NOvA Muon Neutrino and Antineutrino Disappearance Results 2018 - PowerPoint PPT Presentation

FERMILAB-SLIDES-18-091-LBNF-ND This document was prepared by [NOvA Collaboration] using the resources of the Fermi National Accelerator Laboratory (Fermilab), a U.S. Department of Energy, Office of Science, HEP User Facility. Fermilab is managed by Fermi Research Alliance, LLC (FRA), acting under Contract No. DE-AC02-07CH11359. NOvA Muon Neutrino and Antineutrino Disappearance Results 2018 Dmitrii Torbunov University of Minnesota June 19, 2018 1/12

Introduction ◮ NOvA uses NuMI muon (anti)neutrino beam. Thank you Fermilab for this tremendous beam. ◮ The muon beam travels from the Near Detector at Fermilab to the Far Detector in northern Minnesota. ◮ As muons travel, some fraction of ν µ oscillates into ν e and ν τ . ◮ The muon neutrino disappearance analysis aims to determine ( θ 23 , ∆ m 2 32 ) oscillation parameters, by measuring the number of survived muon neutrinos at the Far Detector. 2/12

Numu Analysis, Idea ◮ Muon neutrino oscillations no oscillations produce a dip between 1-2 with oscillations GeV range. 60 Events / 0.1 GeV ◮ Dip depth depends on the mixing angle θ 23 . 40 ◮ Precise position of the dip sin 2 2 θ 23 depends on the ∆ m 2 32 . 20 Δ m 2 ◮ But, presence of systematic uncertainties requires extra 0 work to minimize their 0 1 2 3 4 5 Reconstructed Neutrino Energy (GeV) influence. 3/12

Numu Analysis 1, How to detect a muon neutrino? ◮ Muon neutrinos do not have electric charge, so they do not leave any tracks. ◮ We can only detect neutrinos if they happen to interact with the detector. ◮ The only such reaction, that allows us to identify a muon neutrino is a charged current interaction: Proton ν μ + n → μ + p Muon 4/12

Numu Analysis 2, Muon Neutrino selection 1 0 1000 2000 3000 4000 5000 6000 500 x (cm) 0 − 500 500 y (cm) 0 − 500 0 1000 2000 3000 4000 5000 6000 z (cm) NOvA - FNAL E929 3 4 hits 10 hits 10 2 3 Run: 18791 / 48 10 10 10 2 10 Event: 765587 / -- 10 1 1 UTC Fri Jan 30, 2015 0 100 200 300 400 500 2 3 10 10 10 07:19:18.516289184 µ t ( sec) q (ADC) Where is a muon neutrino here? 5/12

Numu Analysis 2, Muon Neutrino selection 2 ◮ We identify the ν µ CC event by the presence of the muon. ◮ First, we have a great computer vision algorithm CVN, which can identify particle types. ◮ But, we must also use another algorithm which looks at the physical properties of the tracks to distinguish muon tracks from pion tracks (which look similar). ◮ Finally, there are a lot of muons coming from the top – cosmics. To minimize cosmics background we: ◮ use small timing window around the beam pulse ◮ veto muons that come from the edge of the detector ◮ have an algorithm that rejects the remaining muons, by analyzing their kinematics 6/12

Numu Analysis 2, Muon Neutrino selection 3 0 1000 2000 3000 4000 5000 6000 500 x (cm) 0 − 500 500 y (cm) 0 − 500 0 1000 2000 3000 4000 5000 6000 z (cm) NOvA - FNAL E929 3 4 hits 10 hits 10 2 3 Run: 18791 / 48 10 10 10 2 10 Event: 765587 / -- 10 1 1 UTC Fri Jan 30, 2015 0 100 200 300 400 500 2 3 10 10 10 07:19:18.516289184 µ t ( sec) q (ADC) Here is a muon neutrino! 7/12

Numu Analysis 3, Energy reconstruction ◮ Since we are looking at the ν µ CC events of the form: Proton ν μ + n → μ + p Muon ◮ The energy of the muon neutrino is E ν µ = E µ + E had ◮ Energy of the muon E µ is reconstructed from the muon’s track length. ◮ Hadronic energy E had is reconstructed from the visible calorimetric energy. 8/12

Numu Analysis 4, Quartiles split ◮ After, we have reconstructed neutrino’s energies, we split our sample into 4 classes(also known at NOvA as ”quartiles”) ◮ The split is based on the fraction energy in the event, that comes from hadronic activity(i.e. not from muon) Quartile boundaries Quartile 4 for muon (anti)neutrinos Quartile 3 Quartile 2 Quartile 1 9/12

Numu Analysis 5, Exptrapolation Finally, we use the Near Detector data to construct the Far Detector predicted energy spectrum: ND Events/1 GeV 8 FD Events/1 GeV 80 ND data 6 60 Base Simulation 4 40 Data-Driven Prediction 2 20 5 10 0 0 0 1 2 3 4 5 True Energy (GeV) True Energy (GeV) 4 4 3 3 2 2 1 1 0 1 2 3 4 5 0 1 0 2 0 1 12 0 0 1 2 3 4 5 ν → ν P( ) ND Reco Energy (GeV) 5 -3 FD Events FD Reco Energy (GeV) 10 ND Events 10 F/N Ratio µ µ This extrapolation procedure helps us to minimize flux and cross-section uncertainties. 10/12

Numu Analysis 6, Data/Prediction comparison, Details Neutrino beam NOvA Preliminary Antineutrino beam NOvA Preliminary FD Data FD Data 12 8 Prediction Prediction All Quartiles All Quartiles σ σ 1- syst. range 1- syst. range 10 ν ν Wrong Sign: CC Wrong Sign: CC Events / 0.1 GeV Events / 0.1 GeV µ µ 6 Total bkg. Total bkg. 8 Cosmic bkg. Cosmic bkg. 6 4 4 2 2 0 0 0 1 2 3 4 5 0 1 2 3 4 5 Reconstructed Neutrino Energy (GeV) Reconstructed Neutrino Energy (GeV) Beam Neutrinos Antineutrinos 8 . 9 · 10 20 POT 6 . 9 · 10 20 POT Exposure Total Observed 113 65 Best fit prediction 121 50 Cosmic Bkgd. 2.1 0.5 Beam Bkgd. 1.2 0.6 Unoscillated 730 266 Prediction here is based on the best fit values for joint numu-disappearance and nue-appearance analysis 11/12

Comparison with other experiments NOvA Preliminary Normal Hierarchy 90% CL NOvA MINOS 2014 T2K 2017 IceCube 2017 3.0 SK 2017 ) 2 eV -3 (10 2.5 32 2 m ∆ Best fit 2.0 0.4 0.5 0.6 2 θ sin 23 Best fit: sin 2 θ 23 = 0 . 58 ± 0 . 03, ∆ m 2 32 = 2 . 51 +0 . 12 − 0 . 08 · 10 − 3 eV 2 Combined muon neutrino disappearance and electron neutrino appearance results 12/12

Conclusions ◮ We have the first results with neutrino and antineutrino data (sin 2 θ 23 = 0 . 58 ± 0 . 03, ∆ m 2 32 = 2 . 51 +0 . 12 − 0 . 08 · 10 − 3 eV 2 ). ◮ In the future we expect to refine our measurements of θ 23 and ∆ m 2 32 by accumulating higher exposure values. ◮ This will improve NOvA’s sensitivity to the mass hierarchy, θ 23 octant and CP violation. ◮ Please find the full NOvA ν µ disappearance and ν e appearance results in the next talk. 13/12

Backups 1/9

Data/Prediction comparison, Neutrinos ◮ We have observed 113 muon Neutrino beam NOvA Preliminary neutrino like events: FD Data 12 Prediction All Quartiles σ 1- syst. range Total Observed 113 10 ν Wrong Sign: CC Events / 0.1 GeV µ Best fit prediction 121 Total bkg. 8 Cosmic bkg. Cosmic Bkgd. 2.1 6 Beam Bkgd. 1.2 Unoscillated 730 4 2 ◮ Corresponding exposure 8 . 9 · 10 20 0 0 1 2 3 4 5 POT. Reconstructed Neutrino Energy (GeV) 2/9

Data/Prediction comparison, Antineutrinos ◮ We have observed 65 muon Antineutrino beam NOvA Preliminary antineutrino like events: FD Data 8 Prediction All Quartiles σ 1- syst. range Total Observed 65 ν Wrong Sign: CC Events / 0.1 GeV µ Best fit prediction 50 6 Total bkg. Cosmic bkg. Cosmic Bkgd. 0.5 Beam Bkgd. 0.6 4 Unoscillated 266 2 ◮ Corresponding exposure 6 . 9 · 10 20 0 0 1 2 3 4 5 POT. Reconstructed Neutrino Energy (GeV) 3/9

Numu Contours Below are the contours that show 90% confidence contours for the oscillation parameters θ 23 , ∆ m 2 32 No Feldman-Cousins NOvA Preliminary 3.0 NOvA NH 90% CL ν ν ν only + µ µ µ 2.8 ν ν only 2017 µ µ ) 2 eV -3 (10 2.6 32 2 m ∆ 2.4 2.2 0.3 0.4 0.5 0.6 0.7 θ 2 sin 23 Consistency with the combined fit oscillation parameters for the neutrino and antineutrino datasets is better than 4%. 4/9

Octant-Hierarchy Sensitivity Joint ν µ + ¯ ν µ fit prefers higher octant of θ 23 for the normal hierarchy and lower octant for the inverted hierarchy. No Feldman-Cousins NOvA Preliminary 8 ν NOvA NH only µ ν ν NOvA IH + µ µ ) 6 2 ν χ only µ ∆ Significance ( 4 2 0 0.3 0.4 0.5 0.6 0.7 θ 2 sin 23 5/9

Numu Analysis 4, Motivation for Quartile split Why do we split our sample in 4 parts? ◮ The NOvA’s sensitivity towards oscillation parameters depends on how well we reconstruct neutrino energy. ◮ It turns out, that we can reconstruct muon’s energy much better than the hadronic energy. ◮ So, by separating sample into 4 quartiles we can better exploit good energy reconstruction of the quartiles with low fraction of hadronic energy. 6/9