GENIE/Professor Framework for Neutrino Data Global fit a first Application: Global fit of CC 0Π Datasets

Home , Gargamelle

Introduction Genie CC 0π tuning Analysis Conclusion

GENIE/Professor framework for neutrino data global fit A first application: global fit of CC 0π datasets

Marco Roda - [email protected] on behalf of GENIE collaboration

University of Liverpool

19 April 2017 IPPP/NuStec

1 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Outline

Introduction Genie status vs recent datasets Tuning mechanism Tuning results Conclusions

Thanks to IPPP Associateship award

2 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Outline

Introduction Genie status vs recent datasets Tuning mechanism Tuning results Conclusions

Thanks to IPPP Associateship award

3 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Why 0π Why care about 0π?

We want to study neutrinos ﬂavour and mixing Lepton CP violation

Oscillation experiments T2k, NOvA DUNE, HyperK Beam energy ∼ few GeV

CC 0π is the dominant reaction Two body reaction Ideal for ν energy estimation ...... on free nucleons

4 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

0π puzzle CC Quasi-Elastic - 0π on single nucleons

Neutrinos Antineutrinos Theoretically  2003-2017, GENIE - http://www.genie-mc.org  2003-2017, GENIE - http://www.genie-mc.org ] ] 2 understood 2 2 cm cm -38

-38 1 Well constrained by 1.5

experimental 1 information 0.5 CCQE [10 0.5 CCQE [10 µ µ ν Electron scattering ν 0 0 Neutron β decay 10−1 1 10 1 10 102

ANL_12FT,1 ANL_12FT,3 Eν [GeV] BNL_7FT,2 Gargamelle,3 Eν [GeV] Experiments on BEBC,12 BNL_7FT,3 FNAL_15FT,3 Gargamelle,2 Gargamelle,5 NOMAD,3 Hydrogen / Deuterium NOMAD,2 SERP_A1,0 SERP_A1,2 SKAT,9 SERP_A1,1 SKAT,8

5 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

0π puzzle CC Meson Exchange Current - 0π on heavy nuclei

On higher A nuclei things are complicated

CCQE is not enough to describe data MiniBooNE MiniBooNE data

MEC is required nucleons interactions 2p-2h np-nh > 20 % effect On carbon

[Martini et al. PRC 84 055502 (2011)]

6 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

0π puzzle Effect of MEC on energy reconstruction

CCQE is a 2-body reaction Eν depends is just a function of lepton momentum and angle

MEC is not a 2-body reaction low energy tails in recontructed energy distributions

MEC also relevant for CP searches np-nh is different for ν/ν¯

⇒ MEC is important to achieve Martini et al. precise measurements

7 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

0π puzzle Search for 2p-2h

ArgoNEUT

Characteristic events 2 back-to-back nucleons

Nuclear effect can change observed topology migrations in the number of observed protons [Phys.Rev. D90 (2014) 1, 012008] future LarTPCs (or gas TPCs) important role Disentangle FSI from MEC CC 0π samples proton multiplicity

Important dataset that will "soon" be available

[Ulrich Mosel]

8 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

MC generators

What can we do as generator people? Comparing different data and models Being quantitative ⇒ highlight tensions Call for experiments: we need full covariance matrices feedback for experiments ⇒ drive the format of cross section releases ⇒ hint toward key measurements Global ﬁts Model ⇒ Cross sections is not analytic

What is the status of Genie in all of this?

9 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

MC generators

What is the status of Genie in all of this?

10 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

MC generators

What is the status of Genie in all of this?

11 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Models Genie - Models for 0π

Default - G00_00a No MEC CCQE process is LwlynSmith Model Dipole Axial Form Factor - Depending on MA = 0.99 GeV Nuclear model: Fermi Gas Model - Bodek, Ritchie

Default + MEC - G16_01b with Empirical MEC CCQE process is LwlynSmith Model Dipole Axial Form Factor - Depending on MA = 0.99 GeV Nuclear model: Fermi Gas Model - Bodek, Ritchie

Nieves, Simo, Vacas Model - G16_02a Theory motivated MEC CCQE process is Nieves Dipole Axial Form Factor - Depending on MA = 0.99 GeV Nuclear model: Local Fermi Gas Model

G17_02a (not presented in this talk) - G17_02a with Z-Expansion for Axial form factor Get rid of MA

12 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

CCQE MiniBooNE CCQE

 2003-2017, GENIE - http://www.genie-mc.org 2 [GeV]

Both ν and ν¯ µ T Double differential cross section 2 1.5 ﬂux integrated

No correlations 1 1 Preferred model is Nieves Model (G16_02a) 0.5 excellent agreement for ν χ2 = 101/137 DoF 0 −1 −0.5 0 0.5 1 Cosθ worse for ν¯ µ -38 ∂2σ(ν CC 0π)/∂ Cosθ /∂ T [10 cm2/GeV/n] χ2 = 176/78 DoF µ µ µ Data: miniboone_nuccqe_2010

13 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

CCQE MiniBooNE CCQE

Tµ ∈[0.4; 0.5 ] GeV

Both ν and ν¯ 2 /GeV/n] Double differential cross section 2 cm

ﬂux integrated -38 1.5 [10 µ T ∂ / µ

No correlations θ 1 Cos ∂

Preferred model is Nieves Model )/ π (G16_02a) 0.5 CC 0 µ

ν ν excellent agreement for ( σ

2 2

χ = 101/137 DoF ∂ 0 −1 −0.5 0 0.5 1 worse for ν¯ miniboone_nuccqe_2010 Cosθµ 2 2 χ = 176/78 DoF trunk:G00_00a:miniboone_fhc χ = 40.6/17 DoF trunk:G16_01b:miniboone_fhc χ2 = 64.8/17 DoF

trunk:G16_02a:miniboone_fhc χ2 = 14.3/17 DoF

14 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

CCQE T2K ND280 0π

 2003-2017, GENIE - http://www.genie-mc.org

Double differential cross section [GeV] µ 6

P 1 ﬂux integrated

Fully correlated 4

Tensions between datasets 0.5

Preferred model is G16_01b 2 χ2 = 135/67 DoF

all models look reasonable "By eye" 0 0 estimation −1 −0.5 0 0.5 1 Cosθ correlation is complicated µ ∂2σ/∂ Cosθ /∂ P [10-38 cm2/GeV/n] We can’t ignore it! µ µ Data: t2k_nd280_numucc0pi_2015

15 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

CCQE T2K ND280 0π

Double differential cross section ﬂux integrated

Bin ∈ [ 36; 41 ]

Fully correlated /GeV/n] 2 cm -38 [10 µ

P 0.5

Tensions between datasets ∂ / µ θ Cos ∂

Preferred model is G16_01b / σ 2 ∂ Cos Cos Cos Cos Cos Cos 2 0 θ θ θ θ θ θ χ = 135/67 DoF µ = 0.875 P µ = 0.875 P µ = 0.875 P µ = 0.92 P µ = 0.92 P µ = 0.92 P µ = 0.9 GeV µ = 1.25 GeV µ = 15.75 GeV µ = 0.2 GeV µ = 0.45 GeV µ = 0.55 GeV

t2k_nd280_numucc0pi_2015 trunk:G00_00a:t2k_nd280_numu_fhc χ2 = 11.5/6 DoF trunk:G16_01b:t2k_nd280_numu_fhc χ2 = 3.43/6 DoF all models look reasonable "By eye" trunk:G16_02a:t2k_nd280_numu_fhc χ2 = 28.1/6 DoF estimation correlation is complicated We can’t ignore it!

16 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Genie status GENIE

New Models MEC models Empirical Nieves Simo Vacas

Better CCQE model Nieves ...

Nuclear models

Multiple combinations Need to check the balance between each component will have a tuning!

17 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Genie status GENIE

New Models MEC models Empirical Nieves Simo Vacas

Better CCQE model Nieves ...

Nuclear models

Multiple combinations Need to check the balance between each component will have a tuning!

18 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Comparisons The Comparisons

The GENIE suite contains a package devoted to comparing GENIE predictions against publicly released datasets. Crucial technology for new GENIE global ﬁt to neutrino scattering data

Provides the opportunity to improve and develop GENIE models

All sorts of data Modern Neutrino Cross Section measurement nuclear targets typically ﬂux-integrated differential cross-sections MiniBooNE, T2K, MINERvA

Historical Neutrino Cross Section Measurement Bubble chamber experiment

Measurements of neutrino-induced hadronic system characteristics

19 / 38 Introduction Genie CC 0π tuning Analysis Conclusion

Professor Professor

http://professor.hepforge.org Numerical assistant Developed for ATLAS experiment I(p) used instead of a full MC 1 MC runs subset of param space 2 sample bin’s behaviour 3 Parametrization I(p) Polinomial interpolation Repeat for each bin

a parameterization Ij (p) for each bin p Minimize according to~I(p)

∼ 15 parameters