Download presentation
Presentation is loading. Please wait.
Published byJocelin Casey Modified over 8 years ago
1
MIND Systematic Errors EuroNu Meeting, RAL 18 January 2010 Paul Soler
2
2 EuroNu, RAL, 18 January 2011 o A comprehensive analysis for MIND at a Neutrino Factory has been carried out (A. Laing, PhD thesis) o Features of analysis: –Nuance: event generation –GEANT4: detector simulation in 1 T dipole field –Smearing position resolution –Simple detector digitisation and clustering –Smearing of hadronic energy: –Smearing hadron angular resolution: –Full pattern recognition and reconstruction (Kalman filter and Cellular Automaton) –Likelihood function analysis _ –Generation of full detector response matrices for and o Estimation of systematic errors –Hadronic energy and hadronic angular resolution –Relative proportion of QEL, RES and DIS interactions Introduction Full description analysis in talk by Anselmo
3
3 EuroNu, RAL, 18 January 2011 Baseline detector at a Neutrino Factory: MIND o Golden channel signature: “wrong-sign” muons o Far detector: 100 kton at 2000-4000 km o Magic detector: 50 kton at 7500 km o Appearance of “wrong-sign” muons o Segmentation: 3 cm Fe + 2 cm scintillator o 1 T magnetic field iron (3 cm) + scintillators (2cm) beam 15 m B=1 T 50-100kT 50-100 m IDS-NF baseline for 25 GeV NuFact: MIND
4
4 EuroNu, RAL, 18 January 2011 MIND: new analysis oImprovements MIND analysis with full GEANT4 simulation oAdd quasi-elastics and resonance production (NUANCE): Non DIS processes dominate at low energies and should improve efficiency at low energies Benchmark of NUANCE with data
5
5 EuroNu, RAL, 18 January 2011 MIND: new analysis o New analysis with Nuance and GEANT4: CC background Anti-numu as numuNumu as anti-numu
6
6 EuroNu, RAL, 18 January 2011 MIND: new analysis o New analysis with Nuance and GEANT4: NC background NC as numuNC as anti-numu
7
7 EuroNu, RAL, 18 January 2011 MIND: new analysis New analysis with Nuance and GEANT4: e background nue as numuantinue as anti-numu
8
8 EuroNu, RAL, 18 January 2011 MIND: new analysis o New analysis with Nuance and GEANT4: signal efficiencies Numu efficiencyAnti-numu efficiency
9
9 EuroNu, RAL, 18 January 2011 MIND: new analysis o New analysis with Nuance and GEANT4: better efficiencies at low energies, due to addition of QES and RES events
10
10 EuroNu, RAL, 18 January 2011 MIND: new analysis New plots: signal right-sign nutau as numuantinutau as anti-numu
11
11 EuroNu, RAL, 18 January 2011 MIND: new analysis New plots: signal wrong-sign nutau as numuantinutau as anti-numu
12
12 EuroNu, RAL, 18 January 2011 MIND: new analysis o Difference in numu and anti-numu efficiencies: effectively only because of Bjorken y distribution (inelasticity) of neutrinos and antineutrinos -CC
13
13 EuroNu, RAL, 18 January 2011 MIND: new analysis o Systematic errors: hadronic energy & hadron angular resolution ~1% effect
14
14 EuroNu, RAL, 18 January 2011 MIND: new analysis Systematic errors: ratio of QES/DIS, 1 /DIS, “Other”/DIS ~1% effect
15
15 EuroNu, RAL, 18 January 2011 Flux Near Detector at Neutrino Factory oNear Detector sees a line source (600 m long decay straight) oFar Detector sees a point source 1 km 100 m 2500 km ND FD Need to take into account these differences for flux measurement 100 m 1 km
16
16 EuroNu, RAL, 18 January 2011 Flux extrapolation o Extrapolation near-to-far at Neutrino Factory : –Matrix method similar to MINOS: –Fit FD spectrum to predicted spectrum from ND: Laing, PS Fit improves at 3 level Comparison fitted 13 and with true values
17
17 EuroNu, RAL, 18 January 2011 MIND: new analysis sensitivities with NuTS 13 sensitivities: sin 2 2 13 1.2 ×10 -5 1.2 ×10 -3 sin 2 2 13
18
18 EuroNu, RAL, 18 January 2011 MIND: new analysis sensitivities with NuTS CP sensitivities: sin 2 2 13 1.2 ×10 -5 1.2 ×10 -3 sin 2 2 13
19
19 EuroNu, RAL, 18 January 2011 MIND: new analysis sensitivities with NuTS o Mass hierarchy sensitivities: sin 2 2 13 1.2 ×10 -5 1.2 ×10 -3 sin 2 2 13
20
20 EuroNu, RAL, 18 January 2011 How to treat systematics? oGlobes systematics: –We have response (migration) matrices –Should we calculate error in each term response matrix? –Systematics take into account current knowledge –What is expected error in cross-sections by the time NuFACT? –Analysis not optimised for signal –No systematic errors done for analysis cuts yet –Near Detector flux normalisation known to ~1% (R. Tsenov) –Matrix extrapolation method can reduce systematic errors oR&D effort: –Develop realistic B-field –Measure charge mis-ID rate in test beams – can this be used to improve errors?
21
21 EuroNu, RAL, 18 January 2011 Future directions oAnalysis and simulations: –Improve digitisation and optimise geometry –Add toroidal field – how well can we know this? –Move to GENIE for neutrino interactions (nearly done) –Improve hadronic reconstruction: energy and angular resolution –Final sensitivity plots and systematic errors oR&D effort: –Prototype detectors with SiPM and extruded scintillator –Develop realistic B-field –Measure charge mis-ID rate –Develop CERN test beam for neutrino detector R&D – European AIDA proposal to make H8 into low E beam
22
22 EuroNu, RAL, 18 January 2011 Systematics due to knowledge of B-field? Preliminary field map for 14 m plates from ANSYS simulation (Bob Wands, FNAL): ~1 T - 2.2 T with 92 kA-turn Add toroidal field: need to go to MINOS-like geometry to avoid low B-field corners 14 m Need to know how accurately we can determine B-field inside iron
23
23 EuroNu, RAL, 18 January 2011 Conclusions New MIND analysis with Nuance, Geant4, full pattern recognition and reconstruction provides 3 discovery of 13, and mass hierarchy down to 13 ~0.25 o : lower threshold o Preliminary systematic errors include hadronic energy (6%) and angular resolution (50%): affects efficiency by <1% above 5 GeV o Relative weight of QEL, RES and DIS reactions by their known cross-section errors affects efficiency by ~1% below 5 GeV o Near Detector flux normalisation known to ~1% o Matrix extrapolation method can reduce systematic errors on fits by 30% (probably due to correlated errors between near and far?) o Need to understand how Globes will include systematics and what do they need from us
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.