Parameter history

Bari, NuFit and Valencia, on the same axes

How the oscillation parameters have moved across three independent global analyses, and, in finer grain, across every published update of the Bari fit: best fit and 3σ range, for both mass orderings. No point is interpolated, rescaled or read off a figure.

Compared with the other groups

normal ordering · 8 releases from NuFit and Valencia

Reading a comparison across conventions

The three groups do not report the same quantity. We use Δm² = m₃² − ½(m₁² + m₂²). NuFit reports Δm²₃ℓ, which is Δm²₃₁ > 0 for normal ordering and Δm²₃₂ < 0 for inverted. Valencia reports |Δm²₃₁| for both orderings.

From the identity Δm² = Δm²₃₁ − δm²/2, the correction is −δm²/2 for normal ordering in every case. In inverted ordering the two groups differ, and the difference is clearest stated on the modulus, which is what every number on this page is plotted as. NuFit publishes Δm²₃ℓ = Δm²₃₂ < 0, and adding δm²/2 to a negative number makes its modulus smaller: the shift is −δm²/2 for NuFit. Valencia publishes |Δm²₃₁|, already a modulus, and the same addition makes it larger: the shift is +δm²/2 for Valencia. Same identity, opposite effect on the plotted number — which is why this site stores what each paper printed and converts in code, where the rule can be read: tools/make_history.py, function to_our_Dm2.

The 3σ ranges are shifted by the same constant as the best fit, computed with δm² at its own best fit. The papers do not publish the joint δm²–Δm² information a rigorous reprojection would need; the resulting error is far smaller than the width of the ranges shown, but it is an approximation and not a re-analysis.

NuFit prints δCP in degrees only, so it does not appear on the δ/π panel. Where a fit has two quasi-degenerate θ₂₃ minima, the marker is the first one quoted and the 3σ range spans both.

Bari (circle) NuFit (square) Valencia (diamond) Upper limit, level printed on the marker

δm² / 1e-5 eV² · normal ordering

6.51 7.47 8.42 9.38 01 04 06 11 12 13 16 17 18 20 21 24 25 26 Bari, 2006: δm² = 7.92 (1e-5 eV²) Bari, 2011: δm² = 7.58, 3σ 6.99–8.18 (1e-5 eV²) Bari, 2012: δm² = 7.54, 3σ 6.99–8.18 (1e-5 eV²) Bari, 2013: δm² = 7.54, 3σ 6.99–8.18 (1e-5 eV²) Bari, 2016: δm² = 7.37, 3σ 6.93–7.97 (1e-5 eV²) Bari, 2017: δm² = 7.37, 3σ 6.93–7.96 (1e-5 eV²) Bari, 2018: δm² = 7.34, 3σ 6.92–7.91 (1e-5 eV²) Bari, 2021: δm² = 7.36, 3σ 6.93–7.93 (1e-5 eV²) Bari, 2025: δm² = 7.37, 3σ 6.93–7.93 (1e-5 eV²) Bari, 2026: δm² = 7.48, 3σ 7.21–7.78 (1e-5 eV²) NuFit, 2001: δm² = 3.3 (1e-5 eV²) — below this panel's range, drawn at the floor, not to scale 3.3 NuFit, 2004: δm² = 8.1, 3σ 7.2–9.1 (1e-5 eV²) NuFit, 2012: δm² = 7.5, 3σ 7.0–8.09 (1e-5 eV²) NuFit, 2018: δm² = 7.39, 3σ 6.79–8.01 (1e-5 eV²) NuFit, 2020: δm² = 7.42, 3σ 6.82–8.04 (1e-5 eV²) NuFit, 2024: δm² = 7.49, 3σ 6.92–8.05 (1e-5 eV²) Valencia, 2018: δm² = 7.55, 3σ 7.05–8.14 (1e-5 eV²) Valencia, 2020: δm² = 7.5, 3σ 6.94–8.14 (1e-5 eV²)

Best fit with its 3σ range, normal ordering. NuFit 2001 (3.3, in units of 1e-5 eV²) lies far below the rest of this series; drawn at the panel floor and labelled with its value, not to scale, so the later measurements keep their resolution.

|Δm²| / 1e-3 eV² · normal ordering

1.13 1.92 2.7 3.49 04 12 13 16 17 18 20 21 24 25 Bari, 2012: |Δm²| = 2.43, 3σ 2.19–2.62 (1e-3 eV²) Bari, 2013: |Δm²| = 2.43, 3σ 2.23–2.61 (1e-3 eV²) Bari, 2016: |Δm²| = 2.5, 3σ 2.37–2.63 (1e-3 eV²) Bari, 2017: |Δm²| = 2.525, 3σ 2.411–2.646 (1e-3 eV²) Bari, 2018: |Δm²| = 2.455, 3σ 2.355–2.557 (1e-3 eV²) Bari, 2021: |Δm²| = 2.485, 3σ 2.401–2.565 (1e-3 eV²) Bari, 2025: |Δm²| = 2.495, 3σ 2.433–2.558 (1e-3 eV²) NuFit, 2004: |Δm²| = 2.1595, 3σ 1.3595–3.2595 (1e-3 eV²) NuFit, 2012: |Δm²| = 2.4355, 3σ 2.2385–2.6575 (1e-3 eV²) NuFit, 2018: |Δm²| = 2.48805, 3σ 2.39405–2.58505 (1e-3 eV²) NuFit, 2020: |Δm²| = 2.4799, 3σ 2.3979–2.5609 (1e-3 eV²) NuFit, 2024: |Δm²| = 2.47555, 3σ 2.41355–2.54055 (1e-3 eV²) Valencia, 2018: |Δm²| = 2.46225, 3σ 2.37225–2.56225 (1e-3 eV²) Valencia, 2020: |Δm²| = 2.5125, 3σ 2.4325–2.5925 (1e-3 eV²)

Best fit with its 3σ range, normal ordering. Values of other groups converted to our convention.

sin²θ₁₂ / 1e-1 · normal ordering

2.12 2.74 3.36 3.98 04 06 11 12 13 16 17 18 20 21 24 25 26 Bari, 2006: sin²θ₁₂ = 3.14 (1e-1) Bari, 2011: sin²θ₁₂ = 3.06, 3σ 2.59–3.59 (1e-1) Bari, 2012: sin²θ₁₂ = 3.07, 3σ 2.59–3.59 (1e-1) Bari, 2013: sin²θ₁₂ = 3.08, 3σ 2.59–3.59 (1e-1) Bari, 2016: sin²θ₁₂ = 2.97, 3σ 2.5–3.54 (1e-1) Bari, 2017: sin²θ₁₂ = 2.97, 3σ 2.5–3.54 (1e-1) Bari, 2018: sin²θ₁₂ = 3.04, 3σ 2.65–3.46 (1e-1) Bari, 2021: sin²θ₁₂ = 3.03, 3σ 2.63–3.45 (1e-1) Bari, 2025: sin²θ₁₂ = 3.03, 3σ 2.64–3.45 (1e-1) Bari, 2026: sin²θ₁₂ = 3.085, 3σ 2.866–3.303 (1e-1) NuFit, 2004: sin²θ₁₂ = 3.0, 3σ 2.3–3.8 (1e-1) NuFit, 2012: sin²θ₁₂ = 3.02, 3σ 2.67–3.44 (1e-1) NuFit, 2018: sin²θ₁₂ = 3.1, 3σ 2.75–3.5 (1e-1) NuFit, 2020: sin²θ₁₂ = 3.04, 3σ 2.69–3.43 (1e-1) NuFit, 2024: sin²θ₁₂ = 3.08, 3σ 2.75–3.45 (1e-1) Valencia, 2018: sin²θ₁₂ = 3.2, 3σ 2.73–3.79 (1e-1) Valencia, 2020: sin²θ₁₂ = 3.18, 3σ 2.71–3.69 (1e-1)

Best fit with its 3σ range, normal ordering.

sin²θ₁₃ / 1e-2 · normal ordering

-0.452 1.45 3.35 5.25 04 06 08 11 12 13 16 17 18 20 21 24 25 Bari, 2006: sin²θ₁₃ = 0.9 (1e-2) Bari, 2008: sin²θ₁₃ = 1.6 (1e-2) Bari, 2011: sin²θ₁₃ = 2.1, 3σ 0.1–4.4 (1e-2) Bari, 2012: sin²θ₁₃ = 2.41, 3σ 1.69–3.13 (1e-2) Bari, 2013: sin²θ₁₃ = 2.34, 3σ 1.76–2.95 (1e-2) Bari, 2016: sin²θ₁₃ = 2.14, 3σ 1.85–2.46 (1e-2) Bari, 2017: sin²θ₁₃ = 2.15, 3σ 1.9–2.4 (1e-2) Bari, 2018: sin²θ₁₃ = 2.14, 3σ 1.9–2.39 (1e-2) Bari, 2021: sin²θ₁₃ = 2.23, 3σ 2.04–2.44 (1e-2) Bari, 2025: sin²θ₁₃ = 2.23, 3σ 2.06–2.38 (1e-2) NuFit, 2004: sin²θ₁₃ < 4.7 (3σ) (1e-2) NuFit, 2012: sin²θ₁₃ = 2.27, 3σ 1.56–2.99 (1e-2) NuFit, 2018: sin²θ₁₃ = 2.24, 3σ 2.044–2.437 (1e-2) NuFit, 2020: sin²θ₁₃ = 2.219, 3σ 2.032–2.41 (1e-2) NuFit, 2024: sin²θ₁₃ = 2.215, 3σ 2.03–2.388 (1e-2) Valencia, 2018: sin²θ₁₃ = 2.16, 3σ 1.96–2.41 (1e-2) Valencia, 2020: sin²θ₁₃ = 2.2, 3σ 2.0–2.405 (1e-2)

Best fit with its 3σ range, normal ordering.

sin²θ₂₃ / 1e-1 · normal ordering

2.89 4.33 5.78 7.22 04 06 11 12 13 16 17 18 20 21 24 25 Bari, 2006: sin²θ₂₃ = 4.4 (1e-1) Bari, 2011: sin²θ₂₃ = 4.2, 3σ 3.4–6.4 (1e-1) Bari, 2012: sin²θ₂₃ = 3.86, 3σ 3.31–6.37 (1e-1) Bari, 2013: sin²θ₂₃ = 4.37, 3σ 3.74–6.26 (1e-1) Bari, 2016: sin²θ₂₃ = 4.37, 3σ 3.79–6.16 (1e-1) Bari, 2017: sin²θ₂₃ = 4.25, 3σ 3.81–6.15 (1e-1) Bari, 2018: sin²θ₂₃ = 5.51, 3σ 4.3–6.02 (1e-1) Bari, 2021: sin²θ₂₃ = 4.55, 3σ 4.16–5.99 (1e-1) Bari, 2025: sin²θ₂₃ = 4.73, 3σ 4.37–5.81 (1e-1) NuFit, 2004: sin²θ₂₃ = 5.0, 3σ 3.4–6.8 (1e-1) NuFit, 2012: sin²θ₂₃ = 4.13, 3σ 3.42–6.67 (1e-1) NuFit, 2018: sin²θ₂₃ = 5.82, 3σ 4.28–6.24 (1e-1) NuFit, 2020: sin²θ₂₃ = 5.73, 3σ 4.15–6.16 (1e-1) NuFit, 2024: sin²θ₂₃ = 4.7, 3σ 4.35–5.85 (1e-1) Valencia, 2018: sin²θ₂₃ = 5.47, 3σ 4.45–5.99 (1e-1) Valencia, 2020: sin²θ₂₃ = 5.74, 3σ 4.34–6.1 (1e-1)

Best fit with its 3σ range, normal ordering.

δ/π / 1 · normal ordering

0.552 1.1 1.64 2.19 12 13 16 17 18 20 21 25 Bari, 2012: δ/π = 1.08 (1) Bari, 2013: δ/π = 1.39 (1) Bari, 2016: δ/π = 1.35 (1) Bari, 2017: δ/π = 1.38 (1) Bari, 2018: δ/π = 1.32, 3σ 0.83–1.99 (1) Bari, 2021: δ/π = 1.24, 3σ 0.77–1.97 (1) Bari, 2025: δ/π = 1.2, 3σ 0.73–2.03 (1) Valencia, 2018: δ/π = 1.21, 3σ 0.87–1.94 (1) Valencia, 2020: δ/π = 1.08, 3σ 0.71–1.99 (1)

Best fit with its 3σ range, normal ordering.

The 2001 and 2004 entries predate the NuFit name. They are earlier IFIC-Valencia global fits — Gonzalez-Garcia, Maltoni, Peña-Garay and Valle in 2001; Maltoni, Schwetz, Tórtola and Valle in 2004 — grouped here with the later NuFit papers for continuity of lineage, and marked as predecessors rather than presented as NuFit releases. Two of those authors also appear on the Valencia papers listed below, so the three series are less separate before 2012 than three columns suggest.

The releases of the other groups included above, with the convention each one publishes in. Their values are stored exactly as printed and converted only at rendering time.
YearPaperPreprintSourceConvention
2001NuFit (predecessor) — Global three-neutrino oscillation analysis of neutrino dataPhys. Rev. D 63, 033005 (2001)arXiv:hep-ph/0009350Table V/VI, "Solar + ALL-ATM. + CHOOZ" (global combined analysis)tan2(theta), not sin2(theta), for every mixing angle; Dm2_32 = m3^2-m2^2 reported without a normal/inverted split
2004NuFit (predecessor) — Status of global fits to neutrino oscillationsNew J. Phys. 6, 122 (2004)arXiv:hep-ph/0405172Table 1Dm2_3l = Dm2_31 > 0; no inverted-ordering fit performed
2012NuFit — Global fit to three neutrino mixing: critical look at present precisionJHEP 12 (2012) 123arXiv:1209.3023Table 1, first column
Free Fluxes + RSBL
Dm2_3l = Dm2_31 > 0 (NO), Dm2_32 < 0 (IO)
2018NuFit — Global analysis of three-flavour neutrino oscillations: synergies and tensions in the determination of θ₂₃, δCP, and the mass orderingJHEP 01 (2019) 106arXiv:1811.05487Table 1, lower block
with SK-atm
Dm2_3l = Dm2_31 > 0 (NO), Dm2_32 < 0 (IO)
2018Valencia — Status of neutrino oscillations 2018: 3σ hint for normal mass ordering and improved CP sensitivityPhys. Lett. B 782, 633 (2018)arXiv:1708.01186Table I|Dm2_31| quoted for both orderings
2020NuFit — The fate of hints: updated global analysis of three-flavor neutrino oscillationsJHEP 09 (2020) 178arXiv:2007.14792Table 3, lower block
with SK atmospheric data
Dm2_3l = Dm2_31 > 0 (NO), Dm2_32 < 0 (IO)
2020Valencia — 2020 global reassessment of the neutrino oscillation pictureJHEP 02 (2021) 071arXiv:2006.11237Table III|Dm2_31| quoted for both orderings
2024NuFit — NuFit-6.0: updated global analysis of three-flavor neutrino oscillationsJHEP 12 (2024) 216arXiv:2410.05380Table 1, lower block
IC24, with SK atmospheric data
Dm2_3l = Dm2_31 > 0 (NO), Dm2_32 < 0 (IO)

The Bari series

11 releases · best fit and 3σ range, for both mass orderings

How to read these panels

Values are shown in the normalisation used in the papers, given under each title. The vertical rule is the 3σ range; the marker is the best fit — a circle for normal ordering, a square for inverted, a diamond where the analysis quotes a single value for both. Hover a marker for the numbers.

All entries use our convention Δm² = m₃² − ½(m₁² + m₂²) and δm² = m₂² − m₁² > 0. Analyses by other groups use different conventions and are not plotted here: the comparison above converts first, and states the conversion.

Normal ordering (circle) Inverted ordering (square) Quoted for both (diamond) Upper limit, level printed on the marker

δm² / 1e-5 eV²

6.77 7.29 7.81 8.33 06 11 12 13 16 17 18 21 25 26 Normal ordering, 2018: δm² = 7.34, 3σ 6.92–7.91 (1e-5 eV²) Inverted ordering, 2018: δm² = 7.34, 3σ 6.92–7.91 (1e-5 eV²) Both orderings, 2006: δm² = 7.92 (1e-5 eV²) Both orderings, 2011: δm² = 7.58, 3σ 6.99–8.18 (1e-5 eV²) Both orderings, 2012: δm² = 7.54, 3σ 6.99–8.18 (1e-5 eV²) Both orderings, 2013: δm² = 7.54, 3σ 6.99–8.18 (1e-5 eV²) Both orderings, 2016: δm² = 7.37, 3σ 6.93–7.97 (1e-5 eV²) Both orderings, 2017: δm² = 7.37, 3σ 6.93–7.96 (1e-5 eV²) Both orderings, 2021: δm² = 7.36, 3σ 6.93–7.93 (1e-5 eV²) Both orderings, 2025: δm² = 7.37, 3σ 6.93–7.93 (1e-5 eV²) Both orderings, 2026: δm² = 7.48, 3σ 7.21–7.78 (1e-5 eV²)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

|Δm²| / 1e-3 eV²

1.99 2.24 2.49 2.74 06 11 12 13 16 17 18 21 25 Normal ordering, 2012: |Δm²| = 2.43, 3σ 2.19–2.62 (1e-3 eV²) Normal ordering, 2013: |Δm²| = 2.43, 3σ 2.23–2.61 (1e-3 eV²) Normal ordering, 2016: |Δm²| = 2.5, 3σ 2.37–2.63 (1e-3 eV²) Normal ordering, 2017: |Δm²| = 2.525, 3σ 2.411–2.646 (1e-3 eV²) Normal ordering, 2018: |Δm²| = 2.455, 3σ 2.355–2.557 (1e-3 eV²) Normal ordering, 2021: |Δm²| = 2.485, 3σ 2.401–2.565 (1e-3 eV²) Normal ordering, 2025: |Δm²| = 2.495, 3σ 2.433–2.558 (1e-3 eV²) Inverted ordering, 2012: |Δm²| = 2.42, 3σ 2.17–2.61 (1e-3 eV²) Inverted ordering, 2013: |Δm²| = 2.38, 3σ 2.19–2.56 (1e-3 eV²) Inverted ordering, 2016: |Δm²| = 2.46, 3σ 2.33–2.6 (1e-3 eV²) Inverted ordering, 2017: |Δm²| = 2.505, 3σ 2.39–2.624 (1e-3 eV²) Inverted ordering, 2018: |Δm²| = 2.441, 3σ 2.338–2.54 (1e-3 eV²) Inverted ordering, 2021: |Δm²| = 2.455, 3σ 2.376–2.541 (1e-3 eV²) Inverted ordering, 2025: |Δm²| = 2.465, 3σ 2.403–2.527 (1e-3 eV²) Both orderings, 2006: |Δm²| = 2.4 (1e-3 eV²) Both orderings, 2011: |Δm²| = 2.35, 3σ 2.06–2.67 (1e-3 eV²)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

sin²θ₁₂ / 1e-1

2.37 2.82 3.27 3.72 06 11 12 13 16 17 18 21 25 26 Normal ordering, 2018: sin²θ₁₂ = 3.04, 3σ 2.65–3.46 (1e-1) Inverted ordering, 2018: sin²θ₁₂ = 3.03, 3σ 2.64–3.45 (1e-1) Both orderings, 2006: sin²θ₁₂ = 3.14 (1e-1) Both orderings, 2011: sin²θ₁₂ = 3.06, 3σ 2.59–3.59 (1e-1) Both orderings, 2012: sin²θ₁₂ = 3.07, 3σ 2.59–3.59 (1e-1) Both orderings, 2013: sin²θ₁₂ = 3.08, 3σ 2.59–3.59 (1e-1) Both orderings, 2016: sin²θ₁₂ = 2.97, 3σ 2.5–3.54 (1e-1) Both orderings, 2017: sin²θ₁₂ = 2.97, 3σ 2.5–3.54 (1e-1) Both orderings, 2021: sin²θ₁₂ = 3.03, 3σ 2.63–3.45 (1e-1) Both orderings, 2025: sin²θ₁₂ = 3.03, 3σ 2.64–3.45 (1e-1) Both orderings, 2026: sin²θ₁₂ = 3.085, 3σ 2.866–3.303 (1e-1)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

sin²θ₁₃ / 1e-2

-0.416 1.36 3.14 4.92 06 08 11 12 13 16 17 18 21 25 Normal ordering, 2012: sin²θ₁₃ = 2.41, 3σ 1.69–3.13 (1e-2) Normal ordering, 2013: sin²θ₁₃ = 2.34, 3σ 1.76–2.95 (1e-2) Normal ordering, 2016: sin²θ₁₃ = 2.14, 3σ 1.85–2.46 (1e-2) Normal ordering, 2017: sin²θ₁₃ = 2.15, 3σ 1.9–2.4 (1e-2) Normal ordering, 2018: sin²θ₁₃ = 2.14, 3σ 1.9–2.39 (1e-2) Normal ordering, 2021: sin²θ₁₃ = 2.23, 3σ 2.04–2.44 (1e-2) Normal ordering, 2025: sin²θ₁₃ = 2.23, 3σ 2.06–2.38 (1e-2) Inverted ordering, 2012: sin²θ₁₃ = 2.44, 3σ 1.71–3.15 (1e-2) Inverted ordering, 2013: sin²θ₁₃ = 2.4, 3σ 1.78–2.98 (1e-2) Inverted ordering, 2016: sin²θ₁₃ = 2.18, 3σ 1.86–2.48 (1e-2) Inverted ordering, 2017: sin²θ₁₃ = 2.16, 3σ 1.9–2.42 (1e-2) Inverted ordering, 2018: sin²θ₁₃ = 2.18, 3σ 1.95–2.43 (1e-2) Inverted ordering, 2021: sin²θ₁₃ = 2.23, 3σ 2.03–2.45 (1e-2) Inverted ordering, 2025: sin²θ₁₃ = 2.23, 3σ 2.08–2.41 (1e-2) Both orderings, 2006: sin²θ₁₃ = 0.9 (1e-2) Both orderings, 2008: sin²θ₁₃ = 1.6, 3σ 0.6–2.6 (1e-2) Both orderings, 2011: sin²θ₁₃ = 2.1, 3σ 0.1–4.4 (1e-2)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

sin²θ₂₃ / 1e-1

2.91 4.28 5.66 7.03 06 11 12 13 16 17 18 21 25 Normal ordering, 2012: sin²θ₂₃ = 3.86, 3σ 3.31–6.37 (1e-1) Normal ordering, 2013: sin²θ₂₃ = 4.37, 3σ 3.74–6.26 (1e-1) Normal ordering, 2016: sin²θ₂₃ = 4.37, 3σ 3.79–6.16 (1e-1) Normal ordering, 2017: sin²θ₂₃ = 4.25, 3σ 3.81–6.15 (1e-1) Normal ordering, 2018: sin²θ₂₃ = 5.51, 3σ 4.3–6.02 (1e-1) Normal ordering, 2021: sin²θ₂₃ = 4.55, 3σ 4.16–5.99 (1e-1) Normal ordering, 2025: sin²θ₂₃ = 4.73, 3σ 4.37–5.81 (1e-1) Inverted ordering, 2012: sin²θ₂₃ = 3.92, 3σ 3.35–6.63 (1e-1) Inverted ordering, 2013: sin²θ₂₃ = 4.55, 3σ 3.8–6.41 (1e-1) Inverted ordering, 2016: sin²θ₂₃ = 5.69, 3σ 3.83–6.37 (1e-1) Inverted ordering, 2017: sin²θ₂₃ = 5.89, 3σ 3.84–6.36 (1e-1) Inverted ordering, 2018: sin²θ₂₃ = 5.57, 3σ 4.44–6.03 (1e-1) Inverted ordering, 2021: sin²θ₂₃ = 5.69, 3σ 4.17–6.06 (1e-1) Inverted ordering, 2025: sin²θ₂₃ = 5.45, 3σ 4.43–5.83 (1e-1) Both orderings, 2006: sin²θ₂₃ = 4.4 (1e-1) Both orderings, 2011: sin²θ₂₃ = 4.2, 3σ 3.4–6.4 (1e-1)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

δ/π / 1

0.574 1.11 1.65 2.19 12 13 16 17 18 21 25 Normal ordering, 2012: δ/π = 1.08, 3σ 0.77–1.36 (1) Normal ordering, 2013: δ/π = 1.39, 3σ 1.12–1.77 (1) Normal ordering, 2016: δ/π = 1.35, 3σ 1.13–1.64 (1) Normal ordering, 2017: δ/π = 1.38, 3σ 1.18–1.61 (1) Normal ordering, 2018: δ/π = 1.32, 3σ 0.83–1.99 (1) Normal ordering, 2021: δ/π = 1.24, 3σ 0.77–1.97 (1) Normal ordering, 2025: δ/π = 1.2, 3σ 0.73–2.03 (1) Inverted ordering, 2012: δ/π = 1.09, 3σ 0.83–1.47 (1) Inverted ordering, 2013: δ/π = 1.31, 3σ 0.98–1.6 (1) Inverted ordering, 2016: δ/π = 1.32, 3σ 1.07–1.67 (1) Inverted ordering, 2017: δ/π = 1.31, 3σ 1.12–1.62 (1) Inverted ordering, 2018: δ/π = 1.52, 3σ 1.07–1.92 (1) Inverted ordering, 2021: δ/π = 1.52, 3σ 1.07–1.9 (1) Inverted ordering, 2025: δ/π = 1.48, 3σ 1.12–1.83 (1)

Best fit with its 3σ range, by year of publication. Values and sources in the table below.

The releases

11 releases · 393 values, 381 of them checked against the source they name

Every value on this page is transcribed from the table or equations named here. Every best fit, and every range endpoint the paper prints, is checked against that source by tools/tests/test_history_numbers.py, which re-reads each paper on every run. The exceptions are 12 range endpoints in the two earliest releases, where the paper states a central value ± an error and the endpoints are computed from it: the register marks those releases as derived, and the test does not search the source for a number the source never printed. Papers marked as partial updates revise only part of the parameter set.
YearPaperPreprintSource table
2006Global analysis of three-flavor neutrino masses and mixingsProg. Part. Nucl. Phys. 57, 742 (2006)arXiv:hep-ph/0506083Eq. (53)-(57), global oscillation analysis
2008Hints of θ₁₃ > 0 from global neutrino data analysisPhys. Rev. Lett. 101, 141801 (2008)arXiv:0806.2649Eq. (3), all oscillation datapartial update
2011Evidence of θ₁₃ > 0 from global neutrino data analysisPhys. Rev. D 84, 053007 (2011)arXiv:1106.6028Table I (old reactor fluxes)
2012Global analysis of neutrino masses, mixings and phases: entering the era of leptonic CP violation searchesPhys. Rev. D 86, 013012 (2012)arXiv:1205.5254Table I
2013Status of three-neutrino oscillation parameters, circa 2013Phys. Rev. D 89, 093018 (2014)arXiv:1312.2878Table I
2016Neutrino masses and mixings: Status of known and unknown 3ν parametersNucl. Phys. B 908, 218 (2016)arXiv:1601.07777Table 1
2017Global constraints on absolute neutrino masses and their orderingPhys. Rev. D 95, 096014 (2017)arXiv:1703.04471Table I
2018Current unknowns in the three neutrino frameworkProg. Part. Nucl. Phys. 102, 48 (2018)arXiv:1804.09678Table 1
2021Unfinished fabric of the three neutrino paradigmPhys. Rev. D 104, 083031 (2021)arXiv:2107.00532Table I
2025Neutrino masses and mixing: Entering the era of subpercent precisionPhys. Rev. D 111, 093006 (2025)arXiv:2503.07752Table Icurrent release
2026Updated bounds on the (1, 2) neutrino oscillation parameters after first JUNO resultsPhys. Rev. D 114, 016026 (2026)arXiv:2511.21650Table I, row “w/ SNO+ & JUNO 2025”partial update

The solar-sector-only papers of 2002–2003 that precede this series are deliberately not on this page: none of them tabulates the full parameter set this page tracks. They are listed, with the reason, in the excluded section of the source data (site-src/data/history.yaml).

The register as data

same rows, two formats, stable URLs

FileFormatURL
Parameter historyJSON/data/history.json
Parameter historyCSV/data/history.csv

Both files hold exactly the same rows and the same columns — CSV for a spreadsheet or a quick pandas.read_csv, JSON for a script. The URLs are stable: a regenerated release corrects rows in place, it does not move the file. They are written by tools/make_history_data.py from the same history.yaml this page is drawn from, so the page and the files cannot disagree — nothing is retyped between them.

The JSON file is an object, not a bare array: the rows sit under a rows key, and beside them a note says in one sentence what the two value columns mean, so a copy that has been downloaded and passed on still carries its own reading instructions. The CSV has nowhere to put that note: its first line is the header of the 13 field names, and every line after it is one row, in the same order.

The two value columns

Every row carries the value twice, under names that cannot be confused. value_as_published is exactly what the paper printed, in the paper's own convention and normalisation. value_our_convention is the same quantity converted to ours: δm² = m₂² − m₁² > 0 and Δm² = m₃² − (m₁² + m₂²)/2.

Only Dm2 is ever converted. NuFit reports Δm²₃ℓ and Valencia reports |Δm²₃₁| for both orderings — neither is our Δm², and the sign of the correction is not even the same for the two groups, as the note at the top of this page works through. The conversion lives in one function, to_our_Dm2() in tools/make_history.py, and nowhere else. For every other parameter, and for Bari's own Dm2 which already is our convention, the two columns hold the identical number. That repetition is deliberate: a downloader gets one column that is directly comparable across all six parameters and all three groups, without first having to know which value needed the arithmetic.

13 columns, one row per (group, year, parameter, ordering) point. This table is generated from the exporter's own field list, so it cannot describe a column the files do not carry.
FieldMeaning
groupbari, nufit or valencia
yearPublication year of the release, as used on this page
arxivarXiv identifier of the source paper
journalFull journal citation
tableThe table or equation, inside that paper, the value is transcribed from and checked against
parameterOne of dm2 (δm²), Dm2 (Δm²), sin2_th12, sin2_th13, sin2_th23, delta_pi (δ/π)
orderingno (normal), io (inverted) or any (the paper quotes one value for both, or does not split by ordering)
conventionThe release's own stated mass-splitting convention, verbatim from history.yaml
kindmeasurement (a best fit) or limit (a bound with no measured central value)
value_as_publishedThe number as printed in the paper — see the two value columns, above
value_our_conventionThe same quantity in this group's convention — see the two value columns, above
unitThe normalisation the value is written in: 1e-2 means the printed digits are hundredths
levelFor a limit row only: the confidence level the bound holds at (3sigma, 2sigma, 90%CL or 95%CL). A measurement row has none, and each format says so in its own way: null in JSON, an empty field in CSV