We have established the correlation between the atmospheric temperature contribution to the diurnal variation observed by a meson monitor at Deep River and the diurnal variation of two easily and continuously observable atmospheric variables, the ground-level air temperature and the barometric pressure. The atmospheric meson diurnal variation vector is taken to be, on a statistical basis, A = M−RN, where M and N represent the observed meson-monitor and neutron-monitor diurnal variations and R is the factor of proportionality between the meson and neutron monitor responses to the primary anisotropy. It is found that A is proportional in amplitude to T, the ground-level temperature diurnal variation, and, further, that T and the barometric-pressure diurnal variation P are proportional in amplitude. The "best-fit" representation of A in terms of T and P is determined by minimizing the mean-square deviation between the daily vectors RN and (M−A). Where A = CtT + CpP, the best fit occurs when Ct = −0.0052%/ °C, Cp = 0.038%/mb, R = 0.47, and the phase of T is shifted by + 1.0 hour. These values apply to Deep River, where the original hourly meson data have been barometer-corrected using a coefficient of 0.16%/mb.
Coincidence counter measurements of cosmic rays in an aeroplane