The role of final-state interaction modeling in neutrino energy reconstruction and oscillation measurements

Abstract

Next-generation long-baseline neutrino oscillation experiments promise to provide dramatically improved measurements of PMNS neutrino oscillation parameters, including measurements of charge-parity violation in the lepton sector, in addition to searches for new physics. Achieving such precise measurements requires inferring neutrino oscillation probabilities over a wide neutrino energy range, which demands the most accurate neutrino energy reconstruction through precise measurements of all visible final-state particles produced in neutrino interactions. However, any reconstruction will inevitably miss a significant fraction of energy when it is, for example, carried away by neutrons, the nuclear remnant or unidentified charged pions. The size of the subsequent neutrino energy reconstruction bias is affected by many aspects of neutrino interaction physics, but the poorly understood re-interactions of struck hadrons within the nuclear medium, final-state interactions (FSI), are especially important. In this work, we assess how variations of FSI modeling affect neutrino energy reconstruction. As a case study, we use the neutrino flux and baseline of the upcoming DUNE experiment to illustrate that FSI model variations, in the absence of robust near-detector constraints, have the potential to be degenerate with variations of neutrino oscillation parameters at the level of projected precision for future measurements. The results highlight the need for further development of sophisticated FSI models, alongside capable near detectors at next-generation experiments to constrain them.

Figures (4)

• Figure 1: The simulated distribution of neutrino energy bias for muon neutrino (top) and antineutrino (bottom) charged-current interactions, using the DUNE neutrino and antineutrino beam-mode fluxes respectively before oscillations occur, for a variety of different FSI models. Each distribution is generated using 5 million charged-current events. The blue shaded histogram assumes no FSI, while the other histograms correspond to the four FSI models. The expression for $E^{\mathrm{rec}}_{\nu}$ is given in \ref{['eq:enuhad']} and $E_{\nu}$ corresponds to the true neutrino energy. The bottom panel shows the ratios of different FSI models with respect to the hA2018 baseline model.
• Figure 2: Comparison of the simulated $\nu_\mu$ reconstructed energy spectra and an oscillated DUNE experiment flux using different FSI models, to variations of $\Delta m_{32}^2$. This is simulated for neutrino (top) and antineutrino (bottom) beam modes. In each sub-figure, reconstructed energy spectra under different FSI models are shown as histograms and $\pm 0.4$% $\Delta m_{32}^2$ variations are shown with shaded blue (+) and red (-) bands centered on the hA2018 baseline. The error bars are broadly indicative of the DUNE experiment's statistical uncertainty with 10 years of operation. The inset provides a zoomed-in view of the first oscillation maximum. The lower panel displays ratios of FSI model and $\Delta m_{32}^2$ variations relative to the hA2018 baseline. In the lower panel, the bands show 1-$\sigma$ (inner) and 2-$\sigma$ (outer) variations to $\Delta m_{32}^2$.
• Figure 3: Comparison of the simulated $\nu_e$ reconstructed energy spectra and an oscillated DUNE experiment flux using different FSI models, to variations of $\Delta m_{32}^2$. This is simulated for neutrino (top) and antineutrino (bottom) beam modes. The left and right correspond to $\delta_{\rm CP}=0$ and $\delta_{\rm CP}-\pi/2$, respectively. Other figure elements follow the same conventions as described in \ref{['fig:numu_dm32']}, except that only the 1-$\sigma$ uncertainty band is shown in the lower panels.
• Figure 4: Comparison of the simulated $\nu_\mu$ reconstructed energy spectra and an oscillated DUNE experiment flux using different FSI models, to $\pm 1.3$% variations of $\sin^2\theta_{23}$. This is simulated for neutrino (top) and antineutrino (bottom) beam modes. Other figure elements follow the same conventions as described in \ref{['fig:numu_dm32']}.