Lodz University Conference Proceedings. [PS] is available.
The brief version of the paper was published in Nuclear Physics B PS 52B(1997) p.182-184.

Primary Cosmic Ray Spectrum Measured Using Cherenkov Light Reflected from the Snow Surface

Antonov R.A., Chernov D.V., Fedorov A.N., Petrova E.A.

Institute of Nuclear Physics, Moscow State University, Moscow, 119899, Russia

The experimental data obtained using Cherenkov light of EAS reflected from the snow surface of the Big Alma-Ata Lake (Kazakhstan) are presented. This method makes it possible to have a large area using a simple compact detector. About 9000 events were detected within 55 hours. The balloon-borne measurements in the energy range 10¹⁵-10²⁰ eV are planned.

INTRODUCTION

The results of recent cosmic ray energy spectrum measurements in the energy range 10¹⁵-10¹⁹ eV do not agree with each other sufficiently well. The actual problem, for this reason, is to carry out further experiments using various methods.

This work is based on the Prof. A.E.Chudakov's suggestion to detect the Cherenkov light reflected from the snow surface [1]. The intensity of Cherenkov light is proportional to the energy of the primary cosmic ray particle. The wide-angle balloon-borne or airborne small detector makes it possible to have a sensitive area up to some hundred km². The first unaccomplished attempt of such an experiment was undertaken by G.Navarra et al. [2].

DETECTOR ARRAY

SPHERE detector array was elaborated for balloon-borne experiment [3-5]. Fig. 1 shows the scheme of this array. The light spots are detected by 19 photomultipliers (FEU-110) situated on the focal surface of the spherical mirror. Dark violet filters and shifters are used with photomultipliers to decrease the influence of the starlight background. The angular aperture of detector is about 50^o x 50^o. The detector may be covered with a lid.

The first measurements were carried out in the Thien-Shan mountains in winter 1993 (fig. 2). SPHERE detector was situated on the 160 m high mountain ledge nearby the B.Alma-Ata lake (2500 m above sea level) to detect Cherenkov light reflected from the snow surface of the lake. The area of the lake is about 0.7 km². The average inclination angle of the detector optical axis to the horizon was 10^o. Cherenkov light is reflected from the snow surface according to the Lambert reflection law:

where is flux reflected at angle , - normally reflected flux. So only about Cos 80^o = 0.17 of normally reflected light was detected in this experiment.

13 PMTs, enumerated in fig. 1b, were used in the experiment. Fig. 3 shows the photocathode surfaces projections to the lake surface. The numbers indicate areas observed by the corresponding PMTs. It's clear from this figure that the most of the lake surface is observed by central photomultipliers. Therefore the trigger condition was - pulse amplitude in one of central PMTs (condition M1) or in one of central PMTs and one of neighbour PMTs (condition M2) must exceed the threshold. Time stability of the detector and transparency of the atmosphere were monitored by the periodical PMTs current measurements and by the trigger rate of detector. Average registration rate was 2.6 min^-1.

DATA ANALYSIS AND RESULTS

Two series of measurements were carried out under trigger condition M2: with covered and with open detector. In measurements with covered detector pulses were caused by Cherenkov light emission of charged particles in glass of PMT tube and light filter.

The experimental differential spectra of pulse amplitude total over the four central PMTs (5, 6, 7, 8) are shown in fig. 4. In following analysis spectrum of covered data was subtracted from that of open one.

To obtain energy spectrum it was necessary to take into account that some part of light spot falls outside the lake area. It was necessary to determine effective registration area also. For this two purposes Monte-Carlo simulation was done. To derive the absolute primary particle fluxes it was necessary to know the effective space angle of the showers detection. The shower angular distribution was not measured in this experiment but was estimated. It was assumed that the attenuation of the inclined showers was caused only by Rayleigh scattering. The effective space angle was about 3.3 ster (on the base of this estimation).

Primary cosmic ray flux at the energy 10¹⁷ eV obtained in this experiment is in agreement with other experimental data (fig. 5).

Energy threshold in this experiment was due to the large dead time of the device, not by starlight background.

FURTHER MEASUREMENTS

Further measurements will be carried out with fastened balloon. During 1994-95 the detector SPHERE was improved significantly.

The amplitude measurements are complimented by the time analysis of PMT pulses. It will allow us to analyse the detected events more completely.

The trigger rate ability is increased up to 50 Hz by using fast electronics and microcomputer. In balloon-borne experiment the reflected light intensity increases by 6 times according to the Lambert reflection law. This two reasons allow us to decrease energy threshold to 10¹⁵ eV. The size of detector storage is sufficient to store about 3.6 10⁶ events.

The detector electronics measures the integral of light pulse in PMT, pulse duration and intervals between pulses. The 25-ns discreteness allow to reliably reject events simulated by charged particles in the PM tubes and filter glass. It makes possible to determine the arrival direction of EAS too.

The ground-based test runs of apparatus were carried out in winter 1995-96 under conditions similar to those in flight (low temperatures etc.)

We plan to carry out measurements at altitudes 1-3 km above snow surface using the fastened balloon in winter 1996-97. In the future it is desirable to perform the large-scale measurements in the Arctic or Antarctic to detect EAS with energy up to 10²⁰ eV. One such session will be enough to get the amount of data on EAS with E > 10¹⁹ eV comparable with that of Yakutsk array. Table 1 shows the estimated event number to be detected by SPHERE detector for given flight height H and detection time t.

CONCLUSIONS

The experiment showed that the method of detection of Cherenkov light reflected from the snow surface is appropriate to the cosmic ray energy spectrum measurement. It is possible to ensure large sensitive area and to measure cosmic ray energy spectrum in wide range of energies (10¹⁵-10²⁰ eV) with small wide-angle balloon-borne detector.

ACKNOWLEDGEMENTS

The authors are grateful to A.E.Chudakov, G.B.Khristiansen and L.A.Kuzmitchev for valuable discussions and help. This work is supported by Russian Foundation for Basic Research (grant 95-02-04325-a) and ISSEP (Petrova E.A.)

REFERENCES

[1] Chudakov A.E. Trudy conf. po cosm. lutcham, Yakutsk, 69, 1972. (in Russian)
[2] Castagnoli C., et al. Proc. 17 ICRC, Paris, 6, 103, 1981.
[3] Antonov R.A., et al. Proc. 14 ICRC, Munich, 9, 3360, 1975.
[4] Antonov R.A., et al. Izvestiya Academii Nauk, 50, 2217, 1986. (in Russian)
[5] Antonov R.A., et al. Vestnik MGU, ser.3, 36, 4, 102, 1995. (in Russian)
[6] Fomin Yu.A., et al. Proc. 22 ICRC, Dublin, 2, 87, 1991.
[7] Nagano M., et al. J.Phys. G: Nucl.Phys, 18, 423, 1992.
[8] Vildanova L.I., et al. Izvestiya Academii Nauk, 58, 79, 1994. (in Russian)

SPHERE Home page

March 11, 1997.
Elena Petrova