# Full text of "3+1 and 3+2 Sterile Neutrino Fits"

## See other formats

EURDNU-WP6-11-37 arXiv:1107.1452v3 [hep-ph] O (N (N 3+1 and 3+2 Sterile Neutrino Fits Carlo Giuntli] INFN, Sezione di Torino, Via P. Giuria 1, 1-10125 Torino, Italy Marco LavedeiO] Dipartimento di Fisica "G. Galilei", Universttd dt Padova, and INFN, Sezione di Padova, Via F. Marzolo 8, 1-35131 Padova, Italy (Dated: September 5, 2011) We present the results of fits of short-baseline neutrino oscillation data in 3+1 and 3+2 neutrino mixing schemes. In spite of the presence of a tension in the interpretation of the data, 3+1 neutrino mixing is attractive for its simplicity and for the natural correspondence of one new entity (a sterile neutrino) with a new effect (short-baseline oscillations). The allowed regions in the oscillation parameter space can be tested in near-future experiments. In the framework of 3+2 neutrino mixing there is less tension in the interpretation of the data, at the price of introducing a second sterile neutrino. Moreover, the improvement of the parameter goodness of fit is mainly a statistical effect due to an increase of the number of parameters. The CP violation in short-baseline experiments allowed in 3+2 neutrino mixing can explain the positive i/jj — ^ i>e signal and the negative f^ — ^ Ve measurement in the MiniBooNE experiment. For the CP-violating phase we obtained two minima of the marginal x^ close to the two values where CP-violation is maximal. PACS numbers: 14.60.Pq, 14.60.Lm, 14.60.St > o X I. INTRODUCTION The recent agreement of MiniBooNE antincutrino data [ij with the short-baseline i^^ — ?> Pe oscillation signal ob- served several years ago in the LSND experiment Q] has opened an intense theoretical and experimental activity aimed at the clarification of the explanation of these ob- servations in a framework compatible with the data of other neutrino oscillation experiments. Several short- baseline neutrino oscillation experiments did not observe neutrino oscillations and their data constraint the in- terpretation of the LSND and MiniBooNE antineutrino signal. However, there are other positive indications of short-baseline neutrino oscillations that may be taken into account: the reactor antineutrino anomaly [3|, in favor of a small short-baseline disappearance of i>e, the Gallium neutrino anomaly |3l4l2l|. in favor of a short- baseline disappearance of i^e, and the MiniBooNE low- energy anomaly [3, 0, m, 113 ■ ^^ ^^is paper we consider only the reactor antineutrino anomaly, by taking into ac- count the new calculation of reactor antineutrino fluxes in Ref. [1^. We leave the discussion of the effects of the more controversial Gallium anomaly and MiniBooNE low-energy anomaly to a following article |14l |. The results of solar, atmospheric and long-baseline neutrino oscillation experiments led us to the current standard three-neutrino mixing paradigm, in which the three active neutrinos Vg, v^^, v^ are superpositions of three massive neutrinos v\^ 1/2, ^'3 with respective masses TTii, 7712, "^3- The measured solar (SOL) and atmospheric * |giunti@to.infn.it[ also at Department of Theoretical Physic University of Torino, Italy t ,laveder@pd. infn.it, (ATM) squared-mass differences can be interpreted as A777|oL = ^™2i = (7-6 ± 0.2) X lO"''^ eV^ [15] , (1) AnzirpM — \Am'^ 311 = 2.32lJ];J|xlO-^eV^ M, (2) with Arn^ = m,f, — 777^. The completeness of the three-neutrino mixing paradigm has been challenged by the LSND [2| and Mini- BooNE [l| observations of short-baseline 7>^ — )■ De transi- tions at different values of distance (L) and energy (E), but approximately at the same L/E. Since the distance and energy dependences of neutrino oscillations occur through this ratio, the agreement of the MiniBooNE and LSND signals raised interest in the possibility of exis- tence of one or more squared-mass differences larger than about 0.5 eV, which is much bigger than the values of A777gQL and Arn^j,^. Hence, we are lead to the exten- sion of three-neutrino mixing with the introduction of one or more sterile neutrinos which do not have weak in- teractions and do not contribute to the invisible width of the Z boson [I3|. In this paper we consider the simplest possibilities: 3+1 mixing with one sterile neutrino and 3+2 mixing with two sterile neutrinos. The existence of sterile neutrinos which have been ther- malized in the early Universe is compatible with Big- Bang Nucleosynthesis data [I3, [l^ , with the indication however that schemes with more than one sterile neutrino are disfavored [20] , and cosmological measurements of the Cosmic Microwave Background and Large-Scale Struc- tures if the neutrino masses are limited below about 1 eV |2ll - [25J | . Therefore, in this paper we consider squared- mass differences smaller than 10 eV^. Am^i 1^4 Amli "normal" 1^4 "inverted" ^1,2,3 FIG. 1. Schematic description of the two possible 3+1 schemes that we are considering, taking into account that lAmiil < |Amii|< |Amli|. II. 3+1 NEUTRINO MIXING In this section we consider the simplest extension of three-neutrino mixing with the addition of one massive neutrino. In such four-neutrino mixing framework the flavor neutrino basis is composed by the three active neu- trinos Ve, i^fj., Vt and a sterile neutrino Vg. So-called 2+2 four-neutrino mixing schemes are strongly disfavored by the absence of any signal of sterile neutrino effects in solar and atmospheric neutrino data [26| . Hence, we consider only the so-called 3+1 four- neutrino schemes depicted in Fig. [1] Since the inverted scheme has three massive neutrinos at the eV scale, it is disfavored by cosmological data [2l|, [22| over the normal scheme. In both 3+1 schemes the effective flavor tran- sition and survival probabilities in short-baseline (SBL) experiments are given by i?SBL_=sin2 2^„.sin2('^^ 'afi i i.SBL_^ ^ 1 _ 3i^2 2^^^ gi^2 /' AmliL A.E (a^/3), (3) (4) for a, /3 = e, /x, r, s, with sin2 2^,^=4|(7„4r|(704| sin2 2^„„=4|;7„4|'(l- |C/c.4p) (5) (6) 4. CP violation cannot be observed in SBL oscilla- tion experiments, even if the mixing matrix con- tains CP-violation phases. In other words, neutri- nos and antineutrinos have the same effective SBL oscillation probabilities. Before the recent indication of an antineutrino v^,_ — >■ v^, signal consistent with the LSND antineutrino signal, the MiniBooNE collaboration published the results of neu- trino data which do not show a corresponding v^ — >■ v^ signal 27[ . This difference between the MiniBooNE neu- trino and antineutrino data may be due to CP violation. The absence of any difference in the effective SBL oscil- lation probabilities of neutrinos and antineutrinos in 3+1 four-neutrino mixing schemes implies that these schemes cannot explain the difference between neutrinos and an- tineutrino oscillations observed in the MiniBooNE exper- iment. Moreover, the dependence of all the oscillation amplitudes in Eqs. ([5]) and ([B]) on three independent ab- solute values of the elements in the fourth column of the mixing matrix implies that the amplitude of z/^ -^ Vf. transitions is limited by the absence of large SBL disap- pearance of Uf, and v^ observed in several experiments. The results of reactor neutrino experiments constrain the value |C/e4p through the measurement of sin^2z9ee- The calculation of the reactor Uf. flux has been recently improved in Ref. [l3J , resulting in an increase of about 3% with respect to the previous value adopted by all exper- iments for the comparison with the data (see Ref. |28|). Since the measured reactor rates are in approximate agreement with those derived from the old Ve flux, they show a deficit with respect to the rates derived from the new i/g flux. This is the "reactor antineutrino anomaly" ^3^, which is quantified by the value -'*'rcactor anomaly — U.y4u it U.UZ4 (7) Therefore: for the average of the ratios of measured event rates and those expected in absence of i^e transformations into other states. We considered the integral rates of the Bugey-3 ^, Bugey-4 [30|, R0VN091 [3i|, Gosgen ^ and Krasnoyarsk [33| short-baseline reactor antineutrino experiments using the information in Table II of Ref. 3]j. Hence, the reactor antineutrino anomaly is a 2.2a in- dication that there is a small short-baseline disappear- ance of v^ which may correspond to the i>^ — t- v^ sig- nal observed in the LSND and MiniBooNE experiments. 1. All effective SBL oscillation probabilities depend only on the absolute value of the largest squared- mass difference Arnl^. 2. All oscillation channels are open, each one with its own oscillation amplitude. 3. The oscillation amplitudes depend only on the ab- solute values of the elements in the fourth column of the mixing matrix, i.e. on three real numbers with sum less than unity, since the unitarity of the mixing matrix implies ^^ |t^a4p = 1 ^ We do not use the two rates of the Savannah River experiment H in Table II of Ref. Q, RsRP I = 0.952 ± 0.006 ± 0.037 and -RsRP n = 1-018 ± 0.010 ± 0.037 because they are about 5.5o- apart, taking into account that their difference 0.066 ± 0.012 is independent of the correlated systematic uncertainty (0.037). Such a large difference cannot be due to neutrino oscillations averaged over the whole energy spectrum, because the two mea- surements have been done at distances which are not different enough (18 m and 24 m). We also do not use the ROVN088 3^ rates in Table II of Ref. 3 , because the correlation with Bugey-4 and R0VN091 is not clear. 10 > CD 10" 10" 99% C.L. Bugey-3(1995) Bugey-4 (1994) + Rovno (1991) Gosgen (1986) + ILL (1995) Krasnoyarsk (1994) 10 10" 10"' > CD 10" 10" /^^ - 99% C.L. CDHSW(1984):v^, , , 10" 10" sin^2-&, ^^ FIG. 2. Exclusion curves obtained from the data of reactor Ue disappearance experiments (see Ref. [3]). However, the De disappearance is small and large val- ues of sin 2dee are constrained by the exclusion curves in Fig. [2] (as in Ref. \3\ , the Bugey-3 exclusion curve has been obtained by fitting the three integral rates measured a,t L = 15,40,95m and the 40m/15m spectral ratio in Fig. 15 of Ref. [29|). Since values of |C/e4p close to unity are excluded by solar neutrino oscillations (which require large \Uei\ \Ue2?) for small sin 2t9pp we have sin"' 2i?p 41 1/,, (8) The value of sin 21?^^ is constrained by the curves in Fig. 131 which have been obtained from the lack of Vj disappearance in the CDHSW from the requirement of large |?7;iilf.+ |t/„,2p + |t/„aP for atmospheric neutrino oscillations [3 small and FIG. 3. Exclusion curves obtained from the data of the CDHSW u^ disappearance experiment [3y], and from at- mospheric neutrino data (extracted from the analysis in Ref. [32 shown in Fig. [S] the constraint becomes stronger by in- cluding the data of the KARMEN 0, NOMAD [4l| and MiniBooNE neutrino [27[ experiments, which did not ob- serve a short-baseline i^^ -^ v^. signal. Since the parame- ter goodness-of-fit [43] is 6 x 10~^, 3-1-1 schemes are dis- favored by the data. This conclusion has been reached recently also in Refs. [33, [43l-[45l a nd confirms the pre- MiniBooNE results in Refs^UlM [H, SI- However, in spite of the low value of the parameter goodness-of-fit it is not inconceivable to refuse to reject the 3-f 1 schemes for the following reasons: 1. It is the simplest scheme beyond the standard three-neutrino mixing which can partially explain the data. sin-^ 2i?, 4|C/, p4| (9) From Eqs. (0, ([8]) and (0, for the amplitude of v^ i/g transitions we obtain [38l , [39| sin^ 2??p 1 sin^ 2z9pp sin^ 2?9,, (10) Therefore, if sin^ 2^ee and sin^ 2i5^^ are small, sin^ 2^ep is quadratically suppressed. This is illustrated in Fig. [H where one can see that the separate effects of the con- straints on sin^ 2i?ee and sin^ 2^^^ exclude only the large- sin^ 2i9e^ part of the region allowed by LSND and Mini- BooNE antineutrino data, whereas most of this region is excluded by the combined constraint in Eq. (ITUl) . As 2. It corresponds to the natural addition of one new entity (a sterile neutrino) to explain a new effect (short-baseline oscillations). Better fits of the data require the addition of at least another new entity (in any case at least one sterile neutrino is needed to generate short-baseline oscillations). 3. The minimum value of the global ^ is rather good: Xmin = 100.2 for 104 degrees of freedom. Note also that 3+1 schemes are favored with respect to 3+2 schemes by the Big-Bang Nucleosynthesis limit iVcff < 4 at 95% C.L. obtained in Ref. [23|. Therefore, we consider the global fit of all data in 3+1 schemes, which yields the best-fit values of the oscillation param- eters listed in Tab. [Tj 10 > CD 10" 10" 99% C.L Reactors CDHSW + Atm Disappearance LSND + MBv \ J \ 10" 10" 10"-= s/n^2de 10" 10 > CD 10" 10" 99% C.L. MBv + LSNDv Dis + KAR + NOM + MBv 10' 10" 10"-^ s/n^2de 10" FIG. 4. Exclusion curves in the sin 2i9e(j-Am4i plane ob- FIG. 5. Exclusion curve in the sin I'&^^-^m^^ plane ob- tained from the separate constraints in Figs. [2] and [3] (blue dashed line and green dotted line) and the combined con- straint given by Eq. (|10l) (red solid line) from disappearance experiments (Dis). The regions allowed by LSND and Mini- BooNE antineutrino data are delimited by dark-blue long- dashed lines. tained with the addition to the disappearance constraint in Fig.Hof the constraints obtained from KARMEN [il (KAR), NOMAD 41^ (NOM) and MiniBooNE neutrino 27| (MBi^) data (red solid line). The regions allowed by LSND and Mini- BooNE antineutrino data are delimited by dark-blue long- dashed lines. Figures [6] and [7] show the allowed regions in the sin^ 2^e -Am|i, sin^ 2??p -Arn|;^ and sin^ 2??„ -Am^i planes and the marginal Ax s for tym\^^ sin 2i?e^, sin^ 2i^pp and sin^2t?,,,, . The best-fit values of the os- cillation amplitudes are sin^ 2i?eA. = 0.0023 , sin^ 2^ee = 0.098 , sin^ 2i?„„ = 0.091 . (11) (12) (13) 3+1 3+2 Xmin 100.2 91.6 NDF 104 100 GoF 59% 71% Am^i [eV"] 0.89 0.90 |f/e4p 0.025 0.017 \UA^ 0.023 0.019 Ami [eV^] 1.61 \U.6\' 0.017 \u,,f 0.0061 V 1.5l7r Axpg 24.1 22.2 NDFpG 2 5 PGoF 6 X 10"'' 5 X lO""* TABLE I. Values of x^ , number of degrees of freedom (NDF) , goodness-of-fit (GoF) and best-fit values of the mixing pa- rameters obtained in our 3+1 and 3+2 fits of short-baseline oscillation data. The last three lines give the results of the parameter goodness-of-fit test [431: Axpgi number of degrees of freedom (NDFpg) and parameter goodness-of-fit (PGoF). From Fig. [6] one can see that the allowed regions are compatible with those allowed by appearance data (the D^ -^ De data of the LSND \2], KARMEN [H and Mini- BooNE [1 [experiments and the i/^j, — )■ Ve data of the NOMADgH and MiniBooNE [13 experiments) and are slightly pushed towards the left by the disappearance constraints. Future experiments aimed at checking the LSND [1] and MiniBooNE [l| i>^ -^ Pe oscillation signal (as those in Refs. |48l - [5(]| |) should aim at exploring these regions. Figure [7] shows that the allowed regions in the sin 2^ee~Am|]^ and sin 2?9^^-Am|]^ planes lie just on the left of the disappearance constraints, as expected. From the left panel in Fig. [7] one can see that the al- lowed regions in the sin 2'dee~^iTi^i plane are compati- ble with the area indicated by the Gallium anomaly [i2| ■ The allowed region around the best- fit point and the isolated region at Am^i ~ GeV'^are also compatible with the recent results in Ref. [51]. If the 3+1 neu- trino mixing scheme is realized in nature, future exper- iments searching for short-baseline v^ disappearance (as <l > 03 s < ^o-' m 3+1 68.27% C.L(1o) 90.00% C.L. k 95.45% C.L. (2o) 99.00% C.L. 99.73% C.L. (So) 10' 10" 10" 10" e\i 8 10 those in Refs. [Ill, |48|, ISfl, l52|-l57| ) should find a disappear- ance compatible with the reactor antineutrino anomaly in Eq. ([7]). Future experiments searching for short-baseline Vfi disappearance (as those in Refs. [4^, [S^) should find a disappearance just below the current bound, for 0.4 < Amli < 7eV^. Short-baseline v^ and f^ disap- pearance can have observable effects, respectively, also in solar neutrino experiments l59ll6Cl, long-baseline neu- trino oscillation experiments [6ll |62[ and atmospheric neutrino experiments [63, ^gg . III. 3+2 NEUTRINO MIXING The CP-violating difference between MiniBooNE neu- trino and antineutrino data can be explained by in- troducing another physical effect in addition to a ster- ile neutrino: a second sterile neutrino in 3-1-2 schemes [l^iSliiJe^Jel, non-standard interactions '4$, CPT violation [4Jj [Z0|- I^ ^his section we discuss the pos- sibility of 3-1-2 neutrino mixing according to the possi- ble schemes illustrated schematically in Fig. [51 The in- verted and perverted schemes have been called, respec- tively, 2-1-3 and 1-1-3-1-1 in Ref. [71|. Since the inverted and perverted schemes have three or four massive neutri- nos at the eV scale, they are disfavored by cosmological s/n 2iJp, sin'^Z'du FIG. 6. Allowed regions in the sin'^ 2i}e^-Amii plane and marginal A^^'s for sin^ 2i9eM and Am|i obtained from the V global fit of all the considered data in 3+1 schemes. The best- fit point corresponding to Xmin is indicated by a cross. The isolated dark-blue dash-dotted contours enclose the regions allowed at 3a by the analysis of appearance data (the i/fj, — >■ i/e data of the LSND [2, KARMEN ^ and MiniBooNE [J experiments and the i*^ — >■ Ue data of the NOMAD [41[ and MiniBooNE [13] experiments). Allowed regions in the sin'^ 2i?ee-Am|i and -Am4i planes and marginal Ax 's for sin 2'Oee and sin'^ 2i?,t^ obtained from the global fit of all the considered data in 3-1-1 schemes. The best-fit point corresponding to Xmin is indicated by a cross. The line types and color have the same meaning as in Fig. [S] The isolated dark-blue dash- dotted lines are the 3a exclusion curves obtained from reac- tor neutrino data in the left plot (corresponding to the blue dashed line in Fig.|4]) and from CDHSW and atmospheric neu- trino data in the right plot (corresponding to the green dotted line in Fig. [4]). The isolated dark-red long-dashed lines delimit the region allowed at 99% C.L. by the Gallium anomaly [l3 ]. data [21], [22| over the normal scheme. Note also that all 3-1-2 schemes are disfavored by the Big-Bang Nucleosyn- thesis limit Ncs < 4 at 95% C.L. obtained in Ref. [20J . In 3+2 schemes the relevant effective oscillation prob- abilities in short-baseline experiments are given by i?_T<-,=4|C/,4| |C/e4psin2(/)4i 4|C/^5nC/e5psin2 05i + 8\Ufj,iUe4:Ufj,5Ue5\ siu (/>4i siu 05i COs( (+) ^54 - V) , (14) pSBL = 1-4(1 |C/a4| |2 „• 2 Ic/csH X (|C/a4|^ sin^ 041 + \Uc5f sin^ 05i) -4|C/„4r|C/a5rsin2 for a, /3 — e, fj,, with ^54 : (15) 0,, = AmlL/AE , r; = a.rg[U:,U,iUe5U;,] . (16) Note the change in sign of the contribution of the CP- violating phase ry going from neutrinos to antineutrinos, which allows us to explain the CP-violating difference Att^Ij Am',, Am^i '^.2.3 " normal" Ami n Amji ^^1,2,3 "inverted" "perverted" FIG. 8. Schematic description of the three possible 3+2 schemes that we are considering, taking into account that |Amii| < |Amii| < |Amli| < |Amii|. In the perverted scheme the identification of the labels k and j is chosen in order to satisfy the inequality |Ar7i|i| < |Am|i|. between MiniBooNE neutrino and antineutrino data. In our analysis we consider Am|]^ > and Arn^i > 0, with Amgj^ > Am|j^, which implies ^rn^^ > 0. These as- sumptions correspond to the normal scheme in Fig. [51 which is favored by cosmological data, as noted above. In any case, the results of our analysis can be applied also to the inverted scheme (Am|j^ < 0, AyTigj^ < 0, ATO54 < 0) with the change 77 — i- 27r — 77. Instead the perverted schemes, which have been considered in the fit of Ref. [431 , require a separate treatment because in these schemes IATO541 lAmiil IATO44I. For simplicity we do not consider them here, because they are strongly dis- favored by cosmological data, having four massive neu- trinos at the eV scale. Figures [HHH] show the marginal allowed regions in two- dimensional planes of interesting combinations of the oscillation parameters and the corresponding marginal Ax^'s obtained in our 3-f2 global fit of the same set of data used in Fig. [5l The best-fit values of the mixing parameters are shown in Tab. [H The correlation of the allowed regions of r] and 4:\Ue4Ufj,4Ue5Ufj,5\ in Fig.[T2]is due to their presence in the last term in the effective j^^ -^ i^e oscillation probability in Eq. (|14p . The marginal Ax^ for ry has two minima close to the two values where CP-violation is maximal (77 = 7r/2 and 77 = 37r/2), in agreement with what we expected from the need to fit the positive Df^ — >■ D^ signal and negative v^ — ^ i^e measurement in the MiniBooNE experiment in the same range of L/ E. From Fig. [l^ one can also see that the marginal Ax^ for ?? is always smaller than the Ax^ — 7.8 corresponding to a negligi- bly small value of 4|C/e4[/^4[/e5C^^5| (this value is reached for 77 ~ O.Itt and around rj — n). Such a Ax^ is smaller than the difference of the x^ minima in the 3-1-1 and 3-1-2 schemes because the condition for 4|[/e4C/^4C^e5C^^5| to vanish requires that only one of C/gs and C/^5 vanishes. In particular, if only U^^ is practically negligible, the re- actor antineutrino data can be fitted sligtly better than in 3-1-1 schemes, as already noted in Ref. \45^. 3+2 68.27% C.L. 90.00% C.L. 95.45% C.L. (20) 99.00% C.L. 99.73% C.L. (3o) Ami [eV Ax FIG. 9. Allowed regions in the Amii-Amli plane and cor- responding marginal Ax^'s obtained from the global fit of all the considered data in 3+2 schemes. The best-fit point corresponding to Xmin is indicated by a cross. The parameter goodness-of-fit obtained with the com- parison of the fit of LSND and MiniBooNE antineutrino data and the fit of all other data is 5 x 10^^. This is an improvement with respect to the 6 x 10^^ parameter goodness-of-fit obtained in 3+1 schemes. However, the value of the parameter goodness-of-fit remains low and the improvement is mainly due to the increased num- ber of degrees of freedom, as one can see from Tab. [D The persistence of a bad parameter goodness-of-fit is a consequence of the fact that the j?^ — > De transitions ob- served in LSND and MiniBooNE must correspond in any neutrino mixing schemes to enough short-baseline dis- appearance of i^e and 7/^ which has not been observed and there is an irreducible tension between the LSND and MiniBooNE antineutrino data and the KARMEN antineutrino data. The only benefit of 3+2 schemes with respect to 3+1 schemes is that they allow to explain the difference between MiniBooNE neutrino and antineu- trino data through CP violation. In fact, neglecting the MiniBooNE neutrino data we obtain Ax|q = 16.6 with PGoF = 3 X 10"-* in 3+1 schemes and AxpQ = 20.4 with PGoF = 1 X lO^'' in 3+2 schemes. In this case Axpq is even lower in 3+1 schemes than in 3+2 schemes! The tension between LSND and MiniBooNE antineu- trino data and disappearance, KARMEN, NOMAD and MiniBooNE neutrino data is illustrated in Fig.[T31 which is the analogous for 3+2 schemes of Fig. \5\ in 3+1 schemes. In practice, in order to show the tension in a two-dimensional figure we have marginalized the x^ over all the other mixing parameters, including the two Am'^'s. 10^ 10 4\U,/{1-\U,/) 10" 10" lO" 10" FIG. 10. Allowed regions in the 4|f/e4p(l-|C/e4r)-Am|iand4|(7p4r(l- 4|t/e4p |[/^4r-AmIi, -Am|i planes and marginal Ax^'s for 4|l7e4p|t/^4p, 4|[/e4|^(l - \Ue4\'^) and 4|C/^i4p(l — |(7^4p) obtained from the global fit of all the con- sidered data in 3+2 schemes. The line types and color have the same meaning as in Fig. (9] The best-fit point correspond- ing to Xmin is indicated by a cross. FIG. 11. Allowed regions in the 4\Ue5\ \U^5\ -Amli, 4|C/e5p(l-|f/e5p)-Amii and 4|[7p5 P (1" If/^s H-Amli planes and marginal Ax^'s for 4|f/e5p|f/^5p, 4|C/e5p(l - It/esl") and 4|f/^5p(l — |C/^t5p) obtained from the global fit of all the con- sidered data in 3-1-2 schemes. The line types and color have the same meaning as in Fig. [9] The best-fit point correspond- ing to Xmin is indicated by a cross. The results of our 3-1-2 global fit are in reasonable agreement with those presented in Ref. [4^. There is a discrepancy in the location of the best-fit point in the Amli-Am'^i plane, but we obtain similar regions for the local x^ minima. Our allowed regions are larger than those presented in Ref. [43] ■ We think that such differ- ence is probably due to a different treatment of the spec- tral data of the Bugey-3 reactor experiment [22i] which cause the wiggling for Am^ < 1 eV^ of the disappear- ance limit in Fig. 3] and the exclusion curve in Fig. [S] Such wigghng is wider in Fig. 3 of Ref. [45], leading to deeper valleys of the x^ function and smaller allowed re- gions. The compatibility with cosmological data of the allowed regions in the Arn^j^-Arn^^ plane shown in Fig.[S] will be discussed in a separate article 72 1 (an interesting previous study was presented in Ref. [73|). 3+2 68.27% C.L. ( 90.00% C.L. 95.45% C.L. (2o} 99.00% C.L. 99.73% C.L. (3o} FIG. 12. Allowed regions in the r)-A\U eiU ^iU ez,U ^z,\ plane and corresponding marginal Ax^'s obtained from the global fit of all the considered data in 3-f2 schemes. The best-fit Figures [TOHII] show the allowed regions for the ampli- point corresponding to xiin is indicated by a cross. tudes of the oscillating terms in short-baseline z/^ — > Ve transitions and Vp and v^ disappearance, for which we 10" "-IE 10" 10" 10 -4 95%C.L. ■^; MBv + LSNDv J-^'^1- Dis + KAR + NOM + MBv y^ / j i 1 y — _|_LJ_LLLi^ i 1 Mini 1 L + I 1 1 1 1 1 1 1 1 1 \ jJ\ \ w 10" 10" 10" 10" ^\Ue,nu^4\ FIG. 13. Comparison of the 95% C.L. allowed regions in the 4|C/e4p|t/^4|^-4|f/e5p|t/M5|^ plane obtained from LSND and MiniBooNE antineutrino data on the right (green area) and disappearance, KARMEN, NOMAD and MiniBooNE neu- trino data on the left (red area). The corresponding best-fit points are indicated by crosses. have the best-fit values (7e4n(7^4|'= 0.0013, t/e5|'|t/M5|'= 0.00042, C/e4p(l-|C/e4n= 0.068, (7e5|'(l-|C/e5n= 0.068, t/^4|'(l-|f/Mp) = 0.076, t/^5|'(l-|t/M5n- 0.024. (17) (18) (19) (20) (21) (22) Comparingthe valuesof 4|C/e4p|C/^4p, 4|[/e4|2(l-|J7e4p) and 4|[/p4p(l — |t^/j4p) with those obtained in 3+1 mix- ing, given in Eqs pT|) - p^ . one can see that they are lower, but keep the same order of magnitude. In the fit of the data the smaller values of these amplitudes is due to the additional contribution of the amplitudes gener- ated by the mixing of i/g and v^ with 1/5 . (-) (-) i/n disappearance [4 IV. CONCLUSIONS In this paper we presented the results of fits of short- baseline neutrino oscillation data in 3+1 and 3+2 neu- trino mixing schemes. In the framework of 3+1 neutrino mixing schemes in Fig. [TJ we confirm the strong tension between LSND and MiniBooNE antineutrino data and disappearance, KAR- MEN, NOMAD and MiniBooNE neutrino data discussed recently in Refs. |37l. I43l445l | . Since however the minimum value of the global x^ is rather good, one may choose to consider as possible 3+1 neutrino mixing, which can partially explain the data, taking into account its sim- plicity and the natural correspondence of one new entity (a sterile neutrino) with a new effect (short-baseline os- cillations). Following this approach, we presented the re- sults of the global fit in 3+1 neutrino mixing, which leads to the determination of restricted allowed regions in the mixing parameter space which can be explored in future Si-i3'j^e disappearance [H El US IH-il and 58| experiments. We also presented a global fit in the framework of the 3+2 neutrino mixing schemes in Fig. [51 We have shown that the tension between LSND and MiniBooNE antineu- trino data and disappearance, KARMEN, NOMAD and MiniBooNE neutrino data is reduced with respect to the 3+1 fit, but is not eliminated (see Fig. [T3l) . Moreover, the improvement of the parameter goodness of fit with respect to that obtained in the 3+1 fit is mainly due to the increase of the number of oscillation parameters, as one can see from Tab. [T] Hence it seems mainly a statis- tical effect. The results of our 3+2 fit arc compatible with those presented recently in Ref. [45.] . but we obtain a differ- ent indication for the best fit (see Tab. H]). For the CP- violating phase we obtained two minima of the marginal X^ close to the two values where CP- violation is maximal. In conclusion, we think that our results are useful for the discussion of the interpretation of the current ex- perimental indications in favor of short-baseline neutrino oscillations and for the study of new experiments aimed at a clarification of the validity of these indications. [1] MiniBooNE, A. A. Aguilar-Arevalo et al, Phys. Rev. Lett. 105, 181801 (2010), arXi v: 1007.1150] [2] LSND, A. Aguilar et al, Phys. Rev. D64, 112007 (2001), hep-ex/010 4049 , [3] G. Mentio n et al, Phys. Rev. D83, 073006 (2011), ,arXiv:110i:2755| [4] J. N. Bahcall, P. I. Krastev, and E. Lisi, Phys. Lett. B348, 121 (1995),'hep-ph/9411414, [5] M. Laveder, Nucl. Phys. Proc. Suppl. 168, 344 (2007), Talk presented at the Workshop on Neutrino Oscillation Physics (NOW 2006), Otranto, Lecce, Italy, 9-16 Sep 2006. [6] C. Giunti and M. Laveder, Mod. Phys. Lett. A22, 2499 (2007), hep-ph/0610352 [7] C. Giu nti and M. Lave der, Phys. Rev. D77, 093002 (2008), rarXiv:0707.4593] [8] M. A. Acero, C. Giunti, and M. Laveder, Phys. Rev. D78, 073009 (2008), arXiv:0711.4222, [9] C. Giunti and M. Laveder, Phys. Rev. D80, 013005 (20091. . arXiv:0902. 1992 [42 [10] C. Giu nti and M. Lave der, Phys. Rev. D82, 053005 (20101. [axXrv:1005.4599l [43 [11] V. N. Gavrin, V. V. Gorbachev, E. P. Veretenkin, and B. T. Cleveland, (2010), arXiv:1006.2l03 [44 [12] C. Giu nti and M. Laveder, Phys. Rev. C83, 065504 (20111. larXiv:1 006.3244 [45 [13] T. A. Mue ller et al, Phys. Rev. CSS, 054615 (2011), larXiv:1101 .2663 [46 [14] C. Giunti and M. Laveder, (2011), In Preparation. [15] Super-Kamiokande, K. Abe et al, Phys. Rev. D83, [47 052010 (2011), arXiv;1010.0118 [16] MINO S, P. Adamson et al, Phys. Rev. Lett. 106, 181801 [ (20111. larXiv:1103.0340l [49 [17] ALEPH, DELPHI, L3, OPAL, SLD, LEP Electroweak Working Group, SLD Electroweak Group, SLD Heavy [50 Flavour Group, S. Schael et al., Phys. Rept. 427, 257 (2006), hep-ex/0509008, [51 [18] R. H. Cyburt, B. D. Fields, K. A. Olive, and E. SkiUman, [52 Astropart. Phys. 23, 313 (2005), |astro-ph/0408033, [19] Y. I. Izotov and T. X. Thuan, Astrophys. J. 710, L67 [53 (2010), arXiv:1001.4440': [20] G. Mangano and P. D. Serpico, Phys. Lett. B701, 296 [54 (2011), arXiv:1103.1261 [21] J. Hamann, S. Hannestad, G. G. Raflelt, I. Tamborra, [55 and Y. Y. Wong, Phys. Rev. Lett. 105, 181301 (2010), [a:rXiv:1006.527 6 [56 [22] E. Giusarma et al, Phys. Rev. D83, 115023 (2011), larXiv:1102 .4"774 [57 [23] J. R. Kristiansen and O. Elgaroy, (2011), arXiv: 1104.0704, [58 [24] Z. Hou, R. Keisler, L. Knox, M. Millea, and C. Reichardt, (2011), arXiv:1104.2333 [59 [25] A. X. Gonzalez-Morales, R. Poltis, B. D. Sherwin, and L. Verde, (2011), arXiv: 1106.5052 [60 [26] M. Maltoni, T. Schwetz, M. Tortola, and J. Valle, New J. Phys. 6, 122 (2004), hep-ph/0405172 [61 [27] MiniBooNE, A. A. Aguilar-Arevalo, Phys. Rev. Lett. 102, 101802 (20091. ■arXiv:0812. 2243. [62 [28] C. Bemporad, G. Gratta, and P. Vogel, Rev. Mod. Phys. 74, 297 (2002), hep-ph/0107277" [63 [29] Bugey, B. Achkar et al, Nucl. Phys. B434, 503 (1995). [30] Bugey, Y. Declais et al, Phys. Lett. B338, 383 (1994). [64 [31] A. Kuvshinnikov, L. Mikaelyan, S. Nikolaev, M. Sko- rokhvatov, and A. Etenko, JETP Lett. 54, 253 (1991), [65 in russian. [66 [32] CalTech-SIN-TUM, G. Zacek et al, Phys. Rev. D34, 2621 (1986). [67 [33] G. S. Vidyakin et al, Sov. Phys. JETP 71, 424 (1990). [34] Z. D. Greenwood et al, Phys. Rev. D53, 6054 (1996). [ [35] A. I. Afonin et al, Sov. Phys. JETP 67, 213 (1988). [36] CDHSW, F. Dydak et al, Phys. Lett. B134, 281 (1984). [69 [37] M. Mahoni and T. Schwetz, Phys. Rev. D76, 093005 (2007), arXiv:0705.0107 [38] N. Okada and O. Yasuda, Int. J. Mod. Phys. A12, 3669 [70 (1997), hep-ph/9606411 [39] S. M. Bilenky, C. Giunti, and W. Grimus, Eur. Phys. J. [71 CI, 247 (1998),|hep-ph/9607372 [40] KARMEN, B. Tm ibruster et al, Phys. Rev. D65, [72 112001 (2002),'hep-ex/0203021, [41] NOMAD, P. Astier et al, Phys. Lett. B570, 19 (2003), [73 hep-ex/0306037 M. Maltoni and T. Schwetz, Phys. Rev. D68, 033020 (2003), hep-ph/0304176 E. Akhmedov and T. Schwetz, JHEP 10, 115 (2010), arXiv: 1007.4171 C. Giunti and M. Laveder, Phys. Rev. D83, 053006 (2011), arXiv;1012.0267 J. Kopp, M. Maltoni, and T. Schwetz, (2011), ■arXiv:1 103.4570 S. M. Bilenky, C. Giunti, W. G rimus, and T. Sc hwetz, Phys. Rev. D60, 073007 (1999),"hep-ph/9903454[ M. Maltoni, T. Schwetz, M. A. Tortola, and J. W. F. Valle, Nucl. Phys. B643, 321 (2002), hep-ph/0207157 C. Rubbia, (2011), CERN-SPSC-2011-012; SPSC-M-773. S. K. Agarwalla and P. Huber, Phys. Lett. B696, 359 (20111, ■arXiv:1007.3228. S. K. Agarwalla, J. Conrad, and M. Shaevitz, (2011), arXiv:1105.4984' J. Conrad and M. Shaevitz, (2011), arXiv:1106.5"552] S. K. Agarwalla, P. Huber, and J. M. Link, JHEP 01, 071 (2010), arXiv:0907.3145. C. Giunti, M. Laveder, and W. Winter, Phys. Rev. D80, 073005 (2009),"arXiv:0907.5487' Nucifer Collaboration, A. Porta, J.Phys.Conf.Ser. 203, 012092 (2010). S. K. Agarwalla and R. S. Raghavan, (2010), larXiv:1011.4509l J. Vergados, Y. Giomataris, and Y. Novikov, (2011), larXiv:1103.5307l M. Pallavicini, (2011), Short-Baseline Neutrino Work- shop (SBNWU), 12-14 May 2011, Fermilab. SciBooNE-MiniBooNE, K. B. M. Mahn et al, (2011), arXiv:11 06.568_5 . C. Giun ti and Y . Li, Phys. Rev. D80, 113007 (2009), arXiv:0910.5856, A. Palazzo, Phys. Rev. D83, 113013 (2011), arXiv:1105.1705 MINOS, P. Adamson et al, Phys. Rev. Lett. 107, 011802 (2011), arXi v:1104.3922l D. Hernandez and A. Y. Smirnov, (2011), larXiv:1105.5 946 O. Peres and A. Smirnov, Nucl.Phys. B599, 3 (2001), hep-ph/0011054 H. Nunokawa, O. L. G. Peres, and R. Z. Funchal, Phys. Lett. B562, 279 (2003), hep-ph/0302039 S. Choubey, JHEP 12, 014 (2007), arXiv:0709.1937 S. Razzaque and A. Y. Smirnov, JHEP 07, 084 (2011), arXiv:1104.1390 M. Sorel, J. Conrad, and M. Shaevitz, Phys. Rev. D70, 073004 (2004), h ep-ph/030525 5 G. Karagiorgi et al, Phys. Rev. D75, 013011 (2007), hep-ph/0609177, G. Karagiorgi, Z. Djurcic, J. Conrad, M. H. Shae- vitz, and M. Sorel, Phys. Rev. D80, 073001 (2009), arXiv:0906.1997 C. Giunti and M. Laveder, Phys. Rev. D82, 093016 (2010), arXiv: 1010. 1395, S. Goswami and W. Rodejohann, JHEP 10, 073 (2007), arXiv:0706.1462 M. Archidiacono, N. Fornengo, C. Giunti, and A. Mel- chiorri, (2011), In Preparation. A. Melchiorri et al, JCAP 0901, 036 (2009), larXiv:0810.5133l