NOvA: Case for more protons Mark Messier Indiana University Fermilab Physics Advisory Committee 10 November 2016 1
Outline FY2016 Run Summary I. Beam and Detector status II. Physics results • ν μ charged-current disappearance • neutral-current disappearance • ν e charged-current appearance • First look at antineutrinos Looking ahead III. Neutrino oscillations post Neutrino 2016 IV. NOvA Physics milestones and FY17 run plan V. Looking further ahead
FY2016 NuMI / NOvA Protons 600 E18 Integrated Protons On Target Total delivery benefitted from extended run 500 4.75E20 Delivered } 96% 4.58E20 Recorded 400 0.63E20 recorded in antineutrino horn focus 300 DESIGN BASE DELIVERED 200 RECORDED 100 0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 Week • Last year saw routine delivery at 550 kW of proton power. FY16 Beam Performance • Peak of 700 kW demonstrated last year. 3
Far Detector Front End Board Count • 96% beam-weighted uptime in FY16 • 32 on-call incidents in 52 weeks • 10745/10752 FEBs (99.9%) operating within normal parameters • Average noise rate: 203 Hz / channel • Added capability to read out continuously for 60+ seconds in case of supernova trigger log10 (Front End Hit Rate / Hz) Near Detector • 99% beam-weighted uptime in FY16 - includes weekly scheduled downtimes to train on call experts. • 623/631 FEBs (98.7%) operating within normal parameters • Average noise rate: 78 Hz / channel NOvA FY16 Detector Operations 4
O ffl ine software and computing NOvA has aligned its o ffl ine computing model with SCD in a way we think is mutually beneficial • We get access SCD’s computing expertise and computing solutions • SCD gets their solutions “battle tested” by an operating and demanding experiment FNAL-supported packages, tools, and services in use by NOvA Simulation tools: GENIE and GEANT4 • ART analysis framework • Code management, build systems, distribution and documentation: SVN/SRT/CMake/UPS/Jenkins/CVMFS/Redmine • Grid computing and OSG: 24 million CPU hours in FY16: 75% FNAL / 25% o ff -site • Large data storage and cataloging (SAM): 30 million files, ~3+ PB added in FY16 •
3 FY14 FY15 FY16 3.26E20 POT 3.12E20 POT 4.75E20 POT Neutrino 2016 analysis data set 2.5 550 kW - Detector Construction 2 1e18 POT Per Day - 400 kW Recorded 330 kW 1.5 Delivered 290 kW 28-day average 28-day average 1 0.5 0 9/1/13 12/1/13 3/1/14 6/1/14 9/1/14 12/1/14 3/1/15 6/1/15 9/1/15 12/1/15 3/1/16 6/1/16 • Last year saw routine delivery at 550 kW of proton power. Beam Performance • Peak of 700 kW demonstrated last year. • Expect routine operations at 630 kW (700 kW-10%) in early calendar 2017 6
p ν μ μ p ν e e π γ } π 0 π 0 ν p γ 1m 7 1m (actual NOvA events) 10 2 3 10 10 q (ADC)
NOvA Preliminary 20 NO A 6.05 10 POT-equiv. ν × 120 Best fit prediction 100 Events / 0.25 GeV Unoscillated prediction Data 80 60 40 20 0 0 1 2 3 4 5 Reconstructed neutrino energy (GeV) NOvA Far detector 473 events expected before oscillations 78 events observed muon neutrino spectrum 8
NOvA Preliminary FD Data NC 3 Flavor Prediction 25 ν CC Background e ν CC Background Events / 0.25 GeV µ 20 Cosmic Background -3 2 2 ∆ m = 2.44x10 eV 32 15 θ ° θ ° = 8.5 , = 45 13 23 20 × 6.05 10 POT-equiv. 10 5 0 0 1 2 3 4 5 6 Calorimetric Energy (GeV) NC events are a way to count the total First look at Neutral-Current neutrino flux which should be una ff ected by standard oscillations. Events at Far Detector Expect: 61 events signal Measure: 72 events 9
NOvA Preliminary NOvA FD Data 25 Best-fit prediction: -2LL=41.6 ν μ Disappearance ∆ Best maximal: -2LL=48.0 ( =6.4) 20 Events 15 � = 2 . 67 ± 0 . 12 × 10 − 3 eV 2 � � � ∆ m 2 10 32 sin 2 θ 23 = 0 . 40 +0 . 03 − 0 . 02 (0 . 63 +0 . 02 − 0 . 03 ) 5 Excludes maximal 0 0 1 2 3 4 5 mixing at 2.5 σ Reconstructed neutrino energy (GeV) NOvA Preliminary 3.5 Normal Hierarchy, 90% CL NOvA 2016 T2K 2014 ) 2 MINOS 2014 eV 3 -3 (10 32 2 m ∆ 2.5 2 0.3 0.4 0.5 0.6 0.7 θ 2 sin 23 10
NOvA Preliminary Empirical model of Meson 20 2.85 10 P.O.T. × Exchange Current 60000 NOvA ND Data coded into GENIE inspired by MEC JLAB electron scattering measurements and guided by QE Events 40000 MINERvA data RES https://www.jlab.org/highlights/phys.html DIS 20000 Other 0 0 0.2 0.4 0.6 0.8 1 Visible E (GeV) had [1] P.A. Rodrigues et al. (MINERvA), PRL 116 (2016) 071802 (arXiv:1511.05944) [2] S. Dytman, based on J. W. Lightbody, J. S. In first analysis this was a leading systematic for OConnell, Comp. in Phys. 2 (1988) 57, and, mixing angle measurement: Contributed to a 4% T. Katori, AIP Conf. Proc. 1663, 030001 (2015) uncertainty on absolute energy scale [3] P.A. Rodrigues et al. (MINERvA), arXiv: 1601.01888 Now leading systematics are: Major update from first analysis to second 2.2% from muon energy scale analysis was an improvement in our 2.0% from calibration understanding of generator-level hadronic energy distribution 2.0% relative near/far energy scale 11
12 Proc. Int. Conf. High Energy Accelerators and Instrumentation, 1959
Borrow ideas from Computer Vision: Convolutional Neural Networks and Deep Learning ν e Event Identification in NOvA Application to NOvA events: A.~Aurisano et al., A Convolutional Neural Network Neutrino Event Classifier , JINST 11 , no. 09, P09001 (2016) 13
ν e Identification in NOvA : : FEATURE MAPS ELECTRON NEUTRINO 14
NOvA Preliminary Near Detector CVN Identifier on 6 10 ND data Total MC Near Detector Data POT Flux Uncert. 5 10 NC Beam CC ν 20 e CC ν 10 • CVN selects 73% of µ 4 10 × Events / 3.72 pre-selected electron- 3 10 neutrino charged current events 2 10 • Produces a 76% pure 0.0 0.2 0.4 0.6 0.8 1.0 CVN classifier NOvA Preliminary ν sample of electron- Near Detector e neutrino CC events ND data 1000 Total MC POT Flux Uncert. • Improved S/N NC 800 Beam CC ν 20 e CC ν equivalent to 30% more 10 µ × 600 Events / 3.72 exposure over techniques used in our 400 first analysis 200 0.75 0.80 0.85 0.90 0.95 1.00 15 CVN classifier ν e
NOvA Preliminary 0.75 < CVN < 0.87 0.87 < CVN < 0.95 0.95 < CVN < 1 20 NH FD Data Events / 0.5 GeV Bin Total Expected 15 Total Background Cosmic Background 20 × 6.05 10 POT equiv. 10 5 0 1 2 3 1 2 3 1 2 3 Reconstructed neutrino energy (GeV) NOvA Electron Observe 33 events at far detector Expect 8 events of background Neutrino Appearance ± 5% error on signal ± 10% on background 16
NOvA Electron Neutrino Appearance NOvA Simulation 50 Total events expected 2 θ NOvA FD sin =0.4-0.6 23 × 20 6.05 10 POT equiv. 40 30 20 10 NH IH 0 π π π π 0 3 2 δ 2 2 CP 17
NOvA Preliminary Electron Neutrino 0.7 Appearance 0.6 23 • Rule out lower octant, θ 0.5 2 inverted hierarchy at >3 σ sin 0.4 • Resolution of remaining 0.3 ambiguities requires 1 2 σ σ 3 NOvA Preliminary σ NH antineutrino running 0.2 π π 0 3 2 π π 0.7 • Recorded 0.5E20 POT in antineutrinos at end of run. 0.6 Will collect 3E20 POT in 23 0.5 θ neutrinos and 3E30 POT in 2 sin antineutrinos next year 0.4 • Current data sample is 1/6th 0.3 σ σ 1 2 of total planned running. σ 3 IH 0.2 π π π π 0 3 2 δ 2 2 CP 18
6E20 POT = 1 TDR Year Run plan: neutrinos antineutrinos Assumes 83% uptime 32 weeks of running Projected FY17 Beam Delivery 10% of time line to Switch Yard
NOvA Run Plan • Our ν μ data favors non-maximal θ 23 with 2.5 σ significance. Implications: 1. Opportunity to exclude maximal mixing with high confidence: Favors additional neutrino running. 2. Opportunity to resolve the θ 23 octant. Requires antineutrino running if θ 23 is in lower octant 3. If θ 23 is in lower octant antineutrino running is required to resolve hierarchy. • Our run plan seeks to take advantage of these opportunities and to clarify the situation as quickly as possible • FY17 : 3E20 POT additional neutrino data to clarify the ν μ situation. Is θ 23 really non-maximal? Can we push the significance beyond 3 σ ? • FY17 : 3E20POT in antineutrinos helps us achieve the optimal balance between neutrinos and antineutrinos for what appears to be the most likely scenario following Neutrino2016 (normal hierarchy, lower octant). 0.6E20POT collected in antineutrinos in FY16 optimizes our use of analysis time. • FY18: Run more antineutrinos 20
Still a wide range of possibilities open Interesting trend to see large-as- possible CP violation Preference for non-maximal mixing driven by NOvA’s recent results Preference for normal hierarchy and lower octant Upper octant and inverted hierarchy is a viable solution Δ 𝝍 2 =3.7 above normal hierarchy (suppressed in plot) Francesco Capozzi (Lisi et al.) Post Neutrino2016 “global picture” reporting at NOW2016 21
Recommend
More recommend
