocentricpolartide

o Atmosphere:§ Drytroposphericcorrection§ Wettroposphericcorrection§ Ionosphericcorrection



§ Inversebarometriccorrection

• LandIce:o Tides:

§ Oceanloadingtide§ SolidEarthtide§ Geocentricpolartide

o Atmosphere:§ Drytroposphericcorrection§ Wettroposphericcorrection§ Ionosphericcorrection

• Ocean:

o Tides: § Oceantide§ Longperiodequilibriumoceantide§ Oceanloadingtide§ SolidEarthtide§ Geocentricpolartide

o Atmosphere:§ Drytroposphericcorrection§ Wettroposphericcorrection§ Ionosphericcorrection§ Dynamicatmosphericcorrection

o Other: § Seastatebias

2.7.2 SurfaceTypeNotethatalthoughaflagforsurfacetypeexistsatbothL1bsurf_type_01andL2surf_type_20_ku,atL1bitisonlyprovidedat1Hzresolution,whichislowerthanthe20Hzmeasurementresolution.Usersmustbecarefulininterpretingthisflaginboundarieswhenselectingtherangecorrectionsneeded.TheDTM2000staticgridprovidesfourstatesoftheflag:0=openoceansorsemi-enclosedseas;1=enclosedseasorlakes;2=continentalice;3=land.

2.7.3 Level-2ApplicationAt level 2, altitude and height are given, as well as uncorrected range. All corrections are alreadyincorporatedintheheightmeasurement.Theyarelistedinthedatasetsothattheycanberemovedorreplacedifrequired.



2.8 ModelsUsed

2.8.1 StaticAuxiliaryDataFilesThistableliststhestaticauxiliarydatafilesusedinprocessingthedata.Thesecontainthemodelsused,aslistedinreference8fordataprovidedviatheEarthExplorerGeoCorrectionsCFI.

Theslopemodeldata(AUX_SLPMSL)iscreatedfromthefollowingDEMs:

• Helm,Veit;Humbert,Angelika;Miller,Heinz(2014):ElevationModelofAntarcticaderivedfromCryoSat-2 in the period 2011 to 2013, links to DEM and uncertainty map as GeoTIFF.https://doi.org/10.1594/PANGAEA.831392

• Helm,Veit;Humbert,Angelika;Miller,Heinz(2014):ElevationModelofGreenlandderivedfromCryoSat-2 in the period 2011 to 2013, links to DEM and uncertainty map as GeoTIFF.https://doi.org/10.1594/PANGAEA.831393

TheambiguityDEMs(AUX_DEMMSL)arecreatedfrom:

• Bamber,JonathanL.,JoseLuisGomez-Dans,andJenniferA.Griggs.2009.Antarctic1kmDigitalElevation Model (DEM) from Combined ERS-1 Radar and ICESat Laser Satellite Altimetry.Boulder,ColoradoUSA:NationalSnowandIceDataCenter

• HowatI.M.,A.NegreteandB.E.Smith,2014,TheGreenlandIceMappingProject(GIMP)landclassificationandsurfaceelevationdatasets,TheCryosphere,8,1509-1518

Filesof'grid'typehavedatadistributedonageographicalgrid.Filesof'data'typearenon-geographical.

Table3:Staticauxiliaryfiles

TypeString

(inFilename)

Description Type Parameters(gridonly)

Alllongitudecoverageis[0,360]degrees

AUX_GEOID_ EGM96geoidvalues grid Latcoverage[-88,88]degrees

Stepinlatitude:0.25degrees

Stepinlongitude:0.25degrees

AUX_MICOEF Bent ionospheric coefficientsfile

data

AUX_DIPMAP

BentModifiedDipMapfile grid Lat.coverage:[-88,88]degrees

Stepinlatitude:2degrees

Stepinlongitude:2degrees



AUX_MSSURF

UCL13MeanSeaSurfacevalues grid Lat.coverage:[-80,88]degrees



AUX_SDC_nn

UCL04SnowDepthClimatologymodel(nn=monthnumber)

grid Lat.coverage:[0,90]degrees



AUX_SICCnn

UCL04 Sea Ice ConcentrationClimatology (nn= monthnumber)

grid Lat.coverage:[-90,90]degrees



AUX_SLPMSL MSSL High resolution Slopemodels of Greenland andAntarctica

grid CartesiangridscoveringAntarctica,grid step = 1km, and Greenland,gridstep=800m

AUX_DEMMSL MSSL high-resolution DEMs ofGreenlandandAntarctica

grid CartesiangridscoveringAntarctica,grid step = 1km, and Greenland,gridstep=800m

AUX_OCTIDE

FES2004 ocean tide modelvalues




AUX_TDLOAD

FES2004 loading tide modelvalues




AUX_ODLE__ MACESS Ocean Depth/LandElevationvalues

grid Lat.coverage:[-90,90]

StepinLatitude:1/30degrees

Stepinlongitude:1/30degrees

AUX_CARTWR CartwrightandEddentables data

AUX_LS_MAP

Four states surfaceidentification grid created byCLS/CNES from GMT,GlobCover, Modis Mosaic of

grid Lat.coverage:[-90,90]

StepinLatitude:1/30degrees

Stepinlongitude:1/30degrees



Antarctica, and Water bodyoutlinesfromLEGOS.

AUX_PRSS00

AUX_PRSS06

AUX_PRSS12

AUX_PRSS18

Climatologypressuregrids,fourperday,for0h,6h12hand18h.Each grid point contains 12values of climatologicalpressure,oneforeachmonthoftheyear

grid Latcoverage:[-90,90]degrees



AUX_S1AMPL

AUX_S2AMPL

The S1 and S2 tide grids ofmonthly means of globalamplitude. Each grid pointcontains12valuesofamplitude,oneforeachmonthoftheyear




AUX_S1PHAS

AUX_S2PHAS

The S1 and S2 tide grids ofmonthlymeansofglobalphase.Each grid point contains 12values of phase, one for eachmonthoftheyear.




AUX_WNDCHE Abdalla2007windspeedtable data Sigma0coverage:[5,19.6]

StepinSigma0:0.2

2.8.2 DynamicAuxiliaryDataFilesThedynamicauxiliarydatafilesusedaremainlysuppliedbySSALTO,SystèmeauSold'ALTimétrieetd'Orbitographie.Theyhaveawebpresenceat:

http://www.aviso.oceanobs.com/es/data/product-information/ssalto/index.html

Theysupply:

• theDORISUSOdriftcorrection• themeteorologicalfiles,sourcedfromMeteo-FranceviaSSALTO,basedondatafromECMWF• thehistoricalpolelocationfiles,usedtoderivethegeocentricpolartide• thesolaractivityR12indexfiles,usedintheBentmodelionosphericcorrections• thedailyGIMfiles(GIMistheGlobalIonosphericMap)• thedailyMOG2Dfiles(MOG2Disabarotropicmodelusedfordynamicatmosphericcorrection)

Thedynamicsea-iceconcentrationfilesareprovidedbyUniversityCollegeLondon,andareonlineathttp://cryosat.mssl.ucl.ac.uk/dsicTheseconcentrationsarederivedfromtheNRTDMSPSSMIdataofNSIDC.



DORISUSODrift CS_OPER_AUX_DORUSOGPSIonosphericMap CS_OPER_AUX_IONGIM(dailyfile)InstantaneousPolarLocation CS_OPER_AUX_POLLOC(dailyincrement)DynamicSeaIceConcentration CS_OPER_AUX_SEA_IC(eachfilehasavalidityof3days)SolarActivityIndex CS_OPER_AUX_SUNACT(monthlyincrement)SurfacePressure CS_OPER_AUX_SURFPS(dailymeteofilesfor00,06,12,18h)OceanMeanPressure CS_OPER_AUX_SEAMPS(dailymeteofilesfor00,06,12,18h)WEWindComponent CS_OPER_AUX_U_WIND(dailymeteofilesfor00,06,12,18h)SNWindComponent CS_OPER_AUX_V_WIND(dailymeteofilesfor00,06,12,18h)WetTroposphereCorrection CS_OPER_AUX_WETTRP(dailymeteofilesfor00,06,12,18h)Dynamic Meteo Correction fromGravityWaveModel

CS_OPER_AUX_MOG_2D(dailymeteofilesfor00,06,12,18h)

GaussianAltimetricGrid CS_OPER_AUX_ALTGRD(updatedonlywhenthehorizontalspatialresolutionoftheECMWFmodelchanges)



3 Cryosat-2DataSupply

3.1 Cryosat-2GroundSystemThe operational ground system (PayloadData System) is located at the Kiruna receiving station inSweden.Cryosat-2dataproductsaredistributedbyFTPfromthissite.

3.2 Cryosat-2UserProducts

3.2.1 Level-2GeophysicalDataRecordThemainuserproductisthelevel-2GDR(GeophysicalDataRecord),whichfulfilstheneedsofmostscientificresearchers.TheGDRcontainsthetimeofmeasurement,thegeolocationandtheheightofthesurfaceabovethereferenceellipsoid.Theheightisfullycorrectedforinstrumenteffects,propagationdelays,measurementgeometry, andothergeophysical effects suchas atmospheric and tidal effects.AdditionallytheGDRcontainssomemodeandqualityflagginginformationandincludesthegeophysicalcorrectionsthatwereusedtocomputetheheightmeasurement.FormeasurementsmadeinSARmodeafreeboardheight,i.e.theheightbywhichanicefloeextendsabovethewatersurface,iscomputedoversea ice. For measurements made in SINmode the location includes any across-track offset due totopography.Theorbitlocationisalsoincluded.Measurementsarecomputedalongtrackatintervalsofapproximatelyevery300metres.

TheGDRissuppliedinfull-orbitsegmentsstartingandendingattheequator.Thismeansthatallthemeasurementsmadeineachmode,andprocessedbydifferentchains,areconsolidated,incorrecttimeorder,andsharethesamerecordformat.

The frequency of consolidated L2 data records is approximately one per second, but each recordcontainsablockof20higher-ratemeasurements.ConsolidatedL2datafilesarepartitionedintowhole-orbitsegmentscutattheascendingequatorcrossingnode(ANX)andthereforeeachconsolidatedL2productisapproximately4948secondslong.

L2dataarealsoavailablebymode.ThishasexactlythesamefileandrecordformatastheGDRbutthedatafilesaresegmentedbymeasurementmodeandtheirlengthisvariable.

3.2.2 FreeboardProcessingPrevioustobaselineCproducts,twoflavoursofGDRexisted, identifiedbythestringSIR_GDR_2AorSIR_GDR_2Binthefilename.Inthe‘2A’versionthefinalstageofprocessingoverseaicehadnotbeenrunandthefreeboardfieldisfilledwith-9999.The‘2B’versionwasusedtovalidatetheprocessing.

Now the freeboardmeasurements are validated, the identification stringwill switch to SIR_GDR_2_,whichwillbetheonlyflavourofGDRdistributedforbaselineCandlater



3.2.3 SlopeCorrectionIngeneral,overlandicethelevel-2SINandLRMdataareslope-corrected.TheSINslopecorrectioniscalculatedfromtheinterferometricphaseandtheLRMfromadigitalelevationmodel(DEM).Detailsaregiveninreference3.

ForSINmode,theechoinglongitudeandlatitudedifferfromnadiraccordingtotheacross-trackoffsetcomputedfromthephasedifference.InLRMmode,theydifferfromnadiraccordingtotheresultoftheslope-correctionalgorithmandtheslopemodelisusedintheprocessortodeterminewherethefirstreturncamefrom.ThemethodusedforthisissimilartoENVISAT.Pleasenote,inSINmode,theDEMisnotusedtodeterminewheretheechocamefrom.Instead,itisusedtoprovideaflagthatstateswhethertheechomaybeambiguousduetophasewrapping.

The DEM used for LRM slopemodel creation is not the same as the DEM used for SIN ambiguitydetection.

FortheslopemodelintheLRMprocessing,thefollowingDEMsareusedtocreatetheslopemodels:

• Helm,Veit;Humbert,Angelika;Miller,Heinz(2014):ElevationModelofAntarcticaderivedfromCryoSat-2 in the period 2011 to 2013, links to DEM and uncertainty map as GeoTIFF.https://doi.org/10.1594/PANGAEA.831392

• Helm,Veit;Humbert,Angelika;Miller,Heinz(2014):ElevationModelofGreenlandderivedfromCryoSat-2 in the period 2011 to 2013, links to DEM and uncertainty map as GeoTIFF.https://doi.org/10.1594/PANGAEA.831393

FortheambiguitydetectionintheSARinmodeprocessing,thefollowingDEMsareused:

• Antarctica: Bamber,JonathanL.,JoseLuisGomez-Dans,andJenniferA.Griggs.2009.Antarctic1km Digital Elevation Model (DEM) from Combined ERS-1 Radar and ICESat Laser SatelliteAltimetry.Boulder,ColoradoUSA:NationalSnowandIceDataCenter.Digitalmedia.

• Greenland:Howat,I.M.,A.Negrete,B.E.Smith,2014,TheGreenlandIceMappingProject(GIMP)land classification and surface elevation datasetsThe Cryosphere, 8, 1509-1518,doi:10.5194/tc-8-1509-2014

3.2.4 Level-1bDataLevel-1bdataarereceivedbysomeusersinterestedintheSIRALinstrumentperformance,andbysomeinterested in algorithm development. It contains time and geolocation information and the SIRALmeasurementsinengineeringunits.Calibrationcorrectionsareincludedandhavebeenappliedtothewindowdelaycomputations.Signalpropagationdelaysandothergeophysicalcorrectionsareincludedintheproductbuthavenotbeenappliedtotherangeortimedelay.

L1Bdataissegmentedaccordingtomeasurementmodeandeachmodehasaslightlydifferentrecordformattoaccommodatethedifferenceintheechoes.



Inpulse-limited(orLRM)modeaveragedpowerechoesareincludedasforpreviousaltimetermissions.Echoesare128samples(or‘bins’)long.

InSARmodetheechoeswhichresult fromDopplerbeamformationareincludedandconsistof256samples.ThereareadditionalparameterswhichcharacterisethedistributionofenergyacrossthesetofDopplerbeamswhicharedirectedatacommonsurfacelocation.

InSINmodetheechoesresultingfromtheDopplerbeamformationprocessconsistof1024samples.Fromtheinterferometricprocessingthereisalsoanarrayof1024phasedifferences,andanarrayof1024coherenceterms,computedbycomparingtheechoesreceivedbybothantennas.

3.2.5 NearRealTimeDataNear-real-timedataisderivedfromthesciencedatastream.Itcanbeusedforoperationaloceanographyandmeteorology,andtoallowmonitoringofsatellitehealthandsafetybeyondthelimitsofthenormalhousekeeping telemetry. It is sometimes tailored to specific requirements during calibration andvalidationcampaigns,anddetailsmustbeobtainedonacampaignbycampaignbasis.Seereference3.

NearReaTimeproductsderivingfromStarTrackerRefinedproductsareavailableafter24hoursfromthedataacquisition,becauseofthedependencyfromtheStarTrackerRefining.

To allow availability of “real” Near Real Time SAR and SARIN products, a different chain has beendedicatedtothespecificoperationalmodetoallowthegenerationofSAR/SARINNRTproductsinshorttime(2/3hours):theStarTrackerRefinedproductisreplacedbytheStarTrackerLevel0products,availableatthetimeofthedownlink.

Backgroundinformationisgiveninreferences4and12.

3.2.6 FastDeliveryMarineModeDataFastDeliveryMarineMode(FDM)dataproductsareproducedfromLRMdataonly,processedassoonaspossibleafteracquisition.

3.3 FileStructureCryosat-2productsarecomprisedoftwofiles,anXMLheaderfile,alwaysnamedwitha.HDRextension,andaproductfile,alwaysnamedwitha.DBLextension.

Foragenericuser,theDBLfileshouldbeenoughtoperformscientificuseofthedata,withouttheneedofthecorrespondingHDRfile.TheHDRfileisanauxiliaryASCIIfilewhichisdesignedtoallowtheusertoquicklyidentifytheproductwithoutlookinginsidetheDBL,howeveritisnotmandatoryforscientificuseofthedata.

EachCryosat-2productiscomposedoftwofiles:

-XMLHeaderFile(.HDR);thisconsistsofaFixedHeaderandaVariableHeader



-ProductFile(.DBL);thisisinnetCDFformat.

AlloftheinformationprovidedintheFixedHeaderoftheHDRfileislistedbelowandmostofthisisactuallyalsoavailableintheDBLfilesasglobalattributes:

<File_Name>;<File_Description>;<Notes>;<Mission>;<File_Class>;<File_Type>;<Validity_Start>;<Validity_Stop>;<File_Version>;<System>;<Creator>;<Creator_Version>;<Creation_Date>

It should be noted that the fields in the Fixed Header of the HDR are named and orderedslightlydifferentlyintheDBLbutthecorecontentofeachfieldisthesame.

TheVariableHeaderoftheHDRconsistsofanXMLMPHandanXMLSPHwhichhaveactuallybeenderivedfromthecorrespondingattributesfromtheDBLfile,consequentlyalltheinformationprovidedinthisVariableHeaderofbothHDRandDBListhesame.

TheXMLheaderfilehasafixedportionandavariableportion.ThefixedportionhasthesameformatinallCryosat-2files.ThevariableportioncontainsmuchofthesameinformationastheglobalattributesintheproductfilebutinXMLformat.

3.4 FileNamingConventionFilenamesareconstructedaccordingtostandardguidelinesforEarthExplorermissions.Allfollowtheconventionalform

MM_CCCC_TTTTTTTTTT_yyyymmddThhmmss_YYYYMMDDTHHMMSS_vvvv.ttt

where:

MM=themissionidentifier,whichisCSforCryosat-2

CCCC=fileclasswhichcanbe:OFFL(Off-LineSystematicProcessing),NRT_(NearRealTime),RPRO(ReProcessing),TEST(Testing),LTA_(LongTermArchive)

TTTTTTTTTT=filetypeasdefinedinTable4below

yyyymmddThhmmss=starttimewindowasextractedfromJobOrder

YYYYMMDDTHHMMSS=stoptimewindowasextractedfromJobOrder

AvvvorBvvv=theversionnumberofthefile.Thiscanbeincrementedinthecaseofreprocessing.“A”correspondstoIPF1versionJandIPF2versionIorpreviousversions.“B”correspondstoIPF1versionKandIPF2versionJ.

ttt=theextension:tgzforagzippedtarfile,HDRforanextractedheaderandDBLforextractedbinarydata.



Table4:RelevantFileTypes

Filetype Description

SIR_LRM_1B SIRALLRMlevel1b

SIR_SAR_1B

SIR_SARN1B

SIRALSARmodelevel1b

SIRALSARNRTmodelevel1b

SIR_SIN_1B

SIR_SINN1B

SIRALSINmodelevel1b

SIRALSINNRTmodelevel1b

SIR_LRM_2_ SIRALLRMlevel2

SIR_SAR_2_

SIR_SARN2_

SIRALSARmodelevel2withicefreeboardmeasurements

SIRALSARNRTmodelevel2withicefreeboardmeasurements

SIR_SIN_2_

SIR_SINN2_

SIRALSINmodelevel2

SIRALSINNRTmodelevel2

SIR_SID_2_ SIRALdegradedSINmodelevel2

SIR_GDR_2_ SIRALgeophysicaldatarecordwithicefreeboardmeasurements

Thistableshowsonlythefiletypesrelevanttothescientificdata.Calibrationmodesandintermediatefiles(FBRandL2I)areexcluded.

Thegeophysicaldatarecords(GDR)meetmostscientificneeds.AGDRcoversafullorbit,startingandendingattheascendingnodeattheequator,andcontainstime-orderedrecordsofdatafromallSIRALsystematicmodes(hencenoFDM).

3.5 OtherConventionsBackgroundinformationonallsectionsbelowcanbefoundinreference12.

3.5.1 LatitudeandLongitudeLimitsLongitudeisexpressedfrom-180to+180degrees.Positiveeastdoesnotcontain180.0,negativewestdoescontain-180.0.ThisisthesamedefinitionasusedbyEnviSat.



3.5.2 RangeCorrectionSignConventionThesignconventionforallrangecorrectionsisthattheyareaddedtorange.

Sincesatellitealtitude–rangemeasurement=surfaceheight (Eqn3.5.2-1)

itfollowsthat

satellitealtitude–(rangemeasurement+ionosphericcorrection)=surfaceheightcorrectedfortheionosphereeffect (Eqn3.5.2-2)

3.5.3 TimestampFormatThetimestampsinthemeasurementdatasetrecordsareTAItimesinsecondspast00:00:00on1-JAN-2000.

Notethat:

i)ThetimeinthemainandspecificproductheadersisinUTCandnotinTAI

ii)TimecorrelationsbetweenUTCandTAIareprovidedviatheorbitfiles-thesemustaccountforthecorrectoffsetsduetoleapseconds

iii)1Jan2000at00:00UTC=1Jan2000at00:32TAI

See the Hong Kong Time Service website for a detailed discussion of time measurement.http://www.hko.gov.hk/wservice/tsheet/timeserv.htm

3.5.4 TimeinL2GDRRecordsIn L2 products a record contains a group of 20measurements together with a single set of otherparameters.Thisgivesadatarateofapproximatelyonerecordpersecond,or1Hz,withmeasurementsatapproximately20Hz.Timestampsforbothareprovidedasabove.

Itisimportanttonotethatthe1Hztimecorrespondstothefirstofthe20measurementsandnotthemiddle. This is different from previous altimetry missions, such as ERS and EnviSat, but for goodreasons.

Unlike previous missions the instrument telemetry does not create the block of 20 high-ratemeasurementswith the1Hzparameters. In thecaseofCryosat-2 this ispurelya choicewithin thegroundprocessingsystemtoreducetheL2datavolume.ThetimestampforrecordsintheSARandSINmodes is a functionof theDopplerbeamprocessingandnot associatedwith aparticular telemetrypacket.Ifagroupof20measurementsspansameasurementmodechange(oreven2)thentheintervalbetweeneachofthe20blockswillnotberegular.Thiswouldmakeaconsistentdefinitionofa‘middle’timeratherdifficult.Furthermore,ameasurementstreammayendatanypointwithintheblockof20.Inwhichcasetheunusedblocksarezero-filledandonlyservetopadoutthedatarecord.Onecanonlybesurethatthereisatleastonemeasurementinarecordof20.



Forthesereasonsthe1Hztimeisprovidedcorrespondingtothefirstmeasurementofthe20.



4 ReferencesandAcronyms

4.1 ReferenceDocuments[1] Anglemeasurementwith a phasemonopulse radar altimeter, J.R. Jensen, IEEE Transactions onAntennasandPropagation,Vol47,No4,April1999

[2]Cryosat-2IPF1InputandOutputDefinitionDocumentCS-IS-MSL-GS-0001

[3]Cryosat-2MissionandDataDescriptionCS-RP-ESA-SY-0059

[4]CryoSat:AmissiontodeterminethefluctuationsinEarth'slandandmarineicefields,Winghametal,AdvancesinSpaceResearch,Volume37,Issue4,2006,Pages841-871

[5]PDS-SSALTOInterfaceControlDefinitions,CS-ID-ACS-GS-0123

[6]SpecificationD'InterfacesInternesSSALTO:OrbitographieMissionSMM-IF-M6-EA-20174-CN

[7]CryoSat2PreciseOrbitContextCS-TN-ESA-SY-0239

[8]EarthExplorerGeoCorrectionsUserManualCS-MA-CLS-GS-0001

[9]Correctionsforaltimeter low-levelprocessingattheEarthObservationDataCentre,Cudlipetal,InternationalJournalofRemoteSensing,Volume15,Issue4,March1994,pages889-914

[10]IPF1DetailedProcessingModelCS-TN-ACS-GS-5105version5.1

[11]D.E.CartwrightandA.C.Edden,Corrected tablesof tidalharmonics.Geophysical Journalof theRoyalAstronomicalSociety,Volume33,1973,pages253-264

[12]CryoSatIcenetCDFL1BProductFormatSpecificationC2-RS-ACS-ESL-5364

[13]CryoSatIcenetCDFL2ProductFormatSpecificationC2-RS-ACS-ESL-5265

[14]SeaStateBiasinRadarAltimetryRevisited,HausmanandZlotnicki,MarineGeodesy33(S1):336-347,2010

[15]SIRAL,aHighSpatialResolutionRadarAltimeterfortheCryosat-2Mission,Reyetal,GeoscienceandRemoteSensingSymposium,2001.IGARSS2001.IEEE2001International,vol7,pages3080-3082

[16]EarthExplorerMissionCFIOrbitSoftwareUserManualEE-MA-DMS-GS-0004

[17]EarthExplorerFileFormatStandardsPE-TN-ESA-GS-0001

[18]Cryosat-2L2ProcessorDesignSummaryDocumentCS-DD-MSL-GS-2002

[19]TheMeanEchoandEchoCrossProductFromaBeamformingInterferometricAltimeterandTheirApplicationtoElevationMeasurement,Winghametal,IEEETransactionsOnGeoscienceAndRemoteSensing,Volume42,No10,October2004,pages2305-2323



[20]Cryosat-2IPF1DesignSummaryDocumentCS-DD-MSL-GS-0008

[21]BrownModelDerivativesWalter,Smith09June2009;revised19May2011.

[22]Cryosat-2IPF2Sigma-zeroTechnicalNote,CS-TN-MSL-GS-6005

4.2 AcronymsandAbbreviationsAGC AutomaticGainControl

ANX AscendingequatorcrossingNode

API ApplicationProgrammingInterface

ASCII AmericanStandardCodeforInformationInterchange

CAL1-4 Calibrationmodes

CFI CustomerFurnishedItem

CoG CentreofGravity

CoM CentreofMass

DAC DynamicAtmosphericCorrection

DAD DynamicAuxiliaryData

DEM DigitalElevationModel

DFCB Dataformatcontrolbook

DORIS DopplerOrbitandRadiopositioningIntegrationbySatellite

DS DataSet

DSD DataSetDescriptor

DSR DataSetRecord

DTM DigitalTerrainModel

EM Electromagnetic

ESA EuropeanSpaceAgency



ESL ExpertSupportLaboratory

FOS FlightOperationsSystem

GDR GeophysicalDataRecord

GIM GlobalIonosphericMap

GPS GlobalPositioningSystem

IBC InverseBarometricCorrection

IERS InternationalEarthRotationService

IPFDB InstrumentProcessingFacilitydatabase

ITRF InternationalTerrestrialReferenceFrame

L1B Level1b

L2 Level2

LRM LowRateMode(synonymouswithPulseLimitedMode)

LTA LongTermArchive

MDS MeasurementDataSet

MDSR MeasurementdatasetRecord

MJD ModifiedJulianDate

MOG2D Modèle2Dd'OndesdeGravité

MPH MainProductHeader

MSS MeanSeaSurface

MSSL MullardSpaceScienceLaboratory(UCL)

NOAA NationalOceanicandAtmosphericAdministration

NRT NearRealTime

ODLE OceanDepth/LandElevation

PDS PayloadDataSystem



POE DORISPreciseOrbitEphemeris

RMS RootMeanSquared

Rx Receiver

SAD StaticAuxiliaryData

SAR SyntheticApertureRadar

SIN SARInterferometry

SBT SatelliteBinaryTime

SID SINdegradedmode

SIRAL SARInterferometricRadarAltimeter

SP SourcePacket

SPH SpecificProductHeader

SS SpaceSegment

SSALTO SystemeauSold'ALTimétrieetd'Orbitographie

SSHA SeaSurfaceHeightAnomaly

SWH SignificantWaveHeight

TAI InternationalAtomicTime(fromtheFrench,TempsAtomiqueInternational)

TEC TotalElectronContent

UCL UniversityCollegeLondon

USO Ultra-StableOscillator

UTC CoordinatedUniversalTime

XML ExtensibleMarkupLanguage