Cosmic Background suppression for a NuMI electron-neutrino cross- - - PowerPoint PPT Presentation

Cosmic Background suppression for a NuMI electron-neutrino cross- section measurement in MicroBooNE

Colton Hill on behalf of the MicroBooNE Collaboration New Perspectives 5 June 2017

Colton Hill - University of Manchester

Motivation - Why do we care about νe?

• Two key motivations:
• LSND/MiniBooNE νe-like excess - MiniBooNE

was limited by the π0 backgrounds.

• We have limited data on νe interactions

around 1 GeV.

• νe cross section also critical for DUNE.

2

MiniBooNE “Low Energy Excess”

Colton Hill - University of Manchester

νe Cross Section Measurements

• The few measurements of νe CC cross-section

were made by the Gargamelle, Minerva and T2K experiments.

• MicroBooNE is also well-equipped to measure

the νe CC cross section given its excellent spatial resolution and calorimetric capabilities.

• MicroBooNE’s energy range is around 1 GeV.

3

arXiv:1407.7389 arXiv:1509.05729

T2K Minerva

Colton Hill - University of Manchester

MicroBooNE Detector

• MicroBooNE sits along the BNB and

about 8° off-axis from the center of Neutrino Main Injector (NuMI).

• TPC is “slow” so PMTs behind wire

planes are used for triggering.

4

MicroBooNE

Wilson Hall

NuMI Beam

BNB Beam NuMI Beam

BNB Beam

Colton Hill - University of Manchester

The GENIE generated νe angles tell us from where along the NuMI beamline most of our neutrinos originate.

5

NuMI Beam at MicroBooNE

Phi [degrees]

Angular Distribution of NuMI Nue

z x Φ = 0 θ

BNB Beam

z Φ y z θ = 0

BNB Beam

Work In-progress

Target Hall

Absorber

Phi [Degrees] Theta [Degrees]

NuMI Beam

MicroBooNE

NuMI Absorber

BNB Beam

Colton Hill - University of Manchester

Neutrino Energy [GeV]

1 2 3 4 5 6

POT

20

/ 6x10

2

) / 50 MeV / cm ν ( Φ

7

10

8

10

9

10

10

10

11

10

12

10

µ

ν

µ

ν

e

ν

e

ν

at MicroBooNE Off-axis NuMI Flux Neutrino Mode

NuMI Beam at MicroBooNE

• Measurement from NuMI gives result

independent of BNB “low-energy excess”.

• νe fraction is larger for NuMI (~5%) vs BNB

(~0.6%).

6

GENIE Parent NuMI Nue Flux at MicroBooNE (Neutrino Mode) K+ 57.1% K0L 41.2% μ+ 1.6% π+ 0.01%

BNB Flux

Colton Hill - University of Manchester

7

NuMI DATA: RUN 10811, EVENT 2549. APRIL 9, 2017.

Sample Topology

Colton Hill - University of Manchester

True Deposited Shower Energy [GeV]

0.5 1 1.5 2 2.5 3

Events

200 400 600 800 1000 1200 1400

QE Res DIS Coh

True Deposited Shower Energy

• For a cross section measurement

we want to measure the shower energy.

• QE events deposit on-average the

most energy per shower.

• Other mode’s average deposited

shower energy peaks below ~200 MeV.

8

Work In-progress MC Truth - GENIE v2.10.10

Colton Hill - University of Manchester

Cosmic Rejection and νe Selection

• MicroBooNE sits on the surface - as

such cosmic rays are a significant background.

• For every neutrino interaction we expect

around 300 cosmic only events.

• Some of these will produce showers,

looking like lone νe interactions.

• Selection cuts focus on rejecting

cosmics:

• Optical Cuts
• Topological Cuts

9

Colton Hill - University of Manchester

Optical Filter

• The optical filter checks for two

things:

• A flash within the beam spill (the

time we expect neutrinos to arrive).

• One flash of at-least 50

photoelectrons.

• This removes cosmic events which

are “out of time”.

10

5 10 15 20 Time with respect to the NuMI Trigger Time [µs] 0.9 1.0 1.1 1.2 1.3 1.4 1.5 Fractional Flash Count per 0.5 µs with respect to Cosmic Background

Measured Cosmic Rate (Beam-Off) NuMI Trigger Data (Beam-On) [4.83E18 POT]

“Beam Spill”

Colton Hill - University of Manchester

Fiducial Volume Cut

• As expected for the cosmic background,

their vertices are concentrated in the top 30 cm of the detector.

• Fiducial Volume: 10 cm from all sides, 30 cm

from top.

11

5 10 15 20 25 30 35 40

Reco Nue-like Vtx ZY

z [cm]

200 400 600 800 1000

y [cm]

100 − 50 − 50 100

Reco Nue-like Vtx ZY

Cosmic - Nue Candidate Reconstructed Vertex ZY

1 2 3 4 5 6 7

Reco Nue-like Vtx ZY

z [cm]

200 400 600 800 1000

y [cm]

100 − 50 − 50 100

Reco Nue-like Vtx ZY

Nues - Nue Candidate Reconstructed Vertex ZY

Cosmics Simulation Work In-progress νe CC Simulation Work In-progress

Colton Hill - University of Manchester

Vertex-to-Flash

• We can combine optical and topological information to

place a cut:

• 2D distance between νe shower-vertex and largest

flash center - cut at 100 cm.

• Edge in νe spectrum at 100 cm results from PMT

coverage granularity, and uncorrelated cosmic events w.r.t. beam flash position.

• This removes events where the shower is reconstructed

far away from the largest flash in the event.

12

Nue shwr vtx to Flash Center [cm]

20 40 60 80 100 120 140 160 180 200

Counts

5 10 15 20 25 30

Cosmic Simulation

Work In-progress

Nue shwr vtx to Flash Center [cm]

20 40 60 80 100 120 140 160 180 200

Counts

100 200 300 400 500

νe CC Simulation

Work In-progress

νe Shower-vertex to Flash Center [cm]

Colton Hill - University of Manchester

νe-like Topology

• This analysis makes use of automated

reconstruction algorithms (Pandora).

• Classification of a νe-like topology

requires:

• At minimum one reconstructed shower.
• A shower object associated to a

neutrino vertex candidate with the greatest amount of TPC activity.

13

arXiv:1506.05348

Pandora - MICROBOONE-NOTE-1015-PUB

Colton Hill - University of Manchester

Cosmic Proximity Cut

• This cut attempts to remove showers

which result from a cosmic track (gamma / delta rays)

• If a νe shower vertex is within a 5 cm

radius cylinder, the event is cut.

14

2 4 6 8 10 12 14 16

Reco Nue-like Shwr Vtx to Reco Cosmic-like

Distance to nearest cosmic track [cm]

20 40 60 80 100 120

y [cm]

100 − 50 − 50 100

Reco Nue-like Shwr Vtx to Reco Cosmic-like

Cosmic Simulation Work In-progress

Colton Hill - University of Manchester

Conclusion and Summary

• We demonstrate that the open cosmic

background can be reduced by 4 orders

• f magnitude.
• With these cuts we retain an efficiency of
• ver 60%.
• A number of important backgrounds will

soon be included in the analysis (such as NC and muon neutrinos).

• A fully-developed νe selection will be

useful for further investigating the LSND/ MiniBooNE anomaly and future measurements on DUNE.

15

In Progress

Colton Hill - University of Manchester

Backup Slides

16

Colton Hill - University of Manchester

T2K Differential Cross Section

17

arXiv:1407.7389