Constraints on high-energy neutrino emission from SN 2008D

0.48 MB·English

by  IceCube Collaboration

Astronomy&Astrophysicsmanuscriptno.15770 (cid:13)c ESO20111
January21,2011
Constraints on high-energy neutrino emission from SN 2008D
IceCubeCollaboration:R.Abbasi1,Y.Abdou2,T.Abu-Zayyad3,J.Adams4,J.A.Aguilar1,M.Ahlers5,K.Andeen1,
J.Auffenberg6,X.Bai7,M.Baker1,S.W.Barwick8,R.Bay9,J.L.BazoAlba10,K.Beattie11,J.J.Beatty12,13,
S.Bechet14,J.K.Becker15,K.-H.Becker6,M.L.Benabderrahmane10,S.BenZvi1,J.Berdermann10,P.Berghaus1,
D.Berley16,E.Bernardini10,D.Bertrand14,D.Z.Besson17,M.Bissok18,E.Blaufuss16,J.Blumenthal18,
D.J.Boersma18,C.Bohm19,D.Bose20,S.Bo¨ser21,O.Botner22,J.Braun1,S.Buitink11,M.Carson2,D.Chirkin1,
B.Christy16,J.Clem7,F.Clevermann23,S.Cohen24,C.Colnard25,D.F.Cowen26,27,M.V.D’Agostino9,
M.Danninger19,J.C.Davis12,C.DeClercq20,L.Demiro¨rs24,O.Depaepe20,F.Descamps2,P.Desiati1,
G.deVries-Uiterweerd2,T.DeYoung26,J.C.D´ıaz-Ve´lez1,M.Dierckxsens14,J.Dreyer15,J.P.Dumm1,
M.R.Duvoort28,R.Ehrlich16,J.Eisch1,R.W.Ellsworth16,O.Engdegård22,S.Euler18,P.A.Evenson7,O.Fadiran29,
1 A.R.Fazely30,A.Fedynitch15,T.Feusels2,K.Filimonov9,C.Finley19,M.M.Foerster26,B.D.Fox26,
1 A.Franckowiak21,R.Franke10,T.K.Gaisser7,J.Gallagher31,M.Geisler18,L.Gerhardt11,9,L.Gladstone1,
0
T.Glu¨senkamp18,A.Goldschmidt11,J.A.Goodman16,D.Grant32,T.Griesel33,A.Groß4,25,S.Grullon1,M.Gurtner6,
2
C.Ha26,A.Hallgren22,F.Halzen1,K.Han4,K.Hanson14,1,K.Helbing6,P.Herquet34,S.Hickford4,G.C.Hill1,
n
K.D.Hoffman16,A.Homeier21,K.Hoshina1,D.Hubert20,W.Huelsnitz16,J.-P.Hu¨lß18,P.O.Hulth19,K.Hultqvist19,
a
J S.Hussain7,A.Ishihara35,J.Jacobsen1,G.S.Japaridze29,H.Johansson19,J.M.Joseph11,K.-H.Kampert6,
0 A.Kappes1,43,T.Karg6,A.Karle1,J.L.Kelley1,N.Kemming36,P.Kenny17,J.Kiryluk11,9,F.Kislat10,S.R.Klein11,9,
2 J.-H.Ko¨hne23,G.Kohnen34,H.Kolanoski36,L.Ko¨pke33,D.J.Koskinen26,M.Kowalski21,T.Kowarik33,

M.Krasberg1,T.Krings18,G.Kroll33,K.Kuehn12,T.Kuwabara7,M.Labare20,S.Lafebre26,K.Laihem18,
]
E H.Landsman1,M.J.Larson26,R.Lauer10,R.Lehmann36,J.Lu¨nemann33,J.Madsen3,P.Majumdar10,A.Marotta14,
H R.Maruyama1,K.Mase35,H.S.Matis11,M.Matusik6,K.Meagher16,M.Merck1,P.Me´sza´ros27,26,T.Meures18,
. E.Middell10,N.Milke23,J.Miller22,T.Montaruli1,37,R.Morse1,S.M.Movit27,R.Nahnhauer10,J.W.Nam8,
h
U.Naumann6,P.Nießen7,D.R.Nygren11,S.Odrowski25,A.Olivas16,M.Olivo22,15,A.O’Murchadha1,M.Ono35,
p
- S.Panknin21,L.Paul18,C.Pe´rezdelosHeros22,J.Petrovic14,A.Piegsa33,D.Pieloth23,R.Porrata9,J.Posselt6,
o
P.B.Price9,M.Prikockis26,G.T.Przybylski11,K.Rawlins38,P.Redl16,E.Resconi25,W.Rhode23,M.Ribordy24,
r
t A.Rizzo20,J.P.Rodrigues1,P.Roth16,F.Rothmaier33,C.Rott12,T.Ruhe23,D.Rutledge26,B.Ruzybayev7,
s
a D.Ryckbosch2,H.-G.Sander33,M.Santander1,S.Sarkar5,K.Schatto33,S.Schlenstedt10,T.Schmidt16,
[ A.Schukraft18,A.Schultes6,O.Schulz25,M.Schunck18,D.Seckel7,B.Semburg6,S.H.Seo19,Y.Sestayo25,

1 S.Seunarine39,A.Silvestri8,K.Singh20,A.Slipak26,G.M.Spiczak3,C.Spiering10,M.Stamatikos12,40,T.Stanev7,
v G.Stephens26,T.Stezelberger11,R.G.Stokstad11,S.Stoyanov7,E.A.Strahler20,T.Straszheim16,G.W.Sullivan16,
2
Q.Swillens14,H.Taavola22,I.Taboada41,A.Tamburro3,O.Tarasova10,A.Tepe41,S.Ter-Antonyan30,S.Tilav7,
4
9 P.A.Toale26,S.Toscano1,D.Tosi10,D.Turcˇan16,N.vanEijndhoven20,J.Vandenbroucke9,A.VanOverloop2,
3 J.vanSanten1,M.Voge25,B.Voigt10,C.Walck19,T.Waldenmaier36,M.Wallraff18,M.Walter10,Ch.Weaver1,
1. C.Wendt1,S.Westerhoff1,N.Whitehorn1,K.Wiebe33,C.H.Wiebusch18,G.Wikstro¨m19,D.R.Williams42,
0 R.Wischnewski10,H.Wissing16,M.Wolf25,K.Woschnagg9,C.Xu7,X.W.Xu30,G.Yodh8,S.Yoshida35,and
1 P.Zarzhitsky42
1
: (Affiliationscanbefoundafterthereferences)
v
i ReceivedSeptember16,2010;acceptedDecember9,2010
X
r
a
ABSTRACT
Context.SN2008D,acorecollapsesupernovaatadistanceof27Mpc,wasserendipitouslydiscoveredbytheSwiftsatellitethroughanassociated
X-rayflash.Corecollapsesupernovaehavebeenobservedinassociationwithlonggamma-rayburstsandX-rayflashesandaphysicalconnection
iswidelyassumed.Thisconnectioncouldimplythatsomecorecollapsesupernovaepossessmildlyrelativisticjetsinwhichhigh-energyneutrinos
areproducedthroughproton-protoncollisions.ThepredictedneutrinospectrawouldbedetectablebyCherenkovneutrinodetectorslikeIceCube.
Aims.AsearchforaneutrinosignalintemporalandspatialcorrelationwiththeobservedX-rayflashofSN2008Dwasconductedusingdata
takenin2007-2008with22stringsoftheIceCubedetector.
Methods.Eventswereselectedbasedonaboosteddecisiontreeclassifiertrainedwithsimulatedsignalandexperimentalbackgrounddata.The
classifierwasoptimizedtothepositionanda“softjet”neutrinospectrumassumedforSN2008D.Usingthreesearchwindowsplacedaroundthe
X-raypeak,emissiontimescalesfrom100−10000swereprobed.
Results.Noeventspassingthecutswereobservedinagreementwiththesignalexpectationof0.13events.Upperlimitsonthemuonneutrinoflux
fromcorecollapsesupernovaewerederivedfordifferentemissiontimescalesandtheprincipalmodelparameterswereconstrained.
Conclusions.Whilenomeaningfullimitscanbegiveninthecaseofanisotropicneutrinoemission,theparameterspaceforajettedemissioncan
beconstrained.Futureanalyseswiththefull86stringIceCubedetectorcoulddetectupto∼100eventsforacore-collapsesupernovaat10Mpc
accordingtothesoftjetmodel.
Keywords.corecollapsesupernovae–SN2008D–cosmicneutrinos–SN-GRBconnection–high-energyneutrinos
1. Introduction generation,andaspherical,pressure-resistantglasshousing.The
DOMsdetectCherenkovphotonsemittedbyrelativisticcharged
Observationsinrecentyearshavegivenrisetotheideathatcore particlespassingthroughtheice.Inparticular,thedirectionsof
collapsesupernovae(SNe)andlongdurationgamma-raybursts muons (either from cosmic ray showers above the surface or
(GRB)haveacommonoriginormayevenbetwodifferentas- neutrino interactions within the ice or bedrock) can be well re-
pects of the same physical phenomenon, the death of a mas- constructedfromthetrack-likepatternandtimingofhitDOMs.
sive star with M > 8M(cid:12) (for a review, see Woosley, Bloom Identification of neutrino-induced muon events in IceCube has
2006). Like GRBs, SNe could produce jets, though less ener- beendemonstratedinAchterbergetal.(2006)usingatmospheric
getic and collimated and possibly “choked” within the stellar neutrinosasacalibrationtool.Sourcesinthenorthernsky,like
envelope.ObservedassociationsofsupernovaewithXRFs,short SN 2008D, can be observed with very little background since
X-rayflasheswithsimilarcharacteristicstolongGRBs,suggest contaminationbyatmosphericmuontracksiseliminatedbythe
including XRFs in the SN-GRB connection as well. Although shielding effect of the Earth. When SN 2008D was discovered,
XRFs are considered a separate observational category from theinstallationofIceCubewasaboutonequartercompletedand
GRBs,acommonoriginandacontinuoussequenceconnecting thedetectorwastakingdatawith22strings.
them have been suggested (Lamb et al. 2004, Yamazakia et al.
As shown above, a search for cosmic neutrinos from core
2004).XRFcouldbelongGRBswithveryweakjetsorsimply
longGRBsobservedoff-axis.SeveralXRFsororlongduration, collapse SNe is motivated by both observational evidence
and theoretical predictions. While analyses using catalogs of
soft-spectrum GRBs have been observed in coincidence with
SNe/GRBswithtiminguncertainties∼1dasthesignalhypoth-
core collapse SNe thus far: SN 1998bw (Galama et al. 1998),
esishavebeenperformedonarchivedAMANDA/IceCubedata
SN 2003lw (Malesani et al. 2004), SN 2003dh (Hjorth et al.
(seeLennarz2009forSNeandAbbasietal.2010bforGRBs),
2003), SN 2006aj (Pian et al. 2006), and of course SN 2008D
theunprecedentedlyprecisetiminginformationavailableforSN
(Soderbergetal.2008,Modjazetal.2009,Mazzalietal.2008).
2008Dsuggestsadesignatedstudyofthisevent.Whileelectro-
For SN 2007gr (Paragi et al. 2007) and SN 2009bb (Soderberg
magnetic observations provide no conclusive evidence for the
etal.2010),twocorecollapseSNenotassociatedwithanXRF
existenceofhighlyrelativisticjets,soft,hiddenjetscouldbere-
or GRB, recent radio observations provide strong evidence for
jets with bulk Lorentz factors of Γ > 1. If some core collapse vealedbyhighenergyneutrinos,assumingsufficientalignment
withthelineofsight.
SNeindeedformsuch”soft”jets,protonsacceleratedwithinthe
jetcouldproduceTeVneutrinosincollisionswithprotonsofthe The paper is organized as follows: Section 2 discusses the
stellarenvelope(Razzaqueetal.2005,Ando&Beacom2005). assumedmodelforneutrinoproduction.Section3describesthe
The soft jet scenario for core collapse SNe can be probed with experimental and simulated data used for the analyis. The se-
high-energy neutrinos even if the predicted jets stall within the lection criteria used to separate signal events from background
stellar envelope and are undetectable in electromagnetic obser- are detailed in Section 4. Section 5 presents the results of the
vations. searchandconstraintsderivedtherefrom.Finally,theanalysisis
OnJanuary9,2008,theX-raytelescopeaboardtheSWIFT summarizedinSection6.
satellite serendipitously discovered a bright X-ray flash during
a pre-scheduled observation of NGC 2770. Optical follow-up
observations were immediately triggered and recorded the op-
tical signature of SN 2008D, a core collapse supernova of type
Ib at right ascension α = 09h 09m 30.70s and declination 2. Modelneutrinospectrum
δ=33◦08(cid:48)19.1” (Soderbergetal.2008).SN2008Doffersare-
Amodelfortheemissionofhigh-energyneutrinosinjetsformed
alisticchancetodetecthigh-energysupernovaneutrinosforthe
by core collapse supernovae has been proposed by Razzaque,
firsttimesincetheobservedX-raypeakprovidesthemostpre-
Meszaros,andWaxman(2005)andfurtherelaboratedbyAndo
cisetiminginformationeveravailabletosuchasearch.Whether
and Beacom (2005). This model will be referred to as “soft jet
ornottheexistenceofjetsinasphericalexplosionsisevidenced
model”inthefollowing.Abriefsummaryofthephysicalmoti-
inthespectroscopicdataforSN2008Dremainshighlydebated.
vationandaderivationofitsanalyticalformshallbepresented.
WhileSoderbergetal.(2008)“firmlyruleout”anyasphericity
and Chevalier and Fransson (2008) speak of a purely spherical The soft jet model assumes the collapse of a massive star
shock-breakoutemission,Mazzalietal.(2008)andTanakaetal. M (cid:38)8M withsubsequentformationofaneutronstarorblack
(cid:63) (cid:12)
(2009)findevidencethatSN2008Dpossessedjetswhichhave hole,rotatingsufficientlytopowerjetswithbulkLorentzfactors
beenobservedsignificantlyoff-axis. ofΓ ∼ 1−10andopeninganglesθ ≈ 1/Γ = 5◦−50◦.Such
b j b
The IceCube neutrino detector, currently under construc- “soft”jets,tooweaktopenetratethestellarenvelope,wouldnot
tion at the South Pole and scheduled for completion in 2011, beobservableintheelectromagneticspectrum.Therebounding
is capable of detecting high-energy neutrinos (Eν (cid:38) 100GeV) core collapse is assumed to deposit Ej ∼ 3 × 1051erg of ki-
of cosmic origin by measuring the Cherenkov light emitted neticenergyinthematerialejectedinthejets–valuesofupto
by secondary muons with an array of Digital Optical Modules Ej =6×1051erghavebeensuggestedforSN2008DbyMazzali
(DOMs) positioned in the transparent deep ice along vertical et al. (2008). Protons are Fermi accelerated to a E−2-spectrum
p
strings (J. Ahrens et al. 2004). The full detector will comprise and produce muon neutrinos through the decay of charged pi-
4,800 DOMs deployed on 80 strings between 1.5 and 2.5 km onsandkaonsformedinproton-protoncollisions.Theneutrino
deep within the ice, a surface array (IceTop) for observing ex- spectrum, shown in Fig. 1, follows the primary proton spec-
tensiveairshowersofcosmicrays,andanadditionaldensesub- trum at low energies and steepens at four break energies above
array(DeepCore)inthedetectorcenterforenhancedlow-energy which pions (kaons) lose a significant fraction of their energy
sensitivity.EachDOMconsistsofa25cmdiameterHamamatsu inhadronicandradiativecoolingreactions,beforedecayinginto
photo-multipliertube(PMT,seeAbbasietal.2010a),electron- neutrinos.Thesebreakenergiesaredistinctforpionsandkaons
icsforwaveformdigitization(Abbasietal.2009),highvoltage and exihibit a sensitive dependence on the jet parameters (see
N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D 3
Table1.Parametersofthesoftjetmodelusedinthisanalysis.
Fig.1. Assumed E2-weighted muon neutrino and antineutrino
spectrum of SN 2008D according to the soft jet model. For
comparison, the atmospheric muon neutrino flux is shown for Parameter Description Default Parameter
Value Dependence
a100stimewindowandacircularaperturewithopeningangle
ω=10◦. E Total kinetic energy of 1051.5erg -
j
Eνπ, c(1b) Eνπ, c(2b) EνK, c(1b) EνK, c(2b) ejectedmaterial
102 Γ Bulk Lorentz factor of 3 -
−2]m Kaon Decays b thejet
c
eV 10 Pion Decays d DistanceofSN2008D 27Mpc -
G Total Flux
dN2 [ EdE 1 Atm. Neutrinos Bπ(BK) Bπ±ra→nchµi±nνgµr(aKti±o→forµ±νµ) 1(0.63) -
(cid:104)n(cid:105) Pion (kaon) multiplicity 1 (0.1) -
π
((cid:104)n(cid:105) ) inppcollisions
K
10−1 E mininmum proton en- 10GeV -
p,min
ergy
10−2 Ep,max maximumprotonenergy 7×104GeV -
Eπ(1) hadronic cooling break 30GeV ∝
102 103 104 E [GeV] (cid:16)Eν,νKc,bc(b1)(cid:17) energyforpions(kaons) (200GeV) ≈E−jE1−Γ17bΓθ52j
j b
Eπ(2) radiative cooling break 100GeV ∝Γ
(cid:16)Eν,Kcb(2)(cid:17) energyforpions(kaons) (20000GeV) b
ν,cb
Table1).UsingthenotationofAndoandBeacom,thespectrum
Eπ maximum neutrino en- 10500GeV ∝Γ
canbewrittenas: (cid:16)ν,max (cid:17) b
EK ergy from pion (kaon) (21000GeV)
Φ (E )= (cid:88) η × EEν−−23Ei(1) EEip(,1m)in≤≤EEν<<EEi(νi2,()c1b) (1) ν,max decay
ν ν i=π,K i  Eνν−4Eννi,,(cc1bb)Eνi,(c2b) Eννi,,(cc2bb) ≤ Eνν ≤ Eννi,,cmbax 3. DataandSimulation
where The analysis uses experimental data to determine the expected
number of background events for a particular search window.
(cid:104)n(cid:105) B E
ηi = (cid:16) (cid:17) i (cid:16)i j (cid:17) (2) Thesignalexpectationsaswellasthecharactericticsofthesig-
8 2πθ2d2 ln E /E nalarederivedfromsimulations.Rawdataconsistsoftimese-
j p,max p,min
ries of photon detections (henceforth “hits”) for each triggered
Withtheexceptionofthedistanced,weassumethesameparam- DOM. From these hit patterns, track reconstruction algorithms
etersforSN2008DthatarequotedinAndo&Beacom(2005). derive the muon’s direction, measured in zenith θ and azimuth
AsummaryisgiveninTable1. φinafixeddetectorcoordinatesystemwheremuonstravelling
AnoptimisticextensionofthismodelproposedbyKoersand upwards in the ice have θ > 90◦ and downgoing tracks have
Wijers(2007)predictsthatmesonsareagainFermi-accelerated θ < 90◦.TheabsolutetimeofaneventisdeterminedbyaGPS
afterproduction.Thisre-accelerationgivesrisetoasimple E−γ clockwithaprecisionofbetterthan200ns,whichismorethan
neutrinospectrumwithγ = 2.0,...,2.6extendingtomaximum sufficientforthisanalysis.
energiesofE ∼ 10PeVwhereradiativecoolingprocesseslead
ν
toasteepeningandeventualcutoffoftheneutrinospectrum.The
3.1. Backgrounddata
details of this high-energy cutoff are negligible in the context
ofthisanalysis,whereneutrinoswithenergiesof100GeV-10 At trigger level (detailed in Sec. 3.3 below), IceCube data is
TeVareexpectedtoyieldthedominantcontributionofthesignal dominatedbythereduciblebackgroundofatmosphericmuons,
expectation. falselyreconstructedasupgoing,i.e.havingpassedthroughthe
Neutrinos are expected to be emitted in alignment with the Earth.Acomparisonofexperimentaldataandsimulatedmuons
jets. Their energy range is set by the maximum proton energy fromcosmicrayshowersshowsgoodagreement(seeFig.2).In
andreachesfarintothesensitiverangeoftheIceCubedetector addition, background data contains an irreducible background
(E (cid:38) 100GeV).Inordertodetecttheseneutrinos,thejetmust of muons produced by atmospheric neutrinos from the north-
ν
bepointingtowardsEarth(e.g.5%chanceforajetwithanopen- ern hemisphere, at a rate lower by a factor of 105. At the final
inghalfangleof17◦).Duetotheunknownjetpointing,however, cut levels of this analysis (see Tab. 2), data consists of approx-
noconstraintscanbeplacedonthemodelinthecaseofanon- imately equal contributions of reducible and irreducible back-
detection. To do so with a confidence level of e.g. 90% would groundevents.
requirealargesampleof∼200nearbysupernovae.Incontrast,a The data sample used to measure and characterize back-
positivedetectionwouldnotonlyindicatethejet’sdirection,but groundwastakenbyIceCubeinthe22stringconfigurationover
alsoyieldconstraintsonthesoftjetmodel–constraintsentirely 275.72daysofdetectorlivetimebetweenMay2007andMarch
independentofobservationsintheelectromagneticspectrum.If, 2008.ThesampleisidenticaltotheoneusedinthefirstIceCube
inaddition,aresolvedneutrinospectrumcouldberecordedwith search for neutrino point sources (R. Abbasi et al. 2009). On
futureneutrinodetectors,theobservationofspectralbreaksand the day of SN 2008D, IceCube was taking data continuously
a spectral cutoff would place strong constraints on the physical in a time range of [−9.5h, +1.8h] around the observed X-ray
parametersofthesupernovajet. flash. To prevent a bias in the cut optimization, this data was
4 N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D
kept“blind”,i.e.excludedfromthedevelopmentandtestingof Ndir,E Number of direct hits, i.e. photon hits detected
selection criteria, and only “unblinded” in the final step of the within a [−15ns, +250ns] time window of the
analysis. arrival time predicted for unscattered Cherenkov
emissionunderthetrackhypothesis
S Smoothnessofhitdistribution.S = 0indicatesa
all all
3.2. SignalSimulation homogeneousenergydepositionalongthetrack
Toquantifyandcharacterizetheexpectedsignal,extensivesim- θmin Minimum zenith when the 1st and 2nd half of the
ulations of the complex Earth–ice–detector system were con- photon hits (ordered in time) are reconstructed as
ducted. IceCube simulation generates primary neutrinos at the separatetracks
surface of the Earth and propagates them through the Earth, σ Estimator for the uncertainty of the reconstructed
p
tracking charged and neutral current interactions, and record- trackdirection(quadraticaverageoftheminorand
ing all secondary particles which can reach the detector (see majoraxisofthe1σerrorellipse)
Kowalski et al. 2005). All secondary muons are then passed L Value of the negative log-likelihood for the recon-
R
tothemuonpropagationsoftware(seeChirkin&Rhode2008)
structedtrackdividedbythenumberofdegreesof
which simulates their random energy loss and the emission of
freedom in the fit (number of hit optical modules
Cherenkovphotons.Finally,thepropagationofphotonsissim-
minusnumberoffitparameters)
ulated accounting for absorption and scattering according to a
R Ratio of the log-likelihoods with and without a
depth dependent ice model (see Lundberg et al. 2007). In the B
Bayesian prior that favors a downgoing track hy-
laststep,thephotomultiplierresponse,readout,andlocalaswell
pothesis
as global triggers are simulated yielding time series of photon
hitswhicharesubsequentlypassedthroughthesameprocessing RU Ratioofthelog-likelihoodswithandwithoutseed-
pipelineasexperimentaldata. ingthereconstructionwiththeinversetrackdirec-
tion
Inconjunctionwiththeselectionofupgoingtracks,thereduced
3.3. Triggeringanddataprocessing log-likelihood L hasproventobeanefficientvariableforsep-
R
TheIceCubetriggersystemonlyreadsoutaphotonhitataspe- arating upgoing atmospheric neutrinos from misreconstructed
cificopticalmoduleifaneighboringmoduleonthesamestring downgoing atmospheric muons. It exploits the fact that for a
is also hit within 1µs (local coincidence). To initiate the event light pattern originating from a downgoing muon the incorrect
read-out,theglobaltriggerofIceCube22required8suchlocal upgoing track hypothesis yields rather low absolute likelihood
coincidenceswithina5µstimewindow.Thisrequirementlead values. In addition, the likelihood ratios RU and RB allow for a
to trigger rates of ∼550 Hz, dominated by atmospheric muon vetooneventsforwhichinvertingthetrackhypothesisleadsto
events. Data contamination was immediately reduced to ∼25 asignificantrelativeenhancementinthelikelihoodvalue.
Hz by first-guess reconstructions running online at the South Histograms of all selection parameters are shown in Fig. 2
Pole,whichfitasimpletrackhypothesistoeacheventandreject forbackgrounddata,backgroundsimulation,andsimulatedsig-
downgoingtracksinrealtime(Ahrensetal.2004).Eventspass- nalevents.Tocombinealleightparametersefficiently,theywere
ing this online muon filter are transferred to the North, where incorporated into a boosted decision tree (BDT) classifier (see
extensive likelihood track reconstructions are performed. For a e.g.Yangetal.2005andreferencestherein).TheBDTmethod
givenhitpatternandafirstguesstrackhypothesis,thelikelihood classifies an event by passing it through a tree structure of bi-
functioniscalculatedastheproductoftheprobabilitiesforeach narysplitswhicheffectivelybreaksuptheparameterspaceinto
hittimetooccurunderthegiventrackhypothesis.Thelikelihood a number of signal or background-like hypercubes. The classi-
reconstruction algorithm then iteratively searches for the track fier is first trained with background data and simulated signal
which maximizes the value of this likelihood function (Ahrens andthenevaluatedwithindependentdatasets.Theresultingdis-
etal.2004).Forthefinalfitresult,theoptimizationsofwarecom- tribution of classifier scores K for experimental data and sim-
putesqualityparameterswhichcanbeusedforeventselection. ulated signal is shown at the bottom of Fig. 2. The classifier
allows for a simple one-dimensional cut on the classification
score.Extensivetestswereconductedtoassureastableresponse
andtoestimatetheuncertaintyoftheclassification.Thisuncer-
4. Eventselection
taintywasestimatedbycomparingtheclassificationefficiencies
The background event rate is further diminished to ∼3 Hz for several independent experimental data and simulated signal
throughanothercutonthemoreprecisetrackdirectionfromthe samples.Variationsintheclassifierresponseprovedtobenegli-
likelihood track reconstruction selecting events with θ > 80◦. giblecomparedtostatisticaluncertainties.
For this analysis, events outside a circular signal region (10◦
openingangle)aroundthepositionofSN2008Dwereremoved
4.2. SearchWindows
from the dataset to obtain a manageably sized sample. At this
filtering level, the background rate is 0.03 Hz and 0.26 sig-
The search for neutrinos in the on-time data from January 9,
nal events are expected for SN 2008D according to the soft jet 2008wasconductedusingthreesearchwindowsofdifferentdu-
model.
rations, apertures, and selection cuts. A circular aperture was
used in all cases. Since the soft jet model does not explicitly
predict the time profile of the neutrino emission, search win-
4.1. QualityCuts
dows with durations of 100s, 1000s, and 10000s were chosen
Specific cuts tailored to the simulated properties of SN 2008D tocoveralargerangeofemisssiontimescales.Thecorrespond-
werebasedonthefollowingeightqualityparameters: ingopeningangleandqualitycutsforeachsearchwindowwere
determinedbyoptimizingthemodeldiscoveryfactorMaccord-
N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D 5
Table 2. Windows used to search for neutrinos in correlation
Fig.2.Normalizedhistogramsofcutparametersusedforevent
with SN 2008D. For each time scale, quality and angular cuts
selection(top)andresultingdistributionofboosteddecisiontree
wereoptimizedtoyieldamaximummodeldiscoverypotential.
classifier values (bottom). Dashed line indicates experimental
backgrounddata,dottedlinemarksbackgroundsimulation,solid
linerepresentssimulatedsignal. Duration Centering Aperture BDTcut
∆t wrt.X-raypeak ω K
10−1 10−1 Window1 100s −70s, +30s 6.2◦ 0.390
Window2 1000s −500s, +500s 2.6◦ 0.464
Window3 10000s −7000s,+3000s 1.5◦ 0.580
10−2
10−2
10−3 Fig.3. Effective areas for a neutrino spectrum obeying the soft
jetmodel.Eachlinerespresentsoneofthefinalsearchwindows
10−3 0 10 R20 30 40 0 1θ [rad]2 3 usedinthisanalysis.Inset:Cumulativepointspreadfunctionfor
U min
thedirectioninwhichSN2008Dwasobserved.
10−1 10−1
10−2 10−2
10−3 10−3
10−40 0.5 1 1.5 10−40 20 40 60
σ [rad] R
p B
10−1
10−1
10−2
10−2
10−3 10−3
gorithm(Feldman&Cousins1998).Thesignalexpectationisin-
10−40 200 400 600 800 1000 10−40 10 20 30 40 creaseds→ s(cid:63)until50%ofthetrialsyieldadiscovery,thatis,a
L [m] N
dir,E dir,E lowerlimitonthesignalsgreaterthanzero.Whenthiscriterion
10−1 10−1 ismet,themodeldiscoveryfactorisgivenby
s(cid:63)
10−2 10−2 M = (3)
s
10−3 10−3 Foreachwindow,theBDTcutK andtheopeningangleωyield-
ingtheminimalvalueofMweredeterminednumerically.Lower
10−4 8 10 12 10−4−1 −0.5 0 0.5 1 limitsaccordingtotheFeldman&Cousinsorderingschemewere
L S
R all requiredtohaveasignificanceof5σ.Thechoicesofcutsforthe
three search windows which yielded minimal model discovery
factorsaresummarizedinTable2.Theresultingeffectiveareas
10−2 foraneutrinospectrumobeyingthesoftjetmodelareshownin
Fig.3.
10−3 Withthesechoices,twoobservedeventswouldconstitutea
5σdiscoveryinanyofthewindowstakenbyitself.Thesignifi-
10−4 cancesforthecompletemeasurementconsistingofthreesearch
Signal Monte Carlo windowsweredeterminedinasimulationstudywith1010trials.
10−5 Foreachpossibleobservationofn , n , n eventsinwindow1,
Data 1 2 3
2,3,thep-valuewascalculatedasthefractionofequallyorless
10−6 Atm. Muon Monte Carlo likelyobservations.
−1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1
BDT score
5. Results
5.1. Unblinding
ingtoHilletal.(2006).Forthispurpose,aPoissondistribution
with mean b + s is randomly sampled, where b and s repre- Noeventspassingthecutswerefoundintheexperimentaldata.
senttheexpectedbackgroundandsignal,respectively.Foreach AsshowninTable3,thisresultisconsistentwithexpectations,
drawn number of observed events n the lower limit on the evenmoresoifweaccountforthe∼5%probabilityofajetwith
obs
signalcontributioniscomputedusingtheFeldman&Cousinsal- openinghalfangle∼17◦pointingtowardsEarth.
6 N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D
Table 3. Summary of the unblinding results and comparison
Fig.4.SpectrumofSN2008Daccordingtothesoftjetmodelfor
withexpectations. different assumed jet Lorentz factors and under the assumption
thatthejetispointingtowardsEarth.
Window1 Window2 Window3
ObservedEvents n =0 n =0 n =0
1 2 3
ExpectedEvents
Signals 0.13 0.060 0.020
Backgroundb 3.67×10−4 5.52×10−4 5.55×10−4
5.2. LimitsontheSoftJetModel
Intheabsenceofmoreprecisetheoreticalpredictionsonthetime
profileoftheemission,quotinglimitsforparticulartimescales
is the only viable way to constrain the soft jet model. Since
n = n = n = 0 and b ≈ b ≈ b , the signal upper limits
1 2 3 1 2 3
s¯ areidenticalforallthreesearchwindowstothefourthsignif-
i
icant digit: s¯ = s¯ = s¯ = s¯ = 2.44 (at 90% CL). The upper
1 2 3
limit Φ¯(90) on the neutrino flux in terms of the expected flux
ν
Φ isgivenbytheratioofthesignalupperlimit s¯tothesignal
ν
expectations:
Φ¯(90) s¯ Fig.5.Expectednumberofeventsasafunctionoftheassumed
Φνν = s (4) jetLorentzfactorΓbundertheassumptionthatthejetispointing
towardsEarth.Theplottednumberscorrespondtoa10◦-signal-
Due to the different signal expectations in each window, the
regionandcutlevel3atwhichthebackgroundrateis0.03Hz.
fluxupperlimitsdependontheassumedemissiontimescaleτ .
e
Therefore,wequotethelimitsonthesoftjetmodelforcanoni-
cτaelapnadraamtaetreerfser(eTnacbe.e1n)esregpyaoraftEelνy=fo1r0e0aGcheVem: ission time scale al µs 103
Φ¯(νG90e)V(1−010cmG−e2V)=(cid:34)10Md pc(cid:35)2×000...00135758 τττeee ===11100000000s0ss (5) ected Sign 11002
p
x
Eachlimitisonlyvalidundertheassumptionthattheentireneu- E
trino signal is contained in the corresponding time window. In 1 E = 1052
j
otherwords,SN2008Dcouldhaveemittedatmost19(41,122)
E = 1051.5
twimithesdmefoaureltnpeauratrmineotsersthΓanb =ass3umaneddEunjd=er1t0h5e1.5soerfgt.jeAt hmigohdeerl 10−1 Ejj = 1051
fluxwouldhavebeenobservedbyIceCubewithaprobabilityof
90%. 10−2 2 3 4 5 6 7 8 9 10
Γ
The primary systematic uncertainty in these limits stems b
from a possible bias in signal simulation, i.e. the value of s.
Systematics for IceCube 22 have been studied by Abbasi et al.
(2009) and lead to a ∼15% uncertainty in s, corresponding to
a +17 percent shift in the limits. Incorporating the uncertainty
−13
oftheBDTclassificationresponse,thatisdecreasingthesignal
predictionandincreasingthebackgroundexpectationbythecor- wellasstrongerbeaminginmorerelativisticjetsleadstoadras-
respondinguncertaintyresultedinanegligibleshiftof∼0.5%in ticincreaseinthesignalexpectation.IncreasingΓ placesmore
b
thelimits. neutrinosathighenergies(cid:38)1TeVwhereIceCubeismoresensi-
Next,wewishtoconstrainthemainparametersofthemodel, tive,thoughthecorrespondingreductioninthejetopeningangle
thekineticenergyreleaseEjandtheLorentzfactorofthejetΓb. leads to smaller probability of jet detection. The measured sig-
Due tothe significant Γb dependence ofthe hadronicbreak en- nalupperlimits¯=2.44andthesignalpredictionss (cid:16)Γ , E (cid:17)for
i b j
eErνπg,/cyKb(E2)νπ,/∝cKb(Γ1)b,∝theEn−j1umΓ5bbearndantdhespreacdtiraatlivdeisctroiobluitnigonborefapkroednuecrgeyd eΓabcthhrwouingdhowsi(cid:16)cΓabn,Ebej(cid:17)u<sesd¯i.toVaclounesstroafinΓbthaendjetEpjanroatmfeutlefirlsliEngj tahnids
neutrinosdependsstronglyonΓb(seeFig.4).Moreover,theflux relationareruledoutat90%CL.Theselimitsareillustratedin
is scaled with EjΓ2b which accounts for the energy release and Fig.6.
thebeamingoftheneutrinoemission.Athighboostfactors,ra-
Finally, the scenario proposed by Koers and Wijers (2007)
diativecoolingofmesonssetsinatlowerenergiesthanhadronic
(cid:16) (cid:17) shallbeexaminedbriefly.Assumingthatmesonre-acceleration
cooling,i.e.Eνπ,c(1b) > Eνπ,c(2b) EνK,c(b1) > EνK,c(b2) forΓb (cid:38)4(Γb (cid:38)9). leads to a simple power law neutrino spectrum in the relevant
To derive constraints on Γb and Ej, we calculated the sig- energyrange(roughly100GeV-10PeV)thesourcespectrum
nal expectations in the intervals Γb = 1.5 − 10 and Ej = canbeapproximatedbyanE−γ-lawwithahigh-energycutoffat
1051 − 1052erg. As Fig. 5 shows, the less efficient cooling as 10PeV.Forthethreevaluesofthespectralindexγdiscussedby
N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D 7
Fig.6.ConstraintsonthejetparametersE andΓ whereE = Sweden; German Ministry for Education and Research (BMBF), Deutsche
j b 51.5 Forschungsgemeinschaft (DFG), Germany; Fund for Scientific Research
1051.5erg.Foreachassumedemissiontimescaleτe,thecolored (FNRSFWO), Flanders Institute to encourage scientific and technological re-
regionsareruledoutat90%confidencelevel. searchinindustry(IWT),BelgianFederalSciencePolicyOffice(Belspo);the
NetherlandsOrganisationforScientificResearch(NWO);M.Ribordyacknowl-
edgesthesupportoftheSNF(Switzerland);A.KappesandA.Groacknowledge
supportbytheEUMarieCurieOIFProgram;J.P.Rodriguesacknowledgesup-
portbytheCapesFoundation,MinistryofEducationofBrazil.
References
AbbasiR.etal.,IceCubeCollaboration2010,Nucl.Inst.Meth.A,618
AbbasiR.etal.,IceCubeCollaboration2010,Astrophys.J.,701346
AbbasiR.etal.,IceCubeCollaboration2009,Astrophys.J.,701L47
AbbasiR.etal.,IceCubeCollaboration2009,Nucl.Instr.Meth.A,601
AchterbergA.etal.,IceCubeCollaboration2006,Astropart.Phys.,26,155
Ahrens,J.etal.,AMANDACollaboration2004,Nucl.Inst.Meth.A,524
AhrensJ.etal.,IceCubeCollaboration2004,Astropart.Phys.20,5
AndoS.,BeacomJ.2005,Phys.Rev.Lett.,95
ChirkinD.,RhodeW.2008,hep-ph/0407075v2
ChevalierR.,FranssonC.2008,Astrophys.J.,683,2
FeldmanG.,CousinsR.1998,Phys.Rev.D,57
FranckowiakA.etal.2009,Proc.ofthe31stICRC
GalamaT.J.etal.1998,Nature,395
HillG.etal.2006,inStatisticalProblemsinParticlePhysics,Astrophysicsand
Cosmology:PHYSTAT2005,ed.L.Lyons&M.K.Unel(London:Imperial
KoersandWijers,thisanalysisyieldsthefollowingupperlimits CollegePress),108
foranassumedemissiontimescaleofτ =100s: HjorthJ.etal.2003,Nature423
e
KoersH.,WijersR.2007,arXiv:0711.4791v1
EγΦ¯(90) =10..61022 γγ==22.3 (6) KLaomwbalDsk.iQM.e.,tGala.z2i0z0ov4,AN.e2w00A5s,tCroonmomp.yPRhyevs.ieCwosm,4m8.,172,3
GeVγ−1cm−2 21.9 γ=2.6 LLeunnndabrezrgDJ.,.IecteaCl.u2b0e0C7,oNllaubcol.raIntisotnru2m0.0M9,ePthro.cAe,ed5i8n1gsofthe31stICRC
Malesani,J.etal.2004,Astrophys.J.,609
Forlongeremissiontimescales,theselimitsscaleasin(5). MazzaliE.etal.2008,Science321
ModjazM.etal.2009,Astrophys.J.,702
ParagiZ.etal.2010,Nature,463
6. SummaryandOutlook PianE.etal.2006,Nature,442
RazzaqueS.,MeszarosP.,WaxmanE.2005,Mod.Phys.Lett.A,20
We have searched for high-energy muon neutrinos in coinci- SoderbergA.etal.2008,Nature453
SoderbergA.etal.2010,Nature463
dencewithSN2008DusingdatafromtheIceCube22stringde-
TanakaM.etal.2009,Astrophys.J.700
tector.Usingablindanalysisoptimizedwithexperimentalback- WoosleyS.,BloomJ.2006,AnnualRev.Astron.Astrophys.
groundandsimulatedsignaldata,weobservednoeventswhich YamazakiaR.2004,Astrophys.J.607
passedthecuts.Fromthenon-observation,wehavederivedfirst YangH.etal.2005,Nucl.Instrum.Meth.A555
constraintsonthesoftjetmodelforcorecollapseSNeunderthe
condition that the predicted jet was pointing in the direction of
theEarth. 1 Dept. of Physics, University of Wisconsin, Madison, WI 53706,
Given the strong dependence of the signal expectation on USA
themodelparameters,thenon-detectionofneutrinosplacessig- 2 Dept.ofSubatomicandRadiationPhysics,UniversityofGent,B-
nificant constraints on the principal model parameters. A two 9000Gent,Belgium
dimensional parameter scan in Γ and E shows that the jet 3 Dept.ofPhysics,UniversityofWisconsin,RiverFalls,WI54022,
LorentzfactorisgenerallyconstraibnedtoΓj <4forjetenergies USA
b 4 Dept.ofPhysicsandAstronomy,UniversityofCanterbury,Private
E > 1051erg.Asmentionedabove,theconstraintsquotedhere
j Bag4800,Christchurch,NewZealand
onlyholdiftheassumedjetofSN2008Dwaspointingtowards 5 Dept.ofPhysics,UniversityofOxford,1KebleRoad,OxfordOX1
Earth. 3NP,UK
IceCube is now operating in an additional mode, scanning 6 Dept. of Physics, University of Wuppertal, D-42119 Wuppertal,
online data for neutrino bursts, i.e. two nearly collinear neutri- Germany
nos within 100 s, in real time. If a burst is detected, IceCube 7 Bartol Research Institute and Department of Physics and
triggersopticalfollow-upobservationssearchingforaSNinthe Astronomy,UniversityofDelaware,Newark,DE19716,USA
corresponding direction (Franckowiak et al. 2009). Constantly 8 Dept. of Physics and Astronomy, University of California, Irvine,
CA92697,USA
monitoringtheentirenorthernsky,thisapproachhasthepoten-
9 Dept. of Physics, University of California, Berkeley, CA 94720,
tialtogeneralizetheconstraintsobtainedfromstudyingindivid-
USA
ualobjects.
10 DESY,D-15735Zeuthen,Germany
11 LawrenceBerkeleyNationalLaboratory,Berkeley,CA94720,USA
Acknowledgements. We acknowledge the support from the following agen- 12 Dept. of Physics and Center for Cosmology and Astro-Particle
cies: U.S. National Science Foundation-Office of Polar Program, U.S.
Physics,OhioStateUniversity,Columbus,OH43210,USA
National Science Foundation-Physics Division, University of Wisconsin
13 Dept.ofAstronomy,OhioStateUniversity,Columbus,OH43210,
Alumni Research Foundation, U.S. Department of Energy, and National
EnergyResearchScientificComputingCenter,theLouisianaOpticalNetwork USA
Initiative (LONI) grid computing resources; Swedish Research Council, 14 Universite´ Libre de Bruxelles, Science Faculty CP230, B-1050
SwedishPolarResearchSecretariat,andKnutandAliceWallenbergFoundation, Brussels,Belgium
8 N.KemmingfortheIceCubeCollaboration:Constraintsonhigh-energyneutrinoemissionfromSN2008D
15 Fakulta¨t fu¨r Physik & Astronomie, Ruhr-Universita¨t Bochum, D-
44780Bochum,Germany
16 Dept.ofPhysics,UniversityofMaryland,CollegePark,MD20742,
USA
17 Dept.ofPhysicsandAstronomy,UniversityofKansas,Lawrence,
KS66045,USA
18 III. Physikalisches Institut, RWTH Aachen University, D-52056
Aachen,Germany
19 OskarKleinCentreandDept.ofPhysics,StockholmUniversity,SE-
10691Stockholm,Sweden
20 VrijeUniversiteitBrussel,DienstELEM,B-1050Brussels,Belgium
21 Physikalisches Institut, Universita¨t Bonn, Nussallee 12, D-53115
Bonn,Germany
22 Dept.ofPhysicsandAstronomy,UppsalaUniversity,Box516,S-
75120Uppsala,Sweden
23 Dept. of Physics, TU Dortmund University, D-44221 Dortmund,
Germany
24 LaboratoryforHighEnergyPhysics,E´colePolytechniqueFe´de´rale,
CH-1015Lausanne,Switzerland
25 Max-Planck-Institutfu¨rKernphysik,D-69177Heidelberg,Germany
26 Dept.ofPhysics,PennsylvaniaStateUniversity,UniversityPark,PA
16802,USA
27 Dept. of Astronomy and Astrophysics, Pennsylvania State
University,UniversityPark,PA16802,USA
28 Dept. of Physics and Astronomy, Utrecht University/SRON, NL-
3584CCUtrecht,TheNetherlands
29 CTSPS,Clark-AtlantaUniversity,Atlanta,GA30314,USA
30 Dept. of Physics, Southern University, Baton Rouge, LA 70813,
USA
31 Dept.ofAstronomy,UniversityofWisconsin,Madison,WI53706,
USA
32 Dept.ofPhysics,UniversityofAlberta,Edmonton,Alberta,Canada
T6G2G7
33 Institute of Physics, University of Mainz, Staudinger Weg 7, D-
55099Mainz,Germany
34 Universite´deMons,7000Mons,Belgium
35 Dept.ofPhysics,ChibaUniversity,Chiba263-8522,Japan
36 Institut fu¨r Physik, Humboldt-Universita¨t zu Berlin, D-12489
Berlin,Germany
37 alsoUniversita` diBariandSezioneINFN,DipartimentodiFisica,
I-70126,Bari,Italy
38 Dept.ofPhysicsandAstronomy,UniversityofAlaskaAnchorage,
3211ProvidenceDr.,Anchorage,AK99508,USA
39 Dept.ofPhysics,UniversityoftheWestIndies,CaveHillCampus,
BridgetownBB11000,Barbados
40 NASAGoddardSpaceFlightCenter,Greenbelt,MD20771,USA
41 SchoolofPhysicsandCenterforRelativisticAstrophysics,Georgia
InstituteofTechnology,Atlanta,GA30332,USA
42 Dept. of Physics and Astronomy, University of Alabama,
Tuscaloosa,AL35487,USA
43 affiliated with Universita¨t Erlangen-Nu¨rnberg, Physikalisches
Institut,D-91058Erlangen,Germany