Og 6 . 3 - 2 Fly ' S Eye Measurement of the Cosmic Ray Composition Above

We present measurements of the depth of maximum for extensive air showers (EAS) produced by cosmic ray nuclei with energies above 10eV. The air showers were observed using the University of Utah's Fly's Eye detectors operating in stereo mode. The data show an elongation rate of 69 ± 5 g/ cm per decade above 10eV before correction for triggering and resolution effects. These effects contribute approximately +5 g/cm per decade to the apparent elongation rate. The depth of maximum distribution is not consistent with the expectation of a composition dominated by protons. Introduction The Fly's Eye detectors observe the nitrogen fluorescence light produced by large air showers in the atmosphere and are located at an atmospheric depth of 860 g/cm in the western desert of Utah. Details of the experiment are given elsewhere (Baltrusaitis et al 1985). The two Fly's Eye detectors, FE1 and FE2 are separated by a distance of 3.4 km. The data discussed in this report represent events detected simultaneously by both instruments. This 'stereoscopic' view of air showers significantly improves the geometrical reconstruction of the data, important in studies sensitive to position dependencies such as composition. Depth of Maximum Measurements The Fly's Eye detectors directly measure the longitudinal development profiles of large EAS and allow a straightforward extraction of the depth of shower maximum (Xmax), an indicator of primary composition. We have made a thorough investigation of possible systematics involved with this measurement (Cassiday et al 1989), including those associated with geometric reconstruction, atmospheric scattering of light, assumptions about the atmospheric density profile, Cerenkov light contamination, and the form of the assumed development profile. We find that these sources of systematic error amount to shifts of less than ± 20 g/cm in the average value of the depth of maximum, and of less than ± 5 g/cm in the value of the width (ó) of the depth of maximum distribution. Interpretation of the depth of maximum data involves the use of a detector Monte Carlo program which takes into account the triggering efficiencies and reconstruction resolution of the detectors. One input for the detector simulation are files of one dimensional shower development profiles which have been generated using a variety of interaction models and primary masses. Each of these profiles, along with random assignments of geometrical parameters, is passed through the detector simulation program which uses the known efficiencies for nitrogen fluorescence and Cerenkov light production to calculate the light produced by each shower.