Three flavors of neutrino are known: electron, muon, and tau. Three mass states are also known, called ν₁, ν₂, and ν₃. The relationship between the flavors and the mass states is described by the PMNS mixing matrix, and decades of oscillation experiments have measured the mixing parameters to good precision.
What those experiments have not yet settled is something deceptively simple: which mass state is the heaviest.
Two scenarios are still consistent with all the data. They are called normal ordering and inverted ordering, and the choice between them has consequences that ripple from cosmology to the search for neutrinoless double-beta decay to the structure of theory beyond the Standard Model.
What the data shows
Oscillation experiments do not measure the masses of the three neutrinos directly. They measure the differences between the squared masses, which is what determines the rate at which neutrinos oscillate.
Two such differences are now known to good precision:
A small difference, called Δm²₂₁, with a value of about 7.5 × 10⁻⁵ eV². This was measured primarily through solar neutrinos and the KamLAND reactor experiment. It is the difference between the ν₁ and ν₂ mass squared values.
A larger difference, called |Δm²₃₂|, with a magnitude of about 2.5 × 10⁻³ eV². This was measured through atmospheric neutrinos at Super-Kamiokande and through long-baseline experiments like T2K and NOvA. It is the difference between the ν₃ and ν₂ mass squared values.
The catch: only the magnitude of Δm²₃₂ is currently known. The sign is not. If Δm²₃₂ is positive, then m(ν₃) > m(ν₂) > m(ν₁) — the normal ordering. If it is negative, then m(ν₂) > m(ν₁) > m(ν₃) — the inverted ordering.
Why “ordering” rather than “hierarchy”
For years, physicists called this question the “mass hierarchy.” The terminology has slowly shifted to “mass ordering” because hierarchy implies that the masses are spread out by orders of magnitude, the way quark masses are. Neutrino masses, however, are all in the same ballpark — clustered within a factor of a few of one another. Calling that distribution a “hierarchy” is somewhat misleading.
“Ordering” is more accurate. We know two of the three states are very close in mass, separated by Δm²₂₁ — call these the two “clustered” states. The third state, ν₃, is somewhat further away. The question is whether ν₃ sits above the cluster (normal) or below it (inverted).
Why the question matters
Settling the mass ordering would affect at least four distinct parts of physics.
Cosmology. The sum of the three neutrino masses, Σm_ν, contributes to the energy density of the universe and affects how cosmic structure grows. Current cosmological data constrains Σm_ν to roughly less than 0.1 eV. In normal ordering, the minimum allowed sum is around 0.058 eV — comfortably below the bound. In inverted ordering, the minimum is around 0.098 eV — right at the boundary. If cosmology continues to tighten the bound, inverted ordering becomes harder to fit.
Neutrinoless double-beta decay. The effective Majorana mass that 0νββ experiments measure depends strongly on the ordering. In inverted ordering with Majorana neutrinos, the effective mass is bounded from below at around 15 meV — well within the reach of next-generation experiments like LEGEND-1000 and KamLAND2-Zen. In normal ordering, the effective mass can be much smaller, including the possibility of nearly zero. If no 0νββ signal is found at the inverted-ordering level, then either neutrinos are Dirac or the ordering is normal.
Cosmological neutrino background. The cosmic relic neutrinos from the Big Bang have, in principle, observable consequences for cosmic structure and the integrated cosmic energy budget. The ordering affects how much each mass eigenstate contributes to these signals.
Theoretical models. Most modern theoretical models of neutrino mass — see-saw, leptogenesis, anarchic flavor models — have a preferred ordering. A confirmed normal ordering supports the simplest see-saw implementations. A confirmed inverted ordering would push theory in more exotic directions.
How experiments will decide
Several experimental approaches can settle the question, each with its own timeline.
JUNO — the 20,000-ton scintillator sphere in southern China — is the most direct probe. JUNO measures reactor antineutrino oscillations at a baseline tuned to be sensitive to the mass ordering through the fine structure of the oscillation pattern. JUNO is expected to determine the ordering at the 3σ level after about six years of running, possibly the early 2030s. (See our JUNO article for details.)
DUNE and Hyper-Kamiokande — the two flagship long-baseline experiments under construction — will both be sensitive to the mass ordering through matter effects on neutrino versus antineutrino oscillation rates. The matter effect arises because neutrinos passing through the Earth interact differently with electrons than antineutrinos do. The size and sign of this effect depends on the ordering. Both experiments should reach significant sensitivity by the early 2030s.
Atmospheric neutrino experiments — Super-Kamiokande, IceCube-Gen2 with its low-energy extension PINGU, KM3NeT-ORCA — measure ordering effects in atmospheric neutrinos arriving from below. Sensitivity is gradually growing as data accumulates.
Cosmological surveys — particularly the next generation of CMB and large-scale structure experiments — will continue to tighten the bound on the total neutrino mass, indirectly favoring one ordering or the other.
What current data already prefers
As of the mid-2020s, the global combination of all neutrino oscillation data shows a mild preference for normal ordering, at roughly the 2.5σ level. This preference is not yet decisive — neither ordering is ruled out — but it has been slowly strengthening over time.
Cosmological data also leans toward the normal ordering, partly because the cosmological bound on the total mass is approaching the level where inverted ordering becomes uncomfortable.
By the early 2030s, the combination of JUNO, DUNE, and Hyper-Kamiokande should resolve the question conclusively. Whatever the answer, it will reshape several active research programs simultaneously.
Why a small question carries big weight
The neutrino mass ordering is, at one level, just a sign — is Δm²₃₂ positive or negative? At another level, it is one of the most consequential outstanding questions in particle physics. It affects what theoretical frameworks are viable, what experimental searches are likely to succeed, and how cosmological data should be interpreted.
It is also, in some sense, the simplest measurement in the field that has resisted all attempts at resolution for two decades. We have measured the relative sizes of the squared mass differences. We have measured three mixing angles. We are working on CP violation. The ordering, alone among the basic oscillation parameters, has stayed stubbornly indeterminate.
That should change in the next decade. JUNO is running. DUNE is being built. Hyper-Kamiokande is being filled. By the time the next major neutrino conference rolls around, one of the two scenarios will be in serious trouble — and the other will become the new starting point for everything else.
For the experiment most directly hunting the ordering, see JUNO. For the long-baseline counterparts, see DUNE and Hyper-Kamiokande. For why this connects to neutrinoless double-beta decay, see The hunt for 0νββ.
Frequently asked
What is the neutrino mass ordering?
The unanswered question of whether the third neutrino mass state (ν₃) is heavier or lighter than the other two. In the 'normal ordering', ν₃ is the heaviest. In the 'inverted ordering', ν₃ is the lightest. Oscillation experiments measure mass-squared differences, but the sign of Δm²₃₂ — which determines the ordering — remains undetermined.
Why does it matter?
Settling the mass ordering affects four parts of physics: cosmology (inverted ordering pushes the minimum Σm_ν close to the upper cosmological bound), neutrinoless double-beta decay sensitivity, theoretical models like see-saw and leptogenesis, and the interpretation of supernova neutrino bursts. The ordering is one of the central unknowns in modern particle physics.
Which experiments will settle it?
JUNO (medium-baseline reactor antineutrinos) targets a 3σ measurement around 2030 through the fine structure of the oscillation pattern. DUNE (long-baseline accelerator) uses matter effects through Earth's mantle. Hyper-Kamiokande adds atmospheric and long-baseline measurements. The combination of all three should produce a definitive result by the mid-2030s.
What does current data prefer?
As of the mid-2020s, the global combination of neutrino oscillation data shows a mild preference for the normal ordering at roughly the 2.5σ level — interesting but not yet decisive. Cosmological data also leans toward normal ordering, partly because the cosmological bound on Σm_ν is approaching the level where inverted ordering becomes uncomfortable.
Could cosmology settle it before the dedicated experiments?
Possibly. Cosmological measurements of Σm_ν are tightening rapidly. The minimum value for inverted ordering is about 0.098 eV; current cosmological upper bounds are around 0.07-0.12 eV. If CMB-S4 and DESI push the bound below 0.098 eV, inverted ordering will be excluded by cosmology alone, independently of the dedicated oscillation experiments.
Cite this article 5 formats
APA
Neutrino Times Editorial Team. (2025, September 11). Normal or inverted? The neutrino mass ordering question, explained. Neutrino Times. https://neutrino-times.com/articles/neutrino-mass-ordering-normal-inverted/
Chicago
Neutrino Times Editorial Team. "Normal or inverted? The neutrino mass ordering question, explained." Neutrino Times, September 11, 2025. https://neutrino-times.com/articles/neutrino-mass-ordering-normal-inverted/.
MLA
Neutrino Times Editorial Team. "Normal or inverted? The neutrino mass ordering question, explained." Neutrino Times, 11 Sep. 2025, https://neutrino-times.com/articles/neutrino-mass-ordering-normal-inverted/.
BibTeX
@misc{neutrino-times-neutrino-mass-ordering-normal-inverted,
author = {Neutrino Times Editorial Team},
title = {Normal or inverted? The neutrino mass ordering question, explained},
howpublished = {Neutrino Times},
year = {2025},
month = {sep},
url = {https://neutrino-times.com/articles/neutrino-mass-ordering-normal-inverted/},
note = {Accessed: 2025-09-11}
} RIS
TY - GEN TI - Normal or inverted? The neutrino mass ordering question, explained AU - Neutrino Times Editorial Team PY - 2025 DA - 2025-09-11 PB - Neutrino Times UR - https://neutrino-times.com/articles/neutrino-mass-ordering-normal-inverted/ ER -