NOvA: Case for more protons Mark Messier Indiana University Fermilab - - PowerPoint PPT Presentation

SLIDE 1

NOvA: Case for more protons

1

Mark Messier Indiana University

Fermilab Physics Advisory Committee 10 November 2016

SLIDE 2

Outline

FY2016 Run Summary

• I. Beam and Detector status
• II. Physics results
• νμ charged-current disappearance
• neutral-current disappearance
• νe charged-current appearance
• First look at antineutrinos

Looking ahead

• III. Neutrino oscillations post Neutrino 2016
• IV. NOvA Physics milestones and FY17 run plan
• V. Looking further ahead

SLIDE 3

FY16 Beam Performance

• Last year saw routine delivery at

550 kW of proton power.

• Peak of 700 kW demonstrated last

year.

3

100 200 300 400 500 600 4 8 12 16 20 24 28 32 36 40 44 48 52

E18 Integrated Protons On Target Week

FY2016 NuMI / NOvA Protons

DESIGN BASE DELIVERED RECORDED

4.75E20 Delivered 4.58E20 Recorded } 96%

0.63E20 recorded in antineutrino horn focus Total delivery benefitted from extended run

SLIDE 4

NOvA FY16 Detector Operations

4

Far Detector

• 96% beam-weighted uptime in FY16
• 32 on-call incidents in 52 weeks
• 10745/10752 FEBs (99.9%) operating within

normal parameters

• Average noise rate: 203 Hz / channel
• Added capability to read out continuously for

60+ seconds in case of supernova trigger Near Detector

• 99% beam-weighted uptime in FY16 -

includes weekly scheduled downtimes to train on call experts.

• 623/631 FEBs (98.7%) operating within

normal parameters

• Average noise rate: 78 Hz / channel

log10 (Front End Hit Rate / Hz)

Front End Board Count

SLIDE 5

Offline software and computing

NOvA has aligned its offline computing model with SCD in a way we think is mutually beneficial

• We get access SCD’s computing expertise and computing solutions
• SCD gets their solutions “battle tested” by an operating and demanding experiment
• Simulation tools: GENIE and GEANT4
• ART analysis framework
• Code management, build systems, distribution and documentation: SVN/SRT/CMake/UPS/Jenkins/CVMFS/Redmine
• Grid computing and OSG: 24 million CPU hours in FY16: 75% FNAL / 25% off-site
• Large data storage and cataloging (SAM): 30 million files, ~3+ PB added in FY16

FNAL-supported packages, tools, and services in use by NOvA

SLIDE 6

0.5 1 1.5 2 2.5 3 9/1/13 12/1/13 3/1/14 6/1/14 9/1/14 12/1/14 3/1/15 6/1/15 9/1/15 12/1/15 3/1/16 6/1/16

1e18 POT Per Day

Recorded Delivered 28-day average 28-day average

Beam Performance

• Last year saw routine delivery at 550 kW of

proton power.

• Peak of 700 kW demonstrated last year.
• Expect routine operations at 630 kW (700

kW-10%) in early calendar 2017

6

FY14

3.26E20 POT

FY15

3.12E20 POT

FY16

4.75E20 POT

550 kW -

• 400 kW

330 kW 290 kW

Detector Construction Neutrino 2016 analysis data set

SLIDE 7

q (ADC)

10 102

3

10

νμ

e

νe ν

p μ p p π

γ γ 1m 1m

π0

} π0

7

(actual NOvA events)

SLIDE 8

NOvA Far detector muon neutrino spectrum

473 events expected before oscillations 78 events observed

Reconstructed neutrino energy (GeV)

1 2 3 4 5

Events / 0.25 GeV

20 40 60 80 100 120

POT-equiv.

20

10 × A 6.05 ν NO Best fit prediction Unoscillated prediction Data

NOvA Preliminary

8

SLIDE 9

First look at Neutral-Current Events at Far Detector

NC events are a way to count the total neutrino flux which should be unaffected by standard oscillations. Expect: 61 events signal Measure: 72 events

Calorimetric Energy (GeV)

1 2 3 4 5 6

Events / 0.25 GeV

5 10 15 20 25

FD Data NC 3 Flavor Prediction CC Background

e

ν CC Background

µ

ν Cosmic Background

2

eV

• 3

= 2.44x10

32 2

m ∆ ° = 45

23

θ , ° = 8.5

13

θ POT-equiv.

20

10 × 6.05

NOvA Preliminary

9

SLIDE 10

NOvA νμ Disappearance

10

• ∆m2

32

• = 2.67 ± 0.12 × 10−3eV2

sin2 θ23 = 0.40+0.03

−0.02(0.63+0.02 −0.03)

23

θ

2

sin

0.3 0.4 0.5 0.6 0.7

)

2

eV

• 3

(10

32 2

m ∆

2 2.5 3 3.5

NOvA Preliminary

Normal Hierarchy, 90% CL NOvA 2016 T2K 2014 MINOS 2014

Reconstructed neutrino energy (GeV)

1 2 3 4 5 Events 5 10 15 20 25

FD Data Best-fit prediction: -2LL=41.6 =6.4) ∆ Best maximal: -2LL=48.0 (

NOvA Preliminary

Excludes maximal mixing at 2.5σ

SLIDE 11

(GeV)

had

Visible E

0.2 0.4 0.6 0.8 1

Events

20000 40000 60000

NOvA ND Data MEC QE RES DIS Other

P.O.T.

20

10 × 2.85

NOvA Preliminary

Empirical model of Meson Exchange Current coded into GENIE inspired by JLAB electron scattering measurements and guided by MINERvA data

[1] P.A. Rodrigues et al. (MINERvA), PRL 116 (2016) 071802 (arXiv:1511.05944)
 [2] S. Dytman, based on J. W. Lightbody, J. S. OConnell, Comp. in Phys. 2 (1988) 57, and,

• T. Katori, AIP Conf. Proc. 1663, 030001 (2015)

[3] P.A. Rodrigues et al. (MINERvA), arXiv: 1601.01888

11

Major update from first analysis to second analysis was an improvement in our understanding of generator-level hadronic energy distribution

https://www.jlab.org/highlights/phys.html

In first analysis this was a leading systematic for mixing angle measurement: Contributed to a 4% uncertainty on absolute energy scale Now leading systematics are: 2.2% from muon energy scale 2.0% from calibration 2.0% relative near/far energy scale

SLIDE 12

• Proc. Int. Conf. High Energy Accelerators and Instrumentation, 1959

12

SLIDE 13

νe Event Identification in NOvA

Borrow ideas from Computer Vision: Convolutional Neural Networks and Deep Learning Application to NOvA events: A.~Aurisano et al., A Convolutional Neural Network Neutrino Event Classifier, JINST 11, no. 09, P09001 (2016)

13

SLIDE 14

νe Identification in NOvA

14

FEATURE MAPS

: :

ELECTRON NEUTRINO

SLIDE 15

classifier

e

ν CVN

0.75 0.80 0.85 0.90 0.95 1.00

POT

20

10 × Events / 3.72

200 400 600 800 1000

ND data Total MC Flux Uncert. NC CC

e

ν Beam CC

µ

ν

NOvA Preliminary

CVN Identifier on Near Detector Data

classifier

e

ν CVN

0.0 0.2 0.4 0.6 0.8 1.0

POT

20

10 × Events / 3.72

2

10

3

10

4

10

5

10

6

10

ND data Total MC Flux Uncert. NC CC

e

ν Beam CC

µ

ν

NOvA Preliminary

• CVN selects 73% of

pre-selected electron- neutrino charged current events

• Produces a 76% pure

sample of electron- neutrino CC events

• Improved S/N

equivalent to 30% more exposure over techniques used in our first analysis

15

Near Detector Near Detector

SLIDE 16

NOvA Electron Neutrino Appearance

Observe 33 events at far detector Expect 8 events of background

±5% error on signal ±10% on background

Reconstructed neutrino energy (GeV) Events / 0.5 GeV Bin

5 10 15 20 1 2 3 1 2 3 1 2 3

0.75 < CVN < 0.87 0.87 < CVN < 0.95 0.95 < CVN < 1

NH

NOvA Preliminary

FD Data Total Expected Total Background Cosmic Background POT equiv.

20

10 × 6.05

16

SLIDE 17

NOvA Electron Neutrino Appearance

17

CP

δ Total events expected

10 20 30 40 50 2 π π 2 π 3 π 2 POT equiv.

20

10 × 6.05 NOvA FD =0.4-0.6

23

θ

2

sin NH IH

NOvA Simulation

SLIDE 18

Electron Neutrino Appearance

• Rule out lower octant,

inverted hierarchy at >3σ

• Resolution of remaining

ambiguities requires antineutrino running

• Recorded 0.5E20 POT in

antineutrinos at end of run. Will collect 3E20 POT in neutrinos and 3E30 POT in antineutrinos next year

• Current data sample is 1/6th
• f total planned running.

CP

δ

23

θ

2

sin

0.2 0.3 0.4 0.5 0.6 0.7 2 π π 2 π 3 π 2

NOvA Preliminary

σ 1 σ 2 σ 3 IH

23

θ

2

sin

0.2 0.3 0.4 0.5 0.6 0.7 π π π 3 π 2

NOvA Preliminary

σ 1 σ 2 σ 3 NH

18

SLIDE 19

Projected FY17 Beam Delivery

Assumes 83% uptime 32 weeks of running 10% of time line to Switch Yard

6E20 POT = 1 TDR Year neutrinos antineutrinos Run plan:

SLIDE 20

NOvA Run Plan

• Our νμ data favors non-maximal θ23 with 2.5σ significance. Implications:
• 1. Opportunity to exclude maximal mixing with high confidence: Favors additional

neutrino running.

• 2. Opportunity to resolve the θ23 octant. Requires antineutrino running if θ23 is in

lower octant

• 3. If θ23 is in lower octant antineutrino running is required to resolve hierarchy.
• Our run plan seeks to take advantage of these opportunities and to clarify the

situation as quickly as possible

• FY17: 3E20 POT additional neutrino data to clarify the νμ situation. Is θ23 really

non-maximal? Can we push the significance beyond 3σ?

• FY17: 3E20POT in antineutrinos helps us achieve the optimal balance between

neutrinos and antineutrinos for what appears to be the most likely scenario following Neutrino2016 (normal hierarchy, lower octant). 0.6E20POT collected in antineutrinos in FY16 optimizes our use of analysis time.

• FY18: Run more antineutrinos

20

SLIDE 21

Post Neutrino2016 “global picture”

Francesco Capozzi (Lisi et al.) reporting at NOW2016

Preference for normal hierarchy and lower octant Upper octant and inverted hierarchy is a viable solution Preference for non-maximal mixing driven by NOvA’s recent results Interesting trend to see large-as- possible CP violation Still a wide range of possibilities

• pen

Δ𝝍2=3.7 above normal hierarchy (suppressed in plot)

21

SLIDE 22

)

e

ν →

µ

ν P( 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 )

e

ν →

µ

ν P( 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

= 0 δ /2 π = δ π = δ /2 π = 3 δ A ν NO (810 km)

2

eV

• 3

10 × | = 2.4

32 2

m ∆ | ) = 0.96

23

θ (2

2

sin ) = 0.09

13

θ (2

2

sin

<0

32 2

m ∆ >0

32 2

m ∆

Taking that global fit at face value

θ

2 3 <

4 5

• θ

2 3

> 4 5

• 22

+2.6σ +1.9σ

best fit

n

• r

m a l m a s s

• r

d e r i n g i n v e r t e d m a s s

• r

d e r i n g

+2.0σ

SLIDE 23

)

e

ν →

µ

ν P( 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 )

e

ν →

µ

ν P( 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

= 0 δ /2 π = δ π = δ /2 π = 3 δ A ν NO (810 km)

2

eV

• 3

10 × | = 2.4

32 2

m ∆ | ) = 0.96

23

θ (2

2

sin ) = 0.09

13

θ (2

2

sin

<0

32 2

m ∆ >0

32 2

m ∆

Need for antineutrinos:

If we are in lower octant, normal hierarchy, antineutrinos are required.

θ

2 3 <

4 5

• θ

2 3

> 4 5

• 23

n

• r

m a l m a s s

• r

d e r i n g i n v e r t e d m a s s

• r

d e r i n g

If we are here in neutrinos we need antineutrinos to know if we are here or here

*not error bands, just bars for illustration

SLIDE 24

First look at 0.6E20 POT taken in antineutrinos

• We spent most of July 2016 in antineutrino mode
• Goal was to accumulate a sizable data in the near

detector to jump start analysis work for the longer antineutrino run to begin in mid 2017

• A few sample distributions for electron-neutrino

events (below) and muon neutrinos (right) show that while many things are in reasonable agreement, many things (mostly cross-sections) will need to be tuned up — in progress.

24

Hadronic shower invariant mass (GeV)

0.5 1 1.5 2 2.5

Events

3

10

5 10 15

POT

19

10 × ND, 5.97 Antineutrino mode

ND data Total MC QE

µ

ν Res

µ

ν DIS

µ

ν MEC

µ

ν Background

NOvA Preliminary

Reconstructed neutrino energy (GeV)

1 2 3 4 5

Events

3

10

0.05 0.1 0.15

POT

19

10 × ND, 5.97 Antineutrino mode

ND data Total MC QE

e

ν Res

e

ν DIS

e

ν MEC

e

ν Background

NOvA Preliminary

Reconstructed muon energy (GeV)

1 2 3 4 5

Events

3

10

1 2 3

POT

19

10 × ND, 5.97 Antineutrino mode

ND data Total MC QE

µ

ν Res

µ

ν DIS

µ

ν MEC

µ

ν Background

NOvA Preliminary

SLIDE 25

5 10 15 20 25

= 0.403 - Joint Fit

23

θ

2

/2 sin π IH rejection - Fake data: NH 3

2

χ ∆

)

20

10 × Total POT (

10 20 30 40 50

)

20

10 × RHC POT (

10 20 30 40 5 10 15 20 25 FHC = RHC Best combin.

= 0.403 - Joint Fit

23

θ

2

/2 sin π IH rejection - Fake data: NH 3

2

χ ∆

An optimal neutrino / antineutrino mix Normal hierarchy / lower octant

year-by-year a 50/50 allocation of protons is very close to optimal

50/50 mix

today

25

SLIDE 26

Projected NOvA physics reach

50/50 run plan for normal hierarchy lower octant

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 =0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403

23

θ

2

/2, sin π =3

CP

δ Normal

systematic uncertainty improvements 2016 analysis techniques with projected

µ

ν +

e

ν NOvA joint

• Max. mixing

Hierarchy Octant CPV

NOvA Simulation

Assumes uncertainties are reduced to

• νe: 2% signal / 5% background
• νμ: 2% muon energy scale, 3%

hadronic energy, very small NC backgrounds

36E20 POT total = TDR assumes 6E20 POT/yr

SLIDE 27

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 =0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403

23

θ

2

/2, sin π =3

CP

δ Normal

systematic uncertainty improvements 2016 analysis techniques with projected

µ

ν +

e

ν NOvA joint

• Max. mixing

Hierarchy Octant CPV

NOvA Simulation

NOvA Physics Milestones

• The most likely scenario emerging from Neutrino2016 presents Fermilab with the opportunity to

lead in neutrino science.

• NOvA has an opportunity for breakthroughs on all its major physics goals

θ23 2018: >3σ exclusion of maximal θ23 2019: >2σ octant determination 2024: >5σ exclusion of maximal θ23 2024: ~3σ octant determination Mass Hierarchy 2018: >2σ determination 2022: >3σ determination CP violation (sinδ≠0) 2023: >2σ observation of CPV

* opportunities enabled by higher than TDR proton delivery

27

* * * *

SLIDE 28

Mass Hierarchy: JUNO

JUNO Experiment

• 20 kt liquid scintillator
• 20+ GW
• L=50 km

Schedule

• Civil construction: 2013-2017
• Detector construction &

installation 2016-2019

• Filling and data taking: 2020

Mass hierarchy reach

• 3σ in 2 to 5 years: 2022-25
• 5σ in 10 years: 2030

28

SLIDE 29

Mass Hierarchy: ORCA

ORCA / KM3NET Experiment

• 1.8 Mton of instrumented sea

water

• Search for resonance in Earth core

in atmospheric neutrinos Schedule

• Construction through 2020

Mass hierarchy reach

• 3σ in 3 years ~2023, maybe faster
• 5σ possible faster

29

SLIDE 30

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 6

NOvA joint analysis

• Max. mixing

Hierarchy Octant CPV

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

Competition

This opportunity is not unique to Fermilab. There are several projects hoping to capitalize on this opportunity.

Both JUNO and ORCA have construction underway. Nearly identical schedules for mass hierarchy reach:

• 2σ as early as 2021
• 3σ as early as 2022

A Super-K + T2K combination gives roughly 2σ Other competition from, global fits, and cosmology fits.

30

SLIDE 31

Assume NOvA beam delivery goes from 6E20 to 7E20 / year starting in 2019

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 6

NOvA joint analysis

• Max. mixing

Hierarchy Octant CPV

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

gains 1 year hierarchy and CPV milestones

31

top curves: 800 kW starting in 2019 bottom curves: constant 700 kW

SLIDE 32

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 6

NOvA joint analysis

• Max. mixing

Hierarchy Octant CPV

=0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.403,

23

θ

2

/2, sin π NH 3

Competition

T2K has proposed an extended run to get 3σ sigma evidence for CPV

(arXiv:1607.08004v1 [hep-ex] 27 Jul 2016)

Until 2020 NOvA running flat-

• ut and T2K have same CPV

reach. T2K beam power ramps from current 420 kW to 770 kW by 2020 (surpassing NuMI power) and then to 1.1+ MW by 2023. Assumes 5 months / year beam allocation for T2K This plus analysis improvements drives the CPV reach of T2K to 3 sigma in 2024.

T2K sensitivity to CPV

32

SLIDE 33

Projections Summary

• Over the next decade Fermilab has an opportunity to lead the world in neutrino

measurements

• Non-maximal θ23
• θ23 octant
• Neutrino mass hierarchy
• CP violation
• To realize this we continue to
• Operate the detectors at high efficiency
• Push analysis to increase efficiency, reduce backgrounds, and reduce systematics
• Push on beam delivery
• Beam delivery continues to ramp toward TDR design parameters 6E20 POT/yr.
• NOvA can achieve these milestones before 2024:
• 5 sigma exclusion of maximal 23 mixing
• 3 sigma resolution of octant
• 3 sigma mass hierarchy determination
• 2 sigma CPV sensitivity
• Higher rate of beam delivery can advance milestones by 1 year which may be important to

maintain NOvA and Fermilab’s leading role in these measurements in the 2020’s

33

SLIDE 34

SLIDE 35

Proton flux evolution

Mary Convery 3 March 2016 PMG

• Does not take into

account uptime efficiencies, etc

• Assumes 4.4E12

ppp to BNB (last week 4.7E12)

SLIDE 36

Electron neutrino systematic uncertainties

left: Signal uncertainties right: Background uncertainties

Signal uncertainty (%)

20 − 10 − 10 20 Statistical error Total syst. error Detector Response Beam Calibration Cross Sections ν Normalization

Background uncertainty (%)

40 − 20 − 20 40 Statistical error Total syst. error Beam Normalization Cross Sections ν Detector Response Calibration

SLIDE 37

Muon neutrino systematic uncertainties

left: Impact of systematics on current contours right: Table of systematic impacts on mixing and mass splitting

0.3 0.4 0.5 0.6 0.7

)

2

eV

• 3

(10

32 2

m ∆

2 2.5 3 3.5

NOvA Preliminary

Normal Hierarchy POT-equiv.

20

10 × NOvA 6.05 90% C.L. all systs 68% C.L. all systs 90% C.L. stats only 23

θ

2

sin

0.3 0.4 0.5 0.6 0.7

)

2

eV

• 3

(10

32 2

m ∆

3 − 2.5 − 2 − 1.5 −

NOvA Preliminary

Inverted Hierarchy POT-equiv.

20

10 × NOvA 6.05 90% C.L. all systs 68% C.L. all systs 90% C.L. stats only

Systematic Effect on sin2(θ23) Effect on Δm232 Normalisation ± 1.0% ± 0.2 % Muon E scale ± 2.2% ± 0.8 % Calibration ± 2.0 % ± 0.2 % Relative E scale ± 2.0 % ± 0.9 % Cross sections + FSI ± 0.6 % ± 0.5 %

• Osc. parameters

± 0.7 % ± 1.5 % Beam backgrounds ± 0.9 % ± 0.5 % Scintillation model ± 0.7 % ± 0.1 % All systematics ± 3.4 % ± 2.4 %

• Stat. Uncertainty

± 4.1 % ± 3.5 %

SLIDE 38

Year

2016 2018 2020 2022 2024

) σ Significance (

1 2 3 4 5 =0.022

13

θ

2

, sin

2

eV

• 3

10 × =2.5

32 2

m ∆ =0.625

23

θ

2

/2, sin π =3

CP

δ Normal

systematic uncertainty improvements 2016 analysis techniques with projected

µ

ν +

e

ν NOvA joint

• Max. mixing

Hierarchy Octant CPV

NOvA Simulation

NOvA Physics Milestones

• Recompute milestones for best fit parameters in upper octant

θ23 2017: >3σ exclusion of maximal θ23 2017: >2σ octant determination 2022: >5σ exclusion of maximal θ23 2021: ~3σ octant determination Mass Hierarchy 2018: >2σ determination 2019: >3σ determination 2022: >4σ determination CP violation (sinδ≠0) 2023: 1.8σ CPV sensitivity

* opportunities enabled by higher than TDR proton delivery Start from 2016 exposure and extrapolate forward at design proton intensity. Assumes some improvement in systematic uncertainties over current analysis.

38

* * *