Search

T2K is an experiment designed to study the ways in which neutrinos evolve as they propagate through space. The experiment uses an artificial beam of muon neutrinos, produced at the J-PARC facility in Japan. The spontaneous mixing of these beam neutrinos with other flavours is measured after 295 kilometres, at the location of the Super-Kamiokande detector. The initial beam composition is measured after 280 metres, using a combination of two near detectors. T2K is optimised for measurements of electron neutrino appearance, and muon neutrino survival. Essential for these measurements is a detailed knowledge of the initial beam composition, before neutrino oscillations had occurred. The production of intense neutrino beams at accelerator facilities is challenging, and requires exceptional understanding of chains of hadronic interactions initiated within thick targets. Most of T2K neutrinos are produced from in-flight decays of focused pions and kaons, emitted from an extended graphite target (90 cm) bombarded with a 30 GeV proton beam. Besides internal constraints from near detector data, T2K relies on hadron production measurements collected by other experiments, most importantly NA61, for accurate modelling of the initial neutrino flux. This thesis presents a new T2K neutrino flux prediction, using the NA61 replica-target dataset from 2009. The earlier flux prediction relied on NA61 measurements collected with protons incident on a thin graphite target (2 cm). The following analysis incorporates NA61 data collected on the full length replica of the T2K target in 2009. In light of this new data, the flux calculation technique has been modified. A dramatic reduction in the unoscillated muon neutrino fractional flux uncertainty has been achieved, from ~10% to ~5% around the T2K flux peak. The impact of the improved flux prediction on oscillation measurements has been examined, by performing separate fits to ND280 data, and joint ND280 and Super-K data, in the well-established Bayesian Markov Chain Monte Carlo analysis framework.