Cosmic Muon Induced EM Showers in NO v A Detector Nitin Yadav - - PowerPoint PPT Presentation
Cosmic Muon Induced EM Showers in NO v A Detector Nitin Yadav Indian Institute of Technology Guwahati 1 Supervisor: Peter Shanahan , Bipul Bhuyan. 2 2 In collaboration with Hongyue Duyang and Sanjib Mishra New Perspectives 2015 June
Nitin Yadav Indian Institute of Technology Guwahati
Supervisor: Peter Shanahan , Bipul Bhuyan. In collaboration with Hongyue Duyang and Sanjib Mishra Fermilab, University of South Carolina.
1 2 2 1 2 ϯ ϯ New Perspectives 2015 June 8 1
neutrino oscillation experiment, near detector at Fermi Lab and the far detector at Ash River with a baseline of 810 km.
detectors differ in size.
intensity 450 kW currently.
v𝑓 appearance and vμ disappearance mode.
14mrad
New Perspectives 2015 June 8 2
Far Detector 14kton, 15.6 m x 15.6 m x 59.8 m Near Detector 300ton, 4.2 m x 4.2 m x 15.8 m 100 m underground
To 1 APD pixel W D
typical charged particle path
L To 1 APD pixel W D
typical charged particle path
L New Perspectives 2015 June 8 3
New Perspectives 2015 June 8 To 1 APD pixel W D
typical charged particle path
L To 1 APD pixel W D
typical charged particle path
L
Fine-grained, low-Z, highly-active tracking calorimeter allows for differences between “tracky” muons, “showery” electrons, and “gappy” π0′s to be seen.
NOνA detectors are finely segmented (1 plane ~0.15 X0), which makes it well optimized for electromagnetic shower reconstruction.
4
p+
Gap from γ
New Perspectives 2015 June 8 5
p+
Gap from γ
EM shower, signal in v𝑓 appearance mode
New Perspectives 2015 June 8 6
New Perspectives 2015 June 8
Artificial Neural Network (ANN) to identify v𝑓 CC events.
shower-shape based likelihoods for particle hypotheses calculated from longitudinal and transverse dE/dx information.
LID : Longitudinal Identification We use data driven technique to benchmark PID algorithms and simulation of EM shower at NOvA
7
Cosmic-ray induced showers: Why Cosmic Brem Shower:
interaction in media.
Nova FD detector, 72kHz
the v𝑓 appearance mode.
samples of pure EM showers.
algorithm including:
shower reconstruction.
New Perspectives 2015 June 8
A 500μ sec cosmic trigger event display
8
New Perspectives 2015 June 8
energy deposition in planes along the muon track.
deposition in planes.
shower regions.
shower region.
Brem Shower
9
Event display of hits of a cosmic muon candidate with Electromagnetic (EM) Bremsstrahlung (Brem) Shower.
New Perspectives 2015 June 8 10
Event display of hits of the EM shower after the removal of hits associated with the muon track from NOvA simulations.
New Perspectives 2015 June 8 11
shower energy after
between Data and MC.
New Perspectives 2015 June 8
shower angle. Good agreement between Data and MC.
12
and MC comparison
shower energy after
between Data and MC.
shower energy in comparison with v𝑓 MC events. Brems are less energetic to v𝑓 events.
New Perspectives 2015 June 8
shower angle in comparison with v𝑓 MC events. Brems are more perpendicular to direction
beam than v𝑓 events.
13
New Perspectives 2015 June 8
2D reweighting matrix is constructed and used to reweight Brem shower energy and angle to v𝑓 CC events to make for these differences.
events and v𝑓 events comes from difference in energy and angle distributions.
14
electron identification ANN (LID) . Good agreement between data and MC. Most of the Brem are identified as v𝑓 like.
New Perspectives 2015 June 8
to v𝑓 events.
15
as v𝑓 like. But to benchmark the PID and simulations Brem should reasonably be similar to v𝑓 . We achieved this by reweighting.
New Perspectives 2015 June 8
and angle to v𝑓 events, Brems do look more like v𝑓 events. This convinces us that Brem can be used as data driven benchmark for testing PIDs and EM shower simulations at NOvA
16
function of vertex X , Y and Z direction in NOvA. Efficiencies are reasonably flat and data and MC agreement is well within 5 %. The rest of the difference will be considered as a source of systematic uncertainty.
New Perspectives 2015 June 8 17
Using Muon Removal algorithm we find and isolate EM Shower from cosmic data and MC. A good agreement between data and MC. v𝑓 reweight method has been developed to make cosmic EM showers resemble beam events. A data-driven technique to benchmark the particle identifications and simulations of EM showers using Brem sample. PID efficiencies as a function of positions across the detector are pretty uniform, indicating calibration effects are well controlled.
New Perspectives 2015 June 8 18
New Perspectives 2015 June 8 19
Event display of raw hits of a cosmic track candidate with Electromagnetic (EM) Bremsstrahlung (Brem) Shower from NOvA simulation.
New Perspectives 2015 June 8 20
Event display of hits of the EM shower after the removal of hits associated with the muon track from NOA simulation. What left are hits of Brem shower.
New Perspectives 2015 June 8 21
New Perspectives 2015 June 8 22