Sterile Neutrino Search with MicroBooNE's Electron Neutrino Disappearance Data

A sterile neutrino is a well motivated minimal new physics model that leaves an imprint in neutrino oscillations. Over the last two decades, a number of hints pointing to a sterile neutrino have emerged, many of which are pointing near m_{4}∼1  eV. Here, we show how MicroBooNE data can be used to search for electron neutrino disappearance using each of their four analysis channels. We find a hint for oscillations with the highest single channel significance of 2.4σ (using the Feldman-Cousins approach) coming from the Wire-Cell analysis and a simplified treatment of the experimental systematics. The preferred parameters are sin^{2}(2θ_{14})=0.35_{-0.16}^{+0.19} and Δm_{41}^{2}=1.25_{-0.39}^{+0.74}  eV^{2}. This region of parameter space is in good agreement with existing hints from source experiments, is at a similar frequency but higher mixing than indicated by reactor antineutrinos, and is at the edge of the region allowed by solar neutrino data. Existing unanalyzed data from MicroBooNE could increase the sensitivity to the >3σ level.

Grid-based minimization at scale: Feldman-Cousins corrections for light sterile neutrino search

High Energy Physics (HEP) experiments generally employ sophisticated statistical methods to present results in searches of new physics. In the problem of searching for sterile neutrinos, likelihood ratio tests are applied to short-baseline neutrino oscillation experiments to construct confidence intervals for the parameters of interest. The test statistics of the form Δχ2 is often used to form the confidence intervals, however, this approach can lead to statistical inaccuracies due to the small signal rate in the region-of-interest. In this paper, we present a computational model for the computationally expensive Feldman-Cousins corrections to construct a statistically accurate confidence interval for neutrino oscillation analysis. The program performs a grid-based minimization over oscillation parameters and is written in C++. Our algorithms make use of vectorization through Eigen3, yielding a single-core speed-up of 350 compared to the original implementation, and achieve MPI data parallelism by employing DIY. We demonstrate the strong scaling of the application at High-Performance Computing (HPC) sites. We utilize HDF5 along with HighFive to write the results of the calculation to file.