The effects of cosmic rays and solar flares on the

ToappearinProc.SPIE6276,“HighEnergy,Optical,andInfraredDetectorsforAstronomyII”,2006,inpressTheeffectsofcosmicraysandsolarflaresontheIRACdetectors:thefirsttwoyearsofin-flightoperationJosephL.Hora*a,BrianM.Pattena,GiovanniG.Fazioa,WilliamJ.GlaccumbaHarvard-SmithsonianCenterforAstrophysics,60GardenSt.,MS-65,Cambridge,MAUSA02138;bSpitzerScienceCenter,MS220-6,Caltech,Pasadena,CA,USA91125ABSTRACTTheInfraredArrayCamera(IRAC)isafour-channelcameraontheSpitzerSpaceTelescope,oneofthreefocalplanescienceinstruments.IRACusestwopairsof256×256pixelInSbandSi:AsIBCdetectorstoprovidesimultaneousimagingat3.6,4.5,5.8,and8μm.IRACexperiencesafluxofcosmicraysandsolarprotonsthatproducetransienteffectsinscienceimagesfromeachofthearrays,with4–6pixelspersecondbeingaffectedduringeachintegration.Duringextremesolarflares,IRACexperiencesamuchhigherrateoftransientswhichaffectsthesciencedataquality.WepresentcosmicrayratesandobserveddetectorcharacteristicsforIRACduringthefirsttwoyearsofscienceoperation,andratesobservedinaperiodofelevatedsolarprotonfluxduringanintensesolarflareinJanuary2005.WeshowthechangestotheIRACdetectorsobservedsincelaunch,andassesstheirimpactstothesciencedataquality.Keywords:SpitzerSpaceTelescope,IRAC,infrareddetectors,cosmicrays,imageprocessingandartifacts,spaceinstrumentation1.INTRODUCTIONTheInfraredArrayCamera1,2(IRAC)isoneofthreefocalplaneinstrumentsontheSpitzerSpaceTelescope3,4.IRACisafour-channelcamerathatobtainssimultaneousimagesat3.6,4.5,5.8,and8μm.Twoadjacent5.2′×5.2′fieldsofviewintheSpitzerfocalplaneareviewedbythefourchannelsinpairs(3.6and5.8μm;4.5and8μm).Allfourdetectorarraysinthecameraare256×256pixelsinsize,withthe3.6and4.5μmchannelsusingInSbarraysandthe4.5and8μmchannelsusingSi:AsIBCarrays5,6,7,8,9.Bothdetectortypeshavepixelswithasizeof30×30μm,buttheSi:Asdetectormaterialisthicker(50μm)thanintheInSbdevices(7μm).LaunchedonAugust25,2003,SpitzerwasplacedintoanEarth-trailingsolarorbit,whichpositionsthespacecraftwelloutsideanyshieldingeffectsoftheEarth’smagnetosphereorenhancementsinchargedparticlefluxduetotheEarth’sradiationbelts9,11.Followinglaunch,Spitzerunderwenta3-monthevaluationduringtheIn-OrbitCheckout(IOC)andScienceVerificationperiods12,duringwhichthetelescopewasfocused13,14andafunctionalcheckoutoftheinstrumentsandobservationaltemplateswasperformed2.Sincelaunch,IRAChasbeenoperatedformorethan7000hours,collectingawealthofcalibrationandsciencedata.Thecameraisnowbeingusedregularlyasapartofthenominaloperations(nomops)phaseofthemission.Onorbit,thearraysexperienceanenvironmentthatincludeshighenergyionswhichproducetransienteventsinIRACimages.Transienteventsarewhereasignalisproducedinoneormorepixelsoveratimemuchshorterthanthetotalexposuretimefortheimage,andarenotcausedbyphotonsfromastronomicalsources.Thesignalsarerandombothspatiallyandtemporally.Ingeneral,thetransientisnotevidentinthefollowingexposure;however,forsomestrongtransients,aresidualimageispresentinsubsequentframesthatfadesinamannersimilartoresidualimagesfrombrightastronomicalsources.Herewewillrefertoalltransientsas“cosmicrays”(CRs),regardlessoftheirorigin.Inapreviouspaper15,wepresentedthecharacteristicsofCRsastheyappearintheIRACscienceimages.Inthispaper,wepresentCRratesandchangesindetectorpropertiesobservedintheperiodDecember2003throughApril2006,aswellasduringaperiodofelevatedsolarprotonfluxfollowingaseriesofX-classsolarflaresinJanuary2005.*[e-mail];phone[phone];fax[phone];cfa-’sperformance.Thedatasetconsistsofobservationsofastronomicalstandardstars,regionsoflowandhighzodiacalbackgroundlevels,andframesobtainedwiththeinternalfloodcalibrationlamps.Thecalibrationstrategyandresultshavebeendescribedelsewhere16.Sincelaunch,therehasbeennodetectedchangetotheaverageinstrumentresponsivityasdeterminedfrommeasurementsofastronomicalstandards.Inthispaperwewillfocusonchangestoindividualpixelssincelaunchthatarepresumablyaresultofdamageduetoradiationorothereffects.Thedetectorcharacteristicstrackedarethenumberofhot(highdarkcurrent),dead,ornoisypixelsineacharray,andthemediannumberofCR-affectedpixelspersecond.Hotpixelsaredefinedasthosehavingdarkcurrent1e–/sfortheInSbarrays(10e–/sfortheSi:Asarrays),deadpixelsarethosewitharesponsivity50%ofthemedianofnearbypixels,andnoisypixelsarethosewithastandarddeviationoftwicethatofthearraymedianinsuccessiveframesofthelowbackgroundregion.2.1.PerformancemonitoringdatasetandreductionThedatausedforassessmentofthenumberofhotandnoisypixelsareasetof100secdarkframes(50secforchannel4)takenatleasttwicepercampaign.Inmostcases,onlythelastsetofdarkframesinacampaignwereusedforthisanalysis.Thesearecloseintimetothefloodlampframesthatareusedtodeterminethenumberof“dead”orlowresponsivitypixels.MeasurementsfromIOCthroughnomopscampaign30(September2003–April2006)wereanalyzed.Forthestatisticsbelow,onlya250×250pixelregionisconsidered(region[3:252,3:252])toavoidnoisycolumnsandrowsalongthearrayedges.Thedarkframedata