Status of Neutrino Oscillations and Future Prospects

Achim Geiser, II. Institut f. Experimtalphysik, Universitaet Hamburg

copies of the transparencies are available on request.

Summary:

The understanding of the mass pattern of the known elemantary particles is one of the main challenges in contemporary high energy physics. Neutrino Oscillations offer the opportunity to study the possibility that neutrinos have mass,and to explore the mass region of a few eV down to 10^-5 eV.
Massive neutrinos require an extension of the Standard Model, either via the introduction of light righthanded ("sterile") counterparts (Dirac-neutrinos), or via the assumption that neutrinos are their own antiparticles (violation of lepton number), and the addition of extra heavy neutrinos (Majorana-neutrinos).
The neutrino oscillation formalism is based on the assumption that , in complete analogy to the CKM matrix of the quark sector, the weak eigenstates of neutrinos are related to the mass eigenstates via a matrix with nonvanishing off-diagonal elements.
In the simplified case of two flavour oscillations (types a and b) in vacuum, the oscillation probability as a function of neutrino energy E and travelled distance L can then be expressed as

P (nu_a -> nu_b) = sin^2 2theta sin^2 (dm^2/4 L/E) (natural units)

Theta , the mixing angle between the two mass eigenstates, fixes the amplitude of the oscillations, and dm^2 = m1^2 - m2^2, the difference of the squared neutrino masses, determines the oscillation frequency. Obviously, oscillations can only occur if neutrino masses are nonzero.

Currently, experimental indications for the occurrence of neutrino oscillations exist from 3 different sources:

the atmospheric neutrino anomaly, and in particular the measurements of the Superkamiokande collaboration (see below).
the solar neutrino deficit (see also previous talk by Ingo Bloch), which can be interpreted either as vacuum oscillations between the sun and the earth or as matter enhanced oscilllations inside the sun.
a direct signal for nu_mu to nu_e oscillations from the LSND experiment at Los Alamos. This signal has not yet been confirmed by other experiments.

In addition, many experiments with negative results yield limits which constrain the possible oscillation parameter space. An example for such a result is the search for nu_mu to nu_tau oscillations with the NOMAD and CHORUS experiments at CERN. This search is based on the (non)observation of tau-neutrinos in a predominantly muon neutrino beam. Tau neutrinos would be detected through their charged current (CC) interactions yielding tau leptons in the final state. This is possible since the neutrino beam energy (typically 30 GeV) is above tau production threshold. Taus are identified either through the detection of the tau decay kink in nuclear emulsion (CHORUS) or through kinematic criteria based on the escaping final state neutrino(s )from tau decay (NOMAD). No signal has been observed so far. This limits the nu_mu - nu_tau oscillation parameters to sin^22theta < 4 x 10^-4 for large dm^2, and dm^2 < 0.8 eV^2 for maximal mixing.

In the remainder of this talk, emphasis is placed on the atmospheric neutrino signal. Atmospheric neutrinos mainly result from the chain:

primary cosmic ray interaction in atmosphere -> hadronic showers
pi -> mu nu_mu, mu -> e nu_mu nu_e

Ignoring the difference between particles and antiparticles (current atmospheric neutrino detectors do not measure the particle charge) one thus naively expects two muon neutrinos for each electron neutrino. \ Modifications to this simple prediction can be reliably estimated from MC simulations. The original atmospheric neutrino anomaly consisted in a significant deviation of the nu_mu to nu_e ratio observed in underground detectors from the predicted value.
In addition, neutrinos observed in these detectors originate from quite different distances: Neutrinos created in the atmosphere just above the detector have typically travelled about 20 km before being detected, while neutrinos coming from below originate from the other side of the earth, and have therefore travelled thousands of km. From pure geometery, the distance parameter L in the oscillation formula can therefore be expressed as a function of the zenith angle of the observed neutrino. The observation of a significant unexpected zenith angle dependance (suppression of upgoing muon neutrinos by a factor two with respect to downgoing ones) by the Super-Kamiokande collaboration constitutes the most stringent evidence for neutrino oscillations so far. This evidence is supported by less precise measurements from other experiments. The two-flavour nu_mu - nu_e oscillation hypothesis is excluded by limits obtained from the CHOOZ reactor experiment, and disfavoured by the Supaer-Kamiokande data themselves. Two flavour nu_mu - nu_tau oscillations are slightly preferred over two flavour nu_mu - nu_sterile oscillations, but in more complicated oscillation schemes both hypotheses are possible. (In Super-Kamiokande, most neutrino energies are below tau production threshold . Differences between tau and sterile neutrinos can therefore only be observed from neutral current interactions, including matter effects).

The three neutrino anomalies (solar, atmospheric, LSND), if interpreted as neutrino oscillations, yield three different values of dm^2. Since only two independent dm^2 values can be obtained from the known 3 neutrino flavours, it is not possible to simulataneously satisfy all indications in a 3-flavour scheme.There are two ways out.

Either one ignores one of the indications, for instance the one from LSND. Then 3 neutrino oscillation solutions are possible. The most popular ones are to invoke maximal nu_mu-nu_tau oscillations for atmospheriic neutrinos, and small angle matter enhanced nu_e-nu_mu oscillations (single maximal mixing) or one of the large angle mixing solutions (bimaximal mixing) for solar neutrinos. Other variants are possible.
Alternatively, one introduces additionl degrees of freedom, for instance one or more light sterile neutrinos. Then all indications can be accomodated in a multitude of different schemes, some of which also include atmospheric nu_mu-nu_sterile oscillations.

The only way to distinguish between these many possible scenarios is to add additional evidence from new experiments. For solar neutrinos, several new projects are under way, which will not be discussed here. LSND will be checked by the Mini-Boone experiment at Fermilab, starting in 2002.
Concerning atmospheric neutrinos, 3 main questions should be addressed within the next few years.

confirmation of the oscillation hypothesis (other non-standard model explanations are possible)
nu_mu-nu_tau or nu_mu-nu_sterile oscillation?
subdominant nu_mu-nu_e contribution?

The oscillation hypothesis can be confirmed by actually observing a sinusoidal oscillation pattern, or by proving the appearance of a final state, for instance nu_tau, which can only be obtained via neutrino oscillations. Since Super-Kamiokande will only yield marginal results concerning these two points, new projects are needed. Two approaches are possible.

new atmospheric neutrino projects (MONOLITH)
experiments on long baseline neutrino beams (K2K, MINOS, OPERA, ICANOE)

In both of these approaches, three complementary experimental strategies can be used.

measurement of nu_mu disappearance by comparing neutrino rates from far and near sources
an indirect measurement of nu_tau appearance through the corresponding modification of the apparent NC/CC ratio
a direct nu_tau appearance measurement

The MONOLITH project is a new experiment to measure atmospheric neutrinos in the Gran Sasso underground laboratory in Italy. It is a 34 kton magnetized iron tracking calorimeter consisting of 8 cm iron plates interleaved with active detectors (RPC and/or scintillator). Its better efficiency and resolution for high energy neutrino events with respect to Super-Kamiokande allows the resolution of the sinusoidal oscillation pattern, in particular the measurement of the first oscillation minimum. The sensitive range after 4 years of data taking fully covers the region allowed by Super-Kamiokande. This allows the confirmation of the oscillation hypothesis and the exclusion of alternative solutions like neutrino decay. Simultaneously, the measurement of the oscillation parameters can be significantly improved. If approved in 2000, data taking could start in 2003.

Several long baseline neutrino projects are under way. The K2K project in Japan involves a beam from the KEK laboratory to the Super-Kamiokande detector 250 km away (typical neutrino energy: 2 GeV). Data taking has started this year (1999) and one event has been observed in Super-Kamiokande so far. K2K will yield a significant check of the atmospheric neutrino evidence in the nu_mu disappearance mode if dm^2 is above 3 x 10-3 eV^2. If dm^2 is lower, the result might be marginal.
The MINOS project involves neutrino beam from Fermilab to the Soudan mine in Minnesota about 730 km away. The concept is similar to K2K, but the higher beam energy and intensity yields a higher event rate which allows a full coverage of the Super-Kamiokande allowed range. Like in K2K, both a near and far detector are forseen.The MINOS far detector is a 5.4 kton magnetized iron tracking calorimeter, smaller in size but more densely instrumented than MONOLITH, and optimized for beam rather than atmospheric neutrinos. Both the disappearance and the indirect appearance method will be exploited. The project is approved and data taking is scheduled to start in 2003. Several beam options are being considered, with average neutrino energies ranging from 5 to 15 GeV. For full coverage of the Super-Kamiokande allowed region, a low beam energy is preferred.

Finally, the project of a long baseline beam from CERN to Gran Sasso (732 km) is in an advanced stage of planning, and is scheduled for approval in december 1999, and operation in 2005. The goal of this project is the direct detection of nu_tau appearance using methods similar to the ones exploited by the NOMAD and CHORUS experiments. In order to be above tau detection threshold, a high energy beam (typical neutrino energy: 15 GeV) is required. Two experiments are being discussed in this context. The OPERA experiment would detect tau particles "a la CHORUS" in an emulsion cloud chamber. In such a setup,the emulsion is used as a tracking device to locate interactions and decays in a mainly lead target, instead of detecting the kink in bulk emulsion This choice is due to the large required mass (about 1 kton) and is allowed by the much reduced requirement on background rejection (3 instead of 6 orders of magnitude in CHORUS). The ICANOE experiment, a 5 kt liquid argon time projection chamber enhanced by a calorimeter, would detect taus "a la NOMAD" via kinematic criteria.
In the case of nu_mu-nu_tau oscillations the OPERA experiment would detect 18 tau events with an expected background of 0.5 events in four years for the central value of the Super-Kamiokande allowed parameters. The 4 sigma discovery limit completely covers the allowed region, ranging from 6 to 50 expected signal events. This is sufficient to prove tau appearance, and therefore to determine the oscilllation mode

A very important check will be the overlap between the allowed confidence regions from appearance (beam) and disappearance (atmospheric and/or beam) studies. In the case of two-flavour nu_mu-nu_tau oscillations the three regions should overlap in the region of the "true" oscillation parameters. Many models involving more complicated oscillation scenarios, in particular with significant contributions from oscillations into sterile neutrinos, predict non-overlapping confidence regions for these three signals. Therefore, the observation of such an effect would be direct evidence that the simple two-flavour nu_mu -nu_tau oscillation picture is wrong, and that sterile neutrinos play a role.

In conclusion, three different indications for neutrino oscillations have been observed, including the evidence for nu-mu disappearance from Super-Kamiokande. Any current explanation of these anomalies, and in particular the neutrino oscillation interpretation, implies that neutrinos have non-zero mass. The oscilllation hypothesis and the main oscillation modes for each indication will be fully tested over the next 5-10 years, and the main oscillation parameters will be measured.

On a longer time scale, muon storage rings (neutrino factories) can be used to produce highly pure high intenisty neutrino beams from muon instead of pion decays. With such beams, also the subdominant oscillation parameters can be fully tested, and CP-violation in the lepton sector becomes accessible.