1. Imperial College London Morgan O. Wascko Neutrino & Antineutrino Oscillation Searches at MiniBooNE Morgan O. Wascko Imperial College London 2010 11 08 Morgan O. Wascko Imperial College London 2010 11 08

2. Imperial College London Morgan O. Wascko Prologue There’s been a lot of confusion about what the MiniBooNE results mean. There’s even confusion about what they are. I want to explain what we see, and show the detailed evidence we have assembled. Interpretations are a different matter...

3. Imperial College London Morgan O. Wascko Outline Motivation and History MiniBooNE Description MiniBooNE Analysis Methods Summary of past results Antineutrino Results Future Prospects

4. Imperial College London Morgan O. Wascko Motivation & History

5. Imperial College London Morgan O. Wascko Neutrino Oscillation if neutrinos have mass... a neutrino that is produced as a  μ (e.g.  + →  +  μ ) might some time later be observed as a  e (e.g.  e n → e - p) ++  ++ X ee e-e-  source  detector Pontecorvo, Maki, Nakagawa, Sakata

6. Imperial College London Morgan O. Wascko Consider only two types of neutrinos If weak states differ from mass states i.e. (    e )  (  1  2 ) Then weak states are mixtures of mass states Probability to find  e when you started with   Neutrino Oscillation 11 22 ee  P osc (   →  e ) =  〈  e   (t) 〉  2  〉   〉     〉   

7. Imperial College London Morgan O. Wascko 2 fundamental parameters Δm 2 12 (=m 1 2 -m 2 2 ) ↔ period   ↔ magnitude 2 experimental parameters L = distance travelled E = neutrino energy Tune L&E for Δm 2 range, uncertainties determine  sensitivity Neutrino disappearance and appearance

8. Imperial College London Morgan O. Wascko 2 fundamental parameters Δm 2 12 (=m 1 2 -m 2 2 ) ↔ period   ↔ magnitude 2 experimental parameters L = distance travelled E = neutrino energy Tune L&E for Δm 2 range, uncertainties determine  sensitivity Neutrino disappearance and appearance

9. Imperial College London Morgan O. Wascko 2 fundamental parameters Δm 2 12 (=m 1 2 -m 2 2 ) ↔ period   ↔ magnitude 2 experimental parameters L = distance travelled E = neutrino energy Tune L&E for Δm 2 range, uncertainties determine  sensitivity Neutrino disappearance and appearance

10. Imperial College London Morgan O. Wascko 2 fundamental parameters Δm 2 12 (=m 1 2 -m 2 2 ) ↔ period   ↔ magnitude 2 experimental parameters L = distance travelled E = neutrino energy Tune L&E for Δm 2 range, uncertainties determine  sensitivity Neutrino disappearance and appearance

11. Imperial College London Morgan O. Wascko Atmospheric  Δm 2 23 |=(2.43±0.11)  10 -3 eV 2  23 =45±6   Δm 2 23 |=(2.43±0.11)  10 -3 eV 2  23 =45±6  http://hitoshi.berkeley.edu/neutrino

12. Imperial College London Morgan O. Wascko Solar +1.6 Δm 2 12 =(7.59±0.21)  10 -5 eV 2  12 =34.4  -1.5 http://hitoshi.berkeley.edu/neutrino

13. Imperial College London Morgan O. Wascko Δm 2  ~ eV 2    2  Short baseline http://hitoshi.berkeley.edu/neutrino

14. Imperial College London Morgan O. Wascko hep-ex/0104049    →  e oscillation probability: 0.264±0.067±0.045% KARMEN2 and LSND collaborators performed joint analysis on both data sets - allowed regions remain! Δm 2 ~ 1eV 2,  ~ 2° The LSND signal  excess! hep-ex/0203023

15. Imperial College London Morgan O. Wascko Verifying LSND LSND interpreted as 2  oscillation Verification requires same (L/E) and high statistics Different systematics MiniBooNE chose higher L and higher E Strategy: search for  e excess in   beam PRD 64, 112007

16. Imperial College London Morgan O. Wascko MiniBooNE Collaboration

17. Imperial College London Morgan O. Wascko MiniBooNE Description

18. Imperial College London Morgan O. Wascko Fermilab Visual Media Services Overview JL Raaf MiniBooNE Overview →→

19. Imperial College London Morgan O. Wascko JL Raaf MiniBooNE Overview →→ Target & Horn Main components of Booster Neutrino Beam (BNB) (96M and 298M+ pulses)

20. Imperial College London Morgan O. Wascko JL Raaf MiniBooNE Overview →→ Meson Production PRD 79 072002 (2009) External meson production data HARP data (CERN) Parametrisation of cross- sections Sanford-Wang for pions Feynman scaling for kaons Use of HARP data reduces total flux error to ~9%

21. Imperial College London Morgan O. Wascko  Flux 99.5% pure muon flavour 0.5% intrinsic  e Constrain  e content with   measurements  mode contains large  contamination JL Raaf MiniBooNE Overview →→ PRD 79 072002 (2009) Neutrino Mode

22. Imperial College London Morgan O. Wascko  Flux 99.5% pure muon flavour 0.5% intrinsic  e Constrain  e content with   measurements  mode contains large  contamination JL Raaf MiniBooNE Overview →→ PRD 79 072002 (2009) Antineutrino Mode

23. Imperial College London Morgan O. Wascko JL Raaf MiniBooNE Overview →→ Neutrino Detector NIM A 599 (2009) 28-46

24. Imperial College London Morgan O. Wascko CC / NC quasi-elastic scattering (QE) 42% / 16% MiniBooNE is here W+W+   n p  -   W+W+  n ++ -- n Z0Z0  p  p Z0Z0 p,n 00  p,n CC / NC resonance production (1  ) 25% / 7%  Interactions

25. Imperial College London Morgan O. Wascko CC / NC quasi-elastic scattering (QE) 42% / 16% MiniBooNE is here W+W+   n p  -   W+W+  n ++ -- n Z0Z0  p  p Z0Z0 p,n 00  p,n CC / NC resonance production (1  ) 25% / 7%  Interactions

26. Imperial College London Morgan O. Wascko Mineral Oil Optics  Particle track Wavefront Production: Cherenkov and scintillation Secondary: Fluorescence and scattering (Raman, Rayleigh) Molecular energy levels of oil light 

27. Imperial College London Morgan O. Wascko Muons full rings Electrons fuzzy rings Neutral pions double rings Track Images

28. Imperial College London Morgan O. Wascko PMT Hit Clusters PMT hits clusters in time form “subevents”    events have 2 subevents , followed by e   events have 1 subevent Simple cuts on subevents remove cosmic backgrounds “pre-cuts” H.A. Tanaka

29. Imperial College London Morgan O. Wascko Track Reconstruction PMT Charged particles produce Cherenkov and scintillation light in oil PMTs collect photons, record t,Q Reconstruct tracks by fitting time and angular distributions Find position, direction, energy

30. Imperial College London Morgan O. Wascko PMT Calibration 10% photo-cathode coverage Charge Res: 1.4 PE, 0.5 PE Time Res: 1.7 ns, 1.1 ns PMTs are calibrated with a laser + 4 flask system Laser data are acquired at 3.3 Hz to continuously calibrate PMT gain and timing constants Two types of 8” Hamamatsu Tubes: R1408, R5912 R.B. Patterson

31. 13% E resolution at 53 MeV Michel electrons Michel electrons: -set absolute energy scale and resolution at 53 MeV endpoint -optical model tuning use cosmic muons and their decay electrons (Michels) Cosmic  calibration Cosmic muons which stop in cubes: -test energy scale extrapolation up to 800 MeV - measure energy, angle resolution - compare data and MC Muon tracker + cube calibration data continuously acquired at 1 Hz Muon tracker 7 scintillator cubes

32. Imperial College London Morgan O. Wascko Muon tracker 7 scintillator cubes Cosmic  calibration use cosmic muons and their decay electrons (Michels) Energy (MeV)  res (  ) E res (%) 94±45.412 155+53.27.0 229±72.27.5 407±91.44.6 584±91.14.2 771±91.03.4 NIM A 599 (2009) 28-46

33. Imperial College London Morgan O. Wascko Particle Identification e-like μ -like Monte Carlo π 0- like e-like Monte Carlo Reconstruct under 3 hypotheses:  -like, e-like and  0 -like  e particle ID cuts on likelihood ratios chosen to maximise   →  e oscillation sensitivity

34. Imperial College London Morgan O. Wascko 1 decay electron 0 decay electrons e/  Likelihood   CCQE data (with muon decay electron) compared to   data with no decay electrons (“All but signal”) Removes most muon events

35. Imperial College London Morgan O. Wascko e/   Likelihood Data and MC PID uses cuts on likelihood ratio reconstructed    mass Open sidebands before unblinding full data sample Signal Region

36. Imperial College London Morgan O. Wascko Detector Stability

37. Imperial College London Morgan O. Wascko Experiment Stability

38. Imperial College London Morgan O. Wascko MiniBooNE Analysis

39. Imperial College London Morgan O. Wascko Beam Flux Prediction Start with a Geant 4 flux prediction for the  spectrum from  and K produced at the target Event Reconstruction Use track-based event reconstruction Cross Section Model Predict  interactions using the Nuance event generator Optical Model Pass final state particles to Geant 3 to model particle and light propagation in the tank Simultaneous Fit to  μ and  e events Fit reconstructed energy spectrum for oscillations Particle Identification Use hit topology and timing to identify electron-like or muon-like Cherenkov rings and corresponding charged current neutrino interactions  e appearance analysis

40. Imperial College London Morgan O. Wascko RB Patterson stacked signal and backgrounds after ν e event selection Signal & Backgrounds Oscillation  e Example oscillation signal – Δ m 2 = 1.2 eV 2 –SIN 2 2  = 0.003 Fit for excess as a function of reconstructed  e energy Example from neutrino mode

41. Imperial College London Morgan O. Wascko RB Patterson stacked signal and backgrounds after ν e event selection Signal & Backgrounds ν e from K + and K 0 Use fit to kaon production data for shape Use high energy ν e and ν μ in- situ data for normalisation cross-check Example from neutrino mode

42. Imperial College London Morgan O. Wascko RB Patterson stacked signal and backgrounds after ν e event selection Signal & Backgrounds ν e from μ + Measured with in-situ ν μ CCQE sample – Same ancestor π + kinematics Most important background - Constrained to a few %   p+Be  +  e  +    e + Example from neutrino mode

43. Imperial College London Morgan O. Wascko RB Patterson stacked signal and backgrounds after ν e event selection Signal & Backgrounds MisID ν μ ~46% π 0 – Determined by clean π 0 measurement ~16% Δ γ decay – π 0 measurement constrains ~24% other – Use ν μ CCQE rate to normalise and MC for shape ~14% “dirt” – Measure rate to normalise and use MC for shape Example from neutrino mode

44. Imperial College London Morgan O. Wascko Additional Background Antineutrino mode Antineutrino beam contains significant fraction of “wrong sign” neutrino events Stemming from unfocussed pions in secondary beam ~20% of reconstructed events MinBooNE cannot sign select events Need other methods to constrain WS BGs ++ ++

45. Imperial College London Morgan O. Wascko Additional Background Antineutrino mode Antineutrino beam contains significant fraction of “wrong sign” neutrino events Stemming from unfocussed pions in secondary beam ~20% of reconstructed events in nubar mode MinBooNE cannot sign select events Need other methods to constrain WS BGs

46. Imperial College London Morgan O. Wascko  e BG prediction 5.66e20 POT Source200-475475-1250 μ±μ± 13.431.4 K±K± 8.218.6 K0K0 5.121.2 other ν e 1.32.0 NCπ 0 41.612.6 Δ→γ12.43.4 dirt6.22.6 ν μ CCQE4.32.0 other ν μ 7.04.2 TOTAL99.598.0 Intrinsic  e Mis-ID  

47. Imperial College London Morgan O. Wascko Strategy Incorporate in-situ data whenever possible MC tuning with calibration data energy scale PMT response optical model MC tuning with neutrino data CCQE - constrain BG with data  0 rate constraint “Dirt” backgrounds WS backgrounds Constraining systematic errors with neutrino data ratio method:  e from  decay Recurring theme: good data-MC agreement

48. Imperial College London Morgan O. Wascko MC Tuning Good data/MC agreement Basic PMT hit distributions showing details of optical model Also have good agreement in aggregate PMT hit distributions showing gross detector behaviour cos 

49. Imperial College London Morgan O. Wascko   CCQE events Used to measure flux and check E  QE reconstruction 2 subevents: e,  Require e be located near end of  track T. Katori   e np 12 C A.A. Aguilar Arevalo E  QE resolution ~10%

50. Imperial College London Morgan O. Wascko   CCQE tuning Need good flux model to study cross section Data-MC mismatch follows Q2 lines, not Eν Problem is not the flux prediction, but the cross section model Data/MC ratio for T  -cos   plane, before tuning Data/MC ratio for T  -cos   plane,after tuning PRL100(2008)032301

51. Imperial College London Morgan O. Wascko   CCQE tuning Tuned nuclear parameters in Relativistic Fermi Gas model Q 2 fits to MB νμ CCQE data using the nuclear parameters: M A eff - effective axial mass  - Pauli Blocking parameter Relativistic Fermi Gas Model with tuned parameters describes   CCQE data well This improved nuclear model is used in  e CCQE model, too. Q2 distribution before and after fitting PRL100(2008)032301

52. Imperial College London Morgan O. Wascko    Mis-ID Backgrounds  0 s are reconstructed outside mass peak if: asymmetric decays fake 1-ring 1 of 2 photons exits high momentum  0 decays produce overlapping rings  e+e-e+e-  00 12 C

53. Imperial College London Morgan O. Wascko Tuning    MC good data-MC agreement in variables not used in tuning! The MC π 0 rate (flux × xsec) is re-weighted to match the measurement in p  bins. J. Link Phys.Lett. B664:41-46,2008

54. Imperial College London Morgan O. Wascko Tuning      MC Phys.Rev.D 81 013005 (2010) Use same techniques to tune MC model in antineutrino mode Also produced POT-normalised cross sections for NC  0 production by neutrinos and antineutrinos

55. Imperial College London Morgan O. Wascko Neutrinos interacting outside detector can cause BGs n,  enter detector and convert Events pile up at low energy near edge of tank Measure directly with “dirt enhanced” sample results from dirt-enhanced fits visible energy (GeV) dist to tank wall along track (cm) shower dirt   “Dirt” Backgrounds Evis RED: CCQE Nue BLACK: Background R-to-wall distance [cm] New 2D dirt cut

56. Imperial College London Morgan O. Wascko WS backgrounds Use two distinct and complementary data samples to constrain WS fraction  CCQE distribution has different angular distribution than  events helicity is different! CC1  + events stem almost entirely from nu events, not  Result: WS BG prediction reduced by ~30% J. Grange

57. Imperial College London Morgan O. Wascko For each E  bin i, Raster-scan in Δm 2 and sin 2 2   e to calculate -2ln L  over  e and   bins Systematic error matrix includes uncertainties for  e and     data plays role of near detector ( ) ee  ee ee Correlations between E ν QE bins from the optical model: ee  ee eeee Combined fit of   &  e data

58. Imperial College London Morgan O. Wascko Error Matrix Use MC variations to study systematic uncertainties Vary underlying parameters and compare to “central value” MC Total error matrix is sum of individual matrices Example of E  distributions for several MC variations i=1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 N E  QE (GeV) Central Value (MC) First variant (  =1) Second variant (  =2)

59. Imperial College London Morgan O. Wascko Kinematic distributions of π + contributing to ν μ and ν e flux (ν mode) θ π+ (radians) p π+ (GeV/c) θ π+ (radians) p π+ (GeV/c) Fit method example R(ν μ ) = Φ(ν μ ) x σ(ν μ ) x ε(ν μ ) R(ν e ) = Φ(ν e ) x σ(ν e ) x ε(ν e ) strong correlations between  e signal, background, and   CCQE sample    ee ee  

60. Imperial College London Morgan O. Wascko Fit method example    ee ee    e from  decay unconstrained  e from  decay constrained strong correlations between  e signal, background, and   CCQE sample

61. Imperial College London Morgan O. Wascko Fit method example strong correlations between  e signal, background, and   CCQE sample R(ν μ ) = Φ(ν μ ) x σ(ν μ ) x ε(ν μ ) R(ν e ) = Φ(ν e ) x σ(ν e ) x ε(ν e ) Effect of  ν μ CCQE constraint on sensitivity 90% CL sensitivity E ν QE > 200 MeV with  ν μ CCQE constraint without  ν μ CCQE constraint ←improves sensitivity and provides stronger constraint to oscillations    ee ee  

62. Imperial College London Morgan O. Wascko BG systematic errors (%) NeutrinoAntineutino Source200-475475-1100200-475475-1100 Flux from π + /μ + decay0.40.91.82.2 Flux from π - /μ - decay3.02.30.10.2 Flux from K + decay2.24.71.45.7 Flux from K - decay0.51.2-- Flux from K 0 decay1.75.40.51.5 Target and beam models1.73.01.32.5 ν cross section6.513.05.911.9 NC π 0 yield1.51.31.41.9 Hadronic interactions0.40.20.80.3 External interactions (dirt)1.60.70.80.4 Optical model8.03.78.92.3 Electronics & DAQ model7.02.05.01.7 TOTAL (unconstrained)13.516.012.314.2

63. Imperial College London Morgan O. Wascko Reminder:  e Search Above 475 MeV... Excellent agreement with background predictions Find 408 events, expect 386 ± 20(stat) ± 30(syst) Chi-square probability of 40% in 475-1250 MeV Since this is the region of highest sensitivity to and LSND-like 2ν mixing hypothesis, can use it to exclude that model (E > 475 MeV) Low E

64. Imperial College London Morgan O. Wascko Reminder:  e Search Above 475 MeV... Excellent agreement with background predictions Find 408 events, expect 386 ± 20(stat) ± 30(syst) Chi-square probability of 40% in 475-1250 MeV Since this is the region of highest sensitivity to and LSND-like 2ν mixing hypothesis, can use it to exclude that model

65. Imperial College London Morgan O. Wascko Reminder:  e Search Below 475 MeV... Find 544 events, expect 415 ± 20(stat) ± 39(syst) Excess is 128 ± 20(stat) ± 39(syst) events BG SourceBG Counts Increase Needed Syst Error*   CCQE 26.4487%~30% NC  0 181.371%~20% Rad. Δ67.0192%~25%  e (μ) 58.1222%~25%  e (K) 17.4740%~40% dirt23.8544%~15% (E > 475 MeV) Low E How much would BGs need to fluctuate to produce excess?

66. Imperial College London Morgan O. Wascko Several possible explanations 3+2 with CP violation [Maltoni and Schwetz, hep-ph0705.0107 ; G. K., NuFACT 07 conference] Anomaly mediated photon production [Harvey, Hill, and Hill, hep-ph0708.1281] New light gauge boson [Nelson, Walsh, Phys. Rev. D 77, 033001 (2008)] … Some have concrete predictions for MiniBooNE antineutrino mode running low energy excess

67. Imperial College London Morgan O. Wascko PRL103(2009)061802   disappearance ν μ and  ν μ disappearance oscillation test is done by shape-only fit for data and MC with massive neutrino oscillation model. MiniBooNE can test unexplored region by past experiments, especially there is no tests for antineutrino disappearance between Δm 2 =10eV 2 and atmospheric Δm 2

68. Imperial College London Morgan O. Wascko First  e results 3.4E20 POT From 200-3000 MeV excess is 4.8 +/- 17.6 (stat+sys) events. No significant excess E < 475 MeV. Statistically small excess (more of a wiggle) in 475-1250 MeV region Assume neutrinos do not oscillate in fit Stat error too large to distinguish LSND-like from null PRL 103,111801 (2009)

69. Imperial College London Morgan O. Wascko First  e results 3.4E20 POT From 200-3000 MeV excess is 4.8 +/- 17.6 (stat+sys) events. No significant excess E < 475 MeV. Statistically small excess (more of a wiggle) in 475-1250 MeV region Assume neutrinos do not oscillate in fit Stat error too large to distinguish LSND-like from null PRL 103,111801 (2009) E>475 MeV 90% CL limit 90% CL sensitivity  Exclusion Limits: 3.4E20 POT

70. Imperial College London Morgan O. Wascko Antineutrino Results

71. Imperial College London Morgan O. Wascko Training for a blind search MOW c. 2002 (blinded)

72. Imperial College London Morgan O. Wascko   results 5.6E20 POT

73. Imperial College London Morgan O. Wascko Below 475 MeV... Find 119 events, expect 100 ± 10(stat) ± 10(syst) Excess is 18.5 ± 10(stat) ± 10(syst) events Inconsistent with many hypotheses explaining the  mode low E excess Low energy   results BG Source  e Prediction CC bkgs38.6 NC π 0 31 Rad Δ24.9 K0K0 114.3 charged K38 WS neutrinos12 same xsec68  mode  mode

74. Imperial College London Morgan O. Wascko   results anti- results Above 475 MeV... In 475-1250 MeV, excess 20.9 ± 14 events (1.4σ) True significance comes from fit over entire > 475 MeV energy region +   constraint Best fit preferred over null at 99.4% CL (2.7σ) Probability of null hypothesis (no model dep.) is 0.5% in 475-1250 MeV signal region

75. Imperial College London Morgan O. Wascko Comparing to LSND Fit to 2  mixing model Model-independent plot of inferred oscillation probability

76. Imperial College London Morgan O. Wascko Assumes ν e excess should be present for WS ν μ in beam Another check E ν QE [MeV] Bkgd200-475475-12501250-3000 MC100.599.134.2 Data11912038 Excess18.5 ± 10 ± 1020.9 ± 10 ± 103.8 ± 5.8 LSND Best Fit7.622.03.5  Low-E excess 11.6~2~0 LSND + Low-E19.224.03.5

77. Imperial College London Morgan O. Wascko Fit to 2 ν mixing model Outside of Karmen2 & Bugey 90 CL, inside of LSND & MiniBooNE 90CL

78. Imperial College London Morgan O. Wascko Prospects

79. Imperial College London Morgan O. Wascko What now? Step 1:  result is stat limited need more data ! Proposal to FNAL to collect 15e20 POT prior to March 2012 shutdown At 15e20,  significance could grow to 3.7σ… or drop below 95% Possibility for ~20% analysis gain during this time LSND  =3.8 , MB  =3.0 , MB  =2.7 

80. Imperial College London Morgan O. Wascko Decay region 50 m MiniBooNE Detector SciBooNE Booster  beam 100 m 440 m Fermilab Visual Media Services Overview

81. Imperial College London Morgan O. Wascko   disappearance MiniBooNE-SciBooNE combined νμ disappearance oscillation analysis combined analysis with SciBooNE can constrain Flux+Xsec error Flux-> same beam line Xsec->same target (carbon) Decay region 50 m MiniBooNE Detector SciBooNE Booster  beam 100 m 440 m

82. Imperial College London Morgan O. Wascko   disappearance MiniBooNE-SciBooNE combined νμ disappearance oscillation analysis combined analysis with SciBooNE can constrain Flux+Xsec error Flux-> same beam line Xsec->same target (carbon) Decay region 50 m MiniBooNE Detector SciBooNE Booster  beam 100 m 440 m

83. Imperial College London Morgan O. Wascko MiniBooNE outlook Approved for another ~5e20 POT Running right now SciBooNE-MiniBooNE joint analysis ready soon Submitted LOI for second mineral oil Cherenkov detector MicroBooNE under construction, can address low energy excess

84. Imperial College London Morgan O. Wascko Summary

85. Imperial College London Morgan O. Wascko What does MiniBooNE claim? 1. No   excess in   beam above 475 MeV. ➡ Maximal sensitivity if LSND is L/E and CPT invariant. 2. 3σ excess (128 ± 43) of  e candidates in   beam below 475 MeV. ➡ Does not fit well to a 2  mixing hypothesis 3. Small excess (18±14) below 475 MeV in   beam. ➡ Rules out some   beam low-E excess explanations. 4. Small excess (20.9 ± 14) in   beam above 475 MeV. ➡ Null hypothesis in 475-1250 MeV region is 0.5% probable ➡ 2  fit prefers LSND-like signal at 99.4% CL.

86. Imperial College London Morgan O. Wascko Thank you!

87. Imperial College London Morgan O. Wascko Backups

88. Imperial College London Morgan O. Wascko 88 25m Absorber Two periods of running with 1 & 2 absorber plates 1 absorber plate - 0.569E20 POT 2 absorber plates - 0.612E20 POT Good data/MC agreement in high statistics samples (   CCQE, NC  0,...) Data included in this analysis p Dirt ~500m Decay region ~50m π+π+ π-π- νµνµ µ-µ- (antineutrino mode)‏

89. Imperial College London Morgan O. Wascko 3 Flavours flavor mass AtmosphericSolar Cross-Mixing where c ij = cos  ij, etc.

90. Imperial College London Morgan O. Wascko 3 Flavours Mass (eV) 33 22 11 solar atmospheric ? 0.05 0.009 flavour key:  e    

91. Imperial College London Morgan O. Wascko Today’s Open Questions What is the last mixing angle? What is the sign of Δm 2 23 ? Do  s and  s oscillate with the same probability? What is absolute mass scale? Are they Majorana or Dirac particles? i.e.,  =  ? How many species?? Mass Quasi-Degenerate Mass Hierarchical Mass Normal Mass Inverted

92. Imperial College London Morgan O. Wascko Oscillation Summary LSND Δm 2 ~ 1eV 2  ~ 2° Atmospheric oscillations Δm 2 ~ 10 -3 eV 2  ~ 45° Solar oscillations Δm 2 ~ 10 -5 eV 2  ~ 32° Problem: That's too many Δm2 regions! Should find: Δm 2 12 + Δm 2 23 = Δm 2 13 Δ m 2 13 Δ m 2 12 Δ m 2 23 10 -5 + 10 -3  1

93. Imperial College London Morgan O. Wascko Many null result SBL accelerator neutrino experiments Positive result: LSND Experiment at LANL Beam:   decay at rest L/E ~ 1m/MeV L ~ 30m 20< E   < 53 MeV   →  e ? Appearance search Clean detection signal Inverse β decay Accelerator Neutrinos →→ ➥→ee➥→ee

94. Imperial College London Morgan O. Wascko LSND  e Background Estimates Estimate  e /    e BkgdLSND Excess LSND Paper 0.086% 19.5+-3.987.9+-22.4+-6.0 Zhemchugov Poster1 0.071% 16.1+-3.291.3+-22.4+-5.6 Zhemchugov Poster2 0.092% 20.9+-4.286.5+-22.4+-6.2 Zhemchugov Seminar 0.119% 27.0+-5.480.4+-22.4+-7.1 All  e background estimates assume a 20% error. Note that the  e /   ratio determines the background! LSND Paper: A. Aguilar et al., Phys. Rev. D 64, 112007 (2001); (uses MCNP) Zhemchugov Poster1: FLUKA  e /   ratio presented at the ICHEP 2010 Conference, Paris Zhemchugov Poster2: GEANT4  e /   ratio presented at the ICHEP 2010 Conference, Paris Zhemchugov Seminar: FLUKA  e /   ratio presented at CERN on September 14, 2010 Although the analysis of Zhemchugov et al. is not fully understood or endorsed, their  e /   ratios agree reasonably well with the published LSND results. Note that LSND measures the correct rate of   p →  + n interactions, which confirms the p - production and background estimates. Note also, that FLUKA & GEANT4 are not as reliable as MCNP at 800 MeV!

95. Imperial College London Morgan O. Wascko data MC MC Tuning Good data/MC agreement Basic PMT hit distributions showing details of optical model Aggregate PMT hit distributions showing gross detector behaviour

96. Imperial College London Morgan O. Wascko Detector Stability

97. Imperial College London Morgan O. Wascko “Analysis region” defined to be 475-1250 MeV Signal efficiency higher at low energy Backgrounds higher there too... Signal and background Predicted ν e efficiency

98. Imperial College London Morgan O. Wascko 475-1250 MeV  e (  decay)  e (  decay) 132 94 Radiative Δ NC  0 Dirt Other 20 62 17 33 Total358 Signal 163 Signal and background Predicted ν e energy distribution

99. Imperial College London Morgan O. Wascko Tuning CCQE MC Q 2 distribution fit to tune empirical parameters of nuclear model ( 12 C) good data-MC agreement in variables not used in tuning! Data  2 /ndf = 4.7 / 13

100. Imperial College London Morgan O. Wascko Systematic Errors constraint? Neutrino flux predictions meson production cross sections ✓ meson secondary interactions ✓ focussing horn current target and horn system alignment Neutrino interaction cross sections nuclear model ✓ rates and kinematics for relevant processes ✓ resonance width and branching fractions ✓ Detector modelling optical model of light propagation ✓ PMT charge and time response ✓ electronics & DAQ model ✓ neutrino interactions in dirt surrounding detector ✓

101. Imperial College London Morgan O. Wascko Events summary (constrained syst + stat uncertainty) E ν QE range (MeV) ν mode ν mode (3.386e20 POT)(6.486e20 POT) 200-300 300-475 200-475 475-1250 Excess Deficit Data24232 MC ± sys+stat (constr.)27.2 ± 7.4186.8 ± 26.0 Excess (σ)-3.2 ± 7.4 (-0.4σ)45.2 ± 26.0 (1.7σ) Data37312 MC ± sys+stat (constr.)34.3 ± 7.3228.3 ± 24.5 Excess (σ)2.7 ± 7.3 (0.4σ)83.7 ± 24.5 (3.4σ) Data61544 MC ± sys+stat (constr.)61.5 ± 11.7415.2 ± 43.4 Excess (σ)-0.5 ± 11.7 (-0.04σ)128.8 ± 43.4 (3.0σ) Data61408 MC ± sys+stat (constr.)57.8 ± 10.0385.9 ± 35.7 Excess (σ)3.2 ± 10.0 (0.3σ)22.1 ± 35.7 (0.6σ)

102. Imperial College London Morgan O. Wascko 102 Fitting down to 200 MeV Dashed pink and blue lines show fit result down to 475 MeV, solid lines extend fit down to 200 MeV Only nubar are assumed to oscillate No inclusion of low-E expectation Large backgrounds in 200-475 means the region carries little weight in the fit Get same result if 12 low E bkg events are added to low E region.

103. Imperial College London Morgan O. Wascko 103 LSND ν μ result LSND Found 40 events on a bkg of 21 Excluded null at just > 2σ MB 90CL well within LSND 95CL Conclusion...some tension but it will be < 2σ

104. Imperial College London Morgan O. Wascko 104 Comparing  results Nubar beam contains a 20% WS background, fits (above 475 MeV) assume only nubar are allowed to oscillate BG composition fairly similar, BG constraints re-extracted Consistent at 1.5σ level  mode 5.6e20 POT  mode 3.4e20 POT

105. Imperial College London Morgan O. Wascko Background  v e Evis distributions for 5.66E20 POT Gamma-ray Intrinsic  ν e

106. Imperial College London Morgan O. Wascko Other  e kinematic distributions for 5.66E20 POT Visible Lepton Energy Reconstructed Lepton Angle wrt Beam χ2/NDF = 23.8/13 shape only χ2/NDF = 13.6/11 shape only