PHYSICS AND ASTRONOMY WITH SHOWERING UPWARD MUONS IN SUPER-KAMIOKANDE SHANTANU DESAI Ph.D Final Oral Examination December 3 rd 2003
Outline of the Talk: Super-Kamiokande detector PHYSICS Dataset used and its classification (upward thru + stopping muons) Subdivision of upward thrugoing muons (showering + non-showering) Overview of algorithm used to identify upward showering muons Energy Spectrum of all 3 categories of events Oscillation Analayis using upward showering muons ASTRONOMY Overview of astrophysical searches with upward muons Gamma-ray burst (GRB) and Soft-gamma-ray repeater (SGR) searches Diffuse flux searches WIMP searches
Super-Kamiokande Detector ~ 40m X 40 m tank containing 50 kton of ultra pure water Water Cherenkov detector 11146 20-inch PMTs in ID 1885 8-inch PMTs in OD Located 1 km under mountain in a Zn mine in Japan Phase 1 : 1996-2001
Subject of this Thesis = Upward Going Muon Events S-K blown up µ Earth Search for signals from astrophysical sources in upward going muons Background from cosmic ray interactions in atmosphere on opposite hemisphere of earth
Intensity (cm 2 s 1 sr 1 ) Need to go underground to reduce background from cosmic ray muons 10 3 10 4 10 5 10 6 10 7 10 8 Soudan 2 Minos Crouch World Survey, 1987 Crookes and Rastin, 1973 Bergamasco et al., 1971 Stockel, 1969 Castagnoli et al., 1965 Avan and Avan, 1955 Randall and Hazen, 1951 Bollinger, 1950 Clay and Van Gemert, 1939 Wilson, 1938 At Super-K depth µ Flux µ Flux ~ 10-5 10 9 10 10 10 11 10 12 IMB Gran Sasso Homestake Super Kamiokande π,k muons π,k muons + I ν µ Bugaev et al 98 10 13 I ν µ = 2.17x10 13 cm 2 s 1 sr 1 0 2 4 6 8 10 12 14 16 18 Depth (10 3 hg cm 2 )
Upward Going Muons In Super-K E 100GeV THRU STOP dφ/dloge υ ( 10-13 cm -2 s -1 sr -1 ) 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Thrugoing Stopping E 10GeV 0 1 10 10 2 10 3 10 4 10 5 E υ (GeV)
Event display of an Upstop Muon
Event Display of Upthru Muon
Upward Muons originally divided into two categories Upward Stopping Upward Thrugoing 467 events 1892 events Non-Showering Showering
Non-Showering muons Showering muons Total Loss < de/dx (MeV cm 2 g -1 ) > 10 2 10 Radiative Loss Ionization Loss Average < de/dx > from M.C. Muon energy loss in water 1 1 10 10 2 10 3 10 4 10 5 Muon energy (GeV) Muon critical energy in water ~ 1 TeV Monte-Carlo de/dx agrees with theoretical curves (Groom et al 02)
Muon track d track For each PMT in cone find point of projection along muon trajectory Calculate avg corrected charge along each 0.5 m along muon track For an ionizing muon Q corri <Q corri > ~ constant along the muon track d track
Muon track Extra Cerenkov light from e + e - pairs Q corri d track Muon which loses energy through radiative processes initiates an electromagnetic shower A showering muon should show huge upward spikes along muon track d track
Strategy : Apply corrections to PMT charge to account for: PMT acceptance Light scattering Attenuation due to photon travel distance Cherenkov light dpmt dwat PMT Q corr = Q d exp (d /L ) raw wat wat atten F ( ) L atten = Light attenuation length
corrected charge The above corrections not enough Light scattering into the cone increases the effective charge as muon path length increases 35 30 25 20 15 10 Monte-carlo muon length greater than 30 m muon length between 20 and 30 m muon length between 10 and 20 m muon length less than 10 m corrected charge 35 30 25 20 15 10 Monte-carlo muon length greater than 30 m muon length between 20 and 30 m muon length between 10 and 20 m muon length less than 10 m Light Scattering Turned Off 5 5 0 0 5 10 15 20 25 30 35 40 Distance travelled by light (m) 0 0 5 10 15 20 25 30 35 40 Distance travelled by light (m) Fit the mean charge of an ionizing muon to 2 nd order polynomial
Generated Monte-carlo samples with different energies using track directions and entry points from downward muon data
Both muons have same entry point and directions Muon energy = 20 GeV Muon energy = 10 TeV Mean corrected photo-electrons / 0.5 m 90 80 70 60 50 40 30 20 10 0 0 2 4 6 8 10 12 14 16 18 20 Projected Distance Along Track (m) Mean corrected photo-electrons / 0.5 m 90 80 70 60 50 40 30 20 10 0 0 2 4 6 8 10 12 14 16 18 20 Projected Distance Along Track (m) dedx2.2 MeVcm dedx8.5 MeVcm
Show Event Display of the 20 Gev muon event with and without corrections and same for 10 TeV muon
{ { For each upward muon event defined a : χ 2 n2 2 i3 Q corri Q corr 2 Qcorri 2 Q corr ql 2 2 ql Shape Comparison Normalization Comparison where and qlq corr Q corr Event considered showering iff : 2 D.O.F.25 and or Q corr ql4.0 pe n2 2 Q corri Qco rr 3 n2 2 1 Qco rr 3 for a ionizing muon as a function of path-length Q corr ql2.0 pe
Consistency Check: Does the distribution of showering variables agree for data and atmospheric neutrino monte-carlo? 800 700 600 Data Monte-Carlo (no Osc) Monte-Carlo (Osc) ( m 2 =0.002 sin 2 2Θ =1.0) 200 180 160 140 Data Monte-Carlo (no Osc) Monte-Carlo (Osc) No of events 500 400 300 200 No of events 120 100 80 60 40 100 20 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 log10(χ 2 ) 0-4 -2 0 2 4 6 8 10 [ mean q corr - Q(L) ] Yes
EVENT DISPLAY OF AN UPWARD SHOWERING MUON Super-Kamiokande Run 2101 Sub 3 Ev 15239 96-07-05:01:46:37 0013 dee4 0606 Inner: 0 hits, 0 pe Outer: 0 hits, 0 pe (in-time) Trigger ID: 0x0b D wall: 1690.0 cm Upward-Going Muon Mode Charge(pe) >26.7 23.3-26.7 20.2-23.3 17.3-20.2 14.7-17.3 12.2-14.7 10.0-12.2 8.0-10.0 6.2-8.0 4.7-6.2 3.3-4.7 2.2-3.3 1.3-2.2 0.7-1.3 0.2-0.7 < 0.2 2000 1600 1200 800 400 0 0 500 1000 1500 2000 Times (ns)
Showering Detection Efficiency Efficiency = # of events identified by algorithm as showering # of events with E/ X > 2.85 MeV/cm where E/ X = (E -E )/L i f E i E f where E i, E f are true muon energies from monte-carlo } L Efficiency ~ 75 % Amount of background from non-showers ~ 5% Total of 332 events in upthru dataset Total of 3129 events in 40 year upthrumu monte-carlo
Applied Showering Algorithm to 40 yr. Monte-Carlo dφ/dloge υ ( 10-13 cm -2 s -1 sr -1 ) 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Showering Non-showering Stopping 3 categories of upward muons seen in Super-K E 10 GeV E 100 GeV E 1TeV 0 1 10 10 2 10 3 10 4 10 5 E υ (GeV)
Angular resolution of Showering thrumus 225 true No of events 200 175 150 125 100 75 Corresponds to 68 % of total area fit 50 25 0 0 1 2 3 4 5 6 7 8 9 10 Angle between precision fit and truth fit Angular resolution = 1.3 degrees Stops : 1.5 degrees Thrugoing: 1.0 degrees
Our sample not completely background free Mt Ikenoyama Muon direction Super-K Downward near horizon could multiple-scatter and appear as upward This background from horizontal muons needs to be subtracted from the upward showering muon dataset
Mt. Ikenoyama not homogenous and isotropic contamination from thin parts of mountain in dataset Plotted the azimuthal distribution of upward + downward showering muons near horizon 10 3 Azimuth distribution of all showering muons (1) (2) (1) 60 < < 240 Zenith vs Azimuth for showering muons (641.4 days) 0-0.2 No of Events 10 2 cos Θ -0.4-0.6 10-0.8 1 0 50 100 150 200 250 300 350 Φ (degrees) -1 0 50 100 150 200 250 300 350 Φ (degrees)
Estimation of Background Subtraction Number of Events 10 3 10 2 10 1 23.38 / 15 P1 1.785 0.7645 P2 1.594 0.7109 P3 67.23 10.81 Region (1) Φ < 60, Φ > 240 Region (2) 60 < Φ < 240 10-1 -0.1-0.08-0.06-0.04-0.02 0 0.02 0.04 0.06 0.08 cosθ N bkgd = 8 7 for cos > -0.1 out of 80 events
Oscillation Probability for a 2-neutrino beam P( sin (2 )sin 2 [1.27 m 2 (ev 2 ) L(km)/E (GeV)] m 2 = m 1 2 - m 2 2 sin 2 (2 ) = Mixing angle First smoking gun evidence for neutrino oscillation discovered in Super-K with m 0.002sin!2 1(Messier 1999)
Flux ( 10-13 cm -2 s -1 sr -1 ) 1 0.8 0.6 0.4 0.2 Zenith Angle Distribution of Upward Showering Muons Upward Showering Thumus Observed Flux Null Oscillation 2 dof10.810 Flux ( 10-13 cm -2 s -1 sr -1 ) 3.5 3 2.5 2 1.5 1 0.5 All upthrumus Observed Flux Null Oscillation 2 dof23.110 0-1 -0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0 cos Θ 0-1 -0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0 cos Θ " Zenith angle distribution of upward showering muons consistent with no oscillation as expected
10-1 All upthrumus 10-1 68%CL 90%CL 99%CL upstopmus m 2 (ev 2 ) 10-2 m 2 (ev 2 ) 10-2 10-3 10-4 68%CL 90%CL 99%CL 0 0.2 0.4 0.6 0.8 1 sin 2 2θ 10-3 10-4 0 0.2 0.4 0.6 0.8 1 sin 2 2θ m 2 (ev 2 ) 10-1 10-2 showering upthrumus 10-3 10-4 68%CL 90%CL 99%CL 0 0.2 0.4 0.6 0.8 1 sin 2 2θ
Oscillation Analysis with all 3 Datasets 3 2 j1 10 i1 F data F MC 2 Weights applied to monte-carlo events : Non -Showering Upthrumus : (1+#) F nonshower Showering Upthrumus : (1+#) (1+$) F shower Stopping Upmus : (1+#) (1+%) F stop #,%, $ kept as free parameters Following terms added to 2 # # 2 % % 2 $ $ 2 where # 0.22 % 0.14 $ 0.08
Null Oscillation : 2 = 47.3 for 30 dof Best fit Oscillation : 2 = 22 for 28 dof 10-1 m 2 (ev 2 ) 10-2 10-3 10-4 68%CL 90%CL 99%CL 0 0.2 0.4 0.6 0.8 1 sin 2 2θ Best fit point m 2 = 0.0023 sin 2 2 = 1 Combined flux of all 3 samples consistent with oscillations
Flux ( 10-13 cm -2 s -1 sr -1 ) 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 Stopping Upmus Observed Flux Null Oscillation (α=-0.203 ε=0.069 β = -0.37) m 2 = 2.29e-03 sin 2 2Θ =0.986 (α =-0.056 β = -0.09 ε = 0.009) Flux ( 10-13 cm -2 s -1 sr -1 ) 3 2.5 2 1.5 1 0.5 Non-Showering Upmus Observed Flux Null Oscillation ( α=-0.203, ε=0.069) m 2 = 2.29e-03 sin 2 2Θ =0.986 (α =-0.056 ε = 0.009) 0-1 -0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0 cos Θ Upward Showering Muons 0-1 -0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0 cos Θ Flux ( 10-13 cm -2 s -1 sr -1 ) 1.4 1.2 1 0.8 0.6 0.4 Observed Flux Null Oscillation ( α=-0.203, ε=0.069 γ = 0.05) m 2 = 2.29e-03 sin 2 2Θ =0.986 (α =-0.056 γ = -0.02 ε = 0.009) 0.2 0-1 -0.9-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1 0 cos Θ
Overview of Astrophysical Searches Done with Upward Muons Point Sources Searches (both known and unknown) & Space-time correlations with transient astrophysical sources such as GRBs and SGRs Diffuse Flux from plane of Galaxy WIMP searches from center of Earth, Sun, Galactic Center
Angular clustering and bootstrap techniques used to search for point sources in upward muon dataset (Washburn, Clough) Upward showering muon sample suited for high energy ν astronomy because atmospheric neutrino background is reduced Sky map distribution of upward showering muons No evidence for point sources in showering upward muon dataset
Space-Time correlation studies with transient astrophysical sources like GRBs and SGRs GRBs (~1600 bursts) : astro-ph/0205304 t = 1000 sec 5 deg. Result : 1 event found. Expected Background = 0.9 t = 1 day 5 deg and look for more than 1 upward muon No such events found SGRs (~150 bursts): t = 1 day 5 deg. Result : 1 event found. Expected Background = 0.7 ==> Expected coincidences with SGRs and GRBs consistent with background
Look for from cosmic rays coming from ISM Number of Events 250 225 200 175 150 125 100 75 Upward thru muons in galactic coordinates Data (1679.6 days) M.C. without oscillations M.C. with oscillations Number of Events 90 80 70 60 50 40 30 20 Upward stopmus in galactic coordinates Data (1679.6 days) M.C. w/o oscillations M.C. with oscillations 50 25 0-80 -60-40 -20 0 20 40 60 80 Galactic latitude (in Degrees) Number of Events Upward showering muons 100 90 80 70 60 50 40 30 10 0 Data (1679.6 days) M.C. (no oscillations) : Normalized by livetime M.C. with oscillations : Normalized to Data -80-60 -40-20 0 20 40 60 80 Galactic Plane Galactic latitude (in Degrees) 20 10 0-80 -60-40 -20 0 20 40 60 80 Galactic latitude (in Degrees)
Matter/ Energy Density Budget of the Universe (WMAP Collaboration 2003) One possibility for Cold Dark Matter are: Weakly Interacting Massive Particles (WIMPs) (Weinberg & Lee 77) Masses in the GeV to TeV energy range Annihilation cross-sections ~ Electroweak scale Exist in Supersymmetric theories beyond Standard Model
INDIRECT WIMP SEARCHES WIMPs orbiting galactic halo WIMP speed ~ 220 km/sec WIMP Density ~ 0.3 GeV/cc Potential Sources of WIMP annihilation Detector Earth : Only sensitive to WIMPs with scalar interactions Sun : Mainly sensitive to WIMPs with spin interactions Galactic Center : WIMP annihilation due to spike in density distribution near central black hole (Gondolo & Silk 99) + cc - - - -, bb, HH, ZZ, tt. W + W -, ' ( Neutrinos
WIMP Searches with all upward thrugoing muons 80 50 Number of events 70 60 50 40 30 20 10 Earth data no osc with osc Number of events 45 40 35 30 25 20 15 10 5 Sun 0 0 5 10 15 20 25 30 Cone Half-angle (degrees) 0 0 5 10 15 20 25 30 Cone half-angle (degrees) Number of events 60 50 40 30 20 Galactic Center No signatures of WIMP annihilation 10 0 0 5 10 15 20 25 30 Cone Half-Angle (degrees)
WIMP Search with Upward Showering Muons Upward Showering muons sensitive to high mass WIMPs Looked for excess in direction of Earth, Sun and Galactic Center Cone Earth Data Atm ν Bkgd Cone Sun Data Atm ν Bgd 3 0 0.10 5 0 0.30 10 2 2.30 3 5 10 0 0.30 0 0.40 2 2.10 Cone 3 5 10 Galactic Center Data Atm ν Bgd 0 0.50 0 1.20 2 3.30 No possible signatures of WIMP-induced upward showering muons from Earth, Sun or Galactic Center
Limits on WIMP-induced upward muons (cm -2 sec -1 ) 10-11 10-12 10-13 10-14 10-15 10-16 Earth Super-K (all upthrumus) Super-K (showering upthrumus) 10 10 2 10 3 10 4 Mass of Neutralino (GeV) Limit on WIMP-induced upward muons (cm -2 sec -1 ) 10-11 10-12 10-13 10-14 10-15 10-16 Limit on WIMP-induced upward muons (cm -2 sec -1 ) 10-11 10-12 10-13 10-14 10-15 10-16 Galactic Center Super-K (all upthrumus) Super-K (showering upthrums) Sun 10 10 2 10 3 10 4 Mass of Neutralino (GeV) Super-K (all upthrumus) Super-K (showering upthrumus) 10 10 2 10 3 10 4 Mass of Neutralino (GeV)
WIMP nucleon Cross section (cm 2 ) Using the flux limits from the Sun and Earth and results from from Kamionkowski et. al. (1994) allows to compare our results with direct detection WIMP searches using Si, Ge, NaI etc 10 40 10 41 10 42 10 43 Scalar Coupling Super K Edelweiss CDMS ZEPLIN DAMA 10 1 10 2 10 3 WIMP Mass (GeV) WIMP Proton Cross Section (cm 2 ) 10 34 10 35 10 36 10 37 10 38 10 39 Super K (1679.6 days) UKDMC NaI from combined Na and I Elegant V Axial Vector Coupling 10 1 10 2 10 3 WIMP Mass (GeV) We partially rule out the DAMA allowed region Our limits on WIMP-proton spin cross-section ~ 100 times more sensitive than direct detection experiments
CONCLUSIONS A sample of upward through-going muons which lose energy through bremsstrahlung and other radiative processes has been isolated Mean energy of parent neutrinos of upward showering muons ~ 1 TeV Zenith angle distribution of only upward showering muons consistent with null oscillations. However combination of all 3 datasets consistent with oscillations Tried all possible tricks to search for signatures of high energy GeV neutrinos from astrophysical sources. Unfortunately no sources found