Latest Neutrino Oscillation Results From SuperKamiokande XXV Physics

Latest Neutrino Oscillation Results From Super-Kamiokande XXV Physics in Collision Prague 6 -9 July 2005 Hank Sobel University of California, Irvine

The Super-Kamiokande Detector 42 m n 50, 000 ton n 1000 m underground (22, 500 ton fid. ) n 11, 146 20 inch PMTs n 1, 885 anti-counter PMTs 1996 1997 1998 1999 2000 2001 SK-1 2002 2003 accident 2004 2005 2006 SK-2 PMT coverage 40% 6 pe/Me. V ; 1489 live-days 2007 reconstruction SK-3 and water filling 19% 40% 2. 8 pe/Me. V ; 622 live-days

OUTLINE Atmospheric Neutrino Results ●Two-flavor Analyses ●SK-I Combined Analysis ●L/E Analysis ●New Combined Finer-Binned Analysis ●SK-II results ●Tau Analysis ●Three-flavor Analysis- dominant mass scale ●Solar Neutrino Results ●Solar flux SK-I, SK-II ●Time variation SK-I, SK-II ●Spectrum SK-I, SK-II ● 2 -Flavor analysis SK-I ●Latest K 2 K results ●

Three Neutrino Picture MNS Mass Maki-Nakagawa-Sakata Flavor Eigenstates mixing matrix Eigenstates 3 independent parameters + 1 complex phase Experimentalist version: Atmospheric Long baseline 12, 23, 13, d Chooz Long baseline Future reactor Solar n Reactor

Mass Hierarchy? Small e allowed Solar ~8 x 10 -5 ev 2 Atmospheric ~2. 5 x 10 -3 e. V 2 Solar ~8 x 10 -5 ev 2 Normal Inverted

Atmospheric Neutrinos Study Angular Distributions of m and e with Super-Kamiokande About 15 km P ’ = sin 22 qsin 2(1. 27 m 2 L/E ) L ~ E / dm 2 About 13, 000 km Neutrinos produced in the atmosphere at ~15 km altitude. . . travel through the earth and interact in the detector.

Sensitivity of angular distribution to neutrino oscillations Vary dm 2 No oscillations dm 2 = 10 -4 L ~ E / dm 2 10 -3 10 -1 Up-going Sensitivity from ~10 -4 to 10 -1 e. V 2 Down-going

Particle Identification sub-Ge. V Electron-like: e, g, p 0 multi-Ge. V Muon-like: m, p+/Based on likelihood of shape around Cerenkov ring. Mis-ID Prob. : MC, CC events 0. 8 +/- 0. 1% Cosmic-ray m: 0. 4 +/- 0. 1% Michel e’s: 1. 8 +/- 0. 5%

Atmospheric Neutrino Interactions • Contained ( e , m ) • Fully-Contained (FC) E ~ 1 Ge. V Start inside, stop inside • Partially-Contained (PC) E ~ 5 Ge. V Start inside, stop outside • Upward-going muons • Stopping E ~ 5 Ge. V Start outside, stop inside • Through-going E ~ 200 Ge. V Start outside, stop outside p p, k m,

SK I Combined 2 -Flavor Analysis Zenith angle distributions 1489 live-days, 100 yr MC, >15, 000 neutrino events upgoing downgoing

Oscillation Analysis 'Pull method', Fogli et al, hep-ph/0206162 Total 180 bins in zenith angle, momentum observed events in bin MC statistical and systematic uncertainties oscillated expectation fractional change in event rate due to systematic parameters (39 based on flux, x-scn, selection, reconstruction) estimated uncertainty in ej

Allowed Oscillation Parameters SK I standard combined analysis BEST FIT: m 2 = 2. 1 10 -3 e. V 2 sin 2(2Θ) = 1. 0 BEST FIT(UNPHYSICAL): m 2 = 2. 1 10 -3 e. V 2 sin 2(2Θ) = 1. 02 1. 5 x 10 -3 e. V 2< m 2< 3. 4 x 10 -3 e. V 2 sin 22 > 0. 92 (@90%C. L. )

L/E Oscillation Analysis FC/PC events Expected oscillation pattern Poor resolution in L/E (~10's of %) Smears out oscillation pattern

Select Events for Good L/E Resolution (<70%) Low Energy Horizont al PC OD stopping Low Energy Horizont al FC multi ring m-like Low Energy PC OD through-going Horizont al FC single ring m-like Low Energy L is not well measured near horizon and at low energy

Result of Event Selection MC Data Oscillation dip decoherence decay 100 Year MC After event selection Selected subset (2700 events ~1/5 of total data) Neutrino Decay: Δ 2=11. 3 = 3. 4 Decoherence: Δ 2=14. 5 = 3. 8

Best Fit Using High-resolution L/E Sample Overlay with Best Fit Results: m 2 = 2. 4 x 10 -3 e. V 2 sin 2(2 ) = 1. 0 (constrained to physical region) 2 min= 37. 9/40 DOF 1. 9 x 10 -3 < m 2< 3. 0 x 10 -3 zenith angle results e. V 2 0. 90 < sin 22 (@90%C. L. ) Somewhat better m 2 resolution

Preliminary test New Finer-Binned 2 -Flavor Analysis Try to combine advantages of: 2 ● Zenith angle analysis (good sensitivity for sin 2 ) 2 ● L/E analysis (good sensitivity for dm ) • PC events divided into OD stop and OD through-going • New FC multi-ring e-like category • Finer multi-Ge. V momentum binning – sensitive to region of oscillation dip • Total of 370 bins in 10 zenith angle, 37 momentum bins • c 2 incorporates Poissonian uncertainties

Zenith angle distributions Preliminary

Allowed region with finer-binned analysis Preliminary SK-I 370 bin analysis Best Fit Results: (constrained to physical region) m 2 = 2. 5 x 10 -3 e. V 2 sin 2(2 ) = 1. 0 2 min= 375. 2/367 DOF Comparison with 180 bin analysis 2. 0 < m 2 < 3. 0 x 10 -3 e. V 2 sin 22 > 0. 93 (@90%C. L. )

Super-K II results 1. 72 live-years FC/PC ; 1. 67 live-years upmu ; 60 yr SK II MC Preliminary

Super-K II Oscillation Analysis Preliminary L/E Analysis Best Fit Results: m 2 = 2. 6 x 10 -3 e. V 2 sin 2(2 ) = 1. 0 (constrained to physical region) 2 min= 54. 8/40 DOF 370 bin combined zenith angle analysis Best Fit Results: m 2 = 3. 1 x 10 -3 e. V 2 sin 2(2 ) = 0. 98 (constrained to physical region) 2 min= 394. 5/367 DOF

Search for Appearance 100 • • Very difficult in water detector With m 2~(few) 10 -3 e. V 2 10 -1 – Expect ~1 event/kt-yr from atmospheric neutrinos – All CC events upward-going 10 -2 – Deep-inelastic channel dominates CC rate/kt-yr vs. m 2 10 -4 10 -3 m 2 (e. V 2) ~1 event/kt-yr i. e. about 90 in our data now. S/N ~0. 7% 10 -2

Appearance Signature Hadrons m production threshold = 3. 5 Ge. V Electron or muon or hadrons • Simulate cross-sections, topologies • Characteristics: • Mean neutrino energy ~10 Ge. V • > 80% are showering » Combination of electron and hadron showers • Lots of pions • Find -like particle ID estimator • Look for excess in upward direction

Two Tau Analyses Not statistically independent • Concentrate on hadronic decays of Tau – Event shape – Extra pion production • Pick FC events in fiducial volume > 2 m from wall • Pick variables which distinguish Tau’s from atmospheric background and compare with downward-going data • Make likelihood or neural network to select tau-like events relative to atmospheric BG

Input Variable Distributions Max. distance between primary and electron vertex from mu→e Visible energy Down-going data Max. momentum for mu-like ring Clustered sphericity by the energy flow analysis Down-going νμ, νe MC MC Log of sphericity in the lab frame by energy flow analysis Number of ring candidates

Signal Efficiency of Cuts Efficiency after Fiducial, Evis, PID and NN cut 36% Tau Efficiency after Fiducial, Evis, PID and Likelihood cut BG Tau 69% Signal 42% Tau Eff Output of Neural Net -likelihood

Tau Signal MC expectation: 79 31 's (error dominated by flux and cross-section) Two analyses (likelihood and neural network) yield consistent answers Neural Network (36%efficiency) Signal = 152 47 stat 12. 027. 3 Likelihood (42% efficiency) = 145 48 stat 9. 436. 2 Parameters varied: Δm 2: 2. 0 -2. 7 x 10 -3 ev 2 ; sin 22 q: . 93 -1. 0 ; q 13 moved to CHOOZ limit; Systematic error dominated by q 13 error.

3 Flavor Neutrino Oscillation Analysis m 2 m 1 { m 3 atmospheric m 2 m 1 Normal }solar Inverted m 3 If matm 2 >> msolar 2 & d=0 m 2 ~ m 132 ~ m 232 ; q 23 ; q 13 Oscillation can be described by 3 parameters Consider matter effects: neutrino anti-neutrino Normal enhanced suppressed Inverted suppressed enhanced 1. Look for signature of non-zero 13 in enhancement of e for certain angles and energies 2. Look for differences between neutrinos and antineutrinos.

Try to Observe Matter Effect 1. Look for enhancement of high-energy e Take m 2 = 3 x 10 -3 ; sin 2 q 23 = 0. 5 ; sin 2 q 13 = 0. 026 P ( m e) Look at Cosq=0. 6 Cos q P( e e) P( m e) E (Ge. V) P( e m) P( m m) Ge. V 20 yr SK MC Single-ring electrons (2. 5<P<5. 0 Ge. V) Positive m 2 Negative m 2 Null oscillation No significant excess due to matter effect was seen in upward-going multi. Ge. V single ring electron sample

2. Look for Differences Between Neutrinos and Antineutrinos For inverted hierarchy, the resonance happens for antineutrinos, not neutrinos. • There are vs differences in and d /dy, and the relative fraction of vs is different for different subsamples. ● Single-ring e-like NC + others CC e Multi-ring e-like CC e higher e/ e ratio NC + others

For example: more enhancement of upward-going multi-Ge. V e-like vs single rings for normal hierarchy Multi-ring elike(2. 5<P<5. 0 Ge. V) Single-ring e-like (2. 5<P<5. 0 Ge. V) SK 20 years m 2= 0. 002 e. V 2 sin 2 23 = 0. 5 sin 2 13 = 0. 05 higher e/ e ratio m 2 Positive Negative m 2 Null oscillation cos

Data: single-ring Multi-Ge. V e-like events multi-ring cos no excess seen. . . - use fine-binned analysis, Poisson-style uncertainties - treat inverted and normal hierarchy cases separately

3 Flavor Analysis(assume Δm 122=0) Normal hierarchy Preliminary Best Fit: 2 min= 376. 82/368 DOF m 2=2. 5 x 10 -3 e. V 2 sin 2 23 = 0. 5 sin 2 13 = 0. 0 Inverted hierarchy Preliminary Best Fit: 2 min= 376. 76/368 DOF m 2=2. 5 x 10 -3 e. V 2 sin 2 23 = 0. 525 sin 2 13 = 0. 00625

Solar Neutrinos e-type neutrino other active flavors Recoil electron spectrum with detector resolution

Solar Neutrino Flux in SK-1 May 31, 1996 – July 13, 2001 q Best fit Background e Data ● f(8 B)day = 2. 32 0. 03 0. 08 0. 07 f(8 B)night = 2. 37 0. 03 0. 08 D-N (D+N)/2 = -(0. 021 0. 020 0. 013 0. 012 ) 22385 solar events flux : (14. 5 events/day) 2. 35 0. 02 0. 08 [x 106 /cm 2/sec] 8 B

Solar Neutrino Flux in SK-2 7. 0 (stat. ) – 20 Me. V 22. 5 kt Signal 5482622 day +132 -130 7. 0 – 20. 0 Me. V Dec. 24, 2002 – May 19, 2005 622 days SK-II Best fit Background Data ● f(8 B)day = 2. 39 f(8 B) D-N night (D+N)/2 0. 09 0. 08 = 2. 35 0. 08 = (0. 014 0. 049) 5482 solar events (8. 8 events/day) flux : 2. 36 0. 06 [x 106 /cm 2/sec] Preliminary 8 B

Time Variation of Solar Neutrino Flux

Spectrum SK-1 SK-2

SK Excluded and Allowed Regions SK Zenith Spectrum & Rate SK Zenith Spectrum

Changes to Global Analysis I. Full SNO Salt Data SNO Collaboration, nucl-ex/0502021 II. Full Kam. LAND Likelihood Kam. LAND Collaboration, hep-ex/0406035 v 3

Oscillation Parameters from Solar Neutrino and Kam. LAND Experiments Solar & Kam. LAND 99. 73% Solar+Kam. LAND 95% Best fit: +0. 08 tan 2 =0. 437 -0. 05 +0. 4 dm 2=7. 94 -0. 3 x 10 -5 e. V 2 Solar Michael Smy, UC Irvine

Schematic Overview of K 2 K Experiment started in April, 1999 Terminated in November, 2004 GPS Satellite (broadcasts time) Proton Synchrotron 12 Ge. V Protons Neutrinos Tracker 1 kton Pions Decay Pipe Magnetic Horns

K 2 K-I + K 2 K-II Results K 2 K-1+2 DATA FC 22. 5 kt, 112 155. 9 　1 ring 67 99. 0 　　m-like 58 90. 8 　　e-like 9 8. 2 45 56. 8 　Multi Ring MC Single-ring m-like events Best Fit Results: m 2 = 2. 8 x 10 -3 e. V 2 sin 2(2 ) = 1. 0 (constrained to physical region) No-oscillation excluded at >4

Limit on qme Expected BG: 1. 63 Observed: 1 Feldman-Cousins Conventional method (single sided) K 2 K 90%CL upper limit CHOOZ 90%CL upper limit 1. 25 m interactions + 0. 38 beam e

Conclusions • Several SK oscillation results point to oscillations near dm 2 = 2. 4 x 10 -3 e. V 2 – Full analysis 1. 5 x 10 -3 e. V 2< m 2< 3. 4 x 10 -3 e. V 2 sin 22 > 0. 92 (@90%C. L. ) – L over E 1. 9 x 10 -3 < m 2< 3. 0 x 10 -3 e. V 2 sin 22 > 0. 90 (@90%C. L. ) – New fine-binned analysis 2. 0 < m 2 < 3. 0 x 10 -3 e. V 2 sin 22 > 0. 93 (@90%C. L. ) • SK-II atmospheric and solar data is consistent with SK-I • Tau appearance now at 3 s level – consistent with expectations • Improved 3 -flavor analysis limit on q 13 • New K 2 K results consistent m 22 = (2. 55 0. 40)x 10 -3 e. V 2 – Rules out no oscillation at >4 s sin 2 = 1. 19 0. 23