Probing 13 with global neutrino data analysis Antonio Palazzo - - PowerPoint PPT Presentation

TAUP 2009

Probing θ13 with global neutrino data analysis

Antonio Palazzo

CSIC/IFIC, AHEP group, Valencia July 2nd

Based on work done in collaboration with: G.L. Fogli. E. Lisi, A. Marrone, A.M. Rotunno

Ou Outl tlin ine

On the tr On the track of k of θ13

13

Hin Hints fr ts from: m: 1) 1) Atmo mosphe spheric da c data;

• 2)

2) Sol Solar & KamLAND r & KamLAND; 3) 3) MINOS ( MINOS (νe a app ppea earanc ance da e data); a);

• In

Intr troduc ductio ion The s he stan anda dard 3 d 3ν fr fram amew ework;

2

Fu Futu ture p e persp spective an e and c d conc nclusio sions ns

The s he stan anda dard 3 d 3ν fr fram amew ework k

3

Dirac CP-violating phase

The l he leptonic m c mix ixin ing

unknown

Explicit form:

4

ν1

NH NH IH IH

ν2 ν3 ν3

The n he neu eutr trin ino ma

• mass sp

ss spectr trum

?

+Δm2 δm2

• Δm2

ν1 ν2

5

Ex Experim imen ental Sensi l Sensitivi vitie ies

Solar, KamLAND Atmospheric, LBL (disapp.) CHOOZ, MINOS (app.)

sub-l sub-lea eadin ding lea eadin ding

6

Fogli et. al. [PRD 78, 033010 (2008)] 7

A preference for θ13 > 0 at a non-negligible C.L (90%) emerged in 2008

[Fogli, Lisi, Marrone, A.P, Rotunnno, PRL 101, 141801 (2008), arXiv:0806.2649,hep-ph].

This brings us to focus on …

Co Cons nstr train aints fr ts from gl m glob

• bal 3

l 3ν ana analysis a ysis as of 20 s of 2008 08

High precision on both mass splittings, now determined by “artificial” neutrino sources experiments (KamLAND for δm2, MINOS for Δm2). Estimates of the two leading mixing angles is less accurate (especially θ23), and experiments using “natural” ν’s play a crucial role in their determination.

8

• Sea

Searche hed f d for r disa disapp ppea earanc ance of r e of rea eactor νe E ~ few MeV L = 1 km

• L/E range comparable to atmospheric ν

→ probes the same Δm2

• No disappearance signal was found (1998)

The CHOOZ experiment

9

CHOOZ exclusion plot

Exclusion plot in the (Δm2, θ13

13)

plane Atm +LBL Δm2 scale (now)

set with precision by

“His Historical” u upp pper bo r bound o d on n θ13

13 e

establ blishe ished in 1998 d in 1998

10

Sinc Since then the 3 e then the 3ν gl glob

• bal ana

l analyse yses ha s have sho e shown a sl n a slow b w but p t progressi ssive enhanc enhancem emen ent in the sensi t in the sensitivi vity t ty to

• θ13

13.

• Wha

hat ins t instea ead (pl d (plea easan santl tly) su y) surprise ises u s us is tha s is that, f , for the fir r the first t t tim ime, the , the neu eutr trin ino da

• data p

a poin int t t towa wards a n ds a non-z n-zero v

• value of

ue of θ13

13 – p

– provi vidin ding thr g three ee in inde depen enden dent an t and c d convergin ing h g hin ints in th ts in this dir is directio ion. n. In the pa In the past the t they ha y have fir e first c t corrobo

• borated (

d (atmo mosphe spheric ana c analyse yses), an ), and then s then str tren engthen gthened ( d (sol solar+KamLAND ana r+KamLAND analyse yses) the CHOO ) the CHOOZ u Z upp pper l r lim imit. It is then n It is then not a c t a compl mplete su e surprise tha ise that the t they n y now s w start t t to be sensi

• be sensitive

to v

• value

ues of s of θ13

13 bel

below the CHOO w the CHOOZ l Z lim imit.

The role of global analyses in pinning down θ13

11

Hin Hint n. t n. I Da Date: e: 20 2006 06 Da Data: a:

• Atm. + LBL(

+ LBL(disa disapp pp.) + CHOO .) + CHOOZ

12

G.L. Fogli, E. Lisi, A. Marrone, A.P., Prog. Part. Nucl. Phys. 57, 742 (2006)

We f e found a h d a hin int f t for θ13

13 > 0

> 0 in the 3 in the 3ν ana analysis of ysis of

• Atm. + LBL + CHOO

+ LBL + CHOOZ Z Be Best fit t fit ~ 0.015 is ~ 1 sigma 1 sigma a awa way fr y from z m zero

Numerical results of 3ν global analysis (2006)

13

The hint persists after inclusion of the latest disappearance LBL results (MINOS)

14

Tracing the origin of the “atmospheric hint”

Atmospheric data present a small excess of sub-GeV electron-like events. This can be partially explained by 3ν subleading effects driven by the “solar” splitting δm2. Indeed, from an estimate of the order of magnitude of the MSW potential:

we see tha we see that in SK da t in SK data ea a earth ma th matter e r effects a ts are: e:

• Δm2-driven for multi-GeV
• δm2-driven for sub-GeV

14

15

Theoretical expectation for the excess

In the spirit of this qualitative discussion, it is helpful to assume that the density of the earth is constant. Indeed, in this case we have that this observable can be written as the sum of three different terms:

multi-GeV sub-GeV

The excess of expected electron events compared to the no-oscillation case can be expressed as [Peres and Smirnov , Nucl. Phys. B 456, 204 (1999); 680, 479 (2004)] : r = Notice that this observable is particularly useful to describe 3ν sub-leading effects since it exactly equals zero when: bo both th θ13

13 = 0 &

= 0 & δm2 = 0 = 0

“θ13 term” “δm2 term” “Interference term” mixing angles in matter

“swapping”

relations

For the case of:

two (non-equivalent) CP conserving cases

we have:

16

“Exact” numerical examples

“θ13 term” dominant “δm2 term” dominant “Interference term” dominant (only in sub-GeV)

These terms help to fit the small electron excess in Sub-GeV and Multi-GeV

17

Two non-equivalent CP-conserving cases

G.L. Fogli, E. Lisi, A. Marrone, A.P., Prog. Part. Nucl. Phys. 57, 742 (2006)

18

Maltoni & Schwetz (arXiv:0812.3161[hep-ph])

Comparison with other existing analyses

Escamilla , Latimer and Ernst (arXiv:0805.2924 [nucl-th]) Roa, Latimer, Ernst (arXiv:0904.3930 [nucl-th])

• Atm. h

hin int su t supp pported.

• d. Sim

Simil ilar p r preferenc ence f e for c r cosδ sin sinθ13

13 < 0

< 0 Schwetz, Tortola, Valle (arXiv:0808.2016[hep-ph])

• Atm. h

hin int n t not su t supp pported. d.

• Atm. h

hin int n t not o t or pa r partially su y supp pported. d.

19

There exist ongoing 3ν analyses in SK after phase I, presented in recent PhD theses using SK-I+II data. Unfortunately, none of the these analyses allows both θ13 > 0 & dm2 > 0, and thus they do not include the interference term.

Concluding remarks on the “atmospheric hint”

Therefore, it will be very important to see the next official SK oscillation analysis, which should hopefully include a complete treatment of three- flavor oscillations with both δm2 > 0 and θ13 > 0. It is also worth noticing that, at the current level of refinement, it is difficult to reproduce in detail the atmospheric analysis outside the SK collaboration, especially if one is looking for effects at the level of ~ 1 σ.

20

Meanwhile, also in consideration of the partial agreement among different independent analyses, it seems wise to consider the atmospheric hint having a fragile status.

Hin Hint n. t n. II II Da Date: e: June 20 e 2008 08 Da Data: a: Sol Solar & KamLAND r & KamLAND

21

3ν ana analysis of sol ysis of solar an r and KamLAND da d KamLAND data (20 a (2008) 08)

Notice that marginalization

• ver δm2 is practically

equivalent to fix this parameter at its best fit (determined KamLAND) Best fit ~ 0.021 is 1.2 sigma away from zero θ13 Then we have a second, independent hint of θ13 > 0

G.L Fogli, E. Lisi, A. Marrone, A.P., A.M. Rotunno PRL 101, 141801 (2008)

Origin: Tension between solar and KamLAND

For θ13

13 =

= 0

Solar and KamLAND prefer different values of θ12

12

No overlap at 1σ level For θ13

13 > 0

Solar data prefer higher θ12

12

KamLAND prefers lower θ12

12

Disagreement reduced*

*See also Balantekin and Yilmaz, J. Phys. G. 35, 075007 (2008)

θ13

13 >

> 0 θ13

13 =

= 0

23

Interplay is more evident in the plane spanned by the two mixing angles

Sol Solar ( r (S) an ) and KamLAND (K d KamLAND (K) displ ) display di y differen ent c t correl elatio ions ns.

• This r

is ren ende ders their c s their combina mbinatio ion pa n particu cularly sensi y sensitive t e to

• θ13

13

24

Hig High-E sol h-E solar r (adi diabatic MSW) c MSW) KamLAND KamLAND (vacuu cuum) Di Differen ent r t rel elative sig e sign f n for ( r (θ12

12,

, θ13

13) in P

) in Pee

ee

• f Sol
• f Solar (h

r (hig igh-E do h-E domina inated) vs KamLAND d) vs KamLAND

Origin of the different correlations +

• scil

cillatio ion pha n phase se

25

SNO-III

• Central value lower than before
• Error reduced when combined

be best fit of t fit of θ12

12 a

at a t a sl slig ightl tly y lower v value ue ran ange a e allowe wed f d for r θ12 app ppreci ciabl bly y narrowed

SNO-II 2005 2008

• Apparently a small change

but big en t big enoug ugh t h to g

• give

r rise t ise to a sig

• a signific

ifican ant tension w with KamLAND th KamLAND

26

How the tension increased in the past year

Difficult to think to an anomalous statistical fluctuation

SNO-II (2005) SNO-III (2008)

bo

both a th alr lrea eady g y good …an d …and… n d… now e w excel ellen ent t

1) “internal” consistency among SNO (CC,NC) and SK (ES) 2) consistency among NC (now higher than before) and Solar Model

27

The S+K hint is supported by other analyses

Hin Hint f t found a d at t ~1.5 sigma .5 sigma Altho thoug ugh wea h weak, the S+K h k, the S+K hin int ha t has a c s a clea ear o r orig igin an in and the d there is a e is agreem eemen ent amo among in g inde depen enden dent ana t analyse

• yses. The

herefore i e its s ts statu tus c s can be c an be consi nside dered mo d more rob

• bust i

t if c f compa mpared t d to the a

• the atmo

mosphe spheric o c one

Sc Schwe hwetz tz, , Tortol

• la,

a, Valle, a , arXi rXiv:0808. v:0808.201 2016 [ 6 [he hep-ph] p-ph]

28

Ge Ge, Giu , Giunti an i and Liu , a d Liu , arXi rXiv:0810 v:0810.54443 [ .54443 [he hep-ph ] p-ph ]

Hin Hint f t found a d at t ~1.2 sigma 2 sigma

The global hint (2008)

Co Combin mbinin ing a g all da l data we fin we find an o d an overall l preferenc ence f e for r θ13

13>0 a

>0 at t 1.6 sigma (90% CL)

Present solar and KamLAND data prefer θ13>0

• nly in combination,

but in future, they may disclose the preference even taken separately

29

Status of the electron neutrino mixing (2008)

~ 0.307 ~ 0.667 ~ 0.016 30

G.L Fogli, E. Lisi, A. Marrone, A.P., A.M. Rotunno PRL 101, 141801 (2008)

31

Hint n. III Date: February 2009 Data: MINOS (νe appearance)

Febr brua uary 20 y 2009 09, Fir First da t data fr a from MINOS m MINOS νe a app ppea earanc ance e

θ13

13 = 0

disf disfavored a d at t ~ 90% 90% The c he col

• llabo

boratio ion, p , pruden dentl tly, do does n s not g t give pa e particu cular r empha emphasis t sis to th

• this f

is fact. Ho Howe wever, c , combin mbinin ing their g their resu sults w ts with o th ours… s…

32

• M. Sanchez, Talk at FNAL Feb 27th

0.9 σ

σ

1.2 σ

σ

1.6 σ

Hint 1σ range

Hin Hints of ts of θ13

13>0

>0 St Statu tus a s as of J s of July 20 y 2009 09

~ 2 σ σ

33 See Fogli et al. arXiv:0905.3549 [hep-ph]

Fu Futu ture p e persp spectives

34

35

What we may expect from Solar and KamLAND ?

SNO

Lo Low en w energy thr y threshold ana shold analysis u ysis unde derwa

• way. CC an

CC and NC s d NC statis istics incr cs increa ease sed. d.

• New da

ew data e a expected.

• d. Fu

Furthe thermo more, i , it ha t has been n s been noted tha d that in a m t in a multi-r i-rea eactor se setu tup o p one e e expects pa ts partial c l canc ancel ellatio ions of r ns of ran ando dom dis m distr trib ibuted e d errors s (Dju (Djurci cic e c et a t al., 0808.07 ., 0808.0747 [ 7 [he hep-e p-exp]). ]). P Possibl ssible impa e impact o t on the of n the of “S+K h S+K hin int”. t”.

KamLAND The future (qualitative) Borexino

New l ew low-en

• energy m

y mea easu surem emen ents sho ts should be ld become incr e increa easin singl gly r y rel elevan ant… t…

Sol Solar r ν’s s alone ma may be of help y be of help

Hig High-E h-E

ma matter do domina inated

Low-E beha behavio vior is ana r is analogous t s to tha

• that o

t one p e probe

• bed in KamLAND

d in KamLAND, so , so we e we expect a sim t a simil ilar in r interpl play amo y among h g hig igh-E an h-E and l d low-E da

• E data

a

+

• Lo

Low-E

• E

averaged “v “vacuu cuum-l

• lik

ike”

• 36

37

Wha hat we ma t we may e y expect fr t from MINOS ? m MINOS ?

θ13>0 confirmed at ~ 2 sigma

a competitve upper bound

…or…

Data with doubled statistic already taken, analysis underway

• M. Sanchez, Talk at FNAL Feb 27th

The two “extreme” cases

Wha hat we ma t we may e y expect fr t from rea eactor an r and a d accel elerator sea r searche hes ? s ?

Adapted from Huber et. al JHEP 0605, 072 (2006) * Global analysis 2009 Global analysis 2009 38

The present hints will be subject to verification in the near future

Co Conc nclusio sions ns

• All da

l data fit w a fit with thin the s in the stan anda dard 3 d 3ν fr fram amew ework.

• Gl
• Glob
• bal p

l preferenc ence n e now a w at the 2 sigma l t the 2 sigma level (95% C el (95% C.L.) .L.)

• Thr

hree in ee inde depen enden dent an t and c d convergin ing h g hin ints of ts of θ13

13>0

>0.

• Ea

Each of them ma h of them may e y evol

• lve in the n

e in the nea ear fu r futu ture.

39

• Onl

Only fu y futu ture de e dedi dicated e d experim imen ents w ts wil ill de l determin ine i e if the f the p presen sent h t hin ints a ts are a m e a mere s e statis istical fluc l fluctua tuatio ion o n or r the they r y represen sent the fir t the first sig t signs of an em ns of an emergen ent sig t signa nal.

40

41

Back-up slides

42

SK-I data SK-II data

(zenith distributions from Takenaga thesis, 2008)

Trend from SK-I to SK-II:

Sub-GeV electron excess persists in both phases I and II Conversely, slight excess of upgoing MGe present in SK-I but not in SK-II

… however, going to SK-III …

This downward MGe fluctuation may disfavor θ13 > 0 (as noted by Maltoni and Schwetz)

(SK-III data, from

• J. Raaf at Neutrino 2008)

43

…together with a persisting excess of SGe data! …in SK-III data, a slight excess of upgoing MGe seems to be back …