Sovan Chakraborty. MPI for Physics, Munich

Neutrino Mass Hierarchy from Supernova Neutrinos Sovan Chakraborty MPI for Physics, Munich

Outline Supernova (SN) as Neutrino Source Oscillation of SN Neutrinos Signatures of Neutrino Mass Hierarchy Conclusions

Supernova one of the most energe1c events in nature. Terminal phase of a massive star (M > 8~10 M ) Collapses and ejects the outer mantle in a shock wave driven explosion. ENERGY SCALES: ~ 10 53 erg : 99% energy is emioed by Neutrinos (Energy ~ 10 MeV). TIME SCALE: The dura1on of the burst lasts ~10 s.

Neutrino Emission Phases Neutroniza3on burst Accre3on Cooling Shock breakout De-leptonization of outer core layers Duration ~ 25 ms powered by infalling matter Stalled shock Cooling by ν diffusion Accretion: ~ 0.5 s ; Cooling: ~ 10 s [Fischer et al. (Basel Simulations), A&A 517:A80,2010, 10. 8 M sun progenitor mass]

Plan of the talk Supernova (SN) as Neutrino Source Oscillation of SN Neutrinos Signatures of Neutrino Mass Hierarchy Conclusions

SN ν Flavor Evolution The flavor evolution in matter is described by the non-linear MSW equations: In the standard 3ν framework H vac = UM U 2E 2 Kinematical mass-mixing term H = 2G diag( N,0,0) e F e ( ) Hνν = 2 G F (1 cos θpq) ρq ρq dq Dynamical MSW term (in matter) Neutrino-neutrino interactions term (non-linear)

Spectral Splits in the Accretion Phase [Fogli, Lisi, Marrone, Mirizzi, arxiv: 0707.1998 [hep- ph] ] Initial fluxes typical of accretion phase at neutrinosphere (r ~10 km) F : F : F 2.4:1.6:1.0 νe νe ν x = Inverted mass hierarchy (IH) Fluxes at the end of collective effects (r ~200 km) Nothing happens in Normal Hierarchy (NH)

Suppression of Collective effects Dense matter (ne) dominates over nu-nu interaction (nν). Dense Matter effect Suppresses Collective Oscillations [ S.C, Fischer, Mirizzi, Saviano &Tomas PRL 107:151101, 2011 PRD 84:025002, 2011 Sarikas, Raffelt, Hüdepohl & Janka PRL 108:061101, 2012 Dasgupta, P. O'Connor, Ott PRD 85:065008, 2012]

Suppression of Collective effects Predictions are robust when collective effects are suppressed, i.e.: 1) Neutronization burst (t < 20 ms) large ν e excess and ν x deficit [Hannestad et al., astro-ph/0608695] 2) Accretion phase (t < 500 ms) Dense matter term dominates over nu-nu interaction term. [S.C, Fischer, Mirizzi, Saviano &Tomas PRL 107:151101, 2011 PRD 84:025002, 2011]

SN neutrino Flux at Earth Neutronization burst & Accretion Phase:

Plan of the talk Supernova (SN) as Neutrino Source Oscillation of SN Neutrinos Signatures of Neutrino Mass Hierarchy Conclusions

Large Detectors for Supernova Neutrinos MiniBooNE (200) LVD (400) Borexino (80) Super- Kamiokande (10 4 ) KamLAND (330) IceCube (10 6 ) In brackets events for a fiducial SN at distance 10 kpc

Next generation Detectors for Supernova Neutrinos Next-generation large volume detectors might open a new era in SN neutrino detection: Mton Cherenkov LAr TPC Scintillator UNO, MEMPHYS, HYPER-K! GLACIER! LENA

Neutronization Burst : Model Independence Different mass Neutrino Transport Nuclear EoS Standard Candle for SN Neutrino [M.Kachelriess et al, hep-ph/0412082]

Oscillations in the Neutronization Burst Water Cherencov (ν e,x e - ν e,x e - ) Liq Ar TPC I.H. N.H. IH NH 70 kton [M.Kachelriess et al, hep-ph/0412082] Peak is absent NH ( [I.Gil-Botella & A.Rubbia, hep-ph/0307244] ) Peak is seen IH ( )

SN neutrino Flux at Earth Earth Matter Effect: EARTH Core Collapse EARTH Identify wiggles in a signal (but good E-resolution & high statistics required): Liquid scintillators like LENA? [ Dighe, Keil & Raffelt, hep-ph/0304150 ]

SN neutrino Flux at Earth Earth Matter Effect: (No E.M) (No E.M) Identify wiggles in a signal (but good E-resolution & high statistics required): Liquid scintillators like LENA? [ Dighe, Keil & Raffelt, hep-ph/0304150 ]

SN neutrino Flux at Earth Earth Matter Effect: Identify wiggles in a signal (but good E-resolution & high statistics required): Liquid scintillators like LENA? [ Borriello, S.C, Mirizzi, Serpico; PRD 86 (2012) ]

Galactic SN Distribution Possible sub-kpc candidate: Red supergiant Betelguse distance~ 0.2 kpc Event count~ 10s of million Identify wiggles in a signal (but good E-resolution & high statistics required): Liquid scintillators like LENA? [ Mirizzi, Raffelt & Serpico; (2006) ]

Rise time Analysis: Hierarchy Determination High degeneracy of ν e and e, suppresses ν e production. ν e more in equilibrium with environment than ν x. Flux of ν x rises faster than ν e NH: IH: Flux in IH (ν x ) rises faster than NH (ν x, ν e ) [Serpico, S.C, Fischer, Hüdepohl, Janka & Mirizzi PRD 85:085031,2012 ]

L Ν 10 51 erg s E Ν MeV Α 50 40 30 20 10 0 15 10 5 0 4 3 2 1 Rise time Analysis: Hierarchy Determination Ν e Ν x 0 0.00 0.05 0.10 0.15 0.20 t pb s Counts Bin 10 3 4.5 4.0 3.5 3.0 IH NH average noise 15 Solar Mass in Ice-Cube 2.5 0.00 0.02 0.04 0.06 0.08 0.10 0.12 t s Flux in IH rises faster than NH [Serpico, S.C, Fischer, Hüdepohl, Janka & Mirizzi PRD 85:085031,2012 ]

Rise time Analysis: Hierarchy Determination Normalized Count rate : 10 different models (12 M -40 M ) R t R 100 ms 1.0 0.8 0.6 0.4 0.2 0.0 IH NH Blue: IH ; Red: NH 0.02 0.04 0.06 0.08 0.10 t s Flux in IH rises faster than NH [Serpico, S.C, Fischer, Hüdepohl, Janka & Mirizzi PRD 85:085031,2012 ]

Rise time Analysis: Hierarchy Determination Normalized Count rate : 32 different models [C.D. Ott et al. Neutrino 2012, Japan, 1212.4250] It would be important in future to study the robustness of the rise time signature with more and more accurate simulations.

Conclusions Observing SN neutrinos is the next frontier of lowenergy neutrino astronomy. Collective effects are suppressed in early SN phases, implying hierarchy sensitivity at large θ 13. Neutronization phase is the best phase to probe mass hiearchy. Earth Matter effect: Detectable for Sub-kpc SNe. Rise time of SNe signal contains hierarchy information.

Appendix: Matter Suppression Neutrinos emitted from spherical source, travel on different trajectories. Different oscillation phases for neutrinos traveling in different paths. Strong ν ν interaction can overcome trajectory dependent dispersion. Collective conversion requires : ne << nν Collective conversion is matter Suppressed : ne [ Esteban-Pretel, Mirizzi, Pastor, Tomas, Raffelt, Serpico & Sigl, arxiv: 0807.0659 ] > nν

Appendix: SN antineutrino Flux at Earth Earth Matter Effect: Identify wiggles in a signal (but good E-resolution & high statistics required): Liquid scintillators like LENA? [ Borriello, S.C, Mirizzi, Serpico; PRD 86 (2012) ]

R t R 100 ms 1.0 Appendix: 0.00 Rise 0.02 time 0.04 0.06 Analysis: 0.08 0.10 0.12 2.5 0.00 0.02 Hierarchy t s Determination IH t s 0.8 Count rate (normalized to te signal in IceCube assuming a distance of 10 kpc, value based at 100 on the ms) simulation g group. 0.6 FIG. Right 3: panel: Left Panel: illustrative average SNexample count rate ofsignal the binned IceCube signal assuming using a distance 2 ms bi o 10 different models, masses for the signal 15 M plus progenitor photomultiplier mass from the Garching background group. noise, Right panel: whoseillustrative average example value is 0.4 ed here and typical in Poisson the following error estimates (see text accounting for details). for the signal 12 plus to 40 photomultiplier Msun backgr 0.2 as dot-dashed curve. A large ϑ 13 is assumed here and [Blue: in the IH following ; Red: NH] (see text for d NH 0.0 ontribute0.02 to the 0.04 final0.06 shape 0.08 of the 0.10curves. Also, For Statistical note that despite analysis growing energy of ν e shown in Fig. 1 contribute to the final shape of the the curves rel ly timescales large (10-20 differences t s ms, existing as already over very shown early timescales in [26]), introduce (10-20 one can ms, as alreadyexpe show escales 1.0 will be needed to beat statistical Cumulative errors. Time Distribution K(x) integrating the signal over a longer timescales will be needed to beat statistica s behaviors It is for useful the to whole compare setthe ofanalogous models, behaviors a task which for the will wholebe setmade of models ea 0.8 easured at a(nthe irrelevant) end ofrescaling the time to the interval rate measured considered, atxthe = end R(t)/R(t of the time end ). interval Also, c following statistical analysis, it is useful to introduce cumulative time distribut l to0.6 introduce cumulative time distributions K(x), defined in terms of IH 0.4 K x C 0.0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 xtend xtend dt R(t) dt R(t) 0.2 0 K(x) = NH tend 0.0 dt R(t), 0 K(x) = tend dt R(t), 0 0 0.0which 0.2 is a0.4 dimensionless 0.6 0.8function 1.0 satisfying K(0) = 0, K(1) = 1, with x [0 sfying K(0) = 0, K(1) = 1, with x [0, 1]. In Fig. (x) 4, forthedifferent we illustrate mod the rate functionsxr i A(t) and the cumulative functions KA i C

R t R 100 ms K x 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 IH NH Appendix: Rise time Analysis: Hierarchy Determination 0.02 0.04 0.06 0.08 0.10 t s IH NH 0.0 0.2 0.4 0.6 0.8 1.0 x Kolmogorov-Smirnov Statistics : D (K A i,k B j )= max x [0;1] K A i (x) K B j (x). Distance between any randomly picked NH model from average IH one is significantly above the one from average NH ones and well expected statistical errors. Assessing theory/numerical error requires detailed study over other simulations with comparable sophistication.