Charged Current Cross SectionMeasurements at T2K

David Hadley on behalf of the T2K collaborationNuFact 2013

The T2K Experiment

Goals

I discover νe appearance in aνµ beam

I make precise measurements ofνµ disappearance

I study neutrino-nucleusinteractions at Eν ∼ 1GeV

I A high intensity proton beam at J-PARC produces anarrow-band νµ beam with a peak energy of0.6GeVat the far detector,

I The Far detector is Super-Kamiokande, a 50ktonwater Cherenkov detector, located 2.5◦ off-axis and295km from the production point.

I Detectors in the ND280 complex at 280m are used todirectly measure the neutrino beam properties andneutrino-nucleus interaction cross sections.

The T2K Beam

I T2K has collected 6.6 × 1020 POT to date (Run 1-4).

I The analyses presented here are based on 2.7 × 1020 POT (Run 1-3@ ND280).

I Improved analyses under development will include Run 4.

The T2K Beam

I Narrow-band beam at off axis angle.

I Mostly νµ from pion decay.

I Beam has a peak energy ∼ 0.6GeV.

I Close to the quasi-elastic peak.I On-axis near detector INGRID monitors the neutrino beam.

I INGRID will also make measurements of neutrino cross sections.

(GeV)νE0 1 2 3 4 5 6 7 8 9 10

PO

T)

21

/50

MeV

/10

2F

lux

(/c

m

710

810

910

1010

1110

1210 at ND280µν

at ND280µ

ν

at ND280eν

at ND280eν

Off-axis Near Detector (ND280)

DownstreamECal

FGDsTPCs

PØD

UA1 Magnet YokeSMRD

PØD ECal

Barrel ECal

Solenoid Coil

Neutrino beam

x

y

z

TPC2 TPC3

Scintillator

µ-

p

TPC1

FGD1

TPC2 TPC3

Scintillator

TPC1

FGD1

Run #: 4200 Evt #: 24083 Time: Sun 2010-03-21 22:33:25 JST

µ-

FGD2

Scintillator

ND280 Tracking Detector

I The Fine Grained Detector(FGD1) consists of layers of10 × 10mm plastic scintillator barsreadout with Multi-Pixel PhotonCounters (MPPCs).

I FGDs provide target mass andvertex reconstruction.

I The Time Projection Chambers(TPCs) provided PID based ondE/dx in the Argon based gasand momentum measurementfrom track curvature in themagnetic field.

π0Detector

I see NCE talk by D. Ruterbories.T2K, Nucl. Instrum. Meth. A 659, 106 (2011)FGD, Nucl. Instrum. Meth. A 696, 1 (2012)TPC Nucl. Instrum. Meth. A 637, 25 (2011)

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

ND280 Event Selection

dz

Muon candidate

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

I Muon PID

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

I Muon PID

µ

e

π

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

I Muon PID

CC QE Selection (µ+ 0π)

I TPC track veto

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

I Muon PID

CC QE Selection (µ+ 0π)

I TPC track veto

I Michel electron veto

ND280 Event Selection

2 samples selected: a QE enhanced,and an non-QE sample.CC Inclusive Selection (µ+ X )

I 1 good-quality negative trackstarting within the FGD fiducialvolume.

I Upstream veto

I Muon PID

CC QE Selection (µ+ 0π)

I TPC track veto

I Michel electron veto

Non-QE selection

I any event that fails either ofthe final 2 cuts

Flux Integrated CC Inclusive Cross Section Measurement

(MeV/c)µ

p0 500 1000 1500 2000

nucl

eon M

eV/c

2cm

θ

dp d

cos

σ 2

d

0

5

10

15

­4210×

Data

Systematic Error

Statistical Error

NEUT

GENIE

< 0.84 µθ 0.00 < cos

(MeV/c)µ

p0 500 1000 1500 2000

nucl

eon M

eV/c

2cm

θ

dp d

cos

σ 2

d

0

5

10

15

­4210×

Data

Systematic Error

Statistical Error

NEUT

GENIE

< 0.90 µθ 0.84 < cos

(MeV/c)µ

p0 500 1000 1500 2000

nucl

eon M

eV/c

2cm

θ

dp d

cos

σ 2

d

0

5

10

15

­4210×

Data

Systematic Error

Statistical Error

NEUT

GENIE

< 0.94 µθ 0.90 < cos

(MeV/c)µ

p0 500 1000 1500 2000

nucl

eon M

eV/c

2cm

θ

dp d

cos

σ 2

d

0

5

10

15

­4210×

Data

Systematic Error

Statistical Error

NEUT

GENIE

< 1.00 µθ 0.94 < cos

I Combine the QE and non-QE samples.

I Unfold the reconstructed pµ − cos(θµ) distributions to estimate the truemuon kinematics.

I Measure double differential pµ − cos(θµ) distribution and total fluxintegrated cross section.

Phys. Rev. D 87, 092003 (2013)

Reconstructed Kinematics

)µθcos(0 0.2 0.4 0.6 0.8 1

even

ts /

0.1

0

500

1000

1500

2000

2500

3000

Nominal NEUT

background

data

[GeV]QE

ν

reconstructed E0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

even

ts /

0.2

GeV

0

200

400

600

800

1000

1200

1400

1600 Nominal NEUT

background

data

[MeV]µ

p0 500 1000 1500 2000

even

ts /

100 M

eV

0

200

400

600

800

1000 Nominal NEUT

background

data

]2 [GeVQE

2reconstructed Q0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

2ev

ents

/ 0

.2 G

eV0

1000

2000

3000

4000

5000Nominal NEUT

background

data

I CCQE efficiency=40%, purity=72%.

I Dominant background from CC resonant pion production.

I Eν and Q2 calculated from muon momentum assuming QE kinematics.

NEUT CCQE Model

I Smith-Moniz implementation CCQEI Dipole form factor (MQE

A = 1.2GeV)I Initial Nucleon state given by relativistic Fermi gas model

I FSI with semi-classical cascade model

I No additional contribution from multi-nucleon effects

Measurement of the CCQE Cross Section

=0.0

0θ

p=

0M

eV,

=0.8

4θ

p=

0M

eV,

=0.9

0θ

p=

0M

eV,

=0.9

4θ

p=

0M

eV,

=0.0

0θ

p=

40

0M

eV,

=0.8

4θ

p=

40

0M

eV,

=0.9

0θ

p=

40

0M

eV,

=0.9

4θ

p=

40

0M

eV,

=0.0

0θ

p=

50

0M

eV,

=0.8

4θ

p=

50

0M

eV,

=0.9

0θ

p=

50

0M

eV,

=0.9

4θ

p=

50

0M

eV,

=0.0

0θ

p=

70

0M

eV,

=0.8

4θ

p=

70

0M

eV,

=0.9

0θ

p=

70

0M

eV,

=0.9

4θ

p=

70

0M

eV,

=0.0

0θ

p=

90

0M

eV,

=0.8

4θ

p=

90

0M

eV,

=0.9

0θ

p=

90

0M

eV,

=0.9

4θ

p=

90

0M

eV,

even

ts

0

200

400

600

800

1000

1200

1400

1600 Nominal NEUT

Background

Data

The Log Likelihood Ratio is minimised,

−2lnλ(θ) = 2

pµ−cos(θµ)bins∑i=1

[Npredictedi (θ) − Nobserved

i

+Nobservedi ln

Nobservedi

Npredictedi (θ)

+ln

πd (d)πd (dnominal)

+lnπf (f)

πf (fnominal)

+lnπx (x)

πx (xnominal)

(1)which includes,

I standard Poisson statistical terms

I penalty terms for the systematics

Analysis Method

I Simulated template histograms were fit to the observed pµ − cos(θµ)distribution.

I The CCQE cross section was extracted by weighting 5 template histogramsin bins of Eν .

I Systematic uncertainties were accounted for by varying bin contents withnuisance parameters.

I A maximum likelihood fit was used to find the best fit parameters.

Systematic Uncertainties and Implementation

Flux

I Flux prediction based onmeasurements at NA61/SHINEand T2K proton beammeasurements.

I ∼ 10 − 15% uncertainty on νµflux.

Implementation

I MC prediction binned in pµ,cos(θµ), Eν and interactiontype.

I Reweight bin contents based onvalue of nuisance parameters.

I Multi-variate Gaussian priors onnuisance parameters.

Interaction Model Uncertainties

I Vary model parameters egMRES

A , pF etc

I Empirical parameters eg pionbackground normalisationuncertainties.

I Uncertainties from comparisonof NEUT generator with externaldata

Detector Uncertainties

I Estimate uncertainty on rate ofselection in pµ − cos(θµ) bins.

I Dominant uncertainties are fromuncertainties on the TPCmomentum measurement andout of FV backgrounds.

I ∼ 5 − 10% overall detectoruncertainty.

Measurement of the CCQE Cross Section

[GeV]ν

E

0 0.5 1 1.5 2 2.5

]2

cm­3

8(E

) [1

0σ

0.5

1

1.5

2

2.5 ND280 stat

ND280 stat+syst

NEUT MC

NEUT (binned)

A χ2 test comparing the fitted result with the nominal NEUT model,with MQE

A = 1.2GeV, gives a p-value of 17% indicating agreementbetween the data and the cross section model.

T2K Preliminary

Measurement of the CCQE Cross Section

[GeV]ν

E

0 0.5 1 1.5 2 2.5

]2

cm­3

8(E

) [1

0σ

0.5

1

1.5

2

2.5 ND280 stat+syst

NEUT MC

MiniBoone

NEUT (binned)

A χ2 test comparing the fitted result with the nominal NEUT model,with MQE

A = 1.2GeV, gives a p-value of 17% indicating agreementbetween the data and the cross section model.

T2K Preliminary

Measurement of the CCQE Cross Section

[GeV]ν

E

­110 1 10 210

]2

cm­3

8(E

) [1

0σ

0.5

1

1.5

2

2.5ND280 stat+syst

NEUT MC

MiniBoone

NOMAD

NEUT (binned)

A χ2 test comparing the fitted result with the nominal NEUT model,with MQE

A = 1.2GeV, gives a p-value of 17% indicating agreementbetween the data and the cross section model.

T2K Preliminary

MQEA -effective extraction

Fit pµ-cos(θµ)distribution with MQE

Aas a free parameter.Fit with 2 settings,

I MQEA shape-only

I MQEA normalisa-

tion+shape­norm RFG

QE effective

AM ­shape RFGQE effective

AM

Fit

val

ue

[GeV

]

0.6

0.8

1

1.2

1.4

1.6

1.8

2

stats only

stats+syst

GeV­0.18

+0.211.26 GeV

­0.22

+0.281.43

I Previous experiments have observed a discrepancy in the fittedvalues of MQE

A between Deuterium and heavier nuclei.

I A large effective MQEA is believed to account for nuclear effects not

included in the model.

I Both fit results are consistent with the value used in NEUT,MQE

A = 1.2GeV.

T2K Preliminary

New CC Analysis

Additional TPC+FGD Tracks

I Highest momentum negative track is selected as the muon candidate.

I Tag pion/electrons/protons by applying PID to additional tracks.

FGD only tracks

I Use FGD only tracks to tag stopping particles.

I Tag pion from Michel tag or dE/dx in FGD.

New CC Analysis

I Cleaner sample of CC1π to better constrain the π background inCCQE cross section analysis.

T2K PreliminaryT2K Preliminary

T2K Preliminary

New CC Analysis

I Some example results from the fit to these samples for the oscillationanalysis.

I Significant improvement on the uncertainty on these modelparameters.

I Primarily from analysis improvements, not from increased statistics.

I We should expect similar improvements when these samples are usedfor cross section measurements.

I See Oscillation Systematics talk by A. Kaboth

Summary

I T2K’s first CC inclusive cross section publication was publishedearlier this year (Phys. Rev. D 87, 092003 (2013)).

I First Preliminary T2K CCQE Cross Section measurement presentedin this talk.

I Extraction of CCQE energy dependent cross section and modelparameters from fit to ND280 pµ − cos(θµ).

I Expect better performance in future cross section analysesI at least double statistics (current analysis only uses data up to

summer 2012)I new improved reconstructionI selection and analysis

Backup

ND280 CC Selection Efficiency and Purity

The T2K Beam

ND280 TPC PID

Flux Integrated CC Inclusive Cross Section Measurement

(GeV)νE0 0.5 1 1.5 2 2.5 3 3.5

/nucl

eon)

2 (

cmσ

0

0.5

1

1.5

2

2.5

3

3.5

PO

T)

21

/50M

eV/1

02

Flu

x (

/cm

0

0.5

1

1.5

2

2.5

3

3.5

SciBooNE data based on NEUT

BNL 7ft

NEUT prediction for SciBooNE

NEUT prediction for T2K

GENIE prediction for T2K

φ⟩ σ ⟨T2K

12 10×­38

10×

data

NEUT prediction

GENIE prediction

flux µν

I Combine the QE and non-QE samples.

I Unfold the reconstructed pµ − cos(θµ) distributions to estimate the truemuon kinematics.

I Measure double differential pµ − cos(θµ) distribution and total fluxintegrated cross section.