Detection of atmospheric neutrino and neutrino oscillation
50Kton ICAL neutrino Detector
Shamelessly using slides of Yuichi Oyama (KEK/J-PARC) talk at Vietnam school on Neutrino Historical context Detection of atmospheric neutrinos •Markov (1960) suggests Cherenkov light in deep lake or ocean to detect atmospheric interactions for neutrino physics •Greisen (1960) suggests water Cherenkov detector in deep mine as a neutrino telescope for extraterrestrial neutrinos •First reported events in deep mines with electronic detectors, 1965: KGF detector (Menon et al.), CWI detector (Reines et al.); Two methods for calculating atmospheric neutrinos: •From muons to parent pions infer neutrinos (Markov & Zheleznykh, 1961; Perkins) •From primaries to , K and to neutrinos (Cowsik, 1965 and most later calculations) • Essential features known since 1961: Markov & Zheleznykh, Zatsepin & Kuz’min •Monte Carlo calculations follow second method
Stability of matter: search for proton decay, 1980’s backgroundbackground forforpp-decay
• IMB & Kamioka -- water Cherenkov detectors • KGF, NUSEX, Frejus, Soudan -- iron tracking calorimeters • Principal background is interactions of atmospheric neutrinos e e • Need to calculate flux of atmospheric neutrinos
0.1 1 10 KGF Cosmic Ray muon experiments 1956 : First experiment by Sreekantan’s group at depth upto 270m But, initiated with simple measurement of muon lifetime experiment in 1949 Geiger counter based telescope measures Cosmic muons intensities at various depth and its angular distributions B V Sreekantan et al Proc Ind Acad Sci 43 (1956) 113 The MNR Experiments, Miyake, Narasimham and Ramanamurthy 1961 : New Series of Cosmic muon Experiments started by TIFR in KGF depths from 270m to 2760m The MNR group measure flux around 6 different depths and found at 2760m muon flux is attenuated This led to the observation neutrino at deep underground S Miyake, V S Narasimham, P V Ramana Murthy Nuovo Cim 32 (1964) 1505; 32 (1964) 1524 First report on Atmospheric neutrinos
Atmospheric neutrino detector at Kolar Gold Field –1965 Physics Letters 18, (1965) 196, dated 15th Aug 1965
PRL 15,PRL (1965), 15, 429,(1965), dated 429, 30th dated Aug. 30th 1965 Aug. 1965 KGF muon and neutrino experiments Collaboration: INDIA , UK and Japan Inelastic collision of an upward-moving neutrino Neon Flash Tubes Observed neutrino induced muon events predicted
MW > 3 GeV
2.5 Lead
Anomalous Kolar events : Event consists of two or more tracks and vertex of tracks pointing in air/low mass mat !
S Miyake, V S Narasimham, P V Ramana Murthy Nuovo Cim 32 (1964) 1505; ibid 32 (1963) 1524 KGF neutrino and proton decay experiments Collaboration: INDIA , UK and Japan Proton Decay Experiment
140 ton (at 2.3 km) Absorber : Iron ½” Detector : 10cm 10cm 4(6)m Proportional counter 8.41years Prof. NK Mondal’s thesis
5.53years 340 ton (at 2km) 60 horizontal array
V = 6m 6m 6m Underground lab The rate of the atmospheric neutrino interactions is about 200 kt−1 year−1. Rate at the surface due to cosmic ray particles is very frequent, namely ∼200 m−2 s−1,
Deepest experimental halls :
EPR Mine in South Africa : ) 8800 m.w.e KGF : Deepest : 8400m.w.e
(whereas neutrino discovery INO
at 7500m.w.e
neutrino+proton
KGF ( KGF
KGF (cosmic KGF muon flux)
EPR Mine, South AFRICA South Mine, EPR
History of Discoveries by neutrino experiments 1956 Reines and Cowan discovered anti-electron neutrino from a reactor (N) 1962 Muon neutrino beam experiment by L.Lederman et al (N) 1965 KGF observed atmospheric neutrino 1968 Homestake experiment claimed solar neutrino deficit (N) 1973 Neutral Current interaction was discovered by Gargamelle 1987 Kamiokande and IMB detected neutrinos from supernova SN1987A (N) Starting point of neutrino astronomy 1988 Kamiokande claimed atmospheric muon neutrino deficit 1989 Kamiokande confirmed the solar neutrino deficit 1998 Super-Kamiokande observed atmospheric neutrino oscillation (N) 1998 Super-Kamiokande confirmed solar neutrino deficit 2000 DONUT observed tau neutrino 2001 SNO confirmed solar neutrino oscillation by neutral current measurement (N) 2002 Kamland observed deficit of reactor neutrinos 2004 K2K confirmed atmospheric neutrino oscillation by artificial neutrino beam
2006 Completely independent confirmation of -t oscillation by MINOS 2012 Non-zero q13 was confirmed by 3 reactor experiments 2015 Discovery of t signal from - t oscillation by OPERA Solar neutrino oscillation? Homestake experiment claimed solar neutrinos deficit for over 15 years until middle of 1980s. However, not many people consider the result seriously. SSM calculation Total neutrino flux seems to be certainly robust. It can be directly evaluated from the solar luminosity. However, Homestake experiment does not measure pp neutrino. Other neutrino components are small fractions and not robust. It strongly depend on core temperature, chemical composition, cross sections, opacity of the Sun, etc. We cannot believe the calculation ! Homestake experiment Experiments in ~1MeV energy range are not territory of high energy physicists. R. Davis was not a physicist, but a chemist. Radiochemical method is not common technology for high energy physicists. Most of our experiments are "counter experiments". We cannot imagine/understood the experiment !
No experiment followed Homestake. (1 SNU is 10-36 captures per target atom per second) Atmospheric neutrinos
p,He,… Primary cosmic rays (p, He,…) collide with upper atmosphere and neutrinos are produced. ±
+
e + e + e
e : ~ 1 : 2 Particles and anti-particles are not distinguished unless needed. + For example, + means 흅 → + − − 흁 + 흂흁 풂풏풅 흅 → 흁 + 흂ഥ흁 in most of the e experiment except INO-ICAL Kamiokande (1983-1996) A large water Cherenkov detector constructed at 1000 m (2400 meter water equivalent) underground in Kamioka mine, Japan. 3000 tons of pure water are viewed by 1000 20-inchF PMTs. In KAM-I, the trigger threshold of the detector was ~29 MeV, was enough to detect nucleon decay mode pK+(+) Fall 1984 to end of 1986, detector upgraded to observe 8B solar neutrinos • KAM-I: trigger threshold of 110 photoelectrons (p.e.), which corresponds to 30 MeV/c (at 50% efficiency) and 37 MeV/c (90%) for electrons (3.4 p.e.=1 MeV for electrons), and 205 MeV/c (50%) and 220 MeV/c (90%) for muons. • KAM-II : 7.6 MeV/c (50%) and 10 MeV/c (90%) for electrons, and 165 MeV/c (50%) and 180 MeV/c (90%) for muons e/ identification in Kamiokande e e Cherenkov light from electromagnetic shower. Electrons and positrons are heavily scattered. Cherenkov ring edge is fuzzy.
e e
Only direct Cherenkov light from Clear Cherenkov ring edge
Simulation : ± ± 흅 → 흁 ퟓ. ퟒ ± ퟎ. ퟖ % 흅ퟎ → 풆± ퟔ. ퟕ ± ퟏ. ퟐ % e/ misidentification probability is less than 1 %. “Experimental study of the atmospheric neutrino flux” The first Kamiokande paper on atm neutrino : Phys. Lett B205 (1988) 416 2.87 kt*yr data e data 93 data 85 data MC MC 88.5 MC 144.0 Single ring e-like 93 88.5 Single ring -like 85 144.0 Multi ring 87 86.2 total 265 318.7
e-like : good agreement -like : data/MC = 59±7% (stat)
e data 93 data 85 Were unable to explain the data as the MC 88.5 MC 144.0 result of systematic detector effects or uncertainties in the atmospheric neutrino fluxes. Some as-yet-unaccounted-for physics such as neutrino oscillations might explain the data. “Observation of a Small /e Ratio in Kamiokande” The second Kamiokande paper on atm neutrino : Phys. Lett B280 (1992) 146 4.92 kt*yr data, 310 single ring events
data MC1 MC2 MC3 e data 159 1992論文 MC1 164.9 Single ring e-like 159 164.9 146.0 127.7 MC2 146.0 MC3 127.7 Single ring -like 151 260.6 234.2 205.2
data 151 MC1151 260.6 MC2260.6 234.2 MC3234.2 205.2 205.2
data/e data expected/e expected Observation of a Small /e …… (continued) "Neutrino Oscillation" was discussed positively. sin22q-Dm2 plot was officially reported first time.
Absolute atmospheric neutrino flux is ambiguous but /e ratio is robust. -t oscillation is favored, but -e cannot be denied, but ruled out by solar neutrino result. Other experimental curves are exclusion plot
-e -t
-t Best fit (sin22q,Dm2)= (0.87,0.8 10-2eV2) e t
-e
Best fit in 2016 (sin22q,Dm2)= e t (1.00,2.5110- 2eV2) NUSEX experiment (1982-1988) The NUSEX detector is a digital tracking calorimeter of 3.5m 3.5m 3.5m, located in Mont Blanc tunnel at 4800 m.w.e. underground.
It is a sandwich of 134 horizontal iron plates of 1.0 cm thickness, and layers of plastic streamer tubes 3.5 m long and of (9 9) mm2 cross-section. The total active mass is 150 ton.
From 740 ton * yr of exposure, data agree with Monte Carlo expectation.
data 32 e data 18 expected 36.8 expected 20.5
R=(data/MC)/(edata/eMC) +0.32 =0.96 -0.28
Europhys. Lett. 8 (7) (1989) 611 Frejus experiment (1984-1988) The Frejus detector is located in the Frejus highway tunnel connecting France and Italy. ｄata 108 The rock coverage is 1780m. expected 125.8 It consists of 912 flash chamber planes and 113 1.56kton・yr data Geiger tube planes. A flash chamber plane is made of a sandwich of 3 mm thick iron plates and 5 mm thick plane of plastic flash tubes (discharge on Ne-He filled chamber). The fiducial mass is 554 tons. Good agreement is obtained between the data and the simulation within statistics.
e Data 57 expected 70.6
Phys. Lett B227 (1989) 489 IMB-3 experiment Data from 851 days of IMB-3 experiment are analyzed. A total of 935 contained atmospheric neutrino events areaccumulated from 7.7 kton・yr of exposure. PhysRev D46 (1992) 3720 data Monte Carlo Nonshower Nonshowering 182 268.0 300 < p < 1500 MeV ( -like single ring) data 182 MC 268.0 Showering 〇: all data 352 257.3 ×: ->e signal (e-like single ring)
The fraction of nonshowering events is Data: 0.36±0.02(stat)±0. 02(syst) MC: 0.51±0.01(stat)±0. 05(syst) Shower １00 < p < 1500 MeV The discrepancy is 2.6s. data 325 MC 257.3 〇: all data Alternative analysis employing e ×: ->e signal decay signal also shows a similar discrepancy. Atmospheric /e ratio in the multi-GeV energy range The third Kamiokande paper on atm neutrino : Phys. Lett. B335 (1994) 237 Atmospheric neutrino analysis was extended to higher (multi-GeV) energy
range. In addition to fully contained events with Evis > 1.33 GeV, partially contained events are included. Analysis for the sub-GeV neutrinos was also updated.
data M.C. Sub-GeV e-like 248 227.6 ・ (7.7 kt yr, 690 events) -like 234 356.8 Multi-GeV Fully-contained e-like 98 66.5 ・ (8.2 kt yr,195 events) -like 31 37.8 Partially-contained e-like ------・ (6.0 kt yr,118 events) -like 104 124.4
In the multi-GeV energy range,
+0.08 R(/e) = 0.57 - 0.07 (stat.)±0.07(syst.) The result shows a small /e ratio, and is consistent with sub-GeV energy range. Atmospheric neutrino oscillation in early 1990s In the first result, the statistics was poor, and the up/down asymmetry was not clear. The straightforward impression was "Number of muon neutrino events is slightly small..."
Some of the experiment reported negative results. Kamiokande/IMB results are not widely believed. From a review article "Atmospheric Neutrinos" by T. Kajita New Journal of Physics 6, 194(2004) (data after 1995 were deleted) Water Cherenkov Tracker To claim "Neutrino Oscillation” was big and risky challenge. 1989,92 If it is not true, members will loose 1994 their confidence as high energy 1992 1997 physicists. 1989 Hesitated to use word "neutrino 1989 oscillations". Frequently used 1999 "muon neutrino deficit” or 1998 "atmospheric neutrino anomaly", 1998 instead. Toward confirmation of -t oscillation To claim neutrino oscillations, there are some problem to be solved. 1. The capability of e/ particle identification is suspicious. 2. Statistics is definitely poor. Much more data is needed. 3. Large uncertainty of atmospheric neutrino flux. 4. Negative results by other experiments.
Only two water Cherenkov detectors, Kamiokande and IMB, claimed atmospheric muon neutrino deficit. Tracking type detectors, NUSEX and Frejus, could not find the anomaly.
Is it a problem of water Cherenkov detectors ? The e/ identification of water Cherenkov detector has been examined only by Monte Carlo events. The e/ identification capability are certainly critical issue to be examined.
To verify the particle identification capability, a “beam test” was planned at KEK. E261A experiment KEK Proton Synchrotron E261A (1992-1994) 1kt water Cherenkov detector was built in KEK North counter hall. Electrons and muons from 12 GeV proton Synchrotron were injected. A gas Cerenkov counter, TOF counters to identify particle over momentum range 100 e, beam MeV/c - 1000 MeV/c. Rejection of pion : Decayed muon in
opposite disrection, p~0.57P Beam test for the particle identification Fuzzy edge for e event and clear edge for event are confirmed. e
Phys. Lett. B374 (1996) 238
e-likelihood (Le) and m-likelihood (L) are calculated. From a comparison e
between Le and L, particle id are
judged. Number of events of Number
The algorithm clearly separate e beam events and beam events.
events e of It was experimentally verified that the e/ identification capability is better than 99%. Number Super-Kamiokande (1996 - ) • 50 kt water Cherenkov detector with 11146 20- inch F PMTs.The fiducial volume is 22.5 kt. • Located at 1000 m underground in Kamioka mine, Japan • Operation since April 1996. Software upgrade in Super-Kamiokande Number of events is much larger than Kamiokande. Visual scan was impossible any more. Automatic analysis tools were developed. They are; 1)Automatic vertex reconstruction 2)Automatic ring counting 3)Ring separation 4)Determination of particle direction Particle identification program were applied to the result of automatic reconstruction. Cherenkov signal in Super-Kamiokande Discovery of atmospheric neutrino oscillation in Super-Kamiokande At NEUTRINO 1998 Conference
in Takayama, discovery of -t oscillation was reported by Prof. T. Kajita, on behalf of Super-Kamiokande collaboration. “Evidence for Oscillation of Atmospheric Neutrinos” Phys. Rev. Lett. 81 (1998) 1562 Immediately after NEUTRINO 1998, the results are published. A total of 4654 atmospheric neutrino events are employed which are accumulated in 535 days, corresponding to 33.0 kt*yr. Number of electron neutrinos well agrees with expectation, but number of muon neutrinos is clearly smaller than the expectation. Significant zenith angle distributions are also found. Evidence for Oscillation of Atmospheric Neutrinos Phys. Rev. Lett. 81 (1998) 1562 The data are consistent
with two-flavor - t oscillations with sin22q > 0.82 and 5 10 -4 < Dm2 < 6 10 -3 eV2 at 90% confidence level.
The best fit parameters are Dm2 = 2.2 10 - 3eV2 sin22q = 1.0 It agree with the numbers in PDG2019(NH), Dm2 = 2.56 10 - 3eV2 sin2q = 0.425 The results are inconsistent with the Kamiokande data.
17 years later, this paper became a “Nobel Prize paper”. Atmospheric neutrino oscillation
Number of upward-going is significantly smaller than expectation. Numbers of downward-going neutrinos and upward-
going e agree with expectations.
Neutrino oscillation from to t can explain the deficit. Downward-going
2 2 2 P() = 1-sin 2q sin (Dm L/4E)
e t Downward- going Upward-going Upward- going
e t MACRO experiment (1989-2000) UpThrough InUp The MACRO detector is a large rectangular box of 76.6 m 12 m 9.3 m. It was located at 3150 m.w.e. underground in Gran Sasso Laboratory, Italy.
The experiment is optimized for muon track measurement.
Streamer tubes provides track information UpStop InDown of the muons with angular resolution between 0.2o and 1o, depending on the track length.
Liquid scintillator planes records arrival time of particles. The nominal time resolution is 500 ps. The time difference between two separate planes provides the travel direction of muons. MACRO results The muon neutrino deficit was reported in 1998. From 1989 to 1997, 451 UpThough events were found where MC expectation was 612.
R = ()data/(m)MC was calculated to be R = 0.74±0.036(stat.)±0.046(syst.)±0.13(theor.) “The observed zenith distribution for -1.0 cosq 0 does not fit well with the no oscillation expectation,“ Phys. Lett. B478 (2000) 5 In 2000, observation of muon neutrino deficit UpStop+InDown in UpStop+InDown and InUP events followed.
UpThrough InUp
Phys. Lett. B434 (1998) 451 Soudan-2 experiment (1989-2001) Fine-grained iron tracking calorimeter with a honeycomb geometry. The fiducial mass was 770 ton.
The detector was located at a depth of 2070 m.w.e underground in Soudan mine, Minnesota.
R=(/e)data/(/e)MC were; R = 0.72±0.19(stat.) +0.05 (syst.) e-flavor, -flavor, -0.07 157 events 167 events 1.52 kt・yr (1997) R = 0.64±0.11(stat.)±0.06(syst.) 3.9 kt・yr (1999)
R = 0.69±0.10(stat.)±0.06(syst.) 5.90 kt・yr (2003) 5.90 kt・yr (2003)
The muon neutrino deficit was Phys. Lett. B391 (1997) 491 Phys. Lett. B449 (1999) 137 confirmed in the 1999 paper. Phys. Rev. D 68 (2003) 113004 Calculation of atmospheric neutrino flux Calculation of atmospheric neutrino flux is based on: Primary proton (and heavier nuclei) flux Geomagnetic field including effects from solar activity. Travel direction of charged particles are dispersed. Hadronic interaction models Atmospheric temperature. Atmosphere is target material and also affects decay/absorption ratio of pions/kaons.
Primary proton Geomagnetic field (and heavier nuclei) flux
South pole There are large uncertainties of the input data. Energy distribution and ratio of atmospheric neutrino flux
M.Honda et al., Phys. Rev. D83 (2011) 123001
HKKM11
HKKM11
Discrepancies between calculations are thought to be intrinsic systematic uncertainty. Zenith angle distribution of atmospheric neutrino flux M.Honda et al., Phys. Rev. D83 (2011) 123001
downward- 3.2 GeV 0.32 GeV 1.0 GeV going upward- going
It is claimed that even though absolute atmospheric neutrino fluxes have
large uncertainty, e/ ratio is robust. However..... It is difficult to say that atmospheric neutrinos are perfectly understood for studies of neutrino oscillations. SK atmospheric neutrino analysis after 1998 The deficit of upward-going atmospheric muon neutrinos can be explained not only by neutrino oscillations but also by other exotic models such as neutrino decoherence or neutrino decay.
Which is the right answer can be examined from neutrino survival probability as a function of L/E. Neutrino survival probability To keep high L/E resolution (D(L/E) < 70%), oscillation low energy events are removed because the correlation between decoherence neutrino direction and outgoing particle direction decay are poor. Horizontal events are also removed because
of large dL/d(cos qz) L/E analysis A clear dip corresponding to the first oscillation maximum can be found. Neutrino oscillation can well explain the data. Number of events in the dip region
1 bin 4 bins Expect No oscillation ~20 ~131 ation decoherence/ ~10 ~68 decay
oscillation ~7 ~48 data 4 44
Neutrino decoherence and neutrino decay are disfavored Phys.Rev.Lett. 93 (2014) 101801 with more than 3.4s. 1489 live-day data atmospheric neutrino flux Measurement of t appearance t CC When neutrino energy is larger than
Eth~3.5GeV, t can produce t by charged current interaction. Such events are direct evidence Eth=3.5GeV
of -t oscillation. HKKM11 t decay immediately, and high multiplicity events are generated.
t CC event with visible energy 3.3 GeV Measurement of t appearance A t selection algorithm was developed. But, most of the selected events are not t events but e or interactions. Zenith angle distribution of “t-like” events are examined.
Mode Non-t- t-like All like 5326 days
CC e 3071.0 1399.2 4470.2
CC 4231.9 783.4 5051.3
CC t 49.1 136.1 185.2 NC 291.8 548.3 840.1 SK-I - IV combined The data well agree with the expected 3.5 GeV t signal for oscillation parameters obtained from other results. × −38 2 s(t) = (0.940.20) 10 cm (large than t-normalisation (wrt 1) is −38 2 DONUT, (0.370.18) × 10 cm though found to be 1.47 0.32, which av energy is larger in DONUT) is 4.6s from 0 PRD98 (2018) 052006 From 2 flavor oscillation to 3 flavor oscillation Study on neutrino oscillations was extended to three flavor oscillations. If 3 flavor oscillation is assumed, the flavor eigenstates are mixtures of the mass eigenstates, and the correlation can be written as The equation has 6 parameters. 2 square mass differences (Dm2 , Dm2 ), 21 32 2 2 2 Dm ij=m i-m j 3 mixing angles (q12,q23,q13) 1 CP violation phase, dCP. 2 2 Dm 21, q12, Dm 32, q23 were already well studied. q13 and dCP, which appear in mixing between 1st and 3rd generations Mass hierarchy, which is sign of 2 2 Dm 32 ~ Dm 31, is another unknown parameter Present status of neutrino oscillation parameter Best fit value 3 sigma error 2 -1 Sin q12 (10 ) 2.97 2.50-3.54 2 -1 Sin q23 (10 ) 4.25 ( NH), 5.89 (IH) 3.81-6.15 (NH), 3.84-6.15 (IH) 2 -2 Sin q13 (10 ) 2.15 (NH), 2.16 (IH) 1.90-2.40 (NH), 1.90-2.42 (IH) 2 -5 2 Dm 21 (10 eV ) 7.37 6.93-7.96 2 -3 2 ǀDm 31ǀ (10 eV ) 2.56 (NH), 2.54 (IH) 2.45-2.69 (NH), 2.42-2.66 (IH) INO site : Bodi West Hills 90 58’ N, 770 16’ E Pottipuram Village Theni District Tamil Nadu State Warm, low rainfall area, low humidity throughout the year •Good rock quality, Cavern set in massive Charnockite rock under 1589 m peak with vertical rock cover of 1289 m. •Accessible through a 2 km long tunnel •Cavern -1 will host 50 kt ICAL detector. Space available for additional 50 kt. •Cavern-2 & 3 available for other experiments. INO-ICAL detector Fast and precise timing information to identify up-down ambiguity of muon: RPC is a the best choice to have 1ns precision with low cost 3cm strip pitch & strip multiplicity : ~5mm position resolution (along with ~1.3T magnetic filed) to identify bending of +ve and -ve muon, very crucial for the determination of neutrino mass hierarchy Float Glass Float Glass • Total number of RPCs in the ICAL = 3 150 64 = 28,800 • Total gas volume = 28,800 190cm 190cm 0.2cm ~ 200 m3 • Total surface area = 28,800 × 1.9m × 1.9m ~ 105 m2 • Standard gas composition for the avalanche mode: – Freon, R134a(C2H2F4):Isobutane(i-C4H10):Sulphur Hexafluoride(SF6)::95.5:4.3:0.2 Development of RPC detector 1m×1m magnetised RPC stack inside 2m ×2m IRON at VECC 1m×1m RPC stack at TIFR 500AT : ~1.5 Tesla 2m×2m RPC stack at IICHEP, Madurai 2m×2m RPC stack inside 4m × 4m magnetised iron at IICHEP, Madurai as a Running Prototype RPC Stack at TIFR/IICHEP Position Time Multiplicity Inefficiency due to button, dead resolution resolution and position strip, but edge effect also present Zenith angle of muon, measurement of cosmic Distinction of up/down Muon muon flux as well as it angular dependency Mis- identification n I = I0 cos q 45 Input to detector simulation and digitisation Mass hierarchy of neutrinos – sensitivity of ICAL The independent analysis of the T2K and NOνA data shows a preference towards normal hierarchy. Additional knowledge can also be gained by studying the atmospheric neutrinos as they offer wide range of energies and baselines. Therefore, magnetized Iron CALorimeter was proposed to identify mass hierarchy using atmospheric (only ICAL can study the matter oscillation at 13000km) ICAL only JHEP 1410 (2014) 189 ICAL + T2K + NovA S. Agarwalla et al. The next frontier Hyper-Kamiokande 250 kt water Dune 40 kton liquid argon JUNO 20 kton scintillator To further pushP( matter effect ) Hierarchy sensitivity • InstrumentHype Mtonsr-Kaofm iceiokand or e: sea-water • Fine granularity to have low threshold Km3net Orca in the Mediterranean sea After 3 years 3σ for most of the parameter space Recent proof-of-concept with IceCube DeepCore data arXiv:1902.07771 weak preference for NH implications for the IceCube Upgrade arXiV:1401.2046 47 Atmospheric ν Supernova ν Solar ν ν ν ν Super-Kamiokande Nucleon Decays ν J-PARC High intensity neutrino and anti-neutrino beam Hyper-Kamiokande 25 x Super-K fiducial mass as neutrino target and proton decay source DUNE Four 10 kt fiducial mass (17 kt total) liquid argon TPCs modules Dual phase Single homogeneous LAr drift volume Signal amplification in the gas phase leads to improved signal/noise Photosensors under semi-transparent cathode grid 500 V/cm drift field: 600 kV at the cathode IceCube Amundsen-Scott South Pole station50 IceCube Gen2 73 stations Scale : O(100) strings, O(10km3) Scale : 40 strings, extending Identify astroparticle sources of DeepCore : 70k (10k) events in µ (e) neutrino and cosmic ray with E>10GeV. Neutrino mass hierarchy, wide range of L/E, matter effect, WIMP search (from the SUN) KM3NET Measurement of IceCubeflux with different methodology, complementary field of view and improved Neutrino mass hierarchy; Neutrino astronomy including Galactic sources; ×2 Conclusion : A slide of Gianluigi Fogli We close with a recent detailed comparison of the sensitivity of each of the cited experiments, in terms of number of s’s, plotted in terms of the time-scale. M. Blennow, P. Coloma, P. Huber and T. Schwetz, JHEP 02 (2014) 028 [arXiv:1311.1822v2[hep-ph]] NH true IH true Due to the dichotomous character of the neutrino mass ordering, the sensitivity is plotted on the left for rejecting IH if NH is true, and viceversa on the right. The width of each band depends on the range of values of the parameters relevant in the estimates, in particular: dCP and the true q23 for LBL accelerator experiments, NOA and LBNE q23 for atmospheric experiments, INO and PINGU energy resolution for MBL reactor experiment, specifically JUNO XVI International Workshop on Neutrino Telescopes - Venice, March 2nd, 2015 Historical context (cont’d) p Atmospheric neutrino anomaly - 1986, 1988 … • IMB too few decays (from interactions of ) 1986 • Kamioka -like / e-like ratio too small. • Neutrino oscillations first explicitly suggested in 1988 Kamioka paper • Hint of pathlength dependence from Kamioka, Fukuda et al., 1994 Discovery of atmospheric neutrino oscillations by S-K • Super-K: “Evidence for neutrino oscillations” at Neutriino 98 e • Subsequent increasingly detailed analyses from Super-K 1998… • Confirming evidence from MACRO and Soudan • Analyses based on ratios comparing to 1D calculations e Need for precise, complete, accurate, 3D calculations • ~ PT / E is large for sub-GeV neutrinos • Bending of muons in geomagnetic field important for from decay • Complicated angular/energy dependence of primaries (AMS measurement) • Use improved primary spectrum and hadroproduction information p Atmospheric neutrino beam • Up-down symmetric except for geomagnetic effects e • One detector for both – long baseline e – short baseline • 1 < L/E < 105 km/GeV • /e ~ 2 for E < GeV D. Ayres, A.K. Mann et al., 1982 Tom Gaisser Atmospheric Neutrino Fluxes August 20, 2004 Also V Stenger, DUMAND, 1980 Summary of Atmospheric Neutrino Calculations Zatsepin, Kuz’min SP JETP 14:1294(1961) Mu Many calculations ~ 1965 --- ~1990 1D D.H. Perkins Asp.Phys. 2: 249 (1994) Mu Honda, Kajita, Kasahara, Midorikawa PRD 52: 4985 (1995) 1D * FRITIOF Agrawal, Gaisser, Lipari, Stanev PRD 53: 1314 (1996) 1D * Target Battistoni et al Asp.Phys 12:315 (2000) 3D FLUKA Asp.Phys 19:269 (2003) P. Lipari Asp.Phys 14:171 (2000) 3D PL B516:213 (2001) GHEISHA V. Plyaskin hep-ph/0303146 3D CALOR-FRITIOF Tserkovnyak et al Asp.Phys 18:449 (2003) 3D GFLUKA/GHEISHA Corsika: DPMJET Wentz et al PRD 67 073020 (2003) 3D VENUS, UrQMD Liu, Derome, Buénerd PRD 67 073022 (2003) 3D Favier, Kossalsowski, Vialle PRD 68 093006 (2003) 3D GFLUKA Barr, Gaisser, Lipari, Robbins, Stanev PRD 70 023006 (2004) 3D Target Honda, Kajita, Kasahara, Midorikawa PRD 64 053011 (2001) 3D ** DPMJET astro-ph/0404457 to PRD ** Used for analysis of Super-K data in publications before 2004; ** used now Summary of Neutrino oscillations in 2017 2 Dm 21 and q12 were determined by solar/reactor neutrino experiments. 2 -5 2 2 o Dm 21 = (7.53±0.18)x10 eV , sin q12 = 0.307±0.013 q12~33 2 |Dm 32| and q23 were studied by atmospheric/long-baseline experiments. 2 -3 2 2 |Dm 32| = (2.45±0.05)x10 eV , sin q23 = 0.51±0.04 for normal hierarchy o q23~45 2 -3 2 2 |Dm 32| = (2.52±0.05)x10 eV , sin q23 = 0.50±0.04 for inverted hierarchy q13 was studied by long-baseline/reactor experiments. 2 sin q13 = 0.0210±0.0011 At present, the CP violation phase dCP, and mass hierarchy, sign of 2 Dm 32 are still unknown. Normal hierarchy Inverted hierarchy The first hints towards dCP ~ -/2 and normal hierarchy are T2K: obtained. 7.48x1020 POT +7.47x1020 POT Neutrino Oscillations & Mass Hierarchy 1965 Ray Davis measured solar neutrino flux and found deficit of flux (=> solar neutrino puzzle) This led to the birth of Neutrino oscillations The neutrino flavour state can be represented as a super-position of mass eigenstate ν1, ν2 and ν3 with mass m1, m2 and m3 2 2 2 Oscillation probability is related to the differences in the mass squared Dm ij (m i - m j ), L and E. Various experiments have been conducted using atmospheric, solar, reactor and accelerator 2 2 neutrinos to measure ∆m 12 and ∆m 13 2 However, the sign of ∆m 13 is not yet known. SK atmospheric neutrino analysis after 1998 Although atmospheric neutrino “beam” have disadvantage, huge statistics is definitely advantage of Super-Kamiokande. Progress after the discovery of neutrino oscillation after 1998 are as follows. 1. Improvement of the statistics by about 10 times. PDG2016 http://pdg.lbl.gov/ 2. E/L analysis “Evidence for an oscillatory signature in atmospheric neutrino oscillation” Y.Ashie et al., Phys.Rev.Lett. 93,101801(2004) 3. Measurement of t appearance “Measurement of Atmospheric Neutrino Flux consistent with Tau Neutrino Appearance” K.Abe et al., Phys.Rev.Lett. 97,171801(2006) SK zenith angle distribution with 20 years data 4654 events (33 kt・yr, 1998) → 35326 events (306 kt・yr, 2016) “UpStop” and “UpThrough” are included in the analysis. PDG2016 http://pdg.lbl.gov/ Status of neutrino oscillations in the middle of 2000s st Around 2000, the neutrino mixing between e and (between 1 and 2nd generation) was established with solar/reactor neutrinos. There were contributions from Homestake, Kamiokande, Super- Kamiokande, SNO and Kamland. (This is out of scope of my lecture) nd rd The neutrino mixing between and t (between 2 and 3 generation) was discovered with atmospheric neutrinos by Super-Kamiokande in 1998. The -t oscillation was confirmed with artificial neutrino beam by K2K experiment in 2004. The -t oscillation was also confirmed with MINOS experiment in 2006. It was first perfectly-independent confirmation of the Super- Kamiokande results. Experimental study of neutrino oscillations moved to new phase: Three flavor oscillation. MSWMSW effecteffect In 1978, L. Wolfenstein pointed out that the neutrino oscillation parameters are changed in matter. In 1985, A. Smirnov and S. Mikheyev predicted that the neutrino mixing can be resonantly enhanced by slow decrease of the matter density. Especially, the change of the matter density from the core to the surface of the Sun satisfy the condition for the enhancement. MSW effect shed light on "solar neutrino problem" S. Mikheyev A. Smirnov L. Wolfenstein Atmospheric neutrino in multi-GeV energy range Best fit in 2016 (sin22q,Dm2)= Zenith angle distribution of multi-GeV (1.00,2.51x10-2eV2) muon neutrino events shows clear up/down asymmetry. Multi-GeV Sub-GeV and Multi-GeV results are consistent with each other. Strict constraints for oscillation parameters are obtained. e Sub-GeV Combined Super-K atmospheric neutrino data (T. Kajita) CC e CC 300 300 60 Sub-GeV e-like Sub-GeV-like multi-ring Upward stopping P 400 MeV/c P 400 MeV/c Sub-GeV-like 100 Events 200 200 40 75 50 100 100 20 25 Number of Number 0 0 0 0 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 -1 -0.8 -0.6 -0.4 -0.2 0 300 400 400 Sub-GeV e-like P 400 Sub-GeV -like multi-ring Multi- Upward through-going MeV/c P 400 MeV/c GeV -like 300 100 300 Events 200 200 200 100 50 100 100 Number of Number 0 0 0 0 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 -1 -0.8 -0.6 -0.4 -0.2 0 200 150 Multi-GeV e-like 150 Multi-GeV -like PC cosq 150 Events 100 100 1489day FC+PC 100 data + 1646day 50 50 50 upward going Number of Number 0 0 0 muon data -1 Tom-0.5Gaisser0 0.5 1 -1 -0.5 Atmospheric0 0.5 Neutrino1 -1 Fluxes-0.5 0 0.5 1 August 20,cos2004q cosq cosq