Underground muon counters as a tool for composition analyses

Abstract

The transition energy from galactic to extragalactic cosmic ray sources is still uncertain, but it should be associated either with the region of the spectrum known as the second knee or with the ankle. The baseline design of the Pierre Auger Observatory was optimized for the highest energies. The surface array is fully efficient above 3 × 10¹⁸ eV and, even if the hybrid mode can extend this range below 10¹⁸ eV, the second knee and a considerable portion of the wide ankle structure are left outside its operating range. Therefore, in order to encompass these spectral features and gain further insight into the cosmic ray composition variation along the transition region, enhancements to the surface and fluorescence components of the baseline design are being implemented that will lower the full efficiency regime of the Observatory down to ∼10¹⁷ eV. The surface enhancements consist of a graded infilled area of standard Auger water Cherenkov detectors deployed in two triangular grids of 433 m and 750 m of spacing. Each surface station inside this area will have an associated muon counter detector. The fluorescence enhancement, on the other hand, consists of three additional fluorescence telescopes with higher elevation angle (30°–58°) than the ones in operation at present. The aim of this paper is threefold. We study the effect of the segmentation of the muon counters and find an analytical expression to correct for the under counting due to muon pile-up. We also present a detailed method to reconstruct the muon lateral distribution function for the 750 m spacing array. Finally, we study the mass discrimination potential of a new parameter, the number of muons at 600 m from the shower axis, obtained by fitting the muon data with the above mentioned reconstruction method.

Introduction

The cosmic rays energy spectrum extends for about eleven orders of magnitude, starting at energies below 1 GeV up to energies of more than 10²⁰ eV. It presents three main features: the knee, the second knee and the ankle. There is evidence of a fourth feature, the so-called GZK suppression [1], [2], which is originated by the interaction of high energy protons with the photons of the cosmic microwave background. In the case of heavier nuclei, a similar effect is expected due to the fragmentation of the nuclei in their interaction with the photons of the microwave and infrared backgrounds [3].

The knee has been observed by several experiments [4], [5], [6] at around 3–5 × 10¹⁵ eV. At this energy the spectral index changes from −2.7 to −3.1. The KASCADE data shows that the composition at the knee presents a transition from light to heavy primaries in such a way that, at energies above 10¹⁶ eV, the composition is dominated by heavy nuclei. These particles are originated in our Galaxy and what is being detected is, very likely, the end of the efficiency of supernova remnant shock waves as accelerators.

The second knee has been observed at around 4 × 10¹⁷ eV by Akeno [7], Fly’s Eye stereo [8], Yakutsk [9] and HiRes [10]. The physics of this feature is still unknown, it might be due to the end of the efficiency of supernova remnant shock waves as accelerators or a change in the diffusion regime in our Galaxy [11], [12].

The ankle is a broader feature that has been observed by Fly’s Eye [8] and Haverah Park [13], centered at approximately the same energy, ∼3 × 10¹⁸ eV. These results have been confirmed by Yakutsk [9], HiRes [10] and Auger in Hybrid mode [1]. AGASA also observed the ankle but at higher-energy, around 10¹⁹ eV [14]. As in the case of the second knee, the origin of the ankle is still unknown and its physical interpretation is intimately related to the nature of the former. The ankle could be the transition between the Galactic and extragalactic components [15] or the result of pair creation by extragalactic protons in the interaction with the cosmic microwave background [16].

The precise determination of the mean chemical composition of the cosmic rays in the energy range above ∼10¹⁷ eV will allow us to understand the origin of the second knee and the ankle and to know the energy and the speed at which the transition between the Galactic and extragalactic components is given [17]. In particular, it will permit to decide among the three main models: (i) the mixed composition scenario [15], in which the composition injected by the extragalactic sources is assumed to be similar to the one of the Galactic sources and in which the transition takes place in the ankle region, (ii) the dip model [16], in which the ankle is originated by the interaction of extragalactic protons with the cosmic microwave background and the transition is given at the second knee and (iii) a two-component transition from Galactic iron nuclei to extragalactic protons, around the ankle energy [18].

The Pierre Auger Observatory consists of two Observatories situated one in each hemisphere. The Southern Observatory, located in Pampa Amarilla close to the city of Malargüe, Province of Mendoza, Argentina, currently consists of nearly 1600 Cherenkov detectors placed in a 1500 m triangular grid covering an area of 3000 km² plus four fluorescence telescope buildings, with six telescopes each, situated in the periphery of the surface array and overlooking it. The construction started in 2000 and is going to be completed early in 2008. A complementary Northern Observatory will be sited in Colorado, United States of America.

The Southern Observatory, in its original design, is able to measure cosmic rays of energies above 3 × 10¹⁸ eV for the surface array and ≲10¹⁸ eV in hybrid mode. Two enhancements, AMIGA (Auger Muons and Infill for the Ground Array) [19] and HEAT (High Elevation Auger Telescopes) [20], will extend the energy range down to 10¹⁷ eV, encompassing the second knee and ankle region where the Galactic-extragalactic transition takes place.

AMIGA will consist of 85 pairs of Cherenkov detectors and 30 m² muon counters buried ∼2.5 m underground, placed in a graded infill of 433 m and 750 m triangular grids. The AMIGA infill area is bound by two hexagons covering areas of 5.9 km² and 23.5 km² corresponding to the 433 m and 750 m-arrays, respectively. The energy thresholds of the 433 m and 750 m-arrays are ∼10¹⁷ eV and ∼10^17.6 eV, respectively [21]. On the other hand, HEAT will be formed by three additional telescopes of 30°–58° elevation angle located next to the fluorescence telescopes building at Coihueco. They will be used in combination with the existing 3°–30° elevation angle telescopes at Coihueco as well as in hybrid mode with the AMIGA infills.

These enhancements will also allow detailed composition studies based on the combined measurement of the atmospheric depth of maximum shower development, X_max, and the shower muon content. These two parameters are very sensitive to primary mass composition. Other mass sensitive parameters, like the slope of the lateral distribution function, rise-time of the signals in the surface detectors, curvature radius, etc. strongly depend on them [22].

In this paper we will concentrate on the AMIGA muon detectors [19]. These counters will consist of highly segmented scintillators with optical fibers ending on 64-pixel multi-anode photomultiplier tubes (PMT). The scintillator strips will be equal to those used for the MINOS experiment [23]. The current baseline design calls for 400 cm long × 4.1 cm wide × 1.0 cm high strips of extruded polystyrene doped with fluors, POP (1%) and POPOP (0.03%), and co-extruded with TiO₂ reflecting coating. They are covered with reflective Al foil. To extract the scintillation light, a wavelength shifting fiber is glued into a grove which is machined along one face of the scintillator strip. A 10 m² module will consist of 64 scintillator strips with the fibers ending on an optical connector matched to a 64-pixel multi-anode Hamamatsu H7546B PMT of 2 mm × 2 mm pixel size, protected by a PVC casing. Each muon counter will consist of three of these 64-channel modules, totalling 192 independent channels covering an effective area of 30 m² (actually during the engineering array phase one of these 10 m² modules from each counter will be split into two 5 m² modules for further analyses close to the shower core). These muon counters will be buried alongside a water Cherenkov tank. Each of the 192 channels of the electronics will count pulses above a given threshold, with an overall counter time resolution of 20 ns.

We also present a detailed method for the reconstruction of the Muon Lateral Distribution Function (MLDF) from data obtained by the muon counters of the 750 m-array. An associated problem is the pile-up effect due to the finite segmentation of the muon counters. We analyse this problem and propose a correction that considerably improves the reconstruction of the MLDF. The number of muons at 600 m from the shower axis, $N_{\mu }^{\text{Rec}}(600)$, is extracted from MLDF fits using our reconstruction method. Subsequently, the design parameters of the muon counters (segmentation and area) are validated by studying the impact of these parameters on the total $N_{\mu }^{\text{Rec}}(600)$ uncertainty. Finally, we study the mass discrimination power of $N_{\mu }^{\text{Rec}}(600)$ as compared to other parameters normally used in composition analyses: the maximum development of the longitudinal profile, X_max, the curvature radius of the shower front, R, rise-time of the signal at the water Cherenkov detectors, t_1/2, and slope of the lateral distribution function of the total signal deposited in the water Cherenkov detectors, β. Second order effects, like multiple triggering due to electrons scattered by a single muon, will be dealt with in a subsequent work.