fitting the short-baseline anomalies
Post on 24-Apr-2022
2 Views
Preview:
TRANSCRIPT
Fitting the short-baseline anomalies
Joachim Kopp
The 4th neutrino – U Chicago, May 19, 2012
Fermilab
Joachim Kopp Fitting the short-baseline anomalies 1
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 2
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 3
Theoretical motivation for sterile neutrinosStandard Model singlet fermions are a very generic feature of “newphysics” models
I Leftovers of extended gauge multiplets (e.g. GUT multiplets) (typically heavy)I Dark matter (keV . . . TeV or above)
Neutrino–singlet mixing is one of the allowed “portals” between the SMand a hidden sector.SM singlet fermions can live at any mass scale
I Here: Focus on O(eV) sterile neutrinos (accessible to oscillationexperiments)
I Motivated experimentally
Typical Lagrangian:
Lmass ⊃ Yν L̄H∗NR + ms ν̄sNR +12
M NcRNR + h.c.
⇒ mass mixing between active and sterile neutrinos
Joachim Kopp Fitting the short-baseline anomalies 4
Signatures in oscillation experiments
Disappearance of active neutrinos (e.g. νe → νs oscillations)Anomalous transitions Appearance among active neutrinos(e.g. νµ → νs → νe)Oscillation length Losc = 4πE/∆m2
41 different from SM expectation(typically shorter)
Notation: ∆m2jk = m2
j −m2k ; m4,5: mostly sterile, m1,2,3: mostly active
Joachim Kopp Fitting the short-baseline anomalies 5
Sterile neutrino oscillations
Idea:Introduce extra neutrino flavor νs, mixing with the active onesAppearance searches (KARMEN, NOMAD, MiniBooNE . . . ) constrain(_ )ν µ →
(_ )ν s →
(_ )ν e,
(_ )ν τ
Disappearance searches (reactors, CDHS, MINOS . . . ) constrain(_ )ν e,
(_ )ν µ → ν̄s
ν̄e → ν̄s oscillations explain reactor anomaly(_ )ν µ →
(_ )ν s →
(_ )ν e oscillations explain LSND + MiniBooNE ν̄
Joachim Kopp Fitting the short-baseline anomalies 6
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 7
Our fitting procedure
Atmospheric neutrinos: Eight classes of events: Sub-GeV e, µ(p < 400 GeV/c), Sub-GeV e, µ (p > 400 GeV/c), Multi-GeV e, µ,Upward stopping µ, upward throughgoing µ, 10 zenith angle bins eachMINOS: Include NC and CC disappearance search(based on 1001.0336 and Neutrino 2010 talk by P. Vahle)
Reactor experiments: Bugey 3 (incl. spectrum), Bugey 4, Chooz (incl.spectrum), Goesgen 1–3, ILL, Krasnoyarsk 1–3, Palo Verde, RovnoSBL νe appearance experiments: LSND, KARMEN, MiniBooNE (ν (2010)and ν̄ data, consider only E > 475 MeV, i.e. low-E excess in νe samplenot included)Gallium anomaly not includedSBL νµ disappearance experiments: CDHS, NOMADAll codes reproduce the individual fits from the respective experiments.
JK Maltoni Schwetz 1103.4570 and work in progress
Joachim Kopp Fitting the short-baseline anomalies 8
LSND and MiniBooNE
Joachim Kopp Fitting the short-baseline anomalies 9
other
p(ν_
e,e
+)n
p(ν_
µ→ν
_
e,e
+)n
L/Eν (meters/MeV)
Beam
Excess
Beam Excess
0
2.5
5
7.5
10
12.5
15
17.5
0.4 0.6 0.8 1 1.2 1.4
LSND ν̄e
(GeV)QEνE
Even
ts / M
eV
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
0.5
1
1.5
2
2.5
3
Dataµ from eν
+ from Keν
0 from Ke
ν
misid 0πγ N→ ∆
dirtother
Total Background
1.5 3.
MiniBooNE νe
MiniBooNE ν̄e
LSND:I ν̄e appearance in ν̄µ beam from
stopped pion source (3σ)
MiniBooNE:I No significant νe or ν̄e excess in the
LSND-preferred regionI but ν̄e consistent with LSNDI Low-E excess not understood
The reactor anomaly
Recent reevaluation of expected reactor ν̄e flux is ∼ 3.5% higher thanprevious prediction Mueller et al. arXiv:1101.2663, confirmed by P. Huber arXiv:1106.0687
Method: Use measured β-spectra from 238U, 235U, 241Pu fission at ILLand convert to ν̄e spectrum (for single β-decay: Eν = Q − Ee)
Problem: Requires knowledge of Q-values for all contributing decays.→ take from nuclear databases where available, fit to data otherwiseCross check:
I Simulate mock e− spectra using few well-understood β-decaysI Reconstruct ν̄e spectrum using old method: Result is 3% too lowI Reconstruct ν̄e spectrum using new method: Result is exact.
Possible problem: Poorly understood effects in nuclei with large log ftHuber arXiv:1106.0687
Joachim Kopp Fitting the short-baseline anomalies 10
The reactor anomaly
Have short-baseline reactor experiments observed a ν̄e deficit?
NO
BS/(
NE
XP) p
red
,new
Distance to Reactor (m)
Bugey−
4
RO
VN
O91
Bugey−
3
Bugey−
3
Bugey−
3 G
oesgen−
I
Goesgen−
II
Goesgen−
III
ILL
Kra
snoyars
k−
I
Kra
snoyars
k−
II
Kra
snoyars
k−
III
SR
P−
I
SR
P−
II
RO
VN
O88−
1I
RO
VN
O88−
2I
RO
VN
O88−
1S
RO
VN
O88−
2S
RO
VN
O88−
3S
Palo
Verd
e
CH
OO
Z
101
102
103
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
Mention et al. arXiv:1101.2755
red = old reactor ν̄e flux predictionblue = new reactor ν̄e flux prediction
Joachim Kopp Fitting the short-baseline anomalies 11
MINOS NC disappearance search
Based on arXiv:1001.0336 and data shown at Neutrino 2010GLoBES simulationNC data:
I Use spectra and detector response functions based on MINOS MC(courtesy Alex Sousa)
CC data:I NuMI fluxes courtesy Mary BishaiI Backgrounds and efficiencies based on published results
Systematic uncertainties:I Based on published numbers, but simplified treatmentI Some fudging in CC channel
Joachim Kopp Fitting the short-baseline anomalies 12
MINOS NC disappearance search (2)
æ
æ
æ
æ
æ
æ
æ
æ
ææææ æ æ æ æ æ æ æ æ
æ
æ
æ
æ
æ
æ
æ
æ
ææææ æ æ æ æ æ æ æ æ
0 5 10 15 20
0
100 000
200 000
300 000
400 000
500 000
Ereco @GeVD
Events�GeV
Near detector NC
GLoBES
GLoBESBG
GLoBES osc.
ææ MINOS data
MINOSMC
MINOSMC BG
æ
æ
æ
æ
æ
æ
æ
æææææææ æ
æ ææ æ æ
æ
æ
æ
æ
æ
æ
æ
æææææææ æ
æ ææ æ æ
0 5 10 15 20
0
100 000
200 000
300 000
400 000
500 000
Ereco @GeVD
Events�GeV
Near detector CC
GLoBES
GLoBESBG
GLoBES osc.
ææ MINOS data
MINOSMC
MINOSMC BG
æ
æ
æ
æ
æ
æ
æ
æ
æ æ
æ
æ
æ
æ æ æ
æ
æ æ
æ
æ
æ
æ
æ
æ
æ
æ
æ
æ æ
æ
æ
æ
æ æ æ
æ
æ æ
æ
0 5 10 15 20
0
20
40
60
80
100
120
140
Ereco @GeVD
Events�GeV
Far detector NC
GLoBES
GLoBESBG
GLoBES osc.
ææ MINOS data
MINOSMC
MINOSMC BG
æ
æ
æ
æ
æ
æ
æ
æ
æ æ
æ
æ
æ
æææ
æ
æ
æ
æ
æ
æ
æ
æ
æ
æ
æ
æ
æ æ
æ
æ
æ
æææ
æ
æ
æ
æ
0 5 10 15 20
0
20
40
60
80
Ereco @GeVD
Events�GeV
Far detector CC
GLoBES
GLoBESBG
GLoBES osc.
ææ MINOS data
MINOSMC
MINOSMC BG
Joachim Kopp Fitting the short-baseline anomalies 13
MINOS NC disappearance search (3)
30 35 40 45 50 55 60
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Θ23 @degreesD
DΧ2
GLoBES
MINOS
0 10 20 30 40 50 60 70
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Θ34 @degreesD
DΧ2
GLoBES
MINOS
0 2 4 6 8 10 12
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Θ24 @degreesD
DΧ2
GLoBES
MINOS
Joachim Kopp Fitting the short-baseline anomalies 14
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 15
A 3+1 model: 3 active neutrinos + 1 sterile neutrino
Short baseline: Standard oscillations ineffective (∆m221, ∆m2
31 too small)Add extra (sterile) neutrinoFit shows: 3+1 neutrino scheme does not work well
JK Maltoni Schwetz 1103.4570 and work in progresssee also Giunti Laveder 1107.1452 and 1109.4033; Mention et al. 1101.2755; Karagiorgi et al. 0906.1997 and 1110.3735
10- 4 10- 3 10- 2 10-1
10-1
100
101
sin 2 2Θ
Dm2
KARMEN
NOMAD
MB Ν
disappearance
LSND + MB Ν
90%, 99% CL, 2 dof
10-3 10-2 10-1
10-1
100
101
sin 2 2Θ
Dm2
disappearance
appearance
90%, 99% CL, 2 dof
“disappearance” = SBL reactors, CDHS, atmospheric ν, MINOSθ = effective mixing angle for
(_ )ν µ →
(_ )ν s →
(_ )ν e oscillations
Joachim Kopp Fitting the short-baseline anomalies 16
Global fit in a 4-flavor scheme
|∆m241| |Ue4| |Uµ4| χ2/dof
STD 287.6/2563+1 0.48 0.14 0.23 255.5/252
LSND+MB(ν̄) vs rest appearance vs disapp.old new old new
χ2PG,3+1/dof 27.3/2 25.8/2 15.7/2 14.2/2PG3+1 1.2× 10−6 2.5× 10−6 3.9× 10−4 8.2× 10−4
Parameter goodness of fit: Test compatibility of 2 data setsby comparing global χ2
min to χ2min for separate fits
Joachim Kopp Fitting the short-baseline anomalies 17
The Giunti–Laveder fit
Includes the following data sets:(_ )ν µ →
(_ )ν e appearance data:
I LSNDI MiniBooNEI KARMENI NOMAD
νµ disappearance data:I CDHSI MINOS bound on |Uµ4|
ν̄e disappearance data:I Short baseline reactor experimentsI KamLAND bound on |Ue4|I Gallium anomaly
νe disappearance data:I νe–12C CC scattering in KARMEN and LSND
Giunti Laveder arXiv:1111.1069
Joachim Kopp Fitting the short-baseline anomalies 18
The Giunti–Laveder fit (2)
APP/DIS curves: 3σ C.L.Parameter goodness of fit (APP vs. DIS): 3× 10−3
Giunti Laveder arXiv:1111.1069
Joachim Kopp Fitting the short-baseline anomalies 19
The Karagiorgi et al. fit
Includes the following data sets:(_ )ν µ →
(_ )ν e appearance data:
I LSNDI MiniBooNEI KARMENI NOMAD
νµ disappearance data:I CDHSI CCFR84
ν̄e disappearance data:I Short baseline reactor experiments
Karagiorgi arXiv:1110.3735, Karagiorgi Djurcic Conrad Shaevitz Sorel arXiv:0906.1997
Joachim Kopp Fitting the short-baseline anomalies 20
The Karagiorgi et al. fit (2)
Karagiorgi arXiv:1110.3735, Karagiorgi Djurcic Conrad Shaevitz Sorel arXiv:0906.1997
Joachim Kopp Fitting the short-baseline anomalies 21
Global fit in a 5-flavor scheme
Check if more than one sterile neutrino improves the fit:
10-1 100 10110-1
100
101
Dm 41 @eV2D
Dm51@eV2D
90%, 95%, 99%, 99.73%
3+2
1+3+1
ø
ø
JK Maltoni Schwetz 1103.4570 and work in progress
Joachim Kopp Fitting the short-baseline anomalies 22
Global fit in a 5-flavor scheme (2)
|∆m241| |Ue4| |Uµ4| |∆m2
51| |Ue5| |Uµ5| δ/π χ2/dof
STD 287.6/2563+1 0.48 0.14 0.23 255.5/2523+2 1.10 0.14 0.11 0.82 0.13 0.12 -0.31 245.2/247
1+3+1 0.48 0.13 0.12 0.90 0.15 0.15 0.62 241.6/247
LSND+MB(ν̄) vs rest appearance vs disapp.old new old new
χ2PG,3+1/dof 27.3/2 25.8/2 15.7/2 14.2/2PG3+1 1.2× 10−6 2.5× 10−6 3.9× 10−4 8.2× 10−4
χ2PG,3+2/dof 30.0/5 24.8/5 24.7/4 19.5/4PG3+2 1.5× 10−5 1.5× 10−4 5.7× 10−5 6.1× 10−4
χ2PG,1+3+1/dof 24.9/5 21.2/5 19.6/4 10.7/4PG1+3+1 1.5× 10−4 7.5× 10−4 6.0× 10−3 3.1× 10−2
Parameter goodness of fit: Test compatibility of 2 data setsby comparing global χ2
min to χ2min for separate fits
Joachim Kopp Fitting the short-baseline anomalies 23
Does removing one experiment relax the tension?
Removing . . . LSND+MB(ν̄) vs rest appearance vs disapp.PG3+1 PG3+2 PG3+1 PG3+2
KARMEN 1.2× 10−5 6.2× 10−4 8.1× 10−4 1.0× 10−3
NOMAD 2.6× 10−6 1.5× 10−4 7.7× 10−4 6.1× 10−4
MB ν 2.2× 10−5 5.1× 10−4 1.7× 10−4 3.7× 10−4
MB ν̄ 2.9× 10−6 2.0× 10−3 1.9× 10−3 4.5× 10−3
LSND 2.3× 10−2 2.1× 10−2 2.0× 10−1 5.8× 10−2
Reactors 4.4× 10−4 7.9× 10−2 1.3× 10−1 2.9× 10−1
CDHS 1.2× 10−5 3.7× 10−4 3.6× 10−3 1.5× 10−3
Atmospheric 8.1× 10−6 6.9× 10−5 1.2× 10−4 2.3× 10−4
MINOS 1.4× 10−5 1.1× 10−3 4.1× 10−3 4.7× 10−3
Joachim Kopp Fitting the short-baseline anomalies 24
Impact on standard oscillation parameters
0.2 0.3 0.4 0.5 0.6 0.7 0.80.00
0.05
0.10
0.15
0.20
Sin2 Θ23
Sin22Θ13
ø
ç
0.2 0.3 0.4 0.5 0.6 0.7 0.80.0015
0.0020
0.0025
0.0030
0.0035
Sin2 Θ23
Dm312@eV2D
ø
ç
90%, 95%, 99%, 99.73%
Curves, circles: 3 flavors, atm +MINO
Colors, stars: 3+2 flavors, all exps
0.00 0.05 0.10 0.15 0.200.0015
0.0020
0.0025
0.0030
0.0035
Sin2 2Θ13
Dm312@eV2D
ø
ç
Joachim Kopp Fitting the short-baseline anomalies 25
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 26
Relation between appearance and disappearance3 + 1 neutrinosAt large baseline (L � 4πE/∆m2
41, but L � 4πE/∆m231
Pee = 1− 2|Ue4|2(1− |Ue4|2)Pµµ = 1− 2|Uµ4|2(1− |Uµ4|2)Peµ = 2|Ue4|2|Uµ4|2
It follows
2Peµ ' (1− Pee)(1− Pµµ)
In the 3 + 1 case, at large enough baseline, there is a one-to-one relationbetween the appearance and disappearance probabilities.
Joachim Kopp Fitting the short-baseline anomalies 27
Relation between appearance and disappearance3 + 2 neutrinosAt large baseline (L � 4πE/∆m2
41, but L � 4πE/∆m231
Pee = 1− 2[|Ue4|2(1− |Ue4|2) + |Ue5|2(1− |Ue5|2)− |Ue4|2|Ue5|2
]Pµµ = 1− 2
[|Uµ4|2(1− |Uµ4|2) + |Uµ5|2(1− |Uµ5|2)− |Uµ4|2|Uµ5|2
]Peµ = 2
[|Ue4|2|Uµ4|2 + |Uµ4|2|Uµ5|2 + Re(U∗
e4Uµ4Ue5U∗µ5)
]
Joachim Kopp Fitting the short-baseline anomalies 28
Relation between appearance and disappearance3 + 2 neutrinosAt large baseline (L � 4πE/∆m2
41, but L � 4πE/∆m231
Pee = 1− 2[|Ue4|2(1− |Ue4|2) + |Ue5|2(1− |Ue5|2)− |Ue4|2|Ue5|2
]Pµµ = 1− 2
[|Uµ4|2(1− |Uµ4|2) + |Uµ5|2(1− |Uµ5|2)− |Uµ4|2|Uµ5|2
]Peµ = 2
[|Ue4|2|Uµ4|2 + |Uµ4|2|Uµ5|2 + Re(U∗
e4Uµ4Ue5U∗µ5)
]It follows
2Peµ ' (1− Pee)(1− Pµµ)
+ 4[Re(U∗
e4Uµ4Ue5U∗µ5) + 4|Ue4|2|Uµ5|2 + 4|Ue5|2|Uµ4|2
]= (1− Pee)(1− Pµµ)− 2
[|Ue4|2|Uµ5|2 + |Ue5|2|Uµ4|2
]− 2
∣∣Ue4Uµ5 − Ue5Uµ4∣∣2
Joachim Kopp Fitting the short-baseline anomalies 28
Relation between appearance and disappearance3 + 2 neutrinosAt large baseline (L � 4πE/∆m2
41, but L � 4πE/∆m231
Pee = 1− 2[|Ue4|2(1− |Ue4|2) + |Ue5|2(1− |Ue5|2)− |Ue4|2|Ue5|2
]Pµµ = 1− 2
[|Uµ4|2(1− |Uµ4|2) + |Uµ5|2(1− |Uµ5|2)− |Uµ4|2|Uµ5|2
]Peµ = 2
[|Ue4|2|Uµ4|2 + |Uµ4|2|Uµ5|2 + Re(U∗
e4Uµ4Ue5U∗µ5)
]It follows
2Peµ ≤ (1− Pee)(1− Pµµ)
Unlike in the 3 + 1 case, for 3 + 2 models, there is NO one-to-one relationbetween the appearance and disappearance probabilities.
However, there is an inequality, which can be used to set meaningfulconstraints.
Joachim Kopp Fitting the short-baseline anomalies 28
Outline
1 Sterile neutrinos
2 Data sets and fitting procedure
3 Fit results
4 Relation between appearance and disappearance
5 Conclusions
Joachim Kopp Fitting the short-baseline anomalies 29
Global fits — take home message
Substantial tension in the global fit.
Is one (or all) of the positive results not due toneutrino oscillations?Are some of the null results wrong?(one being wrong is not enough!)
Are there more than 2 sterile flavors?Are there sterile neutrinos plus something else?
Joachim Kopp Fitting the short-baseline anomalies 30
Thank you!
top related