The Electron / Muon Ratio in Eas at Andabove the " Knee

New data on the electron/muon ratio measured by the KASCADE extensive air shower experiment are presented. This ratio has been proven to be an important parameter for the determination of the mass of the primary particle. The electron and muon detector responses and reconstruction accuracies of the shower parameters have been studied in detailed simulations. Due to the extended lateral distribution of the muons at ground level, reconstruction of the total muon number is aaected by the limited area of the experiment (200200 m 2). Reconstructed and simulated muon density distributions are, therefore, restricted to the geometrical acceptance range of the experiment and are compared only within this region. Measuring the ratio as a function of the muon number, we observe a trend towards a heavier composition with increasing energy. EXPERIMENTAL SETUP AND OBSERVABLES The KASCADE detector array measures the lateral electron and muon distributions of extensive air showers. One aim is to obtain the electron and muon numbers that give important hints to the mass and energy of the primary particle. The array comprises about 500 m 2 of e==-and 620 m 2 of muon detectors which are distributed over 252 detector stations and are arranged on a grid of 200 200 m 2 size. They are described in more detail elsewhere (Klages et al.,1997a, 1997b). The muon detectors are positioned directly below the e== detectors and are shielded by slabs of lead and iron corresponding to 20 radiation length in total. This imposes an energy threshold of 300 MeV to muons. Because of the geometrical arrangement many muons will deposit energy in the e== detector before entering the muon detector. On the other hand, energetic photons or electrons may penetrate the shielding, thereby faking a muon signal. To correct for these eeects, the observed energy depositions produced by the diierent particle species were simulated for both types of detectors. The simulated air shower events were produced with the CORSIKA event generator (V. 4.6) (Capdevielle, 1992) and all particles at observation level were fully tracked through the KASCADE experiment. Since the particles and their energies on top of a detector station are known in the simulations, comparison to their average simulated energy deposition in both detectors can be used to correct the experimental data. Such a simulation has to take into account possible diierences originating from diierent masses of the primary particle, diierent zenith angles etc.. Figure 1 shows …