Latest Oscillation Results from NOvA Diana Patricia Mendez - - PowerPoint PPT Presentation

1

Latest Oscillation Results from NOvA

Diana Patricia Mendez

University of Sussex

• n behalf of the NOvA Collaboration

Moriond EW 2019 Diana Mendez 2

νe νµ ντ νe µ τ νµ ντ νe νµ ντ νe µ τ νµ ντ

ν3 ν2 ν1 ν2 ν1 ν3

m2

Δm2

32

Δm2

21

Δm2

32

Δm2

21

Normal Hierarchy Inverted Hierarchy

Mixing angles Mass squared difference

Neutrino oscillations

νe νμ ντ = UPMNS ν1 ν2 ν3 θ12, θ13, θ23

δCP

Δm2

21, Δm2 32

CP phase

P(νμ → νμ) ≃ 1 − sin2(2θ23)sin2( Δm2

32L

4E )

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.2 0.4 0.6 0.8 1

Oscillation/No oscillation

Ratio NOvA Simulation

20 40 60 80

Events / 0.1 GeV

Prediction No oscillation

Muon neutrino disappearance

Moriond EW 2019 Diana Mendez 3

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.2 0.4 0.6 0.8 1

Oscillation/No oscillation

Ratio NOvA Simulation

20 40 60 80

Events / 0.1 GeV

Prediction No oscillation

Muon neutrino disappearance

P(νμ → νμ) ≃ 1 − sin2(2θ23)sin2( Δm2

32L

4E )

Amplitude

Moriond EW 2019 Diana Mendez 4

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.2 0.4 0.6 0.8 1

Oscillation/No oscillation

Ratio NOvA Simulation

20 40 60 80

Events / 0.1 GeV

Prediction No oscillation

Muon neutrino disappearance

Moriond EW 2019 Diana Mendez 5

Frequency

P(νμ → νμ) ≃ 1 − sin2(2θ23)sin2( Δm2

32L

4E )

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.2 0.4 0.6 0.8 1

Oscillation/No oscillation

Ratio NOvA Simulation

20 40 60 80

Events / 0.1 GeV

Prediction No oscillation

Muon neutrino disappearance

θ23 = π/4 θ23 < π/4 θ23 > π/4

P(νμ → νμ) ≃ 1 − sin2(2θ23)sin2( Δm2

32L

4E )

Lower octant Maximal mixing Upper octant

Moriond EW 2019 Diana Mendez 6

Electron neutrino appearance

2 4 6 8

%

e

ν →

µ

ν P

2 4 6 8

%

e

ν →

µ

ν P

NOvA: L=810 km, E=2.0 GeV

Neutrino-antineutrino No difference in vacuum

No matter effects

Moriond EW 2019 Diana Mendez 7

2 4 6 8

%

e

ν →

µ

ν P

2 4 6 8

%

e

ν →

µ

ν P

NOvA: L=810 km, E=2.0 GeV

=0 δ /2 π = δ π = δ /2 π =3 δ

Electron neutrino appearance

δCP

CP phase

Neutrino-antineutrino Opposite effects

Moriond EW 2019 Diana Mendez 8

NOvA: L=810 km, E=2.0 GeV

2 4 6 8

%

e

ν →

µ

ν P

2 4 6 8

%

e

ν →

µ

ν P

=0 δ /2 π = δ π = δ /2 π =3 δ

NOvA: L=810 km, E=2.0 GeV

Electron neutrino appearance

Hierarchy

NH, IH

δCP

CP phase

Neutrino-antineutrino Opposite effects

Inverted Hierarchy Normal Hierarchy

Moriond EW 2019 Diana Mendez 9

NOvA: L=810 km, E=2.0 GeV

2 4 6 8

%

e

ν →

µ

ν P

2 4 6 8

%

e

ν →

µ

ν P

=0 δ /2 π = δ π = δ /2 π =3 δ

NOvA: L=810 km, E=2.0 GeV

Electron neutrino appearance

Hierarchy

NH, IH

δCP

CP phase Symmetry

θ23

Neutrino-antineutrino Opposite effects Similar effects

Normal Hierarchy Lower

• ctant

Upper

• ctant

Inverted Hierarchy

Moriond EW 2019 Diana Mendez 10

Neutrino oscillation experiment

• and disapperance
• and appearance

Two detectors separated by 810 km

• Near detector (underground): Beam

composition, unoscillated flux

• Far detector (surface): Oscillated spectra

MN WI IL MI

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

50 100 150 200

POT / 0.1 GeV

20

10 × Events / 8.03

3

10

Data Predicted Events

• syst. range

σ 1- CC

ν Wrong Sign Total Background Neutrino beam Area Normalised

NOvA Preliminary

Neutrino (antineutrino) source: Fermilab’s NuMI

• 14 mrad off-axis beam
• Narrow energy spectrum

Far Detector Near Detector

The NOvA Experiment

NuMI Off-axis Electron Neutrino Appearance

νμ ¯ νμ νe ¯ νe

Moriond EW 2019 Diana Mendez 11

The NOvA Detectors

4.1 m 60 m 3.9 x 6.6 cm 16 m 15.6 m

Near Detector

0.3 kton 214 layers

Beam direction x z y

Far Detector 14 kton 896 layers

Moriond EW 2019 Diana Mendez 12

Neutrino Beam

Target Focusing Horns 2 m

νµ(νµ)

120 GeV p+ from MI π+(-) π-(+)

νµ(νµ)

NuMI: Neutrinos at the Main Injector (Fermilab) Most powerful beam in the World. Constantly operating at ~725 kW

Moriond EW 2019 Diana Mendez 13

New for 2018 analysis

• First antineutrino data: Total analysis exposure

6.90x1020 (antineutrino) + 8.85x1020 (neutrino) POT

• Updated particle classifier
• Neutron uncertainty
• Cross section

¯ ν ν

Moriond EW 2019 Diana Mendez 14

q (ADC)

10 102

10

νμ

e

νe ν

p μ p p π

γ γ 1m 1m

π0

Event topology

Long straight track Short wide fuzzy shower Gap between vertex and shower

νμ CC νe CC NC

Moriond EW 2019 Diana Mendez 15

New

• Simpler architecture
• Updated simulation
• Separate training for

antineutrino and neutrino beams

Classifier

CVN Algorithm based on CNNs to classify interactions

Treats events as images to extracts features Multilayer classifier used between analyses

• A. Aurisano et. al, JINST 11 P0001 (2016)

Moriond EW 2019 Diana Mendez 16

20 − 20

)

• 3

10 × (

23

θ

2

Uncertainty in sin

Statistical Uncertainty Systematic Uncertainty Beam Flux Near-Far Differences Detector Response Normalization Muon Energy Scale Neutrino Cross Sections Detector Calibration Neutron Uncertainty

NOvA Preliminary

0.05 − 0.05

)

2

eV

• 3

10 × (

32 2

m Δ Uncertainty in

Statistical Uncertainty Systematic Uncertainty Beam Flux Near-Far Differences Normalization Detector Response Neutron Uncertainty Muon Energy Scale Neutrino Cross Sections Detector Calibration

NOvA Preliminary

0.5 − 0.5

π /

CP

δ Uncertainty in

Statistical Uncertainty Systematic Uncertainty Neutron Uncertainty Beam Flux Muon Energy Scale Normalization Detector Response Neutrino Cross Sections Detector Calibration Near-Far Differences

NOvA Preliminary

Systematic uncertainties

Largest

• Detector

Calibration

• Neutrino cross

sections

New

• Neutron

uncertainty

Limitations

• Statistics
• Data collection up to 2024
• Test beam program will improve

calibration and detector response

Moriond EW 2019 Diana Mendez 17

18

NOvA - FNAL E929

Run: 23630 / 39 Event: 240802 / -- UTC Mon Jul 25, 2016 04:23:7.211474672

218 220 222 224 226 228

sec) µ t (

1 10

2

10

hits

10

2

10

3

10

q (ADC)

1 10

2

10

hits

4000 4500 5000 5500 6000 400 600 800

x (cm)

4000 4500 5000 5500 6000

z (cm)

200 − 200

y (cm)

Disappearance νμ + ¯ νμ

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

2 4 6 8 10 12

Events / 0.1 GeV

FD Data Prediction

• syst. range

σ 1- CC

µ

ν Wrong Sign: Total bkg. Cosmic bkg.

Neutrino beam NOvA Preliminary All Quartiles

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

2 4 6 8

Events / 0.1 GeV

FD Data Prediction

• syst. range

σ 1- CC

µ

ν Wrong Sign: Total bkg. Cosmic bkg.

Antineutrino beam NOvA Preliminary All Quartiles

Beam Observation 113 65 Beam background 1.17 0.62 Cosmic background 2.07 0.46 Wrong sign signal 7.24 12.58 Total expected signal 121.04 50.44

νµ

¯ νµ

Disappearance νμ + ¯ νμ

Moriond EW 2019 Diana Mendez 19

1 2 3 4 5 1 2 3 4 Quartile 3 1 2 3 4 5 worst resolution Quartile 4 1 2 3 4 5 best resolution Quartile 1 FD Data Prediction

• syst. range

σ 1- CC

ν Wrong Sign: Total bkg. Cosmic bkg.

Neutrino beam NOvA Preliminary

Quartile 2

Reconstructed Energy (GeV) Events / 0.1 GeV

1 2 3 4 1 2 3 4 Quartile 3 1 2 3 4 5 worst resolution Quartile 4 1 2 3 4 best resolution Quartile 1 FD Data Prediction

• syst. range

σ 1- CC

ν Wrong Sign: Total bkg. Cosmic bkg.

Antineutrino beam NOvA Preliminary

Quartile 2

Reconstructed Energy (GeV) Events / 0.1 GeV

νµ

¯ νµ

Beam Observation 113 65 Beam background 1.17 0.62 Cosmic background 2.07 0.46 Wrong sign signal 7.24 12.58 Total expected signal 121.04 50.44

νµ

¯ νµ

Disappearance νμ + ¯ νμ

Events and extrapolation split by energy resolution

Moriond EW 2019 Diana Mendez 20

21

NOvA - FNAL E929

Run: 26110 / 49 Event: 3213 / -- UTC Sun May 7, 2017 04:41:20.910875840

218 220 222 224 226 228

sec) µ t (

1 10

2

10

hits

10

2

10

3

10

q (ADC)

1 10

2

10

hits

5000 5200 5400 5600 5800 6000 500 600 700 800

x (cm)

5000 5200 5400 5600 5800 6000

z (cm)

500 600 700

y (cm)

Appearance νe + ¯ νe

Beam Observation 58 (4.21) 18 (5.3) Beam nue background 6.85 2.57 Cosmic background 3.33 0.71 Wrong sign signal 0.66 1.13 Total prediction 59 (30-75) 15.9 (10-22)

Reconstructed Neutrino Energy (GeV)

5 10 15 20

POT-equiv

20

10 × Events / 8.85

NOvA Preliminary

FD data 2018 Best Fit Wrong Sign Bkg. Total Beam Bkg. Cosmic Bkg.

Low PID High PID

Core Peripheral

1 2 3 4 1 2 3 4 Neutrino beam

Reconstructed Neutrino Energy (GeV)

2 4 6 8 10 12

POT-equiv

20

10 × Events / 6.91

NOvA Preliminary

FD data 2018 Best Fit Wrong Sign Bkg. Total Beam Bkg. Cosmic Bkg.

Low PID High PID

Core Peripheral

1 2 3 4 1 2 3 4 Antineutrino beam

Appearance νe + ¯ νe

Strong evidence of appearance (>4 )

¯ νe σ

νe ¯ νe

Events separated by PID for purity

Moriond EW 2019 Diana Mendez 22

20 30 40 50 60 70 80

Total events - neutrino beam

5 10 15 20 25

Total events - antineutrino beam

NOvA Preliminary

= 0

CP

δ /2 π =

CP

δ π =

CP

δ /2 π = 3

CP

δ =0.082

13

θ 2

2

sin NOvA FD ) ν POT (

20

10 × 9.48 ) ν POT (

20

10 × 6.91

• bserved 18
• bserved 58

Appearance νe + ¯ νe

¯ νe νe

Moriond EW 2019 Diana Mendez 23

Joint result

appearance + disappearance

CP

δ

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2 π π 2 π 3 Feldman-Cousins σ 1 σ 2 σ 3 Best Fit

NOvA Preliminary

NH

CP

δ

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2 π π 2 π 3 π 2 Feldman-Cousins σ 1 σ 2 σ 3 Best Fit IH

CP

δ

1 2 3 4 5

) σ Significance (

2 π π 2 π 3 π 2

NOvA FD ν POT

20

10 × + 6.9 ν POT equiv

20

10 × 8.85

NOvA Preliminary

NH Lower octant NH Upper octant IH Lower octant IH Upper octant

Joint best fit with 15.75 x 1020 POT-equivalent

NH preferred by 1.8 Exclude in IH at 3

σ σ

δCP = π/2

δCP = 0.17π

Moriond EW 2019 Diana Mendez 24

0.4 0.5 0.6

23

θ

2

sin

2.2 2.4 2.6 2.8

)

2

eV

• 3

(10

32 2

m ∆

Feldman-Cousins σ 1 σ 2 σ 3 Best Fit

NOvA Preliminary

NH 0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2.8 − 2.6 − 2.4 −

)

2

eV

• 3

(10

32 2

m ∆

Feldman-Cousins σ 1 σ 2 σ 3 Best Fit IH 0.4 0.5 0.6 0.7

23

θ

2

sin

0.5 1 1.5 2 2.5 3

) σ Significance (

NOvA FD ν POT

20

10 × + 6.9 ν POT equiv

20

10 × 8.85 hierarchy Normal hierarchy Inverted

NOvA Preliminary

Joint result

appearance + disappearance

Joint best fit with 15.75 x 1020 POT-equivalent Δm232 = 2.51 (+0.12 -0.08)x10-3 eV2 sin2θ23 = 0.58 (+0.03 -0.03)

Disfavours maximal mixing at 1.8 Excludes LO at similar level

σ

Moriond EW 2019 Diana Mendez 25

• Appearance + disappearance fit favours UO and NO
• 90% C.L. compatible with other experiments

0.4 0.5 0.6

23

θ

2

sin

2.0 2.5 3.0

)

2

eV

• 3

(10

32 2

m Δ

Best fit

NOvA Preliminary Normal Hierarchy 90% CL NOvA MINOS+ 2018 T2K 2018 IceCube 2017 SK 2017

Joint result

appearance + disappearance

Moriond EW 2019 Diana Mendez 26

• Combine data assuming CPT

invariance

• Separate fit prefers maximal mixing

for neutrinos, and non-maximal for antineutrinos

• Results consistent at 4% level

νμ + ¯ νμ octant preference

Moriond EW 2019 Diana Mendez 27

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2.2 2.4 2.6 2.8 3.0

)

2

eV

• 3

(10

32 2

m Δ

NOvA NH 90% CL

• nly

µ

ν

µ

ν +

µ

ν

• nly

µ

ν

µ

ν 2017 No Feldman-Cousins NOvA Preliminary

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2.2 2.4 2.6 2.8 3.0

)

2

eV

• 3

(10

32 2

m Δ

NOvA NH 90% CL

• nly

µ

ν

µ

ν +

µ

ν

• nly

µ

ν

µ

ν 2017 No Feldman-Cousins NOvA Preliminary

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2 4 6

)

2

χ ∆ Significance (

NOvA 2018 Normal Hierarchy

µ

ν +

µ

ν Inverted Hierarchy

µ

ν +

µ

ν

• Favour upper (lower) octant for

normal (inverted) ordering

• Matter effects introduce asymmetry

in the maximal disappearance point

• ctant preference

νμ + ¯ νμ

Moriond EW 2019 Diana Mendez 28

• Combine data assuming CPT

invariance

• Separate fit prefers maximal mixing

for neutrinos, and non-maximal for antineutrinos

• Results consistent at 4% level

Prospects

2018 2020 2022 2024

Year

1 2 3 4 5

2

χ Δ = σ Significance

=0.082

13

θ 2

2

, sin

2

eV

• 3

10 × |=2.5

32 2

m Δ =0.4-0.6, |

23

θ

2

sin

projected beam exposure improvements 2018 analysis techniques and

NOvA Simulation

Hierarchy resolution /2 π =3

CP

δ NH π =

CP

δ NH =0

CP

δ NH /2 π =

CP

δ NH

• Analysis improvements and accelerator for up to 900 kW
• 2 sensitivity to CP violation for favourable parameters by 2024
• Possible hierarchy determination at 3 in 2020
• Joint NOvA-T2K analysis efforts ramping up

σ σ

Moriond EW 2019 Diana Mendez 29

Prospects

2018 2020 2022 2024

Year

1 2 3 4 5

2

χ Δ = σ Significance

=0.082

13

θ 2

2

, sin

2

eV

• 3

10 × |=2.5

32 2

m Δ =0.4-0.6, |

23

θ

2

sin

projected beam exposure improvements 2018 analysis techniques and

NOvA Simulation

Hierarchy resolution /2 π =3

CP

δ NH π =

CP

δ NH =0

CP

δ NH /2 π =

CP

δ NH

• Switched to neutrino mode at the end of February
• Cumulated extra ~5.6 x 1020 POT in antineutrino

mode between April 2018 and February 2019

Summer top up result with additional antineutrino data

Moriond EW 2019 Diana Mendez 30

Summary

• Released first results including antineutrino beam data

beam 6.9x1020 POT + beam 8.9x1020 POT

• Strong evidence of appearance
• Joint appearance + disappearance result prefers non-maximal

mixing and normal hierarchy

• Possible 3 hierarchy sensitivity by 2020 thanks to accelerator

performance, extended running and analysis improvements with test beam

• Top up results this summer with additional 5 x 1020 POT in

antineutrino mode ¯ νe ν ¯ ν σ

Moriond EW 2019 Diana Mendez 31

Backup

Moriond EW 2019 Diana Mendez 33

Neutrino oscillations

να

ν ν ν

Source Detector

νβ

Neutrino oscillations

Quantum interference Source Detector

νµ νµ νe νµ νe

( νe νμ) = ( cosθ sinθ

• sinθ

cosθ) ( ν1 ν2)

θ

ν2 ν1 νe νµ

Moriond EW 2019 Diana Mendez 34

0.5 1 1.5 2 2.5 3

Neutrino energy (GeV)

0.2 0.4 0.6 0.8 1

Probability

)

e

ν →

µ

ν P( )

µ

ν →

µ

ν P(

P(νμ → νμ) ≃ 1 − [sin2(2θ13)sin2(θ23) + cos4(θ13)sin2(2θ23)]sin2( Δm2

32L

4E ) P(νμ → νe) ≃ Patm + Psol + 2 PatmPsol(cosΔ32cosδCP ∓ sinΔ32sinδCP)

Neutrino oscillations

Patm = sin(θ23)sin(2θ13)Δ31 sin(Δ31 − aL) Δ31 − aL

Disappearance Appearance

θ13, θ23

δCP

Δm2

23

Psol = cos(θ23)sin(2θ12) sin(aL) aL Δ21

Moriond EW 2019 Diana Mendez 35

Moriond EW 2019 Diana Mendez 36

Detector Calibration

Calibration window

Select

• Cosmic muons stopping inside of

the detectors

• Hits in the Bethe-Bloch MIP region

for an estimate of dE/dx Correct

• Threshold, shadowing and

attenuation effects per cell Convert

• Photoelectrons to GeV

Tricell hits

• Reliable estimate of the path length
• Down going particle (y-view)
• Decreases chance of noise (x-view)

Moriond EW 2019 Diana Mendez 37

0.6 0.8 1 1.2 1.4

Y View

W

500 − 500

Cell

100 200 300

Y View

Far Detector

0.6 0.8 1 1.2 1.4

Y View

W

200 − 100 − 100 200

Cell

20 40 60 80

Y View

Near Detector

Threshold & Shadowing Correction Factor

Threshold, shadowing and attenuation effects are corrected per cell

Attenuation Fit per cell

Relative Calibration

Distance from center (cm)

500 − 500

Mean PE / cm

20 40 60

NOvA Preliminary

FD cosmic data - plane 49 (vertical), cell 91

Data

• Atten. Fit

Full Fit

Distance from center (cm)

200 − 100 − 100 200

Mean PE / cm

20 40 60

NOvA Preliminary

ND cosmic data - plane 48 (horizontal), cell 81

Data

• Atten. Fit

Full Fit

Moriond EW 2019 Diana Mendez 38

Energy Scale

PECorrhit ∗ ✓ MeV/cm PECorr/cm ◆ = MeVhit

Absolute Calibration

The detector response is expected to be independent of the position and we convert the corrected response to energy units

PECorr/cm

/

MeV/cm

( )=

MeV/cm Mean simulated energy deposition Mean detector response Energy deposited per hit

Moriond EW 2019 Diana Mendez 39

Cosmic estimation

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.05 0.1 0.15 0.2

Events / 0.1 GeV

total: 2.07 Cosmic prediction

fhc

All Quantiles

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.01 0.02 0.03 0.04 0.05

Events / 0.1 GeV

total: 0.46 Cosmic prediction

rhc

All Quantiles

Moriond EW 2019 Diana Mendez 40

Cosmic estimation

0.01 0.02 0.03 0.04

best resolution Quantile 1 Cosmic prediction Quantile 1: 0.04 Quantile 2: 0.08 Quantile 3: 0.08 Quantile 4: 0.25

rhc

Quantile 2 Cosmic prediction Quantile 1: 0.04 Quantile 2: 0.08 Quantile 3: 0.08 Quantile 4: 0.25

rhc

1 2 3 4 5 0.01 0.02 0.03 0.04

Quantile 3 Cosmic prediction Quantile 1: 0.04 Quantile 2: 0.08 Quantile 3: 0.08 Quantile 4: 0.25

rhc

1 2 3 4 5

worst resolution Quantile 4 Cosmic prediction Quantile 1: 0.04 Quantile 2: 0.08 Quantile 3: 0.08 Quantile 4: 0.25

rhc

Reconstructed Energy (GeV) Events / 0.1 GeV

0.02 0.04 0.06 0.08 0.1 0.12

best resolution Quantile 1 Cosmic prediction Quantile 1: 0.61 Quantile 2: 0.20 Quantile 3: 0.17 Quantile 4: 1.09

fhc

Quantile 2 Cosmic prediction Quantile 1: 0.61 Quantile 2: 0.20 Quantile 3: 0.17 Quantile 4: 1.09

fhc

1 2 3 4 5 0.02 0.04 0.06 0.08 0.1 0.12

Quantile 3 Cosmic prediction Quantile 1: 0.61 Quantile 2: 0.20 Quantile 3: 0.17 Quantile 4: 1.09

fhc

1 2 3 4 5

worst resolution Quantile 4 Cosmic prediction Quantile 1: 0.61 Quantile 2: 0.20 Quantile 3: 0.17 Quantile 4: 1.09

fhc

Reconstructed Energy (GeV) Events / 0.1 GeV

Moriond EW 2019 Diana Mendez 41

NOvA - FNAL E929 Run: 21887 / 54 Event: 793352 / -- UTC Wed Jan 6, 2016 18:12:32.044525452

218 220 222 224 226 228

sec) µ t (

1 10

10

hits

10

10

10

q (ADC)

1 10

10

hits

1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 400 − 200 −

x (cm)

1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600

z (cm)

200 400 600

y (cm)

Neutrino Beam FD top view FD side view

NOvA - FNAL E929 Run: 29344 / 28 Event: 8227 / -- UTC Sun Apr 1, 2018 17:16:32.318265728

218 220 222 224 226 228

sec) µ t (

1 10

10

hits

10

10

10

q (ADC)

1 10

10

hits

500 1000 1500 2000 800 − 600 − 400 −

x (cm)

500 1000 1500 2000

z (cm)

200

y (cm)

Antineutrino FD top FD side view

Moriond EW 2019 Diana Mendez 42

Classifier

CVN Algorithm based on CNNs to classify interactions

Treats events as images to extracts features Multilayer classifier used between analyses

0.8 0.85 0.9 0.95 1

CVN e

0.5 1 1.5 2 2.5

POT

20

10 × Events / 8.03

3

10

NOvA Preliminary

Neutrino Mode

CC

µ

ν CC

µ

ν CC

e

ν CC

e

ν NC

ND data Total MC

0.2 0.4 0.6 0.8 1

CVN NuMu score

0.5 1 1.5

3

10 ×

POT

20

10 × Events / 8.03

3

10

Data Simulation

• syst. range

σ 1- CC

µ

ν Wrong Sign: Total Background Area Normalised

NOvA Preliminary Neutrino beam

Moriond EW 2019 Diana Mendez 43

0.8 0.85 0.9 0.95 1

CVN e

0.2 0.4 0.6

POT

20

10 × Events / 3.1

3

10

NOvA Preliminary

Antineutrino Mode

CC

µ

ν CC

µ

ν CC

e

ν CC

e

ν NC

ND data Total MC

Classifier

CVN Algorithm based on CNNs to classify interactions

Treats events as images to extracts features Multilayer classifier used between analyses

0.2 0.4 0.6 0.8 1

CVN NuMu score

100 200 300

POT

20

10 × Events / 3.10

3

10

Data Simulation

• syst. range

σ 1- CC

µ

ν Wrong Sign: Total Background Area Normalised

NOvA Preliminary Antrineutrino beam

Moriond EW 2019 Diana Mendez 44

Cross-section

0.1 0.2 0.3 0.4 0.5 0.6

20 40 60 80 100 120 140 160

Events

3

10

Neutrino Beam

NOvA ND Data Default GENIE QE Weights RES & DIS Weights Add Tuned MEC 0.1 0.2 0.3 0.4 0.5 0.6 (GeV)

had

Visible E 0.6 0.7 0.8 0.9 1 1.1 MC / Data 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

10 20 30 40 50 60 70

Events

3

10

Antineutrino Beam

NOvA ND Data Default GENIE QE Weights RES & DIS Weights Add Tuned MEC 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 (GeV)

had

Visible E 0.6 0.8 1 1.2 1.4 MC / Data

External theory input

• Valencia RPA model of nuclear

charge screening to applied to QE

• Same RPA applied to resonance

NOvA ND data

• 10% increase in non-resonant

inelastic scattering at high W

• Add MEC interactions, retuning

in momentum transfer space

Moriond EW 2019 Diana Mendez 45

Interaction model - MEC uncertainty

Repeat running for uncertainty

• Shape and rate
• Shifts GENIE model to make it more

QE or RES

• Extract MEC model

0.1 0.2 0.3 0.4 0.5 0.6

50 100 150

Events

3

10

NOvA ND data tune ν + ν 2018 NOvA MINERvA MEC σ NOvA - MEC shape -1 σ NOvA - MEC shape +1 Non-MEC

0.1 0.2 0.3 0.4 0.5 0.6

(GeV)

had

Visible E

0.8 0.9 1 1.1 1.2

MC / data

Neutrino Beam NOvA Preliminary

0.1 0.2 0.3 0.4

20 40 60 80

Events

3

10

NOvA ND data tune ν + ν 2018 NOvA MINERvA MEC σ NOvA - MEC shape -1 σ NOvA - MEC shape +1 Non-MEC

0.1 0.2 0.3 0.4

(GeV)

had

Visible E

0.6 0.8 1 1.2 1.4

MC / data

Antineutrino Beam NOvA Preliminary

• MINERvA has ran the same procedure

with different MEC and Valencia model

• They agree with NOvA’s data as well

as our own +1 sigma error

Moriond EW 2019 Diana Mendez 46

Discrepancies with enriched neutron-like prong samples Scale true deposited energy of some low-energy neutrons Neutrons more prominent in antineutrino than in neutrino interactions

Neutron response systematic

0.1 0.2 0.3

Reconstructed prong energy (GeV)

0.8 1 1.2

Data / MC

NOvA Preliminary

500 1000 1500 2000 2500 3000

Events

NOvA ND Data Charged pion Muon Neutron Photon Proton NOvA ND Data Charged pion Muon Neutron Photon Proton

NOvA Preliminary NOvA Preliminary Antineutrino beam

0.1 0.2 0.3

Reconstructed prong energy (GeV)

0.8 1 1.2

Ratio to nom.

NOvA Preliminary

500 1000 1500 2000 2500 3000

Events

NOvA ND Data Nominal simulation Shifted neutron response NOvA ND Data Nominal simulation Shifted neutron response

NOvA Preliminary NOvA Preliminary Antineutrino beam

1% (0.5%) shift of mean reconstructed antineutrino (neutrino) energy

Moriond EW 2019 Diana Mendez 47

Discrepancies with enriched neutron-like prong samples Scale true deposited energy of some low-energy neutrons Neutrons more prominent in antineutrino than in neutrino interactions

Neutron response systematic

1% (0.5%) shift of mean reconstructed antineutrino (neutrino) energy

1 − 0.5 − 0.5

(Reco - True)/True A.U. (Area normalized) FD MC ) σ 1 + NeutronSyst ( Nominal ) σ 1 − NeutronSyst (

NOvA Simulation

Neutrino Beam

1 − 0.5 − 0.5

(Reco - True)/True A.U. (Area normalized) FD MC ) σ 1 + NeutronSyst ( Nominal ) σ 1 − NeutronSyst (

NOvA Simulation

Antineutrino Beam

Moriond EW 2019 Diana Mendez 48

49

Reconstructed neutrino energy

Zoom in detector Top view Side view

Reconstructed muon energy with track length Hadronic energy with calorimetry

Enu = Emu + Ehad

Beam direction

Energy resolution

Mean resolution

• Muon energy = 3.5%
• Hadronic energy = 40%
• Neutrino energy = 9%

Neutrino energy resolution = E_{Had}/E_{nu} Hadronic fraction = 0.25 —> Uncertainty = 10.0% Eμ = 1.5 ± 0.06 GeV Ehad = 0.5 ± 0.2 GeV Hadronic fraction = 0.70 —> Uncertainty = 30% Eμ = 0.5 ± 0.018 GeV Ehad = 1.5 ± 0.6 GeV

Moriond EW 2019 Diana Mendez 50

ν

/ E

had.

E

0.2 0.4 0.6 0.8 1

Events

200 400 600 800 1000 1200 1400 Limits [0,25)% [25,50)% [50,75)% [75,100]%

NOvA Simulation

1 2 3 4 5

Reconstructed Neutrino Energy (GeV)

0.2 0.4 0.6 0.8 1

ν

/ E

had

E

100 200 300 400

3

10 ×

Neutrino beam Antineutrino beam NOvA Simulation

Quartile 4 Quartile 3 Quartile 2 Quartile 1

Energy resolution

Separate well resolved energies by quantiles of hadronic energy fraction: For each bin of reconstructed neutrino energy

• Make a distribution of hadronic energy fraction
• Find the boundaries to divide the sample evenly

Neutrino energy resolution = Ehad/Enu

Energy resolution Muon = 3.5% Hadronic = 40%

Moriond EW 2019 Diana Mendez 51

Extrapolation

• 1. Reconstructed-to-true energy matrix takes data/MC differences.
• 2. Detector flux and acceptance differences taken in ratio.
• 3. Survival probability at specific parameters
• 4. Convert true to reconstructed energy with matrix.

1 2 3 4 5 2 4 6 8 20 40 60 80 1 2 3 4 5 1 2 3 4 1 2 3 4 5 1 2 3 4 1 12 2 0 1

ND Events/1 GeV

5

10 True Energy (GeV) True Energy (GeV) ND Reco Energy (GeV) FD Reco Energy (GeV) FD Events/1 GeV ND Events

5

10 FD Events F/N Ratio

• 3

10 )

µ

ν →

µ

ν P(

ND data Base Simulation Data-Driven Prediction

Disappearance

Moriond EW 2019 Diana Mendez 52

Extrapolation

• 1. Reconstructed-to-true energy matrix takes data/MC differences.
• 2. Detector flux and acceptance differences taken in ratio.
• 3. Survival probability at specific parameters
• 4. Convert true to reconstructed energy with matrix.

1 2 3 4 5 2 4 6 8 5 10 15 1 2 3 4 5 1 2 3 4 5 10 15 20 25 1 2 3 4 1 12 2 0 0.1

ND Events/1 GeV

5

10 True Energy (GeV) True Energy (GeV) ND Reco Energy (GeV) FD Analysis Bin FD Events ND Events

5

10 FD Events F/N Ratio

• 3

10 )

e

ν →

µ

ν P(

ND data Base Simulation Data-Driven Prediction

Appearance

Moriond EW 2019 Diana Mendez 53

1 2 3 4 5

Reconstructed neutrino energy (GeV)

5 10 15

Events / (0.1 GeV)

Nominal with FD/ND in ND σ + 1 in FD σ + 1 in ND σ

• 1

in FD σ

• 1

RES A

M

NOvA Simulation

Neutrino beam 1 2 3 4 5

Reconstructed neutrino energy (GeV)

0.8 0.9 1 1.1 1.2

Ratio to nominal MC

• extrap. ND shift

σ 1 ± FD shift σ 1 ±

RES A

M

NOvA Simulation

Neutrino beam 1 2 3 4 5

Reconstructed neutrino energy (GeV)

10 − 5 − 5 10

Residual difference (%)

shift in FD minus ND σ +1 shift in FD minus ND σ

• 1

RES A

M

NOvA Simulation

Neutrino beam

Moriond EW 2019 Diana Mendez 54

1 2 3 4 5

Reconstructed neutrino energy (GeV)

5 10 15

Events / (0.1 GeV)

Nominal with FD/ND in ND σ + 1 in FD σ + 1 in ND σ

• 1

in FD σ

• 1

RES A

M

NOvA Simulation

Antineutrino beam 1 2 3 4 5

Reconstructed neutrino energy (GeV)

0.8 0.9 1 1.1 1.2

Ratio to nominal MC

• extrap. ND shift

σ 1 ± FD shift σ 1 ±

RES A

M

NOvA Simulation

Antineutrino beam 1 2 3 4 5

Reconstructed neutrino energy (GeV)

10 − 5 − 5 10

Residual difference (%)

shift in FD minus ND σ +1 shift in FD minus ND σ

• 1

RES A

M

NOvA Simulation

Antineutrino beam

Moriond EW 2019 Diana Mendez 55

0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2.2 2.4 2.6 2.8 3.0

)

2

eV

• 3

(10

32 2

m Δ

NOvA NH 90% CL 2018

µ

ν +

µ

ν

No Feldman-Cousins NOvA Preliminary

Best fit 0.3 0.4 0.5 0.6 0.7

23

θ

2

sin

2.8 − 2.6 − 2.4 − 2.2 − 2.0 −

)

2

eV

• 3

(10

32 2

m Δ

NOvA IH 90% CL 2018

µ

ν +

µ

ν

No Feldman-Cousins NOvA Preliminary

Muon Neutrino + Antineutrino Disappearance

Moriond EW 2019 Diana Mendez 56

Matter effect

density = 2.8 gr/cm3

0.5 0.505 0.51 0.515 0.52

Neutrino energy (GeV)

0.001 − 0.0005 − 0.0005 0.001

)

µ

ν →

µ

ν P(

CP

δ π 0.0 π 0.5 π 1.0 π 1.5

numubar__rho2.8_energy1.625

0.5 0.505 0.51 0.515 0.52

23

θ

2

sin

0.001 − 0.0005 − 0.0005 0.001

)

µ

ν →

µ

ν P(

CP

δ π 0.0 π 0.5 π 1.0 π 1.5

numu__rho2.8_energy1.625

NO = Solid line IO = Dashed line

Matter effects introduce a difference in neutrino and antineutrino oscillations, thus a small asymmetry in the maximal disappearance point , which can mimic CPT violation.

Moriond EW 2019 Diana Mendez 57

Matter effect

0.5 1 1.5 2 2.5 3

Neutrino energy (GeV)

0.05 0.1

)

µ

ν ) - P(

µ

ν P(

23

θ

2

sin 0.42 0.46 0.50 0.52 0.54 0.58 0.62

signdif__nh_rho2.8

density = 2.8 gr/cm3 NH

Matter effects introduce a difference in neutrino and antineutrino oscillations, thus a small asymmetry in the maximal disappearance point , which can mimic CPT violation.

Moriond EW 2019 Diana Mendez 58

Matter effect

0.5 1 1.5 2 2.5 3

Neutrino energy (GeV)

0.05 0.1

)

µ

ν ) - P(

µ

ν P(

23

θ

2

sin 0.42 0.46 0.50 0.52 0.54 0.58 0.62

signdif__nh_rho10.0

density = 10. gr/cm3 NH

The neutrino-antineutrino asymmetry enhances with larger earth crust density

Moriond EW 2019 Diana Mendez 59

Matter effect

0.5 1 1.5 2 2.5 3

Neutrino energy (GeV)

0.05 0.1

)

µ

ν ) - P(

µ

ν P(

23

θ

2

sin 0.42 0.46 0.50 0.52 0.54 0.58 0.62

signdif__ih_rho2.8

density = 2.8 gr/cm3 IH

Moriond EW 2019 Diana Mendez 60

Matter effect

0.5 1 1.5 2 2.5 3

Neutrino energy (GeV)

0.05 0.1

)

µ

ν ) - P(

µ

ν P(

23

θ

2

sin 0.42 0.46 0.50 0.52 0.54 0.58 0.62

signdif__ih_rho10.0

density = 10. gr/cm3 IH

Moriond EW 2019 Diana Mendez 61

62

• Maximal disappearance in

vacuum with theta13 = 0 is at sin2theta23=0.5

• In vacuum, with reactor

value of theta13, it is at 0.511, not at maximal mixing

• In matter:

For neutrinos and NH is at 0.514 For antineutrinos and NH it is at 0.508 For IH it flips from the above values

Prospects

• Running through 2024
• Analysis improvements and accelerator for up to 900 kW
• 2 sensitivity to CP violation for favourable parameters by 2024
• Possible hierarchy determination at 3 in 2020

σ σ

0.5 1 1.5 2

π /

CP

δ

1 2 3

) σ Significance of CP violation(

• Stat. Only

Inverted Normal

=1.00

23

θ 2

2

=0.082, sin

13

θ 2

2

sin POT

20

10 × 36 ν +

20

10 × 36 ν

NOvA Simulation

Moriond EW 2019 Diana Mendez 63