Evidence for a high-energy cosmic-ray spectrum cutoff.

© 1985 The American Physical Society 1875 In 1966, Roll and Wilkinson reported the first confirmation that the intense isotropic radiation observed earlier by Penzias and Wilson 2 was thermal in nature and all pervasive, characteristic of a universe cooled to a temperature of 2.7 K. Given the previous report by Linsley of the detection of a primary cosmic ray with an energy estimated to be in excess of 100 EeV, Greisen astutely realized that the spectrum of such energetic cosmic-ray primaries, if their sources were predominantly of extragalactic origin, should steepen abruptly. Such a steepening results naturally from the rapid onset of opacity to passage through the universe of ultrahigh-energy cosmic rays due to energy degradation from photo-pion production off the 2.7-K blackbody radiation. Thus, the cosmic-ray spectrum would appear to terminate at energies near production threshold at 60 EeV. Here we present evidence confirming the early prediction of Greisen and the more recent calculations of Hill and Schramm which indicate the development of a spectral "bump," due to pileup of the recoil protons, prior to the onset of the "Greisen cutoff." We report here the energy spectrum obtained using the Fly's Eye detector situated at 41°N latitude in the western Utah desert at an atmospheric depth of 852 g cm. The detector differs from conventional extensive-air-shower (EAS) detectors in that it is essentially a calorimetric device which observes the passage of EAS through the atmosphere by means of the nitrogen fluorescence light given off upon ionization and excitation by the relativistic charged particles in the shower. Events have been detected with impact parameters as far away as 22 km. The effective collecting aperture is thus very large. However, since the apparatus must detect a very low light flux, observing periods are restricted to clear, moonless nights which greatly negates the benefits of large aperture. The data collected for this analysis were obtained during a total elapsed time of 33 months and an actual live time of 0.145 yr. The duty cycle is 6.3%. The techniques of track reconstruction and conversion of measured photoelectron yields into shower size versus atmospheric slant depth have been described elsewhere. Shower energies are assigned by integration of the resultant longitudinal shower development curve to obtain the total track length. Track lengths are then converted into total "electromagnetic" energy via the relation