Speaker
Description
Using its 10-year dataset, NOvA makes competitive measurements of neutrino oscillation parameters $\Delta m^2_{32}$, $\theta_{23}$, $\theta_{13}$, and $\delta_{CP}$ using an 810 km baseline ranging from the NuMI beam at Fermilab to the far detector in Minnesota. NOvA uses a method called "extrapolation" to control systematic uncertainties. First, simulation-data discrepancies are measured using the near detector at the neutrino source. Then, the oscillation measurement is performed using corrected far detector simulation.
NOvA is also exploring alternative methods for managing systematic uncertainties by using both near and far detector data to simultaneously fit systematic and oscillation models. One method, the Bayesian two-detector sampling technique, uses Hamiltonian Markov Chain Monte Carlo to efficiently compute joint oscillation and systematic posteriors. Compared to extrapolation, the technique is better-informed of parameter correlations and more transparently reports how well the systematic model describes the near detector data.
Using the Bayesian two-detector technique, I will estimate sensitivity to oscillation parameters, and I will test the technique's robustness to out-of-model systematic variations.