WorksPhysics and Astronomy
Testing the Pseudo-Dirac Nature of Atmospheric Neutrinos
Abstract
Pseudo-Dirac neutrinos are characterized by extremely small mass-squared differences between two nearly degenerate, strongly-mixed, active-sterile states. These differences are associated with new very-long oscillation lengths and provide a probe of the nature of neutrinos. Atmospheric neutrinos can probe these oscillations as they allow access to a broad range of baseline distances, from tens of kilometers to the Earth's diameter. We investigate the sensitivity of an experiment like the Deep Underground Neutrino Experiment (DUNE) to the pseudo-Dirac nature of neutrinos through measurements of the atmospheric neutrino flux. We concentrate on the modifications to the atmospheric neutrino energy and angular spectra induced by one new oscillation frequency associated with the state and determine the parameter space that can be probed with atmospheric neutrinos at a DUNE-like detector. We also comment on the sensitivity of these measurements to the sub-leading "solar" oscillations, which are also characterized by a relatively small mass-squared difference.
Introduction
The discovery of neutrino oscillations in solar, atmospheric, accelerator, and reactor experiments embodies one of the most important breakthroughs in particle physics over the past few decades. It constitutes the first direct evidence of physics beyond the Standard Model (SM), since neutrinos are massless in the original formulation of the SM.
Neutrino oscillations are quantum mechanical interference processes in which neutrinos change flavors after traversing large distances. The flavor-conversion probability depends on the propagation distance , the neutrino energy , the mass-squared differences among the neutrinos and mixing angles (, ) that characterize charged-current weak interactions [1]. The current data have established the three-flavor paradigm of oscillation on a rather firm footing, and the mixing angles and mass-squared differences are measured with impressive precision. The three main features of the standard picture that remain undetermined are the sign of , i.e., the neutrino mass ordering, the octant of the mixing angle , and the Dirac CP phase .
Neutrino oscillation experiments remain insensitive to the absolute neutrino mass scale. Information about this parameter can come from experiments on beta decay and neutrinoless double beta decay as well as from cosmology. Among these, the combined analysis of DESI DR2 plus Planck CamSpec provides the best bound of 0.0642 eV on the sum of neutrino masses at the 95% level, disfavoring the inverted mass ordering [2]. The small mass-squared differences measured in oscillation experiments, together with the cosmological mass bound, imply that the neutrino masses are very small, safely below 0.1 eV.
Understanding the origin of such small neutrino masses remains one of the central open questions in particle physics. Unlike charged fermions, neutrinos are electrically neutral and can be Majorana or Dirac particles. A Dirac neutrino consists of four independent degrees of freedom corresponding to left-handed and right-handed neutrinos and antineutrinos. In this case, total lepton number is a conserved symmetry. Dirac neutrinos require the introduction of right-handed neutrino fields that are singlets under the SM gauge group. In contrast, Majorana neutrinos are their own antiparticles. A Majorana neutrino consists of two independent degrees of freedom: a left-handed and a right-handed one. Majorana neutrino masses violate the lepton number by two units. Such terms can arise through the dimension-five Weinberg operator [3] whose ultraviolet completion might come via one of the various seesaw mechanisms [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] (for an overview, see, for example, [15]). These can be naturally implemented in grand unified theories and provide an elegant formulation for generating small neutrino masses in which the violation of lepton number is related to the small neutrino masses. Neutrino oscillations cannot distinguish between the Dirac and Majorana nature of the neutrinos. Evidence for this can come from lepton-number-violating processes such as neutrinoless double beta decay. Determining the Dirac or Majorana nature of neutrinos is among the primary objectives of particle physics.
An intriguing intermediate possibility is that neutrinos are pseudo-Dirac fermions [16, 17, 18, 19]. In this scenario, the dominant neutrino mass originates from a Dirac mass term, while small Majorana mass terms split each Dirac neutrino into two nearly degenerate Majorana states. Pseudo-Dirac neutrinos are Majorana fermions but behave, as far as most experimental probes are concerned, as Dirac neutrinos.
As far as oscillations are concerned, if the neutrinos are pseudo-Dirac, every active neutrino is accompanied by a sterile partner and a new, extremely small mass-squared difference that satisfies
such that conventional three-flavor oscillation phenomenology remains essentially unchanged at short-enough baselines. However, oscillations into sterile partners are expected to become observable for sufficiently large values of . Hence, the presence of pseudo-Dirac neutrinos modifies the standard oscillation framework by introducing additional oscillation frequencies associated with the tiny active–sterile mass splittings, which changes the flavor transition probabilities [20, 21]. Assuming the mixing between the active–sterile pair to be maximal, robust constraints have been obtained on the mass splitting associated with the first two mass eigenstates, [22, 23, 24].
Existing constraints on the mass splitting associated with the third mass eigenstate from atmospheric, accelerator, and reactor data are considerably weaker: [25]. Studies have also been done to probe the pseudo-Dirac nature of neutrinos in future reactor and accelerator facilities [26, 27, 28, 29] and a study to probe the pseudo-Dirac nature in JUNO using solar neutrinos can be found in [30]. Since these effects accumulate only over extremely large baselines, high-energy astrophysical neutrinos provide unique opportunities to probe such scenarios (see, for example, [31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42]).
Among all available neutrino sources, atmospheric neutrinos span energies from sub-GeV to multi-GeV while traversing baselines ranging from a few tens of kilometers to the Earth’s diameter. Consequently, atmospheric neutrinos probe an enormous range of , making them potentially sensitive to tiny mass-squared splittings. Furthermore, the propagation of atmospheric neutrinos through the Earth experiences matter effects that can significantly modify active–sterile oscillations.
In this work, we investigate pseudo-Dirac neutrinos at a DUNE-like detector. Motivated by existing experimental constraints, we concentrate on the pseudo-Dirac splitting associated with the third mass eigenstate, , which presently remains the least constrained among the pseudo-Dirac splittings and is most relevant for the oscillations of atmospheric neutrinos. Hence, we investigate pseudo-Dirac neutrinos within the framework of a oscillation formalism. We study the modifications to neutrino oscillation probabilities, investigate their impact on atmospheric neutrinos, and assess the sensitivity of a DUNE-like detector to probe these effects via observations of atmospheric neutrinos.
The paper is organized as follows. Section II introduces the pseudo-Dirac neutrino framework together with the relevant theoretical motivation and the oscillation formalism of interest. Section III describes the numerical methodology and the atmospheric neutrino analysis. The results and phenomenological implications are discussed in Section IV. Section V contains a discussion of how sensitive is the atmospheric neutrino data to ”solar” oscillations. Finally, Section VI summarizes our results.
Pseudo-Dirac Neutrinos
We start by postulating that the neutrino masses are a consequence of the existence of gauge-singlet “right-handed-neutrino” fields. In this case, the most general renormalizable Lagrangian consistent with gauge invariance is
where is the minimal SM Lagrangian, is the Higgs weak doublet, , , are lepton weak doublets, are neutrino Yukawa couplings, and are Majorana masses for the right-handed neutrino fields fields that have been added. The number of fields is and, in order to fit the neutrino oscillation data, . Henceforth, when relevant, we assume .
After electroweak symmetry breaking, the neutrino mass matrix in the basis is
where is the usual Dirac mass related to the Higgs vacuum expectation value . yields the Dirac limit (i.e., pairs of mass-degenerate Majorana fermions form Dirac fermions). On the other hand, when , we can integrate out the right-handed neutrinos and end up with light active Majorana neutrinos (i.e., the seesaw mechanism).
The limit yields the pseudo-Dirac case. The neutrino masses are pairwise quasi-degenerate with magnitudes driven by the eigenvalues of . The small mass splittings are proportional to . We refer to these as , , where and are the six different pairwise-quasi-degenerate states.
Current constraints from solar neutrinos have greatly constrained two out of the three new oscillation frequencies associated with the and states [22, 23, 24, 25, 30]. Since solar neutrino experiments primarily involve electron neutrinos that have negligible mixing with , solar data provides comparatively weaker constraints on . Hence, bounds from Earth-bound experiments can compete with those from solar neutrinos in spite of the very long Earth-Sun distance.
is proportional to (see, for example, [22]), where is the average mass of and . In turn, for the normal neutrino mass ordering and ignoring the possibility that all neutrino masses are quasi-degenerate, is of order , the so-called atmospheric mass-squared difference. Hence, we define the relevant new oscillation frequency via
where is a dimensionless quantity. Throughout we use . We will further assume that the active–sterile mixing angle is maximal, as will become clearer in the next subsection.1 We refer to [22] for more details.
The pseudo-Dirac 3+1 oscillation formalism
In the limit , the relevant Hamiltonian that describes neutrino flavor evolution in the presence of matter is,
where
are the elements of the standard PMNS mixing matrix, ,
and
where is the effective contribution in the Hamiltonian due to the matter potential , with being the ambient electron density. We briefly discuss approximate solutions to the neutrino oscillation probabilities in the presence of constant matter in Appendix § 0.2.1.
Atmospheric Neutrinos in a Liquid Argon Detector
In modeling our simulations of atmospheric neutrinos, we obtained the atmospheric neutrino flux from the publicly available Honda files using the solar minimum variant for the Homestake site [43]. Information regarding the total cross sections was obtained through GLoBES [44], adjusting for as the target of a DUNE-like setup. To account for the matter effects that neutrinos experience along the different trajectories they take through the layers of the Earth from the atmosphere to the detector, we utilized the Preliminary Reference Earth Model (PREM) using an approximation with 5 layers: crust, upper mantle, lower mantle, outer core, and inner core [45].
We consider a setup similar to that proposed for DUNE [46]. DUNE is a long-baseline neutrino experiment under construction in South Dakota, with its far detector located approximately 1.5 km underground. The experiment will ultimately employ a 40 kt liquid argon time projection chamber (LArTPC) to detect neutrinos produced at Fermilab and traversing a baseline of 1300 km. In addition to its primary long-baseline neutrino program, the detector will also be sensitive to atmospheric, astrophysical, and solar neutrinos. The large detector mass, together with its excellent energy and angular resolution, makes this DUNE-like configuration well suited for the study presented here.
The analysis of atmospheric neutrino events is performed using an in-house code. The events per energy and bin are calculated using
Here is the neutrino flux as a function of the energy and direction, and is the cross section. We consider the energy range of GeV and with 80 bins each. The energy resolutions used for detected electrons and muons are,
For the angular resolution, we consider
where we consider (in degrees) values equal to 15, 30, 40, 50. The values for are consistent with those published by the DUNE collaboration in [47]. For , the resolution in [47] is most consistent with . The Gaussian resolution functions are defined as
where the normalization constants are
We have used 20 bins each for measured energy and , with the corresponding ranges being the same as the true case.
We used the events per energy and angular bins for true (ex) and test (th) cases to calculate defined as,
where are the pull variables and are the systematic error coefficients corresponding to the pull variables for each flavor. The pull variables are flux error, zenith error, cross-section error, systematic error, and tilt error, proportional to the theoretical event rate , and are listed in Table I. The zenith error is proportional to the measured angle, i.e., . We minimize by varying the oscillation parameters in the test (th) and obtain relative to the value of at the minimum.
We used the true and test values of oscillation parameters listed in Table II, following the global NuFit values [48]. We kept fixed in both true and test. The total exposure considered is 200 kt·yr, which is equivalent to 10 years of runtime with a 20 kton detector module, relevant in the case of a DUNE Phase I-like setup [46].
| Parameters | Percentage |
| Flux | 20% |
| Zenith Angle | 5% |
| Cross Section | 10% |
| Systematics | 5% |
| Tilt | 5% |
Table 1. Different uncertainties used in our analysis.
| Parameters | True (NO) | True (IO) | Test Range |
| , | |||
| , | |||
| , | |||
Table 2. The true and test values of the oscillation parameters
Sensitivity to the pseudo-Dirac Nature of
Before presenting the results for the sensitivity to the pseudo-Dirac nature of , we first illustrate in Figure 1 the differences in the oscillograms for the dominant muon-neutrino disappearance probability between the pseudo-Dirac hypothesis and the standard scenario for equal to 0.001, 0.01, and 0.1. Non-zero values of the parameter lead to distortions at sub-GeV energies and for negative values of . This is not surprising since a large-enough new oscillation phase is required for observable effects and negative values correspond to longer baselines (and larger matter effects). The effect is seen to be very small – below a percent – for but is order one for .

Figure 1. The difference of the disappearance probability between the pseudo-Dirac hypothesis and the standard scenario for (top), (middle), and (bottom), as a function of the neutrino energy and the neutrino direction .
Our estimates for the sensitivity to the pseudo-Dirac nature of are depicted in Figure 2. We highlight four options for the sub-GeV angular resolution: , , , and . The latter is the most conservative choice in our analysis and was inspired by results made public by the DUNE collaboration [47]. We assumed the neutrino mass ordering to be normal (NO); all other assumptions are listed within the panels of Figure 2. We present results for the inverted neutrino mass ordering (IO) in Appendix § 0.9. We include both the muon-neutrino and electron-neutrino “data” in our analyses; the sensitivity, however, is dominated by the muon-neutrino-associated samples.

Figure 2. Sensitivity to of 200 kt·yr of atmospheric neutrino data at a DUNE-like detector, assuming NO. On the left is the sensitivity computed for fixed values of all other oscillation parameters for different values of the angular resolution below 1 GeV. On the right, we illustrate the effect of profiling over (green) and (gold), relative to the sensitivity obtained for fixed values of all oscillation parameters (blue curve), for a fixed value of the angular resolution below 1 GeV.
In the left panel of Figure 2, all standard oscillation parameters are fixed to the values listed in Table II. In this case, the only free parameter is the value of the dimensionless parameter that sets the hypothetical new oscillation frequency. These results indicate that the sub-GeV angular reconstruction significantly impacts the sensitivity. For the most conservative case presented here, a DUNE-like experiment is sensitive to at the level. This corresponds to . For the more optimistic case we considered ( angular smearing below 1 GeV), the sensitivity limit is which corresponds to . For sub-GeV neutrino energies, corresponds to an oscillation length of order the Earth’s diameter.
These results are competitive with similar studies in the literature, most notably a recent DUNE beam study [25]. This aligns with the fact that atmospheric and beam neutrinos probe, roughly, similar scales. We emphasize, however, that our result indicates a slightly worse relative sensitivity. Essentially, while atmospheric neutrino studies can take advantage of longer baselines and thus can potentially probe relatively small new mass-squared differences, most of the sensitivity lies where the detector angular resolution is the poorest (i.e., low neutrino energies). In other words, the access to longer baselines and relatively larger values of is offset by the experiment’s ability to determine precisely. We chose to report different instances of sub-GeV angular smearing for this reason.
To probe the robustness of our analysis and to account for the uncertainties present in the current neutrino oscillation global fit values, we also explored the impact of profiling over the individual oscillation parameters, taking the current uncertainties quoted in [48] into account. In the right-hand panel of Figure 2, we included the effects of profiling, one at a time, over the oscillation parameters that potentially most impact the sensitivity computed here: and . Here, the angular resolution for low-energy neutrinos is fixed ( angular resolution for neutrino energies below 1 GeV). We profile over the so-called atmospheric parameters because they contribute the most to the dominant oscillation amplitude and phase when it comes to the disappearance of atmospheric muon neutrinos.
Our results indicate that profiling over yields the strongest effect overall, followed by . These results are consistent with expectations from a qualitative inspection of the leading-order terms in the oscillation probabilities
presented in Appendix § 0.2.1.
In a little more detail, terms like , , and in the probabilities indicate a correlation between and . This is illustrated in Figure 3. In the left panel of Figure 3, we depict the correlation between and . In particular, the red curve depicts the values of the test at which the minimum is obtained for a given value of . These differ from the true (cyan line) for large-enough values of ; the correlation is clear. Hence, profiling over yields a relatively lower for higher values of . Similarly, using Eqs. (27), (28) one can see that higher values of can be partially compensated by shifting the vale of . This is reflected in the red curve in the right-hand panel of Figure 3. It indicates that the minimum occurs at higher test- values relative to the true value (blue line) as increases. Quantitatively, for , the -related effects are more pronounced.

Figure 3. Sensitivity in the (left), and plane (right) for NO. The red curve shows the position of the minimum value of as a function of . The blue line indicates the true values on the relevant standard oscillation parameter.
Sensitivity to Solar Parameters
Since the sensitivity to the pseudo-Dirac mass splitting is close to that of the solar MSW mass-squared difference, it is worthwhile to investigate the sensitivity of the atmospheric neutrinos at a DUNE-like experiment to the solar parameters and . Similar studies were performed in [49] in the context of SuperKamiokande data.
We use as input for the standard -oscillation parameters the values listed in Table II. All other oscillation parameters, apart from the solar under investigation, have been kept fixed in the analysis.
For atmospheric neutrinos, the leading-order dependence on or is subdominant, suggesting weak sensitivity to these parameters. Figure 4 depicts the sensitivity to (left panel) and (right panel). In the left (right) panel, the value of () is kept fixed at the value tabulated in Table II, and we assume NO. There are two curves in each panel indicating the different values of that set the angular resolution for GeV: and . As expected, the uncertainty in the direction of low-energy neutrinos has a measurable impact. Atmospheric neutrino data can only set a modest, resolution-dependent limit on , even when is assumed to be perfectly known. The hypothesis cannot be ruled out at more than , but values of larger than eV can be ruled out at the level even when we assume the more pessimistic uncertainty on the direction of low-energy neutrinos. On the other hand, the data can place virtually no bound on .

Figure 4. Sensitivity to (left) and (right) for standard oscillations, assuming NO and 200 ktyr of data. The dotted and dashed lines correspond to different choices for the angular resolution when GeV: (most conservative) and (most optimistic), respectively.
Figure 5 depicts the sensitivity to the solar oscillation parameters in the plane for two different angular resolutions associated with low-energy atmospheric neutrinos: on the left, on the right. As before, we have assumed NO. The figure demonstrates strong correlation, as is expected when is small. The sensitivity, even with the most optimistic angular resolution, allows all values of for eV.

Figure 5. Sensitivity to the solar oscillation parameters in the -plane for standard oscillations assuming 200 kt·yr of data. We assume NO. The left panel corresponds to an angular resolution for GeV of (most optimistic), and the right panel corresponds to (most conservative). The star in each panel indicates the best-fit value from NuFit [48].
The simulated data is most sensitive to disappearance, so the information on the solar mixing angle comes, in part, from (or ; they are related to through the unitarity of the mixing matrix). Since we fixed throughout,
and, numerically,
In particular, decreases as increases for .
Conclusions
We investigated the sensitivity of atmospheric neutrinos at a DUNE-like experiment to the pseudo-Dirac nature of the third neutrino mass eigenstate. We considered the scenario in which the active and sterile components are maximally mixed and the new physics is characterized by a dimensionless parameter, , related to the additional mass-squared splitting through . Atmospheric neutrinos are suitable for this search because they cover a broad range of energies (roughly GeV) and propagation distances (roughly – km) and hence provide access to several orders of that are not accessible to other conventional terrestrial neutrino sources.
Assuming an exposure of 200 kt·yr, we studied the sensitivity to , concentrating on the normal neutrino mass ordering. Results for IO can be found in Appendix § 0.9. The sensitivity depends strongly on the angular reconstruction of sub-GeV atmospheric-neutrino events. For a sub-GeV angular resolution of , the most conservative scenario considered here, if the data are consistent with the standard oscillation scenario, we expect it to translate into a upper bound of
Given a more optimistic angular resolution of , future atmospheric data should be able to constrain
The angular resolution at energies below 1 GeV plays a central role in the sensitivity of atmospheric neutrinos to new, long oscillation lengths.
We further examined the robustness of the sensitivity to the pseudo-Dirac nature of against uncertainties in the standard oscillation parameters. Profiling over and within the current allowed ranges weakens the sensitivity relative to the case in which all standard parameters are fixed. The effect is more pronounced for . A more detailed determination of the pseudo-Dirac sensitivity must therefore account for correlations with the parameters governing the dominant atmospheric oscillation amplitude and frequency. Nevertheless, the atmospheric-neutrino reach remains comparable to that of related terrestrial searches and provides complementary information.
We also explored the sensitivity of atmospheric neutrinos in a DUNE-like detector to the solar parameters and . This question is similar to that of the pseudo-Dirac nature of , i.e., it is also related to the sensitivity to new long oscillation lengths. We find that the sensitivity to the solar oscillation parameters is limited. This is consistent with the bounds on summarized above, which not coincidentally, are of order .
Our analysis highlights the potential and limitations of atmospheric neutrinos as complementary probes of the pseudo-Dirac nature of light neutrinos. It underscores the importance of the angular reconstruction of sub-GeV neutrinos, control of systematic uncertainties, and the treatment of correlations among the known oscillation parameters as central ingredients in determining the final experimental reach.
We anticipate that atmospheric neutrinos collected by the Hyper-Kamiokande experiment [50] should have similar sensitivity to the pseudo-Dirac nature of . Ultimately, the combination of atmospheric-neutrino observations with accelerator, reactor, solar, and astrophysical measurements will provide coverage across several orders of magnitude in pseudo-Dirac mass splittings.
Note added: While some of the authors are members of the DUNE collaboration, this work is not officially associated with DUNE; only publicly available detector specifications were used to simulate and analyze hypothetical data in a DUNE-like detector.
Acknowledgments
SG acknowledges Northwestern University, where the work was initiated, for hospitality and the Fulbright-Nehru Academic and Professional Excellence fellowship for funding her visit to the US. SP acknowledges the funding from ANID-Chile under FONDECYT Postdoctorado No. 3250376. The work of OMB and AdG was supported in part by the US Department of Energy (DOE) grants #de-sc0010143 and #de-sc0010813 and in part by the National Science Foundation under Grant Nos. PHY-1630782 and PHY-1748958.
Approximate Analytic Expressions for the Oscillation Probabilities
We can write the total Hamiltonian that governs the 3+1 system of interest as the sum of two Hamiltonians,
is zeroth-order Hamiltonian proportional to and is the perturbative Hamiltonian containing the small terms proportional to and . We use perturbation theory, considering (in units of ) and as small parameters, to calculate the eigenvalues of the Hamiltonian and the oscillation frequencies in matter. Throughout this appendix, we assume the matter potential is constant. The eigenvalues of are,
where the modified mass-squared differences are,
We can write the mixing matrix in matter, obtained from the eigenvectors of the Hamiltonian, as
Along with the “matter” mixing angles and , there is also an effective mixing angle . It turns out (cf. Eq. (32)) that is very small when . Therefore, terms in the oscillation probabilities proportional to will be sub-leading. Using perturbation theory, the modified mixing angles satisfy
The matter equivalent version of is modified relative to its vacuum value by a function of the matter potential and the mixing angles and . On the other hand, the matter equivalent to the atmospheric mass-squared difference is a function of both the matter potential and the size of the pseudo-Dirac splitting, proportional to . A modified mass-squared difference also arises in matter and is a function of and . However, it is a stronger function of , as seen in Eq. (26).
To express the oscillation probabilities in constant matter, we define
The appearance and disappearance probabilities in matter in the presence of a pseudo-Dirac-related are
where .
A few points related to the expressions above are worthy of note:
The new oscillation frequencies , , terms are subleading due to the smallness of .
The presence of terms with both and in the amplitude, e.g., together with in , translate into a correlation between and ; an increase in (and hence ) can be compensated by a decrease in .
There is a correlation between and . These can be in the expressions for and (using ); larger values of can be compensated by an increase in and hence of .
For completeness, we list the expressions for the oscillation probabilities in the standard oscillation scenario – three active flavors – under the same conditions:
where and are
Results for the Inverted Ordering
Here we collect the results obtained assuming that the neutrino mass ordering is inverted. Note that, in this case, the connection between and is not especially natural since, unless all neutrino masses are degenerate, .

Figure 6. Sensitivity to of 200 kt·yr of atmospheric neutrino data at a DUNE-like detector, assuming IO. On the left is the sensitivity computed for fixed values of all other oscillation parameters for different values of the angular resolution below 1 GeV. On the right, we illustrate the effect of profiling over (green) and (gold), relative to the sensitivity obtained for fixed values of all oscillation parameters (blue curve), for a fixed value of the angular resolution below 1 GeV.

Figure 7. Sensitivity in the (left), and plane (right) for IO. The red curve shows the position of the minimum value of as a function of . The blue line indicates the true values on the relevant standard oscillation parameter.

Figure 8. Sensitivity to (left) and (right) for standard oscillations, assuming IO and 200 kt·yr of data. The dotted and dashed correspond to different choices for the angular resolution when GeV: 50° (most conservative) and 15° (most optimistic), respectively.

Figure 9. Sensitivity to the solar oscillation parameters in the -plane for standard oscillations assuming 200 kt·yr of data. We assume IO. The left panel corresponds to an angular resolution for of (most optimistic), and the right panel corresponds to (most conservative). The star in each panel indicates the best-fit value from NuFit [48].
References
- [1]F. Takahashi et al. (Particle Data Group), Review of Particle Physics, Int. J. Mod. Phys. A 41, 2630011 (2026).
- [2]R. Jimenez, C. P. Garay, F. Simpson, and L. Verde, From evidence to evident: Decisive cosmological evidence for the normal neutrino mass hierarchy (2026), arXiv:2606.18987 [astro-ph.CO].arxiv.org/abs/2606.18987
- [3]S. Weinberg, Baryon and Lepton Nonconserving Processes, Phys. Rev. Lett. 43, 1566 (1979).
- [4]P. Minkowski, μ → eγ at a Rate of One Out of 10⁹ Muon Decays?, Phys. Lett. B 67, 421 (1977).
- [5]T. Yanagida, Horizontal Gauge Symmetry and Masses of Neutrinos, in Proceedings of the Workshop on the Unified Theory and the Baryon Number in the Universe, edited by O. Sawada and A. Sugamoto (Tsukuba, Japan, 1979) p. 95.
- [6]M. Gell-Mann, P. Ramond, and R. Slansky, Complex Spinors and Unified Theories, in Supergravity, edited by P. van Nieuwenhuizen and D. Z. Freedman (North-Holland, Amsterdam, 1979) pp. 315–321, arXiv:1306.4609 [hep-th].
- [7]R. N. Mohapatra and G. Senjanovic, Neutrino Mass and Spontaneous Parity Nonconservation, Phys. Rev. Lett. 44, 912 (1980).DOI
- [8]W. Konetschny and W. Kummer, Nonconservation of Total Lepton Number with Scalar Bosons, Phys. Lett. B 70, 433 (1977).DOI
- [9]M. Magg and C. Wetterich, Neutrino Mass Problem and Gauge Hierarchy, Phys. Lett. B 94, 61 (1980).DOI
- [10]T. P. Cheng and L.-F. Li, Neutrino Masses, Mixings and Oscillations in SU(2) x U(1) Models of Electroweak Interactions, Phys. Rev. D 22, 2860 (1980).DOI
- [11]J. Schechter and J. W. F. Valle, Neutrino Masses in SU(2) x U(1) Theories, Phys. Rev. D 22, 2227 (1980).DOI
- [12]G. Lazarides, Q. Shafi, and C. Wetterich, Proton Lifetime and Fermion Masses in an SO(10) Model, Nucl. Phys. B 181, 287 (1981).DOI
- [13]R. N. Mohapatra and G. Senjanovic, Neutrino Masses and Mixings in Gauge Models with Spontaneous Parity Violation, Phys. Rev. D 23, 165 (1981).DOI
- [14]R. Foot, H. Lew, X. G. He, and G. C. Joshi, Seesaw Neutrino Masses Induced by a Triplet of Leptons, Z. Phys. C 44, 441 (1989).DOI
- [15]A. de Gouvêa, Neutrino Mass Models, Ann. Rev. Nucl. Part. Sci. 66, 197 (2016).DOI
- [16]L. Wolfenstein, Different Varieties of Massive Dirac Neutrinos, Nucl. Phys. B 186, 147 (1981).DOI
- [17]S. T. Petcov, On Pseudodirac Neutrinos, Neutrino Oscillations and Neutrinoless Double beta Decay, Phys. Lett. B 110, 245 (1982).
- [18]J. W. F. Valle and M. Singer, Lepton Number Violation With Quasi Dirac Neutrinos, Phys. Rev. D 28, 540 (1983).DOI
- [19]M. Doi, M. Kenmoku, T. Kotani, H. Nishiura, and E. Takasugi, PSEUDODIRAC NEUTRINO, Prog. Theor. Phys. 70, 1331 (1983).
- [20]A. S. Joshipura and S. D. Rindani, Phenomenology of pseudoDirac neutrinos, Phys. Lett. B 494, 114 (2000), arXiv:hep-ph/0007334.
- [21]M. Kobayashi and C. S. Lim, Pseudo Dirac scenario for neutrino oscillations, Phys. Rev. D 64, 013003 (2001), arXiv:hep-ph/0012266.DOI
- [22]A. de Gouvêa, W.-C. Huang, and J. Jenkins, Pseudo-dirac neutrinos in the new standard model, Physical Review D 80, 073007 (2009).DOI
- [23]A. Donini, P. Hernandez, J. Lopez-Pavon, and M. Maltoni, Minimal models with light sterile neutrinos, JHEP 07 (07), 105, arXiv:1106.0064 [hep-ph].
- [24]S. Asarifard and Y. Farzan, Revisiting pseudo-Dirac neutrino scenario after recent solar neutrino data, Phys. Rev. D 107, 075029 (2023), arXiv:2211.09105 [hep-ph].
- [25]G. Anamiati, R. M. Fonseca, and M. Hirsch, Quasi-dirac neutrino oscillations, Physical Review D 97, 095008 (2018).DOI
- [26]G. Amatiati, V. De Romeri, M. Hirsch, C. A. Ternes, and M. Tórtola, Quasi-Dirac neutrino oscillations at DUNE and JUNO, Phys. Rev. D 100, 035032 (2019), arXiv:1907.00980 [hep-ph].
- [27]N.-I. Boudjemaa, F. F. Deppisch, and S. S. Pattanaik, Probing Quasi-Dirac Neutrino Oscillations at Long Baseline Experiments (2026), arXiv:2607.06862 [hep-ph].
- [28]A. Abada, J. P. Pinheiro, and S. Urrea, Investigating dune oscillations sensitivity to sterile pseudo-dirac neutrinos (2025), arXiv:2506.16390 [hep-ph].arxiv.org/abs/2506.16390
- [29]M. Sen, Constraining pseudo-dirac neutrinos from a galactic core-collapse supernova (2023), arXiv:2205.13291 [hep-ph].arxiv.org/abs/2205.13291
- [30]J. Franklin, Y. F. Perez-Gonzalez, and J. Turner, JUNO as a probe of the pseudo-Dirac nature using solar neutrinos, Phys. Rev. D 108, 035010 (2023), arXiv:2304.05418 [hep-ph].arxiv.org/abs/2304.05418
- [31]J. F. Beacom, N. F. Bell, D. Hooper, J. G. Learned, S. Pakvasa, and T. J. Weiler, Pseudo-dirac neutrinos: A challenge for neutrino telescopes, Phys. Rev. Lett. 92, 011101 (2004).pubmed.ncbi.nlm.nih.gov/14753977
- [32]S. Pakvasa, A. Joshipura, and S. Mohanty, Explanation for the low flux of high energy astrophysical muon-neutrinos, Phys. Rev. Lett. 110, 171802 (2013), arXiv:1209.5630 [hep-ph].DOI
- [33]A. S. Joshipura, S. Mohanty, and S. Pakvasa, Pseudo-Dirac neutrinos via a mirror world and depletion of ultrahigh energy neutrinos, Phys. Rev. D 89, 033003 (2014), arXiv:1307.5712 [hep-ph].DOI
- [34]T. Rink and M. Sen, Constraints on pseudo-Dirac neutrinos using high-energy neutrinos from NGC 1068, Phys. Lett. B 851, 138558 (2024), arXiv:2211.16520 [hep-ph].arxiv.org/abs/2211.16520
- [35]K. Carloni, I. Martínez-Soler, C. A. Arguelles, K. S. Babu, and P. S. B. Dev, Probing pseudo-Dirac neutrinos with astrophysical sources at IceCube, Phys. Rev. D 109, L051702 (2024), arXiv:2212.00737 [astro-ph.HE].DOI
- [36]P. S. B. Dev, P. A. N. Machado, and I. Martínez-Soler, Pseudo-Dirac neutrinos and relic neutrino matter effect on the high-energy neutrino flavor composition, Phys. Lett. B 862, 139306 (2025), arXiv:2406.18507 [hep-ph].DOI
- [37]K. Dixit, L. S. Miranda, and S. Razzaque, Searching for pseudo-dirac neutrinos from astrophysical sources in IceCube data, Eur. Phys. J. C 85, 1481 (2025), arXiv:2406.06476 [astro-ph.HE].arxiv.org/abs/2406.06476
- [38]K. Carloni, Y. Porto, C. A. Arguelles, P. S. B. Dev, and S. Jana, Signatures of quasi-Dirac neutrinos in diffuse high-energy astrophysical neutrino data, Phys. Rev. D 113, 115004 (2026), arXiv:2503.19960 [hep-ph].arxiv.org/abs/2503.19960
- [39]L. P. S. Leal, D. R. Machado, and R. Z. Funchal, Cosmogenic neutrinos as probes of new physics, JHEP (08), 057, arXiv:2504.10576 [hep-ph].DOI
- [40]E. Cabrera, M. Jin, C. A. Arguelles, and A. Esmaili, Searching for sub-eV sterile neutrinos in neutrino telescopes, Phys. Rev. D 113, 075015 (2026), arXiv:2509.20142 [hep-ph].
- [41]C. S. Fong and Y. Porto, Constraining the pseudo-dirac nature of neutrinos using astrophysical neutrino flavor data, Phys. Rev. D 112, 063001 (2025).DOI
- [42]M. MacDonald, K. Carloni, C. A. Arguelles, I. Martínez-Soler, and R. Alves Batista, Exploring New Propagation Scales With Galactic Neutrinos, arXiv:2512.10744 [hep-ph] (2025), preprint.arxiv.org/abs/2512.10744
- [43]M. Honda, M. S. Athar, T. Kajita, K. Kasahara, and S. Midorikawa, Atmospheric neutrino flux calculation using the nrmsise-00 atmospheric model, Phys. Rev. D 92, 023004 (2015).DOI
- [44]P. Huber, J. Kopp, M. Lindner, M. Rolinec, and W. Winter, New features in the simulation of neutrino oscillation experiments with globes 3.0, Comput. Phys. Commun. 177, 432 (2007), arXiv:hep-ph/0701187.DOI
- [45]A. M. Dziewonski and D. L. Anderson, Preliminary reference earth model, Physics of the Earth and Planetary Interiors 25, 297 (1981).
- [46]B. Abi et al. (DUNE), Deep Underground Neutrino Experiment (DUNE), Far Detector Technical Design Report, Volume I Introduction to DUNE, JINST 15 (08), T08008, arXiv:2002.02967 [physics.ins-det].
- [47]S. Abdus et al. (DUNE), Reconstruction of atmospheric neutrinos in DUNE’s horizontal-drift far-detector module, Eur. Phys. J. C 86, 904 (2026), arXiv:2601.05697 [hep-ex].
- [48]I. Esteban, M. C. Gonzalez-Garcia, M. Maltoni, I. Martinez-Soler, J. P. Pinheiro, and T. Schwetz, Nufit-6.0: updated global analysis of three-flavor neutrino oscillations, Journal of High Energy Physics 2024, 216 (2024).
- [49]O. L. G. Peres and A. Y. Smirnov, Testing the solar neutrino conversion with atmospheric neutrinos, Phys. Lett. B 456, 204 (1999), arXiv:hep-ph/9902312.DOI
- [50]K. Abe et al. (Hyper-Kamiokande), Hyper-Kamiokande Design Report, arXiv:1805.04163 [physics.ins-det] (2018), design report.arxiv.org/abs/1805.04163