Dual baseline search for muon neutrino disappearance at 0.5 eV^2 < Δm^2 < 40 eV^2

TL;DR

The paper addresses νμ disappearance into sterile states within the Δm^2 region $0.5$–$40~ ext{eV}^2$ by performing a joint analysis of SciBooNE and MiniBooNE data. It employs two complementary strategies—a simultaneous fit of all samples and a spectrum-fit that leverages SciBooNE corrections to MiniBooNE predictions—to constrain oscillations while significantly reducing flux and cross-section systematics. Neither analysis finds evidence for disappearance; both yield 90% CL limits with improved sensitivity in the $10$–$30~ ext{eV}^2$ range, and best-fit-like indications around Δm^2 ≈ 40–44 eV^2 that are still consistent with the null hypothesis given uncertainties. The results demonstrate the power of cross-detector constraints to tighten sterile-neutrino limits and illustrate a methodological framework for future multi-detector oscillation searches.

Abstract

The SciBooNE and MiniBooNE collaborations report the results of a ν_μdisappearance search in the Δm^2 region of 0.5-40 eV^2. The neutrino rate as measured by the SciBooNE tracking detectors is used to constrain the rate at the MiniBooNE Cherenkov detector in the first joint analysis of data from both collaborations. Two separate analyses of the combined data samples set 90% confidence level (CL) limits on ν_μdisappearance in the 0.5-40 eV^2 Δm^2 region, with an improvement over previous experimental constraints between 10 and 30 eV^2.

Figures (9)

• Figure 1: (Top) Predicted number of CC events in the SciBar FV as a function of true $E_\nu$. The number of selected events in each sub-sample are also shown. (Bottom) Detection efficiency as a function of true neutrino energy for each sub-sample Nakajima:2010fp.
• Figure 2: $E_{\nu}^{QE}$ distribution for SciBooNE data (black) with statistical errors, and prediction assuming no oscillations (relatively normalized by the total number of SciBar-stopped and MRD-stopped events) for the SciBar-stopped sample (top) and MRD-stopped sample (bottom). Cosmic background is subtracted. Attached to the prediction are the size of the systematic uncertainty (shaded boxes). The predicted non-CCQE events (dash) events are also shown.
• Figure 3: Top: $E_{\nu}^{QE}$ distribution for MiniBooNE data (black) with statistical errors, and prediction assuming no oscillations for the simultaneous fit analysis (relatively normalized by the total number of SciBar-stopped and MRD-stopped events). Attached to the prediction are the size of the systematic uncertainty (shaded boxes). The predicted non-CCQE events (dash) events are also shown. Bottom: Same as above, for the spectrum fit analysis, where the prediction is scaled by the SciBooNE corrections.
• Figure 4: Correlation coefficients of the total systematic uncertainties on the reconstructed $E_\nu$ distribution for the simultaneous fit analysis. Bins 0-15 and bins 16-31 are respectively for the SciBar-stopped and the MRD-stopped samples from SciBooNE, shown in Fig. \ref{['datamc_sb']}. Bins 32-47 are for the MiniBooNE sample shown in the top panel of Fig. \ref{['datamc_mb']}.
• Figure 5: The fractional size of the systematic uncertainties for each MiniBooNE reconstructed $E_\nu$ bins for the spectrum fit analysis. The dashed lines show the uncertainties using MiniBooNE alone, and the solid lines show the uncertainties after the SciBooNE corrections.