Testing the heavy decaying sterile neutrino hypothesis at the DUNE near detector

TL;DR

The LSND and MiniBooNE anomalies may be explained by a heavy decaying sterile neutrino, not conventional oscillations. The authors evaluate the DUNE ND-LAr near detector’s sensitivity to a heavy decaying sterile neutrino (HDSN) in both Dirac and Majorana realizations, using realistic fluxes, detector response, and a Poisson $\chi^2$ framework with $10\%$ systematics. They find that ND-LAr can probe a substantially larger region in the $(|U_{0|4}|^2, g m_4)$ parameter plane than the SBN program and could confirm or refute MicroBooNE/SBN hints, with the potential to distinguish Dirac from Majorana signatures via appearance spectra for $g m_4$ above a few eV. These results guide near-detector strategies for testing non-oscillation explanations of LSND/MiniBooNE and highlight the broader relevance of heavy-decaying neutrino searches at DUNE.

Abstract

One of the most convincing explanations of the LSND and MiniBooNE anomalies relies on a heavy, mostly sterile neutrino with a small muon neutrino component, which decays to an electron neutrino and an invisible light scalar field. We investigate the possibility to test this hypothesis at the near detector complex of the upcoming DUNE experiment. We find that the DUNE liquid argon near detector (ND-LAr) can probe a larger region of the parameter space than the Fermilab SBN program, and may help to confirm or reject a possible hint of $ν_e$ appearance in future MicroBooNE, SBND or ICARUS data. We also argue that it may be possible to distinguish between Dirac and Majorana neutrinos if this scenario is realized in Nature.

Figures (8)

• Figure 1: Expected muon neutrino (left) and antineutrino (right) fluxes at the movable DUNE liquid argon near detector (ND-LAr), for on-axis and several off-axis locations.
• Figure 2: Appearance (left) and disappearance (right) event spectra expected at the DUNE ND-LAr detector in the Standard Model, assuming a running time of 3.5 years on axis in either neutrino or antineutrino mode. The top and bottom panels correspond to the neutrino and antineutrino modes, respectively. In each plot, the signal and the various background events are represented with different colours (see legend).
• Figure 3: Appearance event spectra expected at the DUNE ND-LAr detector for various choices of the HDSN parameters $|U_{\mu 4}|^2$ and $g m_4$ (Dirac case), assuming a running time of 3.5 years on axis in either neutrino or antineutrino mode. The top and bottom panels correspond to the neutrino and antineutrino modes, respectively, and the colour code is the same as in the left panels of Fig. \ref{['fig:ND_spectra_SM']}.
• Figure 4: Disappearance event spectra expected at the DUNE ND-LAr detector for various choices of the HDSN parameters $|U_{\mu 4}|^2$ and $g m_4$ (Dirac case), assuming a running time of 3.5 years on axis in either neutrino or antineutrino mode. The red (green) curves show the difference between the number of signal (background) events predicted in the SM and HDSN scenario as a function of the reconstructed (anti)neutrino energy. The top and bottom panels correspond to the neutrino and antineutrino modes, respectively.
• Figure 5: Expected ND-LAr sensitivity to the heavy decaying sterile neutrino parameters $|U_{\mu 4}|^2$ and $g m_4$ (Dirac case), assuming 3.5 years of on-axis data taking in both neutrino and antineutrino modes, and 10% signal and background uncertainties. The left plot shows the 95% C.L. sensitivities obtained using appearance data only (green long-dashed curve), disappearance data only (blue short-dashed curve), and both appearance and disappearance data (red solid curve). The right plot compares the appearance + disappearance combined sensitivities at various confidence levels. The regions that are expected to be excluded by future data are enclosed within these curves.