Rosetta Alice Ultraviolet Spectrograph Flight Operations ... · PDF fileRosetta Alice Ultraviolet Spectrograph Flight Operations and Lessons Learned 2 1 The Rosetta Mission and Comet

RosettaAliceUltravioletSpectrographFlight

OperationsandLessonsLearnedJonP.Pineau1,JoelWm.Parker2,AndrewJ.Steffl2,EricSchindhelm2,RichardMedina2,S.AlanStern2,M.Versteeg3,andEmmaM.Birath2

1StellarSolutions,Inc.,PaloAlto,CA,UnitedStates.2SouthwestResearchInstitute,DepartmentofSpaceStudies,Boulder,CO,UnitedStates.3SouthwestResearchInstitute,6220CulebraRoad,SanAntonio,TX78238,UnitedStates.

AbstractThis paper explores the uniqueness of ESA Rosetta mission operations from the Alice instrument point of view, documents lessons learned, and suggests operations ideas for future missions. The Alice instrument mounted on the Rosetta orbiter is an imaging spectrograph optimized for cometary far-ultraviolet (FUV) spectroscopy with the scientific objectives of measuring properties of the escaping gas and dust, and studying the surface properties, including searching for exposed ices. We describe the operations processes during the comet encounter period, the many interfaces to contend with, the constraints that impacted Alice, and how the Alice science goals of measuring the cometary gas characteristics and their evolution were achieved. We provide details that are relevant to the use and interpretation of Alice data and published results. All these flight experiences and lessons learned will be useful for future cometary missions that include ultraviolet spectrographs in particular, and multi-instrument international payloads in general.

RosettaAliceUltravioletSpectrographFlightOperationsandLessonsLearned 2

1TheRosettaMissionandComet67PAtfirstsightofthecomet67P/Churyumov-Gerasimenko,itwasobviousthattheRosettamissionwouldbedifferentfromanythatcamebefore.Whilethiswasexpectedtosomeextentasreflectedinthemissiondesign,theneedforanevenmorecomplexconceptofoperationswasrealizedwhenthehighlyirregularcometbodywasfinallyresolved10.5yearsafterlaunch.Theshapeofcomet67Piscomposedoftwounequallobeswithadeepneckareaconnectingthetwo.Thisaddedcomplexitiesinthegravityfield,thenon-uniformityofmaterialandlightcomingoffthesurface,andthegeolocationmapping.Thecombinationofthesefactorsalongwiththeanticipateddifficultiesoforbitingalowgravity,dynamicbodythatcreatesitsownever-changingenvironmentresultedintrulyuniquemissionchallengesthatrequiredequallyuniqueoperationssolutions.AsamemberoftheJupiterfamilycomets(JFCs),67P’shighlyellipticalorbittakesitjustpastJupiter’sorbit(5.684au)andalmostasclosetotheSunasEarth(1.246au)every6.45years.Theresultisaperiodictransformationbetweenadormant,dark,frozennucleusnearaphelionandaspectacularlydynamicdisplayofsublimatinggasandejecteddustnearperihelion.ThisstrongvariationinactivitydrovemanyRosettamissionattributesandconstraints,addingtothealreadycomplexsystemthatincludedtheRosettaorbiterwith11instrumentpackagesandthePhilaelanderwithitsownsetof10instruments.AsthecometactivityevolvedasitcircledtheSun,theRosettaandAliceoperationsprocessesevolvedthroughthemissiontoprovideasystemthatcouldmeettheever-changingneeds.Toolswereredesigned,interfaceschanged,andprocessesupdatedtooptimizesciencereturn.Afterlaunch,thecircuitouspathofRosettaincludedseveralorbitsintheinnersolarsystemincludingseveralgravitationalassistsfromEarthandMars,thelastofwhichsentRosettatotheoutersolarsystemonarendezvouspathwith67P.AlongthewayinstrumentswereabletomakeobservationsofEarth,cometLINEAR,cometTempel1duringtheDeepImpactevent,Mars,asteroidSteins,andasteroidLutetia,whichallowedtestingoutflightsystemsandgroundprocesses.Shortlyafterwakingfromanunprecedentedpre-primemission2.5yearhibernationperiodthatlastedthroughaphelionwithoutanycommunicationwithgroundstationsonEarth,Rosettabeganstudyingcomet67Pinamannerneverbeforeattemptedatacomet.Wherepreviousmissionshadstudiedcometsfromflybysofseveral100’skmawaytravelingatrelativevelocitiesof10’sofkm/sorperformingabriefsurfaceimpact,Rosetta’stwoyearencounterperiodwasoftenspentat10’sofkmfromthecometnucleuswithrelativevelocitiesofmeters/second.Aslow,controlledsurfaceimpactconcludedthehistoricmissionaddingtothehighsurfaceresolutionmeasurementsthatthePhilaelanderacquiredatthebeginningofencounter.Table1detailssomeoftheimportantRosettamissionandAliceinstrumentevents.


Table1:MajorMissionandAliceEvents2004Mar2 Launch

2004Apr-May ObservationsofcometLINEAR2005Mar4 Earthflyby#12005Jul4 ObservationsofDeepImpact

2007Feb25 Marsflyby2007Nov13 Earthflyby#22008Jul19 Ø Aliceupset/resetevent#12008Sep5 AsteroidSteinsflyby2009Sep29 Ø Aliceupset/resetevent#22009Nov13 Earthflyby#32010Jul10 AsteroidLutetiaflyby2011Jun8 • Spacecrafthibernationentry(4.5au)2012Oct3 Aphelion(5.3au)2014Jan20 • Spacecrafthibernationexit(4.5au)2014Mar18 Ø Alice1stpoweronafterhibernation2014Jun6 Ø Alicebeginscontinuouspoweredoperations2014Aug6 Orbitinsertionatcomet67P,startofencounterperiod2014Nov12 • Landerdelivery(3.0au)2014Nov15 • Landerhibernationbegins2015Mar28 • Spacecraftstartrackersafemode–instrumentsoff2015Apr15 Ø Alicereturnstonominaloperationsaftersafemode2015May10 Equinox(Nvernal,Sspring)

2015Jun13–Jul9 • Intermittentlandercontacts2015Aug13 Perihelion(1.2au)2015Sep4 Solstice(Nwinter,Ssummer)

2016Mar20 Equinox(Nspring,Svernal)2016May28 • Spacecraftstartrackersafemode–instrumentsoff2015Jun1 Ø Alicereturnstonominaloperationsaftersafemode

2016Sep30 Endofmission(3.8au)

Ø AliceEvent;�SpacecraftorLandereventInSection2wedescribetheAliceUltravioletSpectrographdesignandscienceobjectives.InSection3wedescribetheoperationsofAliceandthevariousinstrumentalandmissionconstraintsthataffecttheoperations.Section4coversthedifferenttypesofAliceobservations,supportingobservationsbyotherinstruments,andthespecialissuesofregardingmakingobservationswhileorbitinganactivecomet.InSection5wedescribethebroadercontextanddesignofRosettamissionoperations.Section6providessomelessonslearnedandadditionalinformation.


2TheAliceUltravioletSpectrograph

2.1AliceDesignDescribedbelowisthedesignoftheNASA-fundedRosettaAliceinstrumentaboardtheESARosettaasteroidflyby/cometrendezvousmission.MostofthetextintheAliceDesignsectionisfromSternetal.,2007[1],butrepeatedhereforstand-alonecompletenessofthisdocument.Aliceisalightweight,low-power,andlow-costimagingspectrographoptimizedforcometaryfar-ultraviolet(FUV)spectroscopy[1].ItwasthefirstofafamilyofsimilarinstrumentscurrentlyincludingAliceonNewHorizons,LAMPUVS(LymanAlphaMappingProjectUltraVioletSpectrograph)onLunarReconnaissanceOrbiter(LRO),UVSonJuno,andtheexpectedUVSinstrumentsontheupcomingJUICEandEuropaClipperMissions.RosettaAlicewasthefirstUVspectrographtostudyacometatcloserange.Itwasdesignedtoobtainspatially-resolvedspectrainthe700–2050Åspectralbandwithaspectralresolutionbetween8Åand12Åforextendedsourcesthatfilltheslit.Anopto-mechanicallayoutofAliceisshowninFigure1.Lightentersthetelescopesectionthrougha40×40mm2entranceapertureatthebottomrightofFigure1andiscollectedandfocusedbyanf/3off-axisparaboloidal(OAP)mirrorontotheentranceslitandthenontoatoroidalholographicgrating,whereitisdispersedontoanimagingphoton-countingmicrochannelplate(MCP)detectorthatusesadouble-delayline(DDL)readoutscheme[2].Thetwo-dimensional(1024×32)-pixelformat,MCPdetectorusesdual,side-by-side,solar-blindphotocathodes:potassiumbromide(KBr;forλ<1200Å)andcesiumiodide(CsI;forλ>1230Å)[3].Themeasuredspectralresolvingpower(λ/Δλ)ofALICEisintherangeof70–170foranextendedsourcethatfillstheinstantaneousfield-of-view(IFOV)definedbythesizeoftheentranceslit.


Figure1.(a)Theopto-mechanicallayoutofAlice.(b)AphotographoftheAliceflightunit.[1]Theslitdesignisdescribedasa“dogbone”;itis5.53°longinthespatialdimension,thecentral2°ofwhich(calledthe“narrowcenter”)is0.05°wideinthespectraldimension,andthe2°“widebottom”and1.5°“widetop”portionsoftheslitare0.1°wideinthespectraldimension,showninFigure2.Thepinholefeatureatthetopoftheslitwasneverdetectedinflight.


Figure2.TheAliceentranceslitdesign[1]Tocapturetheentire700–2050Åbandpassand5.53°spatialfieldofview(FOV),thesizeofthedetector’sactiveareais35mm(inthedispersiondirection)×20mm(inthespatialdimension),withapixelformatof(1024×32)-pixels.The5.53°slit-heightisimagedontothecentral20ofthedetector’s32spatialchannels;theremainingspatialchannelsareusedfordarkcountmonitoring.ThepixelformatallowsaNyquistsampledspectralresolutionof∼3.4Åandaspatialresolutionof0.3°.ThedetectorelectronicsamplifyandconvertthedetectedoutputpulsesfromtheMCPZ-Stacktopixeladdresslocations.InsteadofdiscretephysicalpixelsasinaCCDdetector,thetimeofarrivalofachargeatbothendsoftheorthogonaldelaylinesisusedtodeterminethespectral“pixel”onwhichaphotonlanded.Thesensitivityofthedetectorelectronicstoconvertaphotoneventintoanoutputpulsecanbeadjustedbychangingthedetectorhighvoltagelevel.OnlythoseanalogpulsesoutputfromtheMCPthathaveamplitudesaboveasetthresholdlevel,alsoreferredtoasthediscriminatorlevel,areprocessedandconvertedtopixeladdresslocations.Foreachdetectedandprocessedevent,a10-bitxaddressanda5-bityaddressaregeneratedbythedetectorelectronicsandsenttotheAlicecommand-and-datahandling(C&DH)electronicsfordatastorageandmanipulation.Inadditiontothepixeladdresswords,thedetectorelectronicsalsodigitizestheanalogamplitudeofeachdetectedeventoutputbythepreamplifiersandsendsthisdatatotheC&DHelectronics.Histogrammingthis“pulse-height”datacreatesapulse-height


distributionfunctionthatisusedtomonitorthehealthandstatusofthedetectorduringoperation.Abuilt-in“stim-pulser”isalsoincludedintheelectronicsthatsimulatesphotoneventsintwopixellocationsonthearray.ThispulsercanbeturnedonandoffbycommandandallowstestingoftheentireAlicedetectorandC&DHelectronicsignalpathwithouthavingtopoweronthedetectorhigh-voltagepowersupply.Inaddition,thepositionofthestimpixelsprovidesawavelengthfiducialtoaccountforshiftsthatcanoccurwithoperationaltemperaturechanges.

2.2AliceScienceModesTheAliceinstrumenthasthreemodesoftakingdata:histogram,pixellist,andcountrate.Thefirsttwoacquisitionmodesusethesameeventdatareceivedfromthedetectorelectronics,butthedataareprocessedinadifferentway.Thethirdacquisitionmodeonlyusesthenumberofeventsreceivedinagivenperiodoftime–nospectralorspatialinformationisprovided.

Inhistogrammode,theinstrumentcontinuouslycollectsdetectedeventsbypixellocationthroughouttheexposure,muchlikehowaCCDcountsandaccumulatesdetectedphotons.Duringthecourseofahistogramobservation,eachdetectedeventissavedintomemory,wherethememorylocationsmapone-to-onetopixelsonthedetector.Inthisway,anintegratedspatial-spectralimageisaccumulatedoverthecommandedexposuretime.Thismodeisusefulforobservationswithhighcountrates,andwasthemostcommonlyusedmodeduringthemission.Theresultingimagesarethesameformatasthedetector,1024x32pixels,withthelongaxiscorrespondingtowavelengthandtheshortaxisisthespatialdimension.Sincetheouteredgesoftheimagearean“unexposed”partofthedetector,thesoftwareplacesthepulseheightarrayinthisregion,anditalsoprovidesmonitoringofbackground/cosmicrayanddarkcounts.Thehistogramvalues(counts)saturateat65535counts.

Inpixellistmode,theinstrumentseriallyrecordsthex-ylocationofeachphotonandinsertstimetags(“hacks”)inthislistatregular,user-definableintervalstoprovidethetiminginformationforthephotons.Duetodatavolumeconsiderationsandthetimerequiredformemoryreadout,thismodewasusedprimarilyatlowcountrates.Themaximumnumberofentriesis32767(includingphotonsandtimetags),butmaybelessdependingonthecombinationofbrightnessofthetargetandtheexposuretimeoftheparticularobservation.Oncethememoryisfilledtothislimit,nomoredataarerecordeduntiltheexposureisstoppedandthememoryisreadout.ThismodewasemployedprimarilyduringtheBi-MonthlyCrossSlitScanCalibrationduetotheneedfortimeresolutiontodetectwhenthestarcrossedtheedgesoftheslit.

Incountratemode(alsocalled“photometer”mode),thespatialandspectralinformationforeachphotonisignored.Thenumberofphotonsdetectedineachuser-definedtimeintervalaresummedandsavedasasinglenumberinmemory.Thismodewasnotwidelyused,butcouldbeusefulforobservingverybrightultravioletstars.Themaximumnumberofintervalsis32678,butmaybelessdependingonthecombinationoftimeintervalandtotalexposuretimeoftheparticularobservation.(Notethatthenumberofintervalsisindependentofthe


brightnessofthetarget.)Thesummedcountspertimeintervalsaturateat65535counts.[4]

2.3AliceScienceObjectivesThescientificobjectivesoftheAliceinvestigationweredefinedasfollowsandfurtherdetailsfoundin[1]:

1. Searchforanddeterminetheevolvedraregascontentofthenucleustoprovideinformationonthetemperatureofformationandthermalhistoryofthecometsinceitsformation.

2. Determinetheproductionratesoftheparentmoleculespecies,H2O,COandCO2,andtheirspatialdistributionsnearthenucleus,therebyallowingthenucleus/comacouplingtobedirectlyobservedandmeasuredonmanytimescales.

3. StudytheatomicbudgetofC,H,O,N,andSinthecomaasafunctionoftime.4. StudytheonsetofnuclearactivityinwaysRosettaotherwisecannot.5. Spectralmappingoftheentirenucleusof67PatFUVwavelengthsinorderto

bothcharacterizethedistributionofUVabsorbersonthesurface,andtomaptheFUVphotometricpropertiesofthenucleus.

6. Studythephotometricandspectrophotometricpropertiesofsmallgrainsinthecomaasanaidtounderstandingtheirsizedistributionandhowtheyvaryintime.

7. MapthespatialandtemporalvariabilityofO+,N+,S+andC+emissionsinthecomaandiontailinordertoconnectnuclearactivitytochangesintailmorphologyandstructurenearperihelion.

3AliceOperationsandConstraints

3.1ContaminationThematerialcomingoffthecometposedaseriousmissionriskasthefinedustandgascouldcoatinstrumentopticsandsolarpanelswhichcouldleadtoseveresciencedegradation,decreasespacecraftpowerproduction,andpreventspacecraftattitudeandlocationknowledgedetermination.Furthermore,duringperiodsofhighcometactivitytherewereunpredictablestrongoutburststhatquicklypropelledconcentratedjetsofmaterialoutward.LargepiecesofmaterialwereseenorbitingthecometandattimesimagesshowedthatRosettawasflyingthroughablizzardofdustparticlesasshowninFigure3.


Figure3.(a)DustandgasescapingthecometnucleusonMarch272016[5](b)ThedustenvironmentaroundRosettacapturedbyRosetta’sOSIRIS(Optical,Spectroscopic,andInfraredRemoteImagingSystem)cameraonJuly62015[6]Itwasdifficulttostriketherightbalancebetweenwhatwasperceivedtobeasafedistancefromthematerialcomingoffthecometandbeingclosetothenucleustoobtainhighresolutionmeasurementsandsignificantcountsforthedustandgasinstruments.ForAlice,thelocalgaspressureanddustcountmeasurementsneverreachedsustainedlevelsthatwouldcausegreatconcernofcontamination,andthelevelswerealwaysbelowlimitsthatwouldhavetriggeredanAlicesafingeventthatclosestheaperturedoorandturnsoffthedetectorhighvoltage.RegularcalibrationobservationswiththeAliceinstrumentthroughtheescortphasetrackedsensitivitydegradation,whichwasminimal,duetocontaminationandothersources.However,onesystemaffectedbythedirtyenvironmentwasthespacecraftstartrackersthatoccasionallywouldgetconfusedbythebrightdustparticlesandloseattitudeknowledge.Thatresultedintwospacecraftsafeeventsthathaltedallactivitiestoregainattitudeknowledge,andtherewereseveralotherclosecalls(asfurtherdetailedinTable1).Sincesuchspacecraftsafetyeventscouldjeopardizetheentiremission,theprogramerredonthesideofcautionandusuallykeptahealthymarginonthedistancetothecometbasedonactivity.ThistypicallyhelpedAlicesciencebecausethespectrograph’slongslit(5.53°)attheselargerdistancesallowedAlicetohavesomespectralelementsonthenucleusandotherslookingoff-limbattheescapinggasesintandem(asa“ride-along”)withobservationsdesignedbyotherinstrumentteamsthatpreferredsurfacepointing(seetheFigure4example).AnotherfactorthatpossiblyhelpedpreservetheAlicethroughputisthattheinstrumentoperatedincontinuousdecontaminationmodeformostoftheencounterafterverifyingitdidnotimpactinstrumentcalibration.Thismeantthatthemirrorandgratingheaterswereconstantlyenabledtokeepthoseopticalelementswarmsotheywouldnotbecoldtrapsbydrivingoffanydepositedcontamination.However,giventhattherewereveryfewissuesforcomponentsontheRosettaorbiterthroughencounterphasedemonstratesthatencounteringhighconcentrationsofmaterialandlargedestructivechunksofmaterialwasrareeven


thoughtheenvironmentwasmuchdirtierthantypicalwhenorbitinglessactivebodiessuchasplanets,moons,andasteroids.

Figure4.ExampleofanOSIRIStargetedsurfaceobservation(landersearch)thatcouldbeusedbyAliceasaride-alongtoviewoff-cometgasses[7].

3.2GainSagAddingsomeuniqueconstraintstotheAliceoperations,theinstrumenthasanun-scrubbedmicrochannelplate(MCP)detector.ThescrubbingprocessexposesthedetectortoUVlightreducingthepixelcharge,aconditioningtechniquethat“burnsin”thepixels,eventuallystabilizingthedetectorresponse.However,ascrubbeddetectorimposessignificantconstraintsinpre-flightgroundhandlingincludingrequiringbeingkeptinconstantvacuum,whichwasnotanoptioninthiscaseastheAlicebandpassintheEUVrequiredanopen-faceddetector.Theconsequenceofflyinganun-scrubbeddetectoristhatitsresponsecontinuallydecreasesthroughthemissionfromexposuretoUVlight,atypicalcharacteristicofMCPscalled“gainsag”.Thegainsagisalsospatiallydependentasafunctionofthetotalfluenceonagivenareaofthedetector,leadingtonon-uniformsensitivitydegradationacrossthedetector.Forexample,thevaryingbrightnessofthecometsurface,theobservationofbrightUVstarsonspecificdetectorrows,ortheconstantexposuretoLyman-alphaemissionfromtheinterplanetarymediumwillcausesomepartsofthedetectortodegradedifferentlyfromotherparts(seeFigure5).Thedegradationwastracked,accountedforinthedatacalibration,andperiodicallycompensatedforbychangingdetectorsettingssuchashighvoltagelevel(adetectorsensitivityadjustment)anddiscriminatorlevel(asignalcutoffadjustment).CharacterizationofthedetectorperformancewasconductedperiodicallyusingmoderatelybrightUVstarsascalibrationsources.


Figure5.Theprogressionofdetectorsciencecountsthroughencounter.Theconcentrationofcountsnearcolumn600(1215A)istheLy-αlineandtheregionaroundcolumn~850(~1025Å)correspondstothechameleonartifactregion.Thehighconcentrationofcountsinrows9,15,and21areduetostellarcalibrations.Wavelengthdecreasestotheright,i.e.,tohigherrownumber.


Theanticipatedlongtermgainsagdrovestrategicplanningforthedetectorusagetoensurethatthedetectorwouldmaintainadequatesensitivitythroughoutthemissionincriticalbandsanddetectorregionstosatisfythesciencegoals.Thisinitiallydroveanalysis-intenseobservationconstraintsofconductingonlyhighprioritycometmeasurementsandactivelyavoidingexposuretorelativelybrightUVstars.

3.3BadDogsandChameleonContributiontoGainSagSomestarsweresobrightintheUVthattheycouldcauseexcessivecountratesontheAlicedetector.TheseUV-brightstarswerenicknamed“baddogs”.TheAliceteamwouldreceivethepredictedspacecraftattitudeandlocationinformationanduseitdodeterminewhenbaddogstarswouldcrosstheAliceFOVplusamargintoaccountforpointinguncertainty.TheteamwouldthenadjustinstrumentcommandingtosuspendtheAliceobservationsandclosetheaperturedooraroundtheforecastedbaddogencounters.Makingtheprocessevenmorecomplexwasthattheplannedspacecraftattitudecouldchangelateintheproductdevelopmentprocessduetootherinstrumentteamsadjustingtheirprimeobservationsorbythemissionflightdynamicsteamadjustingthecometorbitplans.ThiswouldrequireagilemonitoringandresponsefromtheAliceteamtoquickly(sometimesasshortaswithinacoupledays)identifyandmakecorrespondingAliceobservationchanges.Asthemissionprogressed,itwasdeterminedthatthedetectorgainsagratewaslowerthaninitiallyprojected,whichallowedthegainsagavoidancetechniquestobecomeprogressivelylessconservative.BytheendofthemissionAlicewasobservingalmostcontinuouslyandbrightUVstarswerenolongermanuallyavoided.Insteadofavoidanceplanning,adetectorcountratemonitorwithaconfigurablethresholdwasreliedupontoreactwhenanytoo-brightstarscameintheFOV,autonomouslysafingtheinstrumentbytemporarilypausingtheobservingandclosingtheaperturedoor.Themomentaryexposuretothebrightsourcedidnotsignificantlycontributetotheoveralldetectorgainsagandwasawelcometradeoffbetweensomelossofexposuretimeinexchangeforsignificantlyreducingthelabor-intensivestaravoidanceplanning.Theotherprimarycontributortobrightsafetyeventscamefromasporadicissuereferredtoasthe“chameleon”,whichisthoughttobeduetochargedparticlesenteringtheinstrument,gettingthroughthedetectorelectrongrid,andcausingstrange,high-countmeasurementartifacts(seeFigure5)[8].Becausethechameleonhadstrongtemporalvariations,usuallythebrightsafetyactionoftemporarilyclosingtheaperturedoorandwaitingthesafetytimeoutperiod(typically10minutes)wassufficienttodealwiththeoccasionalhigh-countratechameleons.

3.4AliceObservingStylesThroughoutthemissionAlicewouldobserveduringbothAlice-planned(termed“prime”)observationsaswellasobservationsplannedbyotherinstrumentteams(termed“ridingalong”).Eventhoughmanyobservationsdesignedbyotherteamsweren’toptimizedforAlicesciencegoals,somesciencereturnwasalwayspossible


whenobservingthecometanditssurroundings.However,potentialsciencereturnalwayshadtobebalancedwithavailablespacecraftresources,developmenteffort,andtheexpectedgainsagimpact.Aliceobservingconsistedoftwomainstyles.Thefirstwascontinuousobservingwherehistogramsweretakenoneafteranother(withshort~40soverheadforimagereadoutbetweenexposures)forhoursatatime,notcorrelatedtoaspecificsceneorpointingchanges.Thiswassimpletoimplement,butcouldleadtodifficulttousedatasuchaswhenhistogramsbridgedchangesinspacecraftstaringandscanningmotionsormajorscenechanges.Theothertypewastomanuallytimehistogramstocorrelatetheirstart/stoptimeswithspecificviewingconditions.Althoughthistechniquecouldoftenreturnbetterscience,thisstylehadahighcostintermsofdevelopmenteffortandcommandingcomplexity.Thiswasespeciallytruewhentiminghistogramsduringridealongsbecauseotherinstrumentteamswouldoftenchangetheirpointingplansduringthedevelopmentperiod,whichusuallyrequiredcorrespondingadjustmentstoAliceobservations.AtthebeginningofthemissiontheAliceteamfocusedonitsprimesciencerefrainingfromobservingduringpointingplannedbymostoftheotherteams.However,asthemissionprogressed,ridealongsincreasedwhendatavolumeandotherresourceswereavailabletoapointwhereAlicewasobservingalmostcontinuously.Thiswasespeciallytrueafteritwasevidentthatthegainsagprogressionratewaslessthanexpectedandsoftwaretoolswerecreatedtoautomateobservationdevelopmentbasedonplannedpointingcharacteristics.

3.5ApertureDoorUsageandPowerCyclingAliceinstrumentoperationsareunlikealotofotherspaceinstrumentsintermsofpowercyclingandmechanicaldooroperations.Duetotheriskofpowerupfailures,theactionofpowercyclingspaceelectronicshaslongbeenavoidedandonsomemissionsonlydonewhenforcedtobecauseofafailureononestringofaredundantsystem.BynecessitythisoperationalmethodwasnotanoptiononRosetta.Powerconstraintsduringcertainperiodsofthemissionrequiredinstrumentsandsomenon-essentialspacecraftcomponentstoalternatepoweredactivitiesleadingtomanyunavoidablecomponentpowercycles.Infact,duringanunprecedented2.5yearhibernationperiodwhenRosettawasattheoutermostpartofitsorbit,whichtookitouttotheorbitofJupiter,itonlyhadenoughsolarpowertohaveafewvitalheatersandelectronicunitson[9].Therewasnocontactwiththespacecraft,notevenbeaconsduringthathibernationperiod.AnonboardtimerwassettotellRosettatowakeuponJanuary20,2014whentherewouldbeenoughpowertoturnsomesystemsonandcontactEarth.Everyonewasontheedgesoftheirseatsonhibernationexitdaytosaytheleastandamazinglyallsystemsturnedonandproceededwiththecometencounter10.5yearsintothemission.Duringtheescortphase(August2014–September2016)standardAliceoperationsweretopowercycletheinstrumentonaweeklybasis.AnAliceupset/reseteventoccurredtwiceduringthecruiseperiodgettingtothecomet(seeTable1forthedates).TheanomalywouldpowercycleAliceandrestartinadefaultstatedefinedbyhard-codedparameters,andalthoughitwouldreturntothenominalactivitytimeline,sciencedataqualitycollectedafterwardcouldbenegativelyimpacteduntilmanualadjustmentswerecommandedtoseveraloperationalparameters.Therootcauseoftheupset/reseteventswasnotidentified,butitwasthoughtthat


periodicallypower-cyclingtheinstrumentperhapscouldpreventsuchaproblemfromoccurringorwouldreturnAlicetoitsnominalstateifanupsetoccurred.WithsomepowercyclesalreadyrequiredandfourseparateEEPROMflightsoftwareimagestobootfrom,itwasdecidedthattheriskwouldbelowertoperformregularpowercyclesratherthanbeingoncontinuously.Additionally,tomitigatetheimpactofanupset/resetevent,activitiesweresplitintosmallindependentsegmentsandseveraloperationalparameterswerereassertedtwicedaily.TheAliceaperturedooroperationsweresimilarlydifferentfrommostotherspacemechanicalsystems.Whereasmostmechanicalsystems,anddoorsespecially,aredesignedforonetimeuse,theAliceaperturedoorwasdesignedtobeopenedandclosedmanytimes,ratedfor10,000such“flaps”,whichgaveafactorof2marginrelativetogroundtests.Alicereliedonarobustlydesignedaperturedoormechanismprovenbyextensivegroundqualificationtestingandsubsequentlybylaterversionsoftheinstrument.TheLAMPUVS,whichisaversionofAliceontheLunarReconnaissanceOrbitermission,hasperformedover95,000cycles.Asasafeguard,Alicealsohadafailsafedoorthatcouldbeopenedpermanentlytoretainsomescienceabilityiftheaperturedoorfailedshut.Drivingallthedoorcycleswastheinstrument’ssusceptibilitytocontaminationanddetectorgainsag.Toavoidcontaminationofinstrumentopticsfromspacecraftthrusterbyproducts,theaperturedoorwasclosedduring,andfor30minutesafter,everyspacecraftpropulsivemaneuver.Toavoidcontaminationfromcometarymaterialtheaperturedoorwasalsoclosedforgapsinobservationsofmorethan15minutesthroughmostoftheencounterandevenshortergapswhenthecometwasespeciallyactive.Monthlydoorperformancetestsconfirmedthemechanismexhibitednomeasurementtrendsthatwouldcauseconcernforcontinuedregularuse.

4AliceObservations

4.1OrbitingaCometTherewasapreliminaryencounterphasetimelineoutliningastrawmanplanofRosetta’sorbitsaroundthecomet,butduetotheuniquecometaryenvironmentittooksometimeforthemissionteamstogetusedtooperatingaroundacometandrefinetheplan.Inadditiontothestandardengineeringandenvironmentmodelsneededforinterplanetarytravel,manyunusualfactorsneededtobeaccountedforincluding:orbitingalowgravitybodywithanasymmetricgravityfieldcreatedbythetwolobedcometshape,variableaerodynamicdragduetotemporalandspatialvariabilityofcometaryactivity,navigationwithstartrackersinanenvironmentwithhighvisualdustconfusion,andcontaminationrisks.Theorbitsvariedquitealottoaccountfortheseconstraintsandtoaccommodatetherequestedobservationalconditions,butthefollowingdescribesthetypicalobittypes:Thespacecraftflewlow(<32kmradius)circularorbitswhenthecometwaslessactive.Thisallowedtheinstrumentstoobtainhigh-resolutionsurfaceinformationwhenthecontaminationriskwaslow,andtoincreasesourcefluxfortheinsituinstruments.


Ascometactivityincreased,thespacecraftorbitwasexpandedtoseveralhundredkilometerstoamaximumradiusofabout~600kmaroundperihelionasasafetyprecaution,buttoaccommodatescienceneeds,themissiontriedtokeeptheorbitheightaslowassafetywouldallow.Boundorbitswerenotusedwhenorbitingabove~32kmbothduetothedifficultyofbalancingaccelerationfactorswhenorbitingalow-gravitybodyduringperiodsofhighercometactivity.Abovethatheight,thespacecrafttransitionedtojoinedsegmentsofhyperbolicarcsthattypicallyresultedinfullorbitsthatwereroughlycircularshape.Thismaintainedregularobservationconditionsforscienceplanningwhilealsoprovidingthesecurityofasafetrajectorythatwouldavoidcollisionwiththecometifcontrolofthespacecraftwerelost[9].Theorbitstypicallywerealignedwiththeterminatortominimizegasdragonthespacecraftandavoideclipses.TheorbitorientationallowedthesolarpanelstobecontinuouslypointedtotheSunandresultedintheminimumsurfaceareapointinginthecometdirection.However,angleoffsetswereoccasionallyusedtoobtainavarietyofilluminationconditions.Closecometflybys,ascloseas7kmfromthesurface,wereusedforhighsurfaceresolutionmeasurements.Distant“excursions”to1000kminthetaildirectionand1500kmintheSundirectionwereusedtomeasurethelocalmagneticfieldproperties.Finally,thelowgravitycometbodyallowedforsomeunusualorbittypesthatwouldhaverequiredanexorbitantamountoffueltoaccomplishifRosettaorbitedalargerbodyrequiringhigherrelativespeeds.Forinstance,sharpturnswereusedforspecialactivitiesincludingthePhilaeLanderreleasetrajectoryandRosettasweptbackandforthoverasingleareaabovethesurfaceinanefforttolocatethePhilaeLander(seeFigure6).Becauseoftheloworbitalspeedsofmeterspersecond,alargechangeindirectionactuallycouldinvolveinarelativelylowchangeinmomentum,andthus,notasignificantuseoffuel.


Figure6.Rosettaorbitdiversityexample[10].ThedarkbluecolordisplaystheapproachbeginningJuly312014andearlyencountercloseorbits.ThelightbluecolordisplaysmoredistantorbitsaroundperihelionthroughAugust92016andthelandersearch.

4.2PointingUncertaintyAnotheruniquechallengethismissionhadtocontendwithwhileplanningobservationswasfairlylargepointinguncertainties.Asisthecaseformostspacemissions,thespacecraftflightdynamicsteamprovidedthescienceteamsmanyorbitalparameterstouseforplanningobservations.However,uniquetothismissionwasthetaskofunderstandingthecraft’srelativelocationandattitudewithrespecttothecometinlightofsignificantnon-gravitationalforces.ThespacecraftinertiallocationandattitudewasknownverywellfromthestartrackerdataandenabledcommunicationwithgroundstationsonEarthandprecisepointingatstarsforinstrumentcalibrations.However,therewereoftenlargeuncertaintiesforthespacecraftpointingrelativetoacometreferencepointsuchasnadir,limb,orsurfacelocationsandthishadasignificantimpactonobservationplanningatseverallevels.Thepointinguncertaintysourcecamefromthecombinationofenvironmentandcometmodeluncertainties,thrusterefficiencyuncertaintiesforupcomingorbitcorrectionmaneuvers,andthequalityofthecomet-relativelocationknowledgeasascertainedfromnavigationimagesofthecomet.Thepointinguncertaintycontinuallychanged,beingbestjustafterthecurrentspacecraftlocationwasdeterminedandupdatedonthespacecraftcomputerand


thenitwouldgrowuntilthenextupdateastheuncertaintiescompoundedinthemodels.Theprojectedcomet-relativepointinguncertaintyrangewastypicallybetween0.1°–4.0°formostofthemission,butwhenthespacecrafttraveledveryclosetothecometsurface(under10km)theuncertaintyforecastcouldgrowmuchlargerupto~16°.Inpracticeitwasfoundthatthedifferencebetweenthedesiredandactualpointingcouldoccasionallygrowlargeforshortperiodsoftime,howeverconservativefactorsinthemodelstypicallyresultedinalargeoverestimateforthepointinguncertainty,typicallybyafactorof0.5ormore.Evenifobservationswereplannedusingpointinguncertaintyvaluesofhalftheforecastedlevel,thescaleoftheuncertaintieswasverydifficultforsomeoftheinstrumentteamstoincorporateintheirplanning.Negotiationsforprimeobservationslotsbytheremote-observinginstrumentteamsoftenprioritizedslotswithlowpointinguncertaintyvaluestoreduceobservationsmear.However,Aliceplanningwasabletobemoreflexiblethanthatforotherinstrumentteamsregardingpointingconstraintsandacquiredalotofprimeobservationswithhighpointinguncertaintyperiods.MakinguseoftheseslotswasmadepossiblesincetheAliceFOVwasmuchlargerthanmostoftheotherinstruments(5.53°long)andittookrelativelylongexposures(5-10minutes)makingobservationsofspecificlocationslessimportantthangettinggoodregionalcoverage.Also,severalAliceobservationtypeswereoff-comet,includinginertialor360°GreatCirclesthatwerenotimpactedbythecomet-relativepointinguncertainty.ThismadeAliceobservingmuchlessdependentonthepointinguncertaintyvaluescomparedtotheothersandAlicewasabletoputmanyoftheworsepointinguncertaintyblockstogooduse.

4.3ObservationTypesSpecificpointingcouldbedevelopedoncetheorbittypewasdefinedandthetrajectoryandsub-spacecraftparameterdetailsfrozen,suchasorbitheight,phaseangle,latitude,longitude,andpointinguncertainty.Afternegotiationsforobservationtimewerecompleted,theinstrumentteamsturnedtheirattentiontoimplementingtheplans.Eachinstrumentteamwouldtailorthepointingduringtheirprimeobservationstooptimizesciencereturn.Teamscouldverifyanditeratethepointingproductsusingmissionsimulatortools,thenaftertheyweredeliveredtheywereverifiedandapprovedbytheRosettaScienceGroundSegment(RSGS)groupatESA’sEuropeanSpaceAstronomyCentre(ESAC)inSpainandtheRosettaMissionOperationsCenter(RMOC)groupattheEuropeanSpaceOperationsCentre(ESOC)inGermany.Therewerequitearangeofobservationtypesimplementedbythevariousinstrumentteams,butmostwerevariationsonthese:

• Staringatthesurfacewiththecometrotatingunderneaththespacecraft• Matchingthecomet’srotationratetotrackasurfacelocation• Scanningacrossthesurface• Staringatoneofthecometlimbs• Pointingoffcomet

BelowaredetailsofsomeofthevariantsoftheseobservationtypesthatAliceutilized.ForreferencetheobservationnamesarealsousedinAlicedatanomenclature,whichisdiscussedfurtherattheendofthepaperintheAliceLogbookSection.


4.4AliceObservationTypes:GasObservationsAprimaryAlicescienceobjectivewastocharacterizethegassescomingoffthecomettobetterunderstandthesolarsystem’sprimordialconstituentsandenvironment.Thisrequiredmeasuringthedifferentgastypes,theirrelativeabundances,andhowtheyvariedthroughtheevolutionofthecometduringitsorbitaroundtheSun.Todothis,avarietyofobservationtypesweredesignedtoviewthegassesagainstadarkbackground(eitherlookingoffcometorashadowedregiononthecomet)wherethecometsurfacereflectancewouldn’tinterferewiththemeasurements.Thestylesvariedandwererefinedbasedondistancetothecomet,gasabundance,andtheunderstandingofthecometenvironment.VolatileAbundanceCampaign.AprimarystyleofgasobservationusedforAliceandotherinstrumentswasnamedVolatileAbundanceCampaign(VAC),whichwasdesignedtomeasureilluminatedgassesagainstthedeepspacebackground(a“DeepVAC”wasalongexposureversionofsuchpointing).ThiswasastablestaringobservationwheretheAliceslitwouldbepositionedsuchthatitwasapproximatelyhalfonthecometsurfaceforlocationcontextandhalfoffthesunwardlimb,thoughtheexactpositionvariedduetopointingerrorandnegotiationswithotherinstrumentteams.InnerComaRaster.Whenthecometwasmoreactive,InnerComaRasterobservationswouldbeused,whichwouldstareinseverallocationssteppingoffthecometinaradialline,typicallysunward,totrackhowthegasabundancefallsoffwithrespecttodistancefromthecomet.GreatCircle.SinceRosettawasnearthecenterofthecomet’scomathroughouttheencounterperiod,theAliceinstrumenttookfull360°“GreatCircle”observationstomeasurethedistributionofthegassesitwasimmersedinandhowtheychangedthroughouttheorbitaroundtheSun.NightStares.Theseobservationsweredesignedtolookattheilluminatedgasbetweenthespacecraftandthecometsurfaceeitheronthenightsideofthecometoroverlargeshadowedregions,whichhadtheusefuleffectofcuttingouttheskybackground.ComaRide-Alongs.TheAliceteamalsorodealongonsimilarobservationsplannedbyotherinstrumentteamssuchastheOSIRIS-ledVACs(termedDiurnalVACs)thatscannedbackandforthalongthesunlitcometlimblookingforconcentrationsofactivityandtheVIRTIS(VisibleandInfraredImagingSpectrometer)instrumentSnowflakesthatwereacombinationmanyoff-cometstares.StellarOccultations.AlthoughobservingstellaroccultationswasalwayspartoftheobservationconceptforAliceandotherinstrumentstomeasurecomagasesinabsorption,planningthoseprovedtobeunfeasibleduetotheuncertaintiesinthepredictedspacecraftorbitandattitude.However,asUVstarsweremonitoredpassingthroughtheAliceFOV,itwasrealizedthatitwasstillvaluabletoobservestellarappulsesevenifthestardidn’tgettotheabsolutelowestlayersofthecomajustabovethenucleus.TheAliceteamwasabletomodel(withinuncertainties)the


tangentialdistancesfromthenucleusofvariousUV-brightO-andB-typestarstotargetthosethatcouldbeobservedthroughthecolumnofgasnearthesunwardlimbofthecomet.Thelargepointinguncertaintiesandfrequentpointingplanchangesmadeplanningtheseobservationsdifficult,butwhensuchobservationsweresuccessful,thesamestarwasobservedagainatalatertimefarawayfromthecometasabaselinemeasurement.Thesetwospectrawerecomparedtomeasurethelevelofattenuationinspecificbandstocharacterizegastypeabundance.Thetechniquewasinitiatedlateinthemissionafterperihelion,soatimehistoryofthesemeasurementsthroughoutthecomet’sactivitycyclecouldnotobtained,butH2OandO2abundancesweremeasuredfromseveralofthe29stellarappulsemeasurementsets[11].

4.5AliceObservationTypes:SurfaceStudiesAsecondaryAlicescienceobjectivewastocharacterizethecometsurfacereflectanceintheUVandsearchforsurfaceicestobetterunderstandthecomet’scompositionandmorphologicalprocesses.Fortheseinvestigationstwotypesofobservationsoftheilluminatedcometsurfacewereplanned.SurfaceCenter.Thisstyleofobservationwouldstareatthesurfaceeitherdirectlynadirorwithadefinedoffsetfromnadirandtakecontinuoushistogramsasthecometrotatedunderneaththespacecraftwithaperiodof12hours24minutes.Ifthecomet’srotationaxiswasnotalignedsothatthescenewouldvarythroughtheobservationperiod,theobservationtypewouldbechangedtoaSurfaceScan.SurfaceScan.Thisstylewouldscanveryslowlyat0.01°/sbackandforthfromthesunwardlimbtotheanti-sunwardlimb(spacecraftY-axis,perpendiculartotheALICEslit)orasfarascouldbescannedwithinanobservationwindowwhenclosetothecomet.SurfaceRide-Alongs.AlicerodealongonanassortmentofotherinstrumentobservationsespeciallyonlongstaresandtheTargetsofOpportunity(ToO),wherespecificinterestingsurfacefeatureswerestaredatandmostinstrumentsrecordeddatatoprovideacomprehensivestudyofthearea.TherewereafairamountofobservationsoptimizedforotherinstrumentsthatAlicedidnottypicallyridealongbecausetheywouldn’tyieldhighqualityAlicedata.Forinstance,observationsthatincludedhighscanratesorshortstaresoflessthan5minutesforfastmappingofthesurfacewouldsmeartoomuchsurfaceareatobeofmuchuseforAliceanalysis.Alicetypicallyused5-10minutehistogramsforsurfaceobservationsand10-20minutehistogramsforgasobservationstobalancemeasurementsignalandspatialresolution.

4.6AliceObservationTypes:CalibrationsItwasveryimportantforseveralRosettainstrumentsincludingAlicetocharacterizetheirperformancewithregularcalibrations.Inadditiontothetypicalinstrumentopticsdegradationandalignmentchanges,eachAliceobservationcontributedtoanon-uniformdetectorgainsageffectthatneededtobetrackedandaccountedfor.ModeratelybrightUVstarswereusedasstablecalibrationsourcesforAlice,howevertheytypicallywerefaroff-cometanditwouldtakealotoftimeto


slewtoacalibrationstar’sposition.Sincealltheinstrumentteamspreferredtospendasmuchtimeaspossibleobservingthecomet,calibrationobservationswerecoordinatedtominimizeoff-cometpointingtime.OSIRIS,VIRTIS,andAlicewereabletocombinestellarcalibrationrequirementsreducingthetotalnumberofinstances.TheAliceinstrumentperformedseveraltypesofstellarcalibrationstotrackchangesinperformance:Bi-WeeklyStellarCalibrationswereusedtomeasuretheeffectivearea(instrumentsensitivity)byperiodicallystaringatcalibrationstarsandmeasuringchangesinresponseforthreerepresentativelocationsonthedetector.Bi-MonthlyFlatFieldCalibrationswereconductedasarasterscanalongthelengthoftheslitofacalibrationstartomeasurerelativesensitivitydifferencesacrossthe2Ddetector.Thissetofbi-weeklystellarandbi-monthlyflatfieldmeasurementsallowedfortheeffectiveareachangestobeappliedacrossthefulldetectorprovidingamethodofcorrectingfordetectorgainsagonaperpixelbasis.Bi-MonthlyCrossSlitScanCalibrationswereperformedthatconsistedofrasterscansperpendiculartotheslitofacalibrationstarusingpixellistmeasurementstopreciselydefinetheinstrumentfieldofview.Thiswasimportanttobeabletotrackandaccountfortheinstrumentflexureduetochangesinthermalenvironmentthroughoutthemission.Bi-MonthlyHighVoltage/DiscriminatorCalibrationswereperformedbystaringatacalibrationstarandmakingobservationswithasetofdetectorhighvoltage(HV)levelsanddiscriminatorlevels,settingsthatcanbeadjustedtochangethedetectorsensitivity.Theseinstrumentsettingscouldthenbeoptimizedtomitigategainsageffects.ThestandardoperationalhighvoltageanddiscriminatorvaluewerechangedseveraltimesduringthecourseofthemissionasshowninTable2.Thesechangesaretakenintoaccountinthecalibrateddata,butmaybeofinterestifworkingwiththerawAlicedata.

Table2:NominalAliceHighVoltageandDiscriminatorSettingsfortheRosettaMission

InitialSettings HV-3.8kV,Discriminator0.09V

2006Dec3 HV-3.7kV,Discriminator0.34V2007Feb23 HV-3.8kV,Discriminator0.09V2007Sep13 HV-3.9kV,Discriminator0.09V2014Apr2 HV-4.0kV,Discriminator0.09V

2014May12 HV-4.0kV,Discriminator0.45V2015June25 HV-4.1kV,Discriminator0.45V2016Feb23 HV-4.2kV,Discriminator0.45V


Bi-WeeklyDarkCalibrationswerealsoperformedwherehistogramsweretakentomeasureandaccountforthedetectorresponsewiththeaperturedoorclosed.

4.7NCIRandOCIRContextImagesEventothetrainedeye,Alicehistogramscanbedifficulttointerpret,especiallywhenviewingpartiallyshadowedterrainandthevariableconcentrationsandstructuresofgasanddustcomingoffthecomet.Toaidintheinterpretationofhistograms,itwasveryusefulexamineNavigationCamera(NavCam)andOSIRISWide-AngleCamera(WAC)imagestakenaroundthesametimetoprovidecontexttothesceneasshowninFigure7.ThetimingoftheimagesplannedbytheSpacecraftandOSIRISteamswereoftennotwellcorrelatedtoimportantAliceobservations.Soinanotherexampleofcooperationwithinthemissiontoachievethebestsciencepossible,theAliceteamwaspermittedtoplanadditionalNavCamandOSIRISImagesRequests(NCIRsandOCIRs)withtheconstraintthattheimagedatavolumewascoveredbyAliceresourceallocations.TheNavCamwasusedsolelyforAlicecontextimaginguntilMay11th2016,wheredatavolumeconstraintsstartedtosignificantlyincrease.AtthatpointtherewasatransitiontouseOSIRISWACimages,whichcouldbecompressed,savingalargeamountofdatavolumetoberepurposedasadditionalAlicescienceobservations.

Figure7.ExampleofanAlicehistogram(left)of67PandaNavCamcontextimage[12]withanoverlaidAliceFOV(right)acquiredon08/18/2014(2014DOY230).Thecontextimagecanhelpidentificationofscenecharacteristicstocorrelatehistogramfeatures.

4.8HEETMeasurementsAsasecondarysciencepursuit,Alicewasperiodicallyconfiguredtomeasurethehigh-energyelectron,or“HEET”,componentofthespaceenvironment.TheAlicedetectorcouldsensepenetratingelectronsandwouldrecordtheminthedetectorcountratedata.ThiswasaccomplishedatlowdatavolumecostsincethecountrateisincludedintheAlicehousekeepingdatapacket.HEETmeasurementsweremadewiththeAliceaperturedoorclosed,thedetectorhighvoltageon,andatanelevated


housekeepingrateof1Hzinanattempttocaptureshort-termelectrondensitychanges.

5MissionPlanningandProductDevelopment

5.1MissionPlanningCoordinationAtfirstglancetheprospectofbalancingtheneedsofthe21Rosettamissioninstrumentsseemsdaunting,especiallywitheachbasedatseparateinstitutionsscatteredacrossEuropeandtheUS,indifferenttimezones,andstaffedbyteamspassionateabouttheiranticipatedscience.Nonetheless,themissionwasaresoundingsuccessanddatawererecordedtosatisfytheAliceinstrumentsciencegoals.Thiswasachievedthroughtirelessoperationsdesign,development,andplanningsupportfromalltheinstrumentteamsandtheESAscienceoperationsteam.Additionally,thefreedomtoself-harmonizescienceplanningwithintheRosettasciencecommunityinsteadofbeingdirectedbymanagementledtoholisticallybalancedsolutions.Anobservationimportancehierarchybetweeninstrumentswasnotalwaysclearasmissionsciencegoalsweren’trequirementsororderedbypriority.However,teamsusuallycametoanagreementontheirownwhenscienceprioritiesweremadebasedondatasetuniquenessandcontinuity.Oftentheteamsmadecompromisestomergeobservationtypesforhighlydesiredperiodssuchasaddingperiodicstaresduringscans,adjustingscanrates,starelocations,etc.Fortunately,thistaskbecamemoremanageableasonlyasubsetofinstrumentteamsactivelyparticipatedinresourcenegotiationsincludingAlice,MIRO(MicrowaveInstrumentfortheRosettaOrbiter),OSIRIS,ROSINA(RosettaOrbiterSpectrometerforIonandNeutralAnalysis),RPC(RosettaPlasmaConsortium),VIRTIS,andsometimesthePhilaeLanderTeam.Theresourcesthattheremaininginstrumentstypicallyrequiredwerelowenoughthattheywereautomaticallygrantedortheywereabletoachievetheirdesiredobservationsbyridingalongonthepointingdesignedbytheotherteams.Therewereexceptionstothetypicalnegotiationprocess,thebiggestbeinghowtobalancethepotentialresourceneedsofthePhilaeLanderandits10instrumentsifandwhenitbecamefunctionalagaininthemonthsafterthelanding.Sometimes,thisledtoresourcesandschedulingtosearchfortheLanderorrecoveraLandersignaltakingpriorityoverotherRosettaspacecraftplans.Thatinturnresultedinlatechangestothetrajectorytooptimizecommunicationandviewopportunities,addedmultiple-caseplanning,droveadditionalpointingconstraints,andrequiredresourcestobereserved.Unfortunately,amongthemanycontactattemptstherewereonlybriefadditionalcontactswithPhilaeroughly7andthen8monthsafterlanding(seeTable1forthedates),thecontactswerenotsufficienttocommandextendedadditionallanderscienceactivities.

5.2Long,Medium,andShortTermPlanningPhasesEachinstrumenthaditsownmasterscienceplanthatdefinedmeasurementgoals,theirrelativeimportance,andobservationmethodstoaccomplishthem.Toachievethesegoalseachteamwouldfirstmakerequestsfortheneededresourcesatthe


LongTermPlanning(LTP)phaseincludingspacecraftpointing,datavolume,power,andtelecommandcount.Thiswasdonewitharesourceandrequesttrackingspreadsheettypicallycovering4weeksofoperationsandwassubdividedintoshortblocksoftimedefiningspacecraftactivitiesorblocksavailableforinstrumentteamstorequest.Thistypicallyresultedin3-4hourinstrumentobservationblocksseparatedby30-70minutespacecraftactivityblocksthatinstrumentobservingmayormaynotneedtoavoidduetopropulsionandnon-optimalpointingactivities.Everyobservationblockfortheentiretwo-yearencounterwasreviewedandnegotiatedforbytheinstrumentandspacecraftteams.HarmonizationofplanstookplacewithRSGSliaisonscientistsmoderatingtheprocessfromESA’sESACfacilityinSpainwhilethevariousinstrumentteamsdialedinfromaroundtheglobe.Harmonizingtheobservationrequestsoftentookalotoftimeandcompromises,andoccasionallygotheatedwhendesignscouldn’tachieveeveryone’sneeds.However,itwasatestamenttointernationalcollaborationandcooperationthatmostnegotiationswereattainedsmoothlyandwithenoughtimetoimplement.AfterharmonizationtheliaisonscientistswouldhandofftheplantotheMediumTermPlanning(MTP)counterpartswherecommandingproductswouldbebuilt,refined,andvalidated.MTPperiodsweretypically4weekslong,correspondingtotheLTPplanningspreadsheetperiods.Finallytherewasopportunityfornon-standardoremergencyactivitiesandchangestobeaddedattheShortTermPlanning(STP)phase,typically1-weekduration.

5.3EvolutionofOperationsPlanningDuringthefirst9monthsofencounterthecometaryactivitywaslowandthespacecraftorbitaltrajectoryplanningwasverystable,excludingthevariousPhilaelandingscenarios.Thisallowedforlongdevelopmentprocessdurations,forexampletheLTPphasestarted6monthsbeforetheexecutionoftheplannedactivities,theMTPphasekickedoff2.5monthsbeforeexecution,andtherewereseveralweeksforSTPadjustments.ThisledtoconcurrentplanningofseveralMTPperiods,butplentyoftimeforleisurelyplanning.However,thisluxurywasshort-livedasthecometbecameactiveandtheorbitaltrajectoryneededtobeadjustedonamuchshortertimeframetokeepasafedistancefromthecometandparticularlyactiveareassuchasthecometneckandthesideofthecometpointedtotheSunwherethegasanddustdensitieswerehigher.Planningphaseswerehighlycompressedafterthispoint,butthemissionwasabletomaintaintheabilitytoplanobservationsthatcorrelatedtospecificcometphaseangles,locations,andothercharacteristics,whichresultedinmuchbetterformulateddatasetswhencomparedtoarbitrarypointingscheduling.Theremaining16monthsofthemissionplanningproceededatafranticpacewiththeLTPphasestarting2monthsbeforeexecutionandtheMTPandSTPphasesmergedandkickedoffamonthbeforeexecution.Unfortunately,lastminutechangestopointingattheSTPlevelbyotherinstrumentteamswerenotrareandusuallyrequiredascrambletoaccommodateinthecommandingforAliceandotherinstruments.However,theAliceteamwasappreciativefortheopportunitytomakeSTPlevelchangestothepointingaswelltooptimizesomeobservations,especiallystellarappulses,whicharediscussedinthe“AliceObservationTypes:GasObservations”Section.


Thecompressionoftheplanningtimeframeimpactedallfacetsofoperationsfromobservationdesign,resourceharmonization,implementation,developmenttools,productreviewtiming,andproductvalidationtiming.Itrequiredthatprocessesandsystemsbedevelopedthatcouldbeefficient,adaptable,reliable,andmodular.AmongtheAliceteam,awikiwasusedforefficientcommunicationofproductstatusandSubversion(svn)wasusedforproductversiontracking.ModularPython,Perl,andbashscriptswereusedforproductdevelopmentandverificationthatallowedforquickadjustmentsandtestingasmissioninputsandformatschangedandnewcapabilitywasneeded.ESAmissionsimulationmodels(includingESA'sMappingAndPlanningPayloadScience(MAPPS)software[7]),SatelliteToolKit(STK)scenarios,andanAliceEngineeringQualificationUnitwerealsousedforproductvalidation.

6MissionPerspectives

6.1LessonsLearnedSummaryCooperationandjointplanningbetweenteamsleadtobetteroverallscience.ToOs,contextimages,starcalibrations,andothercasesofcorrelatingseveralinstrumentmeasurementsledtoamorecompleteunderstandingofthecometprocessesandobservedscenes.Transparencyintheoperationsplanningforallinstrumentteamshelpedwithunderstandingeachother’sgoalsandhelpedmaketheharmonizationprocessmoreefficient.Thiswasachievedbydocumentingobservingstyles,labelingobservationsinastandardizedway,andcollectingobservationrequestsinoneplaceforside-by-sidereviewandharmonizing.Itisimportanttoautomateplanningandcommandingasmuchaspossibletominimizemanualeffort.Thisleadstomoreefficientplanningandreducesoperationalrisk.Itwasfoundthatrollingoutcapabilityincrementallyanduseofopensourcesoftwarequickenedimplementationofautomationsoftware.ArecommendationforfutureAlice-likeinstrumentswouldbetoupdateflightsoftwaresoitisflexibleenoughtocreateadditionalinstrumentsequencesthatcombinesetsofcommandsusedregularlytoreduceriskincommandloadsandmakeoperationsdevelopmentmoreefficient.Intermsofpersonnelitwasfoundthatitwasworththeresourcestohavesomeonewhoisworkingonoperationsdevelopmentwithintimatefamiliaritythesystemsanditsneedswithsufficienttimetobuildandtestoperationstoolstoimprovetheprocess.Thereisalsostrongvalueinhavingaliaisonbetweenoperationsteamandscienceteamwhocanhelptranslaterequirements,constraints,jargon,etc.betweenthetwoteams.Havingainstrumentflexibleinpointing(madepossiblewithalargeFOVandlongexposuretimes)meanttheAliceinstrumentcouldtakeadvantageofobservation


blocksthatwereoflittleusetootherinstruments,givingAlicemoreprimepointingthanitwouldhaveotherwisesecured.Implementinganongoingprocessofassessingoperationalconstraintsandassumptionshelptoidentifyareastooptimize.Forinstance,thegainsagratewasreassessedresultinginadjustmentsinBadDogavoidanceandridealongobservationplanning.Thefreedomtoself-harmonizescienceplanningwithintheRosettasciencecommunityinsteadofusingarbitraryschedulingorsolutionsbeingdirectedbymanagementledtoholisticallybalancedsolutionsandadditionalcollaborations-notjustcompromises.Beingclearwithcommunicationandschedulingdeadlineswasveryimportanttoreduceissuesforaninternationalmissionwithteammembersspanningmanytimezonesandnativelanguages,butrequiringmanyweeklymeetingsforplanningcoordinationandquick-turnaroundresponses.OneareaofcommunicationsthatcouldbeimprovedonRosettawasoccasionallyESAmissionsoftwareusedforproductdevelopmentandmodelingwouldchangewithoutaheadoftimenoticeimpactingongoingdevelopment.Thiswouldrequirechangestointerfaceanddevelopmentsoftwareinaveryshorttimeframe.Bettertransparencyintomissionsoftwarechangesbeforetheyareimplementedwouldhavereducedrushchanges,softwareworkaroundsandrollbacks,andriskofcommanderrors.Havingmission-commonvisualizationandresourcetrackingtoolsareimportanttoprovideasingleframeofreferenceforallteamsfacilitatingcompatiblesolutionstoend-to-endoperationsdevelopmentfromobservationdesignthroughimplementation.Participationinmeetingsisnecessaryforallteamsthatdesiretohaveavoiceindecisionsbeingmade.Revisitingdecisionscanwastealotofpeople’stime.

6.2AliceLogbookForfurtherAliceoperationaleventdetailspleaserefertotheRosettaAliceLogbookthatwillbeintheRosettaAliceArchive.ItcontainsdetailsforAliceevents,majorspacecraftevents,allobservationsbynameandbydate,andotherinformationvaluabletousingandunderstandingAlicedata.

7AcknowledgementsRosettaisanESAmissionwithcontributionsfromitsmemberstatesandNASA.WewouldliketothankthemembersoftheRosettaScienceGroundSegmentandMissionOperationsCenterteamsfortheirexpertanddedicatedhelpinplanningandexecutingtheAliceobservations.WewouldliketothankthededicatedscienceandengineeringstaffatSwRIforalltheircontributionsthroughtheyears.TheAliceteamacknowledgescontinuingsupportfromNASAviaJetPropulsionLaboratorycontract1336850totheSouthwestResearchInstitute.


8References[1]S.A.Stern,D.C.Slater,J.Scherrer,J.Stone,M.Versteeg,M.F.A’Hearn,J.L.Bertaux,P.D.Feldman,M.C.Festou,JoelWM.ParkerandO.H.W.Siegmund,“ALICE:THEROSETTAULTRAVIOLETIMAGINGSPECTROGRAPH”,SpaceScienceReviews(2007)128,pp.507–527[2]Siegmund,O.W.H.,Stock,J.,Raffanti,R.,Marsh,D.,andLampton,M.,“UVandX-RaySpectroscopyofAstrophysicalandLaboratoryPlasmas”,Proceedingsfromthe10thInternationalColloquium,Berkeley,CA3–5February1992,inSilver,E.andKahn,S.(eds.),p.383.[3]Siegmund,O.W.H.,Everman,E.,Vallerga,J.,andLampton,M.,“OptoelectronicTechnologiesforRemoteSensingfromSpace”,ProceedingsoftheSPIE,SPIE,Bellingham,Washington,USA,1987Vol.868,p.17.[4]JoelWm.Parker,AndrewJ.Steffl,S.AlanStern.“ROSETTA-ALICETOPLANETARYSCIENCEARCHIVEINTERFACECONTROLDOCUMENT”RosettaPSADocumentNo.8225-EAICD-01,March2016Rev4Chg0[5]Cometon27March2016–NavCam,CopyrightESA/Rosetta/NavCam–CCBY-SAIGO3.0,Id358019,Released01/04/2016,http://www.esa.int/spaceinimages/Images/2016/04/Comet_on_27_March_2016_NavCam[6]Comet’sdustyenvironment,CopyrightESA/Rosetta/MPSforOSIRISTeamMPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA,Id345498,Released13/08/2015,http://www.esa.int/spaceinimages/Images/2015/08/Comet_s_dusty_environment[7]MikeAshman,MaudBarthélémy,LaurenceO'Rourke,MiguelAlmeida,NicolasAltobelli,MarcCostaSitjà,JuanJoséGarcíaBeteta,BernhardGeiger,BjörnGrieger,DavidHeather,RaymondHoofs,MichaelKüppers,PatrickMartin,RichardMoissl,ClaudioMúñozCrego,MiguelPérez-Ayúcar,EduardoSanchezSuarez,MattTaylor,ClaireVallat,“RosettascienceoperationsinsupportofthePhilaemission”,ActaAstronautica,Volume125,August–September2016,Pages41-64[8]JohnNoonan,EricSchindhelm,JoelWm.Parker,AndrewSteffl,MichaelDavis,S.AlanStern,ZuniLevin,SaschaKempf,MihalyHoryani,“AninvestigationintopotentialcausesoftheanomalisticfeatureobservedbytheRosettaAlicespectrographaround67P/Churyumov–Gerasimenko”,ActaAstronauticaVolume125,August–September2016,Pages3-10[9]AndreaAccomazzo,SylvainLodiot,VicenteCompanys,“Rosettamissionoperationsforlanding”,ActaAstronautica,Volume125,August–September2016,Pages30-40


[10]Rosetta’sjourneyaroundthecomet,CopyrightESA,Released:05/08/2016,http://www.esa.int/spaceinvideos/Videos/2016/08/Rosetta_s_journey_around_the_comet[11]BrianA.Keeney,S.AlanStern,MichaelF.A’Hearn,Jean-LoupBertaux,LoriM.Feaga,PaulD.Feldman,RichardA.Medina,JoelWm.Parker,JonP.Pineau,EricSchindhelm,AndrewJ.Steffl,M.Versteeg,andHaroldA.Weaver,“H2OandO2AbsorptionintheComaofComet67P/Churyumov-GerasimenkoMeasuredbytheAliceFar-UltravioletSpectrographonRosetta”,2017,MonNotRAstronSocstx1426.doi:10.1093/mnras/stx1426[12]Cometon18August2014–NavCam,ROS_CAM1_20140818T200718,CopyrightESA/Rosetta/NavCam–CCBY-SAIGO3.0,horizontallyinvertedtomatchAlicehistogramorientation,https://imagearchives.esac.esa.int/picture.php?/7204/categories/created-monthly-calendar-2014-8-18

Documents

Rosetta Alice Ultraviolet Spectrograph Flight Operations ... · PDF fileRosetta Alice Ultraviolet Spectrograph Flight Operations and Lessons Learned 2 1 The Rosetta Mission and Comet