1: \documentclass{icrc}
2:
3: \usepackage{times}
4: \usepackage{graphicx} % when using Latex and dvips
5: % % (the latter best with option -Pcmz, if available,
6: % % to invoke Type 1 cm fonts)
7: %\usepackage[pdftex]{graphicx} % when using pdfLatex (preferred)
8:
9: \begin{document}
10: \title{Anisotropy analysis of EAS data in the knee region}
11: \author{S.V. Ter-Antonyan}
12: \affil{Yerevan Physics Institute, 2 Alikhanian
13: Br. Str., 375036 Yerevan, Armenia}
14:
15: \correspondence{samvel@jerewan1.yerphi.am}
16:
17: \firstpage{1}
18: \pubyear{2001}
19:
20: \maketitle
21: \begin{abstract}
22: Based on MAKET-ANI EAS data the distributions of
23: equatorial coordinates of EAS core directions are obtained in the knee
24: region. Anisotropy of primary cosmic rays is displayed
25: only by declination equatorial coordinates ($\delta_a\simeq20^0\pm3^0$) at
26: primary energies more than 5-10 PeV. The fraction of anisotropic component
27: turns out $\sim10\%$ in the knee region.
28: \end{abstract}
29:
30: \section{Introduction}
31: The behavior of EAS size spectra in the knee region points out
32: a possibility of a multi-component nature of the primary
33: nuclear flux \citep{JL, TG, PB, SBG, EW, TH}.
34: As follows from \citep{SBG,TH,BONNY}, the best description of EAS size
35: spectra in the knee region may be achieved providing at least two
36: components in the primary cosmic ray flux. Confirmation of
37: the multi-component origin of cosmic rays can be obtain by
38: investigations of distribution of EAS arrival directions.
39: Especially, it is interesting to measure of the anisotropy
40: in the vicinity of the knee region ($10^{15}-10^{16}$ eV) where
41: the
42: accuracies of EAS experiments in last years have significantly
43: increased. \\
44: In this paper, based on MAKET-ANI \citep{ANI} EAS data
45: ($\sim10^6$ events with $N_e>10^5$) the equatorial coordinate
46: distribution of EAS arrival directions are investigated.
47: Violations of isotropy ($\sim5-10\%$) are obtained
48: at energies $E>10$ PeV.
49:
50: \section{EAS anisotropy}
51: Principal complications of the
52: investigation of the EAS ani- sotropy are:
53: \begin{enumerate}
54: \item EAS attenuation in the atmosphere (the intensity of shower
55: size $F(N_e,\theta)$ strongly depends on a zenith angle of incidence
56: ($\theta$) at the given observation level);
57: \item dependence of the EAS size on the energy ($E$) and primary nucleus
58: ($A$);
59: \item interruptions of exposition local time, which are inevitable in EAS
60: experiments.
61: \end{enumerate}
62: Factors 1-3 do not allow in principle to apply the traditional (on-off)
63: methods of measuring the cosmic-ray anisotropy and here,
64: a simulation method is applied
65: taking into account above factors in the frameworks
66: of QGSJET interaction model \\
67: \citep{QGS} and predictions of the 2-component
68: origin of cosmic rays \citep{PB}. The test of this models by modern
69: EAS size spectra at different zenith angles and different
70: observation
71: levels one can find in papers \citep{TH,BONNY}.\\
72: Assume that the primary nuclear flux $I(E)$ consists of isotropic
73: $I_0(E)$
74: and anisotropic $\Im(E)$ components \citep{G}
75: \begin{equation}
76: I(E)=I_0(E)+\Im(E,\alpha_a,\delta_a)\;\;,
77: \end{equation}
78: where $\alpha_a, \delta_a$ are equatorial coordinates of anisotropic
79: direction.
80: \begin{figure*}[t]
81: \includegraphics[width=17.0cm,height=12cm]{icrcc1.eps} %
82: \caption{Equatorial coordinate distributions of all showers
83: at different primary energies ($E$).
84: Symbols are the Maket-ANI data. Solid lines are the
85: expected distributions obtained by simulations of the isotropic primary
86: component
87: (off-component) taking into account the local time of each detected
88: EAS event. Dashed lines are the isotropic components
89: at the uniform local time distribution of each EAS detected event.}
90: \end{figure*}
91: Let also the factor of anisotropy $\varepsilon=\Im(E)/I(E)$ is small
92: ($\varepsilon\ll 1$)
93: and does not distort the EAS zenith angular distribution due to
94: the Earth rotation and independence of EAS attenuation on the azimuthal
95: angle of incidence ($\varphi$).\\
96:
97: Then, providing each detected EAS event from $I(E)$ flux with
98: horizontal coordinates $\theta,\varphi$ and local time $\vec{T}$
99: by adequate simulated event but from isotropic
100: $I_0(E)$ flux with corresponding coordinates $\theta^*,\varphi^*$ and
101: the same $\vec{T}$ one can compare (after
102: transformation $\theta,\varphi,\vec{T}$ and $\theta^*,\varphi^*,\vec{T}$
103: $\Rightarrow$
104: $\alpha,\delta$ and $\alpha^*,\delta^*$) the obtained equatorial
105: coordinate
106: distributions of real EAS data and isotropic simulated data.
107: Evidently, if the anisotropic
108: part of intensity $\Im(E,\alpha_a,\delta_a)=0$ then the distribution
109: obtained from the experiment must overlap with the simulated one. Any
110: considerable
111: discrepancies (out of statistical errors) will point out to the
112: existence of anisotropy.\\
113: The presented method completely solves the problems 1 and 3, however, the
114: same normalization of real and simulated distributions
115: (due to a flux $\Im(E,\alpha_a,\delta_a$) is unknown) does not give a
116: direct possibility to determine the absolute intensity of
117: the anisotropic component.\\
118:
119: Taking the above into account,
120: each detected EAS event is characterized by a 9-dimensional vector
121: $\vec{D}(E,\theta,\varphi,\vec{T}(y,m,$ $d,h,u,s))$,
122: where $E\equiv E(N_e,\theta)$ is the evaluation of a primary energy
123: on the basis of the measured EAS size $N_e$ and zenith angle $\theta$,
124: $\vec{T}(y,m,d,h,u,e)$
125: is a local time vector with components $y$~-year, $m$~-month,
126: $d$~-day,
127: $h$~-hour, $u$~-minute, $e$~-second of a detected event. Here we used
128: the vector definition only for briefness and convenience. \\
129: \begin{figure*}[t]
130: \includegraphics[width=17.0cm,height=12cm]{icrcc2.eps} %
131: \caption{Detected normalized all-particle primary energy spectra at
132: different equatorial
133: declination $\delta$-coordinates. Black symbols
134: correspond to ANI EAS data. Open
135: circles are the expected energy spectra of an isotropic component
136: (off-component) according to \citep{PB} model.}
137: \end{figure*}
138:
139: The estimation of the primary energy $E$ by the measured EAS size $N_e$
140: is performed by inverse interpolation of function
141: $\overline{N_{e}}\equiv \overline{N_{e}}(E,\overline{A},\theta)$
142: at average primary nucleus $\overline{A}(E)=$ $\exp(\overline{\ln
143: A})$. $\overline{A}$ values are determined by
144: 2-component primary spectra \citep{PB}. A tabulated function
145: $\overline{N_{e}}(E,A,\theta)$ at given $E$, $A$, $\theta$
146: parameters
147: of a primary nucleus was preliminary calculated by means of
148: EAS simulation using the
149: CORSIKA code \citep{COR} at QGSJET interaction model
150: \citep{QGS}. Some details of this simulation one can find in the
151: paper \citep{TH}. Statistical errors did not exceed $3-5\%$
152: and the high altitude location ($700$g/cm$^2$) of the ANI experiment
153: \citep{ANI}
154: allowed to obtain $\Delta E<20\%$.
155:
156: Because the exposition time of the MAKET-ANI array is not
157: continuous
158: and the EAS zenith angular distribution does not adequate to the angular
159: distribution of primary nuclei
160: (factors 1,3 above), we created
161: $i=1,\dots,m$ simulated 9-dimensional vectors
162: $\vec{R}_i(E^*_i,\theta^*_i,\varphi^*_i,\vec{T})$
163: for each detected vector $\vec{D}$ which differ from
164: $\vec{D}$ only by simulated primary energy ($E^*$), zenith
165: ($\theta^*$) and azimuth ($\varphi^*$) angles.\\
166: The values of angles $\theta^*$ and $\varphi^*$ are simulated according to
167: distributions: $dF/d\cos\theta=\cos\theta$ and $d\Phi/d\varphi=1/2\pi$
168: respectively. The energy ($E^*$) of simulated events is obtained from 2-component
169: primary energy spectra \citep{PB,BONNY} at additional condition
170: of detected events $N_{e}^*(E^*,A^*,\theta^*)>10^5$.
171:
172: Transformation of $\vec{D}$ and $\vec{R}_i$ vectors
173: to simple 3-dimensional vectors $\vec{d}(E,\alpha,\delta)$ and
174: $\vec{r}_i(E^*,\alpha^*,\delta^*)$ with
175: $\alpha$ -right ascension and $\delta$ -declination
176: equatorial coordinates are performed by astronomical soft
177: \citep{SK}.
178:
179: \section{Results}
180: The results of investigation of the EAS anisotropy in the knee region
181: by ANI EAS data are shown in Fig.~1,2. Equatorial coordinate distributions
182: $dN/d\alpha$ and $dN/d\delta$ of all showers
183: ($N_{tot}=10^6$ events at $N_e>10^5$) at six primary energy
184: ($E$) intervals
185: are given in Fig.~1. Symbols are the Maket-ANI data
186: $dN/d\alpha$ and $dN/d\delta$ in terms of ($100/N_{tot}$). Solid
187: lines are the expected adequate distributions
188: $dN/d\alpha^*$ and $dN/d\delta^*$ obtained by simulations of
189: isotropic primary component (off-component). Local times
190: $\vec{T}$ of simulated events
191: and detected EAS events are equal. The total number of events
192: is the same ($N_{tot}=N^{*}_{tot}$).
193: Dashed lines reflect the expected behavior of the isotropic
194: component at an uniform local time distribution of each EAS detected
195: event.\\
196: It is seen, that the distribution of a $\delta$-coordinate at all
197: primary energies is practically
198: independent on a local time distribution. Moreover, a good
199: agreement of expected and detected distributions for equatorial
200: $\alpha$-coordinate is observed in the energy range $E>0.7$ PeV.\\
201: However, the discrepancies of detected and expected distributions
202: of the equatorial $\delta$-coordinate increase with the energy.
203: Strong disagreement is observed at $E>10$ PeV and
204: $\delta\simeq20^0$. It is necessary to note, that the real disagreement
205: is a little larger than it is seen in Fig.~1 because EAS and
206: simulated data have the same normalizations.
207: Further investigations of the $\delta$-coordinate distribution
208: are given in Fig.~2 where detected normalized all-particle
209: primary energy spectra at different declinations ($\delta$) are presented.
210: Black symbols are ANI data. The open circles are expected
211: energy spectra of isotropical simulated components (off -
212: component) with 50 times higher statistics
213: ($N^{*}_{tot}=5\cdot10^7$).\\
214: Spectral shape in $E<1-2$ PeV range for all declination
215: intervals is determined by the lower limit of the detected EAS size
216: ($N_e>10^5$). However, the observed fine structure of
217: all-particle spectra at $E>2$ PeV and declinations $\delta<30^0$ and
218: $50<\delta<60$ can hardly be explained by trivial statistical
219: fluctuations.
220: All data in Fig.~1,2 are obtained in the framework of QGSJET
221: interaction model \citep{QGS} and the 2-component origin of cosmic rays
222: \citep{PB,BONNY}.
223: \balance
224:
225: \section{Conclusion}
226: From above analysis it follows that there is a partial anisotropic flux
227: with a
228: declination coordinate equal to $\delta_a=20^0\pm3^0$ at primary energies
229: $E>10$ PeV.
230: The distribution of equatorial $\alpha$-coordinates agrees with the
231: hypothesis of the isotropic distributions of cosmic ray.\\
232: It is interesting to note, that the nearest neutron stars PSR0950 and
233: PSR1133 disposed at $l\simeq30-100\;ps$ distance from the Earth have
234: declination coordinates $\delta=30.6^0$ and $\delta=27.45^0$ respectively.
235: These values are close to the results obtained here ($\delta_a\sim20^0$).
236: Are these neutron stars a reason of anisotropic flux or not
237: can be elucidated only with significantly large EAS statistics.
238:
239: \begin{acknowledgements}
240: It is a pleasure to thank Ashot Chilingaryan for many
241: helpful discussions about these matters. I thank
242: Lilith Haroyan and Hamlet Martirosyan for their assistance during the
243: work. I thank the members of the MAKET-ANI group for providing EAS
244: data.\\
245: The work has been partially supported by the research grant N 00-784
246: of the Armenian government,
247: NATO NIG-975436 and CLG-975959 grants and ISTC A216 grant.\\
248: This publication is based on experimental results of the ANI
249: collaboration.
250: MAKET ANI installation has been set up as a collaborative project of
251: the Yerevan Physics Institute, Armenia and the Lebedev Institute,
252: Russia. Continuous contribution of Russian colleagues
253: is thankfully acknowledged.
254: \end{acknowledgements}
255:
256: \begin{thebibliography}{99}
257: \bibitem[Berezinsky et al.(1984)]{G}Berezinsky~V.S., Bulanov~S.V.,
258: Ginzburg~V.L.,
259: "Cosmic ray astrophysics", Izd. Nauka, Moscow, (1984) (in Russion).
260: \bibitem[Biermann(1993)]{PB}
261: Biermann~P.L.: Astron. Astrophys. {\bf{271}} (1993) 649./\\
262: Preprint MPIfR, Bonn, No 700 (1996) 6.
263: \bibitem[Chilingaryan et al.(1999)]{ANI}Chilingaryan~A. et al.,
264: Proc. 26th ICRC, Salt Lake City, {\bf{1}} (1999) 240.
265: \bibitem[Chilingaryan et al.(1997)]{Chili}
266: Chilingaryan~A.A. et al., Nucl.Phys.~B
267: (Proc.Suppl), 52B (1997) 237.
268: \bibitem[Erlykin and Wolfendale(1997)]{EW}
269: Erlykin~A.D., Wolfendale~A.W., Astropart. Phys. 7 (1997) 1./\\
270: Astropart. Phys. 7 (1997) 203.\\Astropart. Phys. 8 (1998) 265.
271: \bibitem[Gaisser and Stanev(1993)]{TG}
272: Gaisser~T.K. and Stanev~T., Proc. 22nd ICRC, Dublin (1991).
273: \bibitem[Heck et al.(1998)]{COR}Heck~D., Knapp~J.,
274: Capdevielle~J.N., Schatz~G.,
275: Thouw~T., Forschungszentrum Karlsruhe Report, FZKA 6019 (1998) 90p.
276: \bibitem[Kalmykov and Ostapchenko(1993)]{QGS}Kalmykov~N.N.,
277: Ostapchenko~S.S., Yad. Fiz. {\bf{56}} (1993) 105.
278: \bibitem[Linsley(1983)]{JL}
279: Linsley~J., Proc. 18th ICRC, Bangalore (1983) {\bf{12}}, 135.
280: \bibitem[Stanev et al.(1993)]{SBG}
281: Stanev~T., Biermann~P.L., Gaisser~T.K., Astrophys. {\bf{274}}
282: (1993) 902.
283: \bibitem[Super-Kamiokande(1999)]{SK}
284: Super-Kamiokande collaboration, 1999,
285: http:// www.phys. washington. edu /~young/superk/coordinates/
286: \bibitem[Ter-Antonyan and Biermann(2001)]{BONNY}Ter-Antonyan~S.V.
287: and Biermann~P.L., 27th ICRC, Hamburg (2001).
288: \bibitem[Ter-Antonyan and Haroyan(2000)]{TH}
289: Ter-Antonyan~S.V, and Haroyan~L.S., Preprint, arXiv:hep-ex/0003006.
290: \end{thebibliography}
291:
292:
293: \end{document}
294:
295: