Upper limit / exclusion
The largest value of a quantity consistent with the data at a given confidence level when no signal is seen.
Examples include KATRIN's mβ < 0.45 eV (90% CL) and GERDA's lower half-life limit. Exclusion plots show regions of parameter space (for example sterile Δm² vs sin²2θ) ruled out by an experiment.
In plain words
When an experiment looks for something and finds nothing, the result is still useful: it says how large the effect could be without having been noticed. That boundary is an upper limit. Most knowledge of the absolute neutrino mass and of neutrinoless double beta decay has so far come in this form.
The simplest case
With no background and zero events observed, Poisson statistics excludes any true mean above 2.30 events at 90% CL, and above 3.00 at 95% CL. Dividing by the expected events per unit of signal turns this into a limit on a rate, a cross section or a coupling. With background, methods such as Feldman–Cousins or CLs avoid unphysically strong limits.
Examples
- KATRIN published mβ < 0.45 eV at 90% CL in 2025, the tightest direct laboratory limit on the neutrino mass.
- For double beta decay a lower limit on the half-life becomes an upper limit on the effective Majorana mass. KamLAND-Zen reported T½ > 3.8 × 10²⁶ years for ¹³⁶Xe, corresponding to mββ below about 28–122 meV depending on the nuclear matrix element.
- Exclusion contours show which combinations of parameters are ruled out, for example regions of Δm² and sin²2θ for a sterile neutrino.
Limit versus sensitivity
A limit depends on how the background fluctuated. If fewer background events occurred than expected, the limit is stronger than the experiment's design reach; if more, it is weaker. Experiments therefore report the observed limit together with the median sensitivity. LEGEND-200, for instance, observed T½ > 0.5 × 10²⁶ years against a median sensitivity of 1.0 × 10²⁶ years.
Frequently asked questions
Does an upper limit mean the quantity is zero?
No. It only means the data cannot distinguish values below the limit from zero. The neutrino mass is known to be non-zero from oscillations, yet direct experiments only give upper limits.
Why can a half-life limit be called a lower limit?
Because a longer half-life means a rarer decay. Seeing no decays rules out short half-lives, so the bound is on the low side of the half-life and on the high side of the decay rate.
Original papers
- TheoryG. J. Feldman and R. D. Cousins, A Unified approach to the classical statistical analysis of small signals, Phys. Rev. D 57, 3873 (1998)DOI 10.1103/PhysRevD.57.3873arXiv:physics/9711021
Related terms
People
On the timeline
- 2019KATRIN's first limit: below 1.1 eV
- 2022KATRIN breaks the 1 eV barrier
- 2025KATRIN: neutrinos weigh less than 0.45 eV
Mentioned in 11 pages
Lessons 1
Milestones 3
People 1
Last updated: · Editorial standard
Cite this page
Text
Neutrino Atlas (2026). “Upper limit / exclusion”. Retrieved 22 November 2026, from https://neutrino-atlas.com/terms/upper-limit/BibTeX
@misc{neutrinoatlas2026termsupperlimit,
title = {Upper limit / exclusion},
author = {{Neutrino Atlas}},
year = {2026},
url = {https://neutrino-atlas.com/terms/upper-limit/},
note = {Accessed 2026-11-22},
language = {english}
}