Absence of Quadratic-Order Sensitivity to Small Neutrino Mass Splittings in Disappearance Measurements

Sanjeev Kumar Verma · 2026

Read Absence of Quadratic-Order Sensitivity to Small Neutrino Mass Splittings in Disappearance Measurements by Sanjeev Kumar Verma (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-27681.txt.

Introduction

Verma shows a blind spot in neutrino disappearance searches. Binned energy spectra may hide small mass splittings. In the small-phase limit, the signal shift is quadratic. That shift also looks smooth across energy bins. Fit nuisance terms can mimic the same smooth shape. The chi-squared then stays flat at quadratic order. Sensitivity must come from higher-order effects. Or it must come from tighter limits on spectral freedom.

Catalog record

CREATORSanjeev Kumar Verma
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE14777
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.27681

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Absence of Quadratic-Order Sensitivity to Small Neutrino Mass Splittings in Disappearance Measurements Sanjeev Kumar Verma Abstract Neutrino disappearance measurements using binned reconstructed-energy spectra exhibit a regime in which small mass-squared splittings become unidentifiable at quadratic order when smooth spectral shape uncertainties are represented by profiled nuisance parameters in the fit. In the small-phase limit, the oscillation-induced modification of the detected spectrum is quadratic in the mass-squared splitting and produces a smooth deformation of the reconstructed-energy distribution. If the nuisance deformation functions used in the fit can reproduce this energy dependence across the fitted bins, the quadratic oscillation-induced distortion can be absorbed by the systematic deformation space and the profiled chi-squared remains unchanged at this order. Sensitivity to the mass-squared splitting then arises only from higher-order oscillation effects or from restrictions imposed on the allowed smooth spectral freedom. I Introduction Neutrino disappearance measurements probe neutrino mass-squared splittings through energy-dependent distortions of detected event spectra at fixed source–detector baselines Pontecorvo1968 ; GiuntiKim . In practice, these measurements are performed using binned event rates as a function of a reconstructed observable, such as visible energy, charged-lepton energy, or recoil energy, rather than the neutrino energy itself. Sensitivity to oscillation parameters is extracted by fitting the reconstructed-energy spectrum, with correlated systematic uncertainties incorporated through nuisance parameters that allow smooth spectral deformations Fogli2002 . Spectral disappearance analyses of this type are standard in both long- and short-baseline experiments MINOS2013 . A natural question therefore arises: what determines sensitivity to very small mass-squared splittings when the oscillation phase remains small across the energies contributing to the observed spectrum? In this regime, the leading oscillation-induced modification of the detected spectrum is smooth and quadratic in the mass-squared splitting, suggesting the possibility that it may be degenerate with smooth spectral deformations allowed by systematic uncertainties. This work analyzes the origin of sensitivity to small neutrino mass-squared splittings in disappearance measurements. The focus is on the regime in which the oscillation phase remains small over the neutrino energies contributing to the reconstructed-energy bins used in the fit. In this limit, the survival probability admits a systematic expansion in powers of \Delta m^{2} , and the resulting modification of the reconstructed-energy spectrum takes a particularly simple form. The analysis shows that, once smooth energy-dependent systematic variations are allowed to vary freely, the leading oscillation-induced distortion becomes indistinguishable from an allowed systematic deformation. As a consequence, the profiled goodness of fit remains unchanged relative to the no-oscillation hypothesis at quadratic order in the mass-squared splitting. The central result of this work is that, in the small-phase regime, the quadratic oscillation-induced spectral distortion becomes unidentifiable whenever it lies within the span of smooth deformation functions allowed by the nuisance-parameter model used in the fit. This reflects a general identifiability condition: an oscillation parameter cannot be constrained by spectral data when the leading oscillation-induced distortion lies within the space of spectral variations allowed by the systematic model. The result presented here rests on a small number of clearly defined assumptions. The analysis applies to disappearance measurements performed at fixed baseline using binned spectra in a reconstructed-energy observable, with sensitivity extracted through a chi-squared fit that profiles over nuisance parameters representing smooth, correlated spectral shape uncertainties Fogli2002 . The oscillation phase is assumed to remain small over the neutrino energies contributing to the fitted bins, so that the survival probability admits a systematic expansion in \Delta m^{2} . At quadratic order in this expansion, the oscillation-induced modification of the detected spectrum is smooth and non-oscillatory in reconstructed energy. Finally, it is assumed that the nuisance deformation functions used in the fit can represent this leading quadratic energy dependence. No assumptions are made about the microscopic origin of the detected spectrum, the detailed detector response, or the functional form of higher-order oscillation effects. II Observed Spectrum and Fit Framework Detected events are binned according to a reconstructed-energy observable E . The data consist of observed event counts N_{i}^{\rm obs} in bins labeled by index i , with representative reconstructed energies E_{i} . All event counts refer to detected final-state events and are functions of the reconstructed-energy observable. The no-oscillation prediction N_{i}^{0} denotes the expected event count in bin i in the absence of oscillations. When oscillations are included but systematic deformations are not yet applied, the predicted spectrum is denoted by N_{i}^{\rm osc}(\Delta m^{2},\theta) , where \Delta m^{2} is the relevant neutrino mass-squared splitting and \theta the associated mixing angle. In the limit of vanishing oscillations, this reduces to the no-oscillation prediction, N_{i}^{\rm osc}(0,\theta)=N_{i}^{0} . Smooth energy-dependent systematic uncertainties are incorporated through nuisance parameters \xi_{k} multiplying a set of smooth deformation functions f_{k}(E) defined in reconstructed energy. These functions represent the correlated spectral shape variations allowed in the fit. The full predicted spectrum entering the statistical analysis is therefore written as Pontecorvo1968 ; GiuntiKim N_{i}(\Delta m^{2},\theta,\boldsymbol{\xi})=N_{i}^{\rm osc}(\Delta m^{2},\theta)\left(1+\sum_{k}\xi_{k}f_{k}(E_{i})\right). (1) The agreement between prediction and data is quantified using a binned chi-squared statistic, \chi^{2}=\sum_{i}\frac{(N_{i}^{\rm obs}-N_{i})^{2}}{\sigma_{i}^{2}}+\sum_{k}\frac{\xi_{k}^{2}}{\sigma_{\xi_{k}}^{2}}. (2) For each value of \Delta m^{2} , the chi-squared is minimized with respect to \boldsymbol{\xi} . The test statistic is \Delta\chi^{2}(\Delta m^{2})=\chi^{2}(\Delta m^{2})-\chi^{2}_{\rm min}. (3) III Small-Phase Expansion For neutrino disappearance at a fixed baseline L , the two-flavor survival probability as a function of the neutrino energy E_{\nu} is Pontecorvo1968 ; GiuntiKim P_{\rm surv}(E_{\nu};\Delta m^{2},\theta)=1-\sin^{2}2\theta\,\sin^{2}\!\left(\frac{\Delta m^{2}L}{4E_{\nu}}\right). (4) The oscillation phase is therefore \phi(E_{\nu})=\frac{\Delta m^{2}L}{4E_{\nu}}. (5) The analysis focuses on the small-phase regime defined by |\phi(E_{\nu})|\ll 1 . This condition is understood to hold over the neutrino-energy range contributing significantly to each reconstructed-energy bin, so that the quadratic expansion remains valid after convolution with detector response and binning. In this limit, \sin^{2}\phi(E_{\nu})=\phi(E_{\nu})^{2}+\mathcal{O}(\phi^{4}). (6) We define the oscillation-induced change in the predicted number of detected events in reconstructed-energy bin i , prior to allowing any systematic deformations, as \delta N_{i}\equiv N_{i}^{\rm osc}(\Delta m^{2},\theta)-N_{i}^{0}. (7) At quadratic order in \Delta m^{2} , the fractional distortion takes the generic form \frac{\delta N_{i}}{N_{i}^{0}}=-\alpha(\Delta m^{2},\theta)\,h(E_{i})+\mathcal{O}\!\left((\Delta m^{2})^{4}\right), (8) where \alpha(\Delta m^{2},\theta)=\sin^{2}2\theta\left(\frac{L}{4}\right)^{2}(\Delta m^{2})^{2}. (9) The quadratic dependence arises from expanding the oscillation probability in the small-phase parameter inside the bin-level event-rate integrals that define the reconstructed-energy spectrum. The function h(E_{i}) is defined by Eq. (9) as the detector-level coefficient of (\Delta m^{2})^{2} in the fractional spectral distortion. Physically, h(E_{i}) corresponds to a weighted average of the inverse-square neutrino-energy dependence over the neutrino energies contributing to reconstructed-energy bin i . At this order, the distortion is non-oscillatory and varies smoothly with reconstructed energy. Smooth distortions of the neutrino-energy spectrum produce correspondingly smooth distortions of the reconstructed-energy spectrum after detector smearing. The quadratic oscillation-induced deformation therefore remains smooth in reconstructed energy. IV Degeneracy with Smooth Spectral Systematics Let \{f_{1},\ldots,f_{K}\} denote the set of smooth spectral deformation functions used in the fit. We assume that these deformation functions can approximate h(E) at the level required by the quadratic expansion in \Delta m^{2} . That is, h(E_{i})=\sum_{k}a_{k}f_{k}(E_{i})+r_{i}, (10) where r_{i} denotes a possible residual. In realistic analyses, individual nuisance modes such as overall normalization or a single energy-tilt deformation will not generally reproduce h(E) . The degeneracy becomes exact only when the combined space of allowed smooth shape variations is sufficiently flexible to approximate this dependence, in which case the residual r_{i} controls the remaining sensitivity. The first term represents the

Read the complete text