Researchers have been worried about the discrepancy between the experimental data and simulations at Super-Kamiokande for some time. According to the simulations, the experiment was detecting too few muon neutrino events or too many electron neutrinos. Boezio and others decided to tackle the discrepancy by making measurements in the atmosphere.

Atmospheric neutrinos are created by so-called secondary cosmic rays in the Earth's atmosphere. These are created when galactic cosmic rays - called primaries - collide with nuclei in the atmosphere. Approximately 85% of the primary cosmic rays are protons and 12% are alpha particles. The rest are electrons or the nuclei of heavier atoms. As they enter the Earth's magnetic field, the primary cosmic rays are accelerated to kinetic energies over 1 GeV. Both the primary and secondary cosmic rays create muons, which decay into neutrinos. Between a quarter and a third of the muon's momentum is transferred to the neutrino. Boezio and his colleagues therefore measured the number of muons in the Earth's atmosphere and compared their results with the number and energy of the neutrinos detected by Super-Kamiokande. Their results were closer to the experimental number of neutrinos measured by Super-Kamiokande than the result of the simulations. It seems as if the simulations have been overestimating the number of muon neutrinos that should be detected by Super-Kamiokande by 36%.