Transcript Atmospheric Neutrino Oscillations in SK-I

S
K
Atmospheric Neutrino
Oscillations in SK-I
An Updated Analysis
Alec Habig, Univ. of Minnesota Duluth
for the Super-Kamiokande Collaboration
With much help from
Masaki Ishitsuka & Mark Messier
Updated Analysis
S
K
• All “SK-I” data (April 1996-July 2001) reanalyzed
(1489 live-days)
– Ring selection, Particle ID, multi-ring fits improved
– Up-m reduction automated and fitting improved (1646
live-days)
• Monte Carlo predictions improved
– New 2001 Honda 3D n flux (was Honda 1995)
– Fermi Momentum, Axial Mass changed to better match
K2K near detector n interaction data
• (pF now flat, MA for QE, single p from 1.01.1)
– New calibs. improve Outer Detector, H2O parameters in
detector simulation (GEANT 3 based)
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 2
Flux Changes
S
K
• Honda 1995 1D to
Honda 2001 3D
– Absolute
normalization lower
– “3D” enhancement
• At low energies
• Near the horizon
• But at low E, nm
following angle is
large
– Smears out the peak
near horizon
– So 3D-ness changes
little for Super-K (see
next slide…)
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 3
Sub-GeV Data
S
K
(note no “3D” horizon peak)
No cos(q) shape information
at the lowest energies, only
flavor ratio is useful
Sub-GeV
(<1.33 GeV)
e-like
m-like
3353 (Data)
3013.9 (MC)
3227 (Data)
4466.9 (MC)
( m / e)data
Sub-GeV
(stat.)
(syst.)
( m / e) MC
 0.649  0.016  0.051
Key:
28th ICRC, 2 Aug.
2003 , Tsukuba
 Data
 MCAlec
(no Habig
osc.)
 MC (best fit)
At higher energies, nm directionality
better preserved plus
shorter L nm no longer oscillate:
cos(q) shape information very useful
Page 6
Multi-GeV data
K
S
m-like
e-like
Multi-GeV 746 (Data)
+ PC
700.4 (MC)
1562 (Data)
2098.0 (MC)
( m / e)data
Multi GeV+PC
( m / e) MC
(stat.)
(syst.)
 0.699  0.032
0.030 0.083
n baseline L: 12800 6200
At even higher energies, n flux
up/down symmetric and low-L nm
do not have time to disappear.
Key:
 Data
2 Aug.
28th
MCICRC,
(no osc.)
Tsukuba
2003
MC ,(best
fit)
700
40
15 km
 Nup - Ndown
A 
N N
down
 up
(stat.)
(syst.)


 0.289  0.028  0.004

μlike
Compare to Ae-like= -0.0200.0430.005
MC Am-like= -0.0030.0050.009
Page 7
Alec Habig
Observed
Am-like 9.5s from no-oscillation prediction!
SK
S
En ~10 GeV
m
nm
K
More Data
Up through going m 
Measured flux:
nm+Nmnp
1.70  0.02  0.04  1013 cm2s 1sr 1
(stat.) (syst.)
Theoretical calc:
SK
m
En~100 GeV
nm
1.57  0.35  1013 cm2s 1sr 1
(theo.)
Up stopping m 
Measured flux:
0.41 0.02  0.02  1013 cm2s 1sr 1
nm+Nmnp
(stat.) (syst.)
Theoretical calc:
Data
MC
Sub-GeV
208
346.4
Multi-ring m
28th ICRC, 2 Aug.
Multi-GeV
439
739.4
2003 , Tsukuba
Multi-ring m
0.61 0.14  1013 cm2s 1sr 1
Key:
 Data
 MC (no osc.)
Alec Habig
 MC (best fit)
(theo.)
• More nm, different En
and systematicsPage 8
S
K
New Oscillation
Results
• For nmnt
oscillation:
• Best fit:
sin2(2q)=1.0,
Dm2=2.0x10-3 eV2
– c2 = 170.8/170 dof
Contours represent
oscillation hypotheses
which fit the observed data
less well with a Dc2
corresponding to:
28th ICRC, 2 Aug.
2003 , Tsukuba
• 90% c.l. region:
– sin2(2q)>0.9
– 1.3 < Dm2 < 3.0x10-3
eV2
Alec Habig
Page 9
S
Difference from
Previous Results
K
Old result
@2.5x10-3
90% CL
regions
• Small
improvements +
the same data:
– but the end result
has changed by
more than you
might expect
New result
@2x10-3
• What happened?
– (Note this figure
is highly zoomed)
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 11
S
K
Effects of
Improvements on Fit
• Changes each of which caused Dm2 region to
move slightly down:
–
–
–
–
n flux change (Honda 19952001)
n interaction model (pF flat, MA 1.01.1)
Improved detector simulation (OD, H2O calib.)
Improved event reconstruction (Particle ID, ring
selection, up-m fitting)
• Net effect on c2 surface of several small
changes in same direction is larger
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 12
S
K
Sub-Sample
Consistency
• Check oscillation fits using different classes of data
independently – allowed regions all overlap best fit
• The low energy sub-sample’s only handle on
oscillations is the m/e flavor ratio
– Used to be high (alone!), is now consistent with other subsamples
Note open-ended
“swoosh” shape of
a one-parameter
flavor ratio fit to
two osc. parameters
(lowest E event
sub-sample)
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 14
Unusual Models
S
K
Mode
Best Fit
c2
P(c2)
Dc2
s
nm-nt
sin22qsin2(1.27Dm2L/E)
sin22q=1.00
Dm2=1.9x10-3 eV2
189
50%
0.0
0s
nm-ne
~sin22qsin2(1.27Dm2L/E)
sin22q=0.98
Dm2=4.2x10-3 eV2
304
0%
111
10.5s
nm-ns
~sin22qsin2(1.27Dm2L/E)
sin22q=0.93
Dm2=2.5x10-3 eV2
231
2%
42.2
6.5s
LxE (L.I. violation)
sin22qsin2(aLxE)
sin22q=0.89
a=5.1x10-4 GeV/km
329
0%
103
10.1s
nm decay (short t)
sin4q+cos4q(1-e-aL/E)
cos2q=0.49
a=3.2x10-3 GeV/km
287
0%
98.1
9.9s
nm decay (long t)
(sin2q+cos2q e-aL/2E)2
cos2q=0.33
a=9.8x10-3 GeV/km
207
19%
18
4.2s
• Ways to make nm
disappear without nm,nt
flavor oscillations
include:
– Lorentz inv. violation
– n decay, decoherence
• Fits using all available
SK n data strongly
constrain many such
models
– Hard for model to get
good fit over 5 orders of
Null Hypothesis
469 0%
280
16.7s
mag. in E and 4 in L
– Long t nm decay and nm
Data Used: (FC+PC (cut into 2 samples @Evis =
decoherence disfavored
(diff. from std.) 5 GeV)+NC+multiring+up-m, 195 bins,
Page 15
28th ICRC, 2 Aug.
Alec Habig
but not eliminated
190 d.o.f.)
nm decoherence
0.5sin22q(1-e-gL/E)
2003 , Tsukuba
sin22q=0.98
g=6.6x10-3 GeV/km
198
33%
9.4
3.1s
S
K
nm to nsterile?
• High energy n experience
matter effects which
suppress oscillations to
sterile n
– Matter effects not seen in
up-m or high-energy PC data
– Reduction in neutral current
interactions also not seen
– constrains ns component of
nm disappearance
oscillations
• Pure nmns disfavored
– ns fraction < 20% at 90% c.l.
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 16
CPT Violation
S
K
• Do nm oscillate differently
than nm?
• SK cannot tell the difference
between nm and nm eventby-event
– But we see the sum of the two
– One behaving very differently
would show up in the total
28th ICRC, 2 Aug.
2003 , Tsukuba
Alec Habig
Page 17
Summary
K
S
•
nmnt oscillations fit the data better than other
means of making nm disappear
– Best fit value is (Dm2 = 2.0x10-3 eV2, sin2(2q) = 1.0)
– 1.3 < Dm2 < 3.0x10-3 eV2, sin2(2q) > 0.9 @ 90% c.l.
• Analysis improvements to
– n interaction & flux models
– Detector simulation
– Event reconstruction
• No one improvement drove the changes to the final fit
– Each contributed a little in the same direction
– All data sub-samples now individually consistent with the
overall fit
28th ICRC, 2 Aug.
2003 , Tsukuba
The presenter gratefully acknowledges support for this presentation
20
AlecScience
Habig Foundation via its RUI grant Page
from the National
#0098579,
and from The Research Corporation’s Cottrell College Science Award