1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: \documentclass[12pt]{article}
3:
4: \usepackage{amssymb}
5: \usepackage{amsmath}
6:
7: \usepackage{epsfig}
8:
9: \begin{document}
10:
11: \title{\bf Interactions of cosmic neutrinos with nucleons in the RS model}
12:
13: \author{A.V. Kisselev\thanks{Email address: alexandre.kisselev@ihep.ru} \\ \\
14: \em Institute for High Energy Physics, 142281 Protvino,
15: Russia}
16:
17: \date{}
18:
19: \maketitle
20:
21: \thispagestyle{empty}
22:
23: \bigskip
24:
25: \begin{abstract}
26: We consider the scattering of the brane fields due to $t$-channel
27: massive graviton exchanges in the Randall-Sundrum model. The
28: eikonal amplitude is analytically calculated and both differential
29: and total neutrino-nucleon cross sections are estimated. The event
30: rate of quasi-horizontal air showers induced by cosmic neutrinos,
31: which can be detected at the Pierre Auger Observatory, is
32: presented for two different fluxes of cosmogenic neutrinos.
33: \end{abstract}
34:
35: %%%%%%%%%%%%%
36: % Main text %
37: %%%%%%%%%%%%%
38:
39: \section{Introduction}
40:
41: The detection of air showers induced by ultra-high energy
42: neutrinos may help to solve many important problems, such as
43: propagation of cosmic neutrinos to the Earth and their
44: interactions with the nucleons at energies around tens (hundreds)
45: of TeV. In this energy region, neutrino-nucleon interactions may
46: be strong due to a new physics. There is a large class of models
47: in a space-time with extra spacial dimensions which result in a
48: new TeV phenomenology. In the present paper we will consider an
49: approach with non-factorizable metric proposed in
50: Refs.~\cite{Randall:99,Randall:99*}, and study a scattering of the
51: SM fields in this scenario.
52:
53: The RS model~\cite{Randall:99,Randall:99*} is a model of gravity
54: in a slice of a 5-dimensional Anti-de-Sitter space (AdS$_5$) with
55: a single extra dimension compactified to the orbifold $S^1/Z_2$.
56: The metric is of the form:
57: \begin{equation}\label{01}
58: ds^2 = e^{-2 \kappa |y|} \, \eta_{\mu \nu} \, dx^{\mu} \, dx^{\nu}
59: + dy^2.
60: \end{equation}
61: Here $y = r_c \theta$ ($0 \leqslant \theta \leqslant \pi$), $r_c$
62: is a "radius" of extra dimension, and parameter $\kappa$ defines
63: the scalar (negative) curvature of the space.
64:
65: From a 4-dimensional action one can derive the relation:
66: \begin{equation}\label{02}
67: \bar{M}_{Pl}^2 = \frac{M^3}{\kappa} \left( 1 - e^{-2 \pi \kappa r}
68: \right) \simeq \frac{M^3}{\kappa},
69: \end{equation}
70: which means that $\kappa \sim \bar{M}_{Pl} \sim M$, with $M$
71: being a 5-dimensional Planck scale.
72:
73: We will consider so-called RS1 model~\cite{Randall:99} which has
74: two 3-dimensional branes with equal but opposite in sign tensions
75: which are located at the point $y = \pi r_c$ (called the TeV
76: brane) and at the point $y = 0$ (referred to as the Planck brane).
77: All SM fields are constrained to the TeV brane, while the gravity
78: propagates in the bulk (all spacial dimensions).
79:
80: From the point of view of an observer located on the TeV brane,
81: there exists an infinite number of graviton Kaluza-Klein (KK)
82: excitations with masses:
83: \begin{equation}\label{04}
84: m_n = x_n \kappa \, e^{-\pi \kappa r_c}, \qquad n=1,2 \ldots,
85: \end{equation}
86: where $x_n$ are zeros of the Bessel function $J_1(x)$:%
87: \footnote{The first four values of $x_n$ are 3.83, 7.02, 10.17,
88: and 13.32.}
89: %
90: \begin{equation}\label{05}
91: J_1(x_n) = 0, \qquad n=1,2 \ldots
92: \end{equation}
93:
94: By using a linear expansion of the metric, one can derive the
95: interaction Lagrangian:
96: \begin{equation}\label{06}
97: \mathcal{L} = - \frac{1}{\bar{M}_{Pl}} \, T_{\mu \nu} \,
98: h^{(0)}_{\mu \nu} - \frac{1}{\Lambda_{\pi}} \, T_{\mu \nu} \,
99: \sum_{n=1}^{\infty} h^{(n)}_{\mu \nu},
100: \end{equation}
101: where
102: \begin{equation}\label{08}
103: \Lambda_{\pi} = \bar{M}_{Pl} \, e^{-\pi \kappa r_c}
104: \end{equation}
105: is a physical scale on the TeV brane. It can be chosen as small as
106: 1 TeV for a thick slice of the AdS$_5$, $r_c \simeq 12/\kappa
107: \simeq 60 \, l_{Pl}$. We see from \eqref{06} that couplings of all
108: massive states are only suppressed by $\Lambda_{\pi}^{-1}$, while
109: the zero mode couples with usual strength defined by the reduced
110: Planck mass $\bar{M}_{Pl} = M_{Pl}/\sqrt{8 \pi}$.
111:
112: The main phenomenological parameters of the model are the scale
113: $\Lambda_{\pi}$ and the ratio
114: \begin{equation}\label{10}
115: \mu = \frac{\kappa}{\bar{M}_{Pl}}.
116: \end{equation}
117: The present experimental data together with theoretical bounds on
118: the curvature of the AdS$_5$ restrict an allowed region for the
119: variable $\mu$ (see, for instance, Fig.~\ref{fig:bounds} taken
120: from Ref.~\cite{Davoudiasl:01}):
121: \begin{equation}\label{11}
122: 0.01 \lesssim \mu \lesssim 0.1.
123: \end{equation}
124: The allowed value of $\mu$ is restricted by so-called naturalness
125: and requiring a 5-dimensional curvature to be small enough to
126: consider a linearized gravity on the brane. Thus, the lightest
127: masses of the KK graviton modes, $m_n = x_n \mu \Lambda_{\pi}$,
128: are of the order of 1 TeV.
129:
130: Our paper has the following structure. In the next Section we
131: consider interactions of the SM fields on the brane in the RS
132: model induced by an exchange of massive gravitons. The eikonal
133: amplitude is calculated and an elastic cross section for different
134: sets of the RS parameters and/or invariant energy is estimated. In
135: Sec.~3 we use these results to study the scattering of ultra-high
136: energy cosmic neutrino off the atmospheric nucleons. The
137: neutrino-nucleon cross section is calculated and an event rate of
138: quasi-horizontal neutrino events expected at the Auger Observatory
139: is presented. Our conclusions and discussions is a topic of
140: Sect.~4.
141:
142:
143: \section{Eikonal amplitude in the RS model}
144:
145: In what follows, we will employ the zero width approximation for
146: the graviton KK resonances. The Born amplitude corresponding to
147: $t$-channel exchange looks like (both the massless mode and KK
148: gravitons contribute):
149: \begin{equation}\label{12}
150: A_B(s,t) = \frac{8 \pi G_N s^2}{-t} + \frac{s^2}{\Lambda_{\pi}^2}
151: \sum_{n=1}^{\infty} \frac{1}{-t + m_n^2}.
152: \end{equation}
153: The sum in \eqref{12} converges very rapidly in $n$, since $x_n =
154: \pi \, (n + 1/4) + \mathrm{O(n^{-1})}$~\cite{Watson}. We consider
155: the scattering of two particles \emph{living on the TeV brane}.
156: Thus, in Eq.~\eqref{12} $t$ means a 4-dimensional momentum
157: transfer which is well-defined and conserved.
158:
159: Let us underline that in \eqref{12} we sum spin-two particles with
160: different KK numbers $n$ (non-reggeized KK gravitons). In more
161: general approach, one should sum Regge trajectories $\alpha_n(t)$
162: which are numerated by $n$ (KK-charged gravireggeons). For the ADD
163: model, it was done in Refs.~\cite{Kisselev:04}. The results of
164: Refs.~\cite{Giudice:02,Kisselev:04} can be reproduced in the limit
165: $\alpha_n(t) \rightarrow 2$. As for the RS model, results on a
166: gravireggeon contribution to the eikonal amplitude will be
167: presented in a forthcoming paper~\cite{Kisselev:05}.
168:
169: Generally, the massive KK states may decay to a pair of SM
170: particles. The partial widths are proportional to
171: $m_n^3/\Lambda_{\pi}^2$, where $m_n$ is the mass of the resonance.
172: In particular, the partial decay widths to massless gauge bosons,
173: fermions, and a pair of Higgs are%
174: \footnote{These expressions can be obtained by a replacement
175: $\bar{M}_{Pl}^{-2} \rightarrow \Lambda_{\pi}^{-2}$ in
176: corresponding formulae derived for large extra dimensions in
177: Ref.~\cite{Han:99}. We have also neglected masses of the SM
178: particles, since $m_{_{SM}} \ll m_n$.}
179: %
180: \begin{eqnarray}\label{13}
181: \Gamma (h^{(n)} \rightarrow VV) &=& N_V \, a \, \frac{m_n^3}{40
182: \pi \Lambda_{\pi}^2},
183: \nonumber \\
184: \Gamma (h^{(n)} \rightarrow f \bar{f}) &=& N_f \, \frac{m_n^3}{160
185: \pi \Lambda_{\pi}^2},
186: \nonumber \\
187: \Gamma (h^{(n)} \rightarrow H \bar{H}) &=& \frac{m_n^3}{480 \pi
188: \Lambda_{\pi}^2}.
189: \end{eqnarray}
190: Here $N_V = 1(8)$ for photons and electroweak bosons (gluons),
191: $N_f = 1(3)$ for a lepton (quark pair) mode, and $a=1/2$ for
192: identical particles. Then for the total width of the massive KK
193: graviton in the RS model, $\Gamma_n$, we get the estimate (see
194: also \cite{Dvergsnes:03}):
195: \begin{equation}\label{14}
196: \Gamma_n \simeq m_n \, (0.31 \, \mu \, x_n)^2.
197: \end{equation}
198: Since the sum which we are interested in converges very rapidly in
199: $n$ (see a comment after Eq.~\eqref{12}), we conclude from
200: \eqref{14} and \eqref{11} that effectively $\Gamma_n/2 \ll m_n$.
201:
202: The sum~\eqref{12} can be calculated analytically by the use of
203: the formula~\cite{Watson}:
204: \begin{equation}\label{15}
205: \sum_{n=1}^{\infty} \frac{1}{ z_{n, \nu}^2 - z^2} =
206: \frac{J_{\nu + 1}(z)}{2 \, z\,J_{\nu}(z)},
207: \end{equation}
208: where $z_{n, \, \nu}$ ($n=1,2 \ldots$) are zeros of the function
209: $z^{-\nu} J_{\nu}(z)$. As a result, we obtain:
210: \begin{equation}\label{16}
211: A_B(s,t) = \frac{8 \pi G_N s^2}{-t} + \frac{s^2}{2 \mu
212: \Lambda_{\pi}^3} \, \frac{1}{\sqrt{-t}} \, \frac{I_2(v)}{I_1(v)}
213: \end{equation}
214: Here $I_i(z)$ ($i=1,2$) are modified Bessel functions and $v =
215: \sqrt{-t}/\mu \Lambda_{\pi}$. Taking into account properties of
216: $I_i(x)$, we conclude from \eqref{16} that a contribution of the
217: massive graviton modes dominates at large $|t|$:
218: \begin{equation}\label{17}
219: A_B(s,t) \Big|_{|t| \gg \mu \Lambda_{\pi}} \simeq \frac{s^2}{2 \mu
220: \Lambda_{\pi}^3} \, \frac{1}{\sqrt{|t|}}.
221: \end{equation}
222: Note, we would get another asymptotics in $t$, namely, $A_B(s,t)
223: \sim |t|^{-1}$, if we sum only finite number of the massive
224: gravitons.
225:
226: As it was shown in Ref.~\cite{Giudice:02}, it is ladder diagrams
227: that makes a leading contribution of the KK gravitons to an
228: amplitude and results in the eikonal representation for the
229: amplitude ($q^2 = -t$):
230: \begin{equation}\label{18}
231: A^{eik}(s,t) = 2is \! \int d^2 b \, e^{i q \, b} \left[ 1 - e^{i
232: \chi(s,b)} \right],
233: \end{equation}
234: with the eikonal given by
235: \begin{equation}\label{19}
236: \chi(s, b) = \frac{1}{4\pi s} \int dq \, q \, J_0(q \, b) \,
237: A_B(s, -q^2).
238: \end{equation}
239:
240: The proper accounting for the massless mode has been presented in
241: Ref.~\cite{Kisselev:03}. The result is the following:
242: \begin{eqnarray}\label{20}
243: && A^{eik}(s,t) = e^{i \phi_4} \, \Bigg\{ \frac{8\pi G_N s}{-t}
244: \, \frac{\Gamma(1 - i G_N s)}{\Gamma(1 + i G_N s)}
245: \nonumber \\
246: && + \, 4\pi i s \, (-t)^{-i G_N s} \! \int\limits_0^{\infty} \,
247: db \, b^{1 -2i G_N s} \, J_0 \left( b \sqrt{|t|} \right) \Big[ 1 -
248: e^{i \chi_{mass} (s,b)} \Big] \Bigg\}.
249: \end{eqnarray}
250: Here $\chi_{mass}(s,b)$ denotes a contribution the massive modes
251: to the eikonal, $G_N$ is the Newton constant, and $\phi_4$ is a
252: 4-dimensional (infinite) phase. The first term in the RHS of
253: Eq.~\eqref{20} is well-known 4-dimensional result derived by
254: different methods in Refs.~\cite{Hooft:87}. It is negligible at
255: any conceivable energy and momentum transfer, and we can write (up
256: to a phase factor):
257: \begin{equation}\label{21}
258: A^{eik}(s,t) \simeq 4\pi i s \! \int\limits_0^{\infty} \, db \, b
259: \, J_0 \left( b \sqrt{|t|} \right) \Big[ 1 - e^{i \chi_{mass}
260: (s,b)} \Big].
261: \end{equation}
262:
263: It follows from \eqref{16} that the eikonal depends on two
264: dimensionless variables, $s/\Lambda_{\pi}^2$, and
265: \begin{equation}\label{22}
266: u = b \mu \Lambda_{\pi},
267: \end{equation}
268: and it looks like
269: \begin{equation}\label{24}
270: \tilde{\chi}_{mass}(s,u) \equiv \chi_{mass}(s,\frac{u}{\mu
271: \Lambda_{\pi}}) = \frac{1}{8\pi} \, \frac{s}{\Lambda_{\pi}^2} \!
272: \int\limits_0^{\infty} \, dv \, J_0(u v) \, \frac{I_2(v)}{I_1(v)}.
273: \end{equation}
274: The eikonal~\eqref{24} is very well approximated by the following
275: expression (see Appendix for details):
276: \begin{equation}\label{25}
277: \tilde{\chi}_{mass}(s,u) \simeq \frac{\sqrt{3}}{16\pi} \,
278: \frac{s}{u \Lambda_{\pi}^2} \, \exp(-2 \sqrt{3} \, u).
279: \end{equation}
280:
281: At $\sqrt{s} \gg 5 \, \Lambda_{\pi}$, the eikonal is
282: exponentially small outside the region
283: \begin{equation}\label{26}
284: b \lesssim b_0(s) = \frac{1}{\sqrt{3} \, \mu \Lambda_{\pi}} \, \ln
285: \frac{\sqrt{s}}{\Lambda_{\pi}}.
286: \end{equation}
287: At $b \rightarrow 0$, it is proportional to $b^{-1}$. Thus, we can
288: roughly estimate the high-energy behavior of elastic cross
289: section:
290: \begin{equation}\label{28}
291: \sigma_{el}(s) \simeq \frac{\pi}{3\,(\mu \Lambda_{\pi})^2} \,
292: \ln^2 \frac{\sqrt{s}}{\Lambda_{\pi}} \thickapprox
293: \frac{\pi}{m_1^2} \, \ln^2 \frac{s}{\Lambda_{\pi}^2},
294: \end{equation}
295: where $m_1$ is a mass of a lightest KK graviton.
296:
297: Let us underline that the Froissart-Martin like formula \eqref{28}
298: describes the contribution of the massive graviton modes. The
299: presence of the \emph{massless} graviton in the theory should
300: result in \emph{infinite} elastic and total cross
301: sections~\cite{Petrov:02}. However, its contribution can be safely
302: neglected in our further calculations.
303:
304:
305: We can rewrite Eq.~\eqref{21} in the form
306: \begin{equation}\label{30}
307: A^{eik}(s,t) \simeq 4\pi i \, \frac{s}{(\mu \Lambda_{\pi})^2} \!
308: \int\limits_0^{\infty} \, du \, u \, J_0 \left( u
309: \frac{\sqrt{-t}}{\mu \Lambda_{\pi}} \right) \Big[ 1 - e^{i
310: \tilde{\chi}_{mass} (s,u)} \Big].
311: \end{equation}
312: Correspondingly, the differential cross section in dimensionless
313: variable
314: \begin{equation}\label{32}
315: y = \frac{-t}{s}
316: \end{equation}
317: is defined by
318: \begin{equation}\label{34}
319: \frac{d \sigma_{el}}{dy} = \frac{1}{16 \pi s} \,
320: |A^{eik}(s,-ys)|^2,
321: \end{equation}
322: and we get the estimate
323: \begin{equation}\label{36}
324: \frac{d \sigma_{el}}{dy} \Big|_{y=0} \simeq \frac{\pi \, s}{36
325: \,(\mu \Lambda_{\pi})^4} \, \ln^4 \frac{\sqrt{s}}{\Lambda_{\pi}}.
326: \end{equation}
327:
328: It follows from \eqref{34}, \eqref{30} that $d\sigma_{el}/dy$
329: depends only on variable $y$~\eqref{32}, parameter $\mu$, and the
330: ratio $\sqrt{s}/\Lambda_{\pi}$, in addition to the dimensional
331: factor $(\mu \Lambda_{\pi})^{-2}$ which defines a magnitude of the
332: cross section. In particular, we have $d\sigma_{el}/dy |_{y=0} = s
333: \, (\mu \Lambda_{\pi})^{-4} \, f(s/\Lambda_{\pi}^2)$, and
334: $\sigma_{el} = (\mu \Lambda_{\pi})^{-2} \, g(s/\Lambda_{\pi}^2, \,
335: \mu)$, were $f(x)$ and $g(x,y)$ are dimensionless functions
336: defined via the eikonal.
337:
338: The results of our calculations with the use of formulae
339: \eqref{30} \eqref{34}, and \eqref{24} are presented in
340: Fig.~\ref{fig:dsigma elas}-\ref{fig:lambda-mu total dependence}.
341: The curves in Fig.~\ref{fig:dsigma elas} which show an energy
342: dependence of the cross section were obtained for $\Lambda_{\pi}=$
343: 2 TeV, and $\mu=0.05$. The dependence of the differential cross
344: section on the parameter $\Lambda_{\pi}$ at $s=2 \cdot 10^{11}$
345: GeV$^2$, $\mu = 0.1$ is presented in Fig.~\ref{fig:Lambda
346: dependence}. Next Fig.~\ref{fig:mu dependence} demonstrates the
347: dependence of $d\sigma/dy$ on the parameter $\mu$ at $s=2 \cdot
348: 10^{10}$ GeV$^2$, $\Lambda_{\pi} = $ 1 TeV. Finally, in
349: Fig.~\ref{fig:lambda-mu total dependence} the reduced differential
350: cross section (namely, multiplied by the factor $(\mu
351: \Lambda_{\pi})^2$) is shown for several sets ($\Lambda_{\pi},
352: \mu$).
353:
354:
355: \section{Neutrino-nucleon cross section and neutrino induced
356: air showers}
357:
358: Let us now estimate a neutrino-nucleon differential cross section
359: as a function of variable $y$. The neutrino scatters off quarks
360: and gluons which are distributed inside the nucleon. Thus,
361: neutrino-nucleon cross section is presented by
362: \begin{equation}\label{38}
363: \frac{d\sigma^{\nu N}(s)}{dy} = \int\limits_{x_{min}}^1 dx \sum_i
364: f_i(x,M^2) \, \frac{d\sigma^{\nu i}(\hat{s})}{dy},
365: \end{equation}
366: where $f_i(x,M^2)$ is a distribution of parton $i$ in momentum
367: fraction $x$, and $\hat{s} = xs$ is an invariant energy of a
368: partonic subprocess. The partonic differential cross section,
369: $d\sigma^{\nu i}(\hat{s})/dy$, is defined via the eikonal
370: \eqref{25} taken at the energy $\sqrt{\hat{s}}$.
371:
372: We use a set of parton distribution functions (PDFs) from
373: Ref.~\cite{Alekhin:02} based on an analysis of existing deep
374: inelastic data in the next-to-leading order QCD approximation in
375: the fixed-flavor-number scheme. The extraction of the PDFs is
376: performed in \cite{Alekhin:02} simultaneously with the value of
377: the strong coupling and high-twist contributions to structure
378: functions. The PDFs are available in the region $10^{-7} < x < 1$,
379: $2.5 \text{ GeV}^2 < Q^2 < 5.6 \cdot 10^7 \text{
380: GeV}^2$~\cite{Alekhin:02}. So, no extrapolation in variable $x$ is
381: needed.
382:
383: We put $x_{min} = \Lambda_{\pi}^2/s$ in \eqref{36}. Since the
384: eikonal is effectively cut at $b = b_0(\hat{s})$~\eqref{26}, we
385: take the mass scale in PDFs to be $M = 1/b_0(\sqrt{\hat{s}})$. The
386: effective impact parameter $b_0$ is much smaller than the size of
387: the nucleon. Thus, our assumption that the neutrino interacts with
388: the constituents of the nucleon and probes its inner structure, is
389: well justified.
390:
391: The differential cross section as a function of $y$, the energy
392: fraction deposited from the neutrino to the nucleon, is presented
393: in Fig.~\ref{fig:dsigma_10EeV} for the neutrino energy $E_{\nu} =
394: 10$ EeV and three sets of parameters of the RS model.
395:
396: In order to estimate an effective range of variable $y$ which
397: contributes to the neutrino-nucleon cross section, we have
398: calculated a quantity
399: \begin{equation}\label{40}
400: \sigma^{\nu N} (y > y_0) = \int\limits_{y_0}^1 \frac{d \sigma^{\nu
401: N}}{dy},
402: \end{equation}
403: where $y_0$ is a minimum fraction of energy lost by the neutrino
404: (deposited to the nucleon). The dependence of the quantity
405: $\sigma_{el} (y > y_0)$ on $y_0$ at different values of the
406: neutrino energy $E_{\nu}$ is shown in Fig.~\ref{fig:sigma_y0_1}
407: for $\Lambda_{\pi} = 2$ TeV, $\mu = 0.05$. Next two figures,
408: Fig.~\ref{fig:sigma_y0_2} and Fig.~~\ref{fig:sigma_y0_3}, show the
409: dependence of $\sigma_{el} (y
410: > y_0)$ on the parameters $\mu$ and $\Lambda_{\pi}$.
411:
412: Ultra-high energy cosmic neutrinos have not yet been detected (see
413: non-observation of neutrino-induced events reported by the Fly's
414: Eye~\cite{Fly's Eye}, the AGASA~\cite{AGASA} and the
415: RICE~\cite{RICE} collaborations). A number of experiments under
416: construction will allow to measure fluxes of such neutrinos within
417: the next few years. Among them are Pierre Auger Observatory,
418: IceCube neutrino telescope at the South Pole, Anita radio detector
419: for a balloon flights around the South Pole, as well as EUSO,
420: SalSA and OWL proposals. We will consider the first
421: possibility~\cite{Auger}.
422:
423: The number of horizontal hadronic air showers with the energy
424: $E_{sh}$ larger than a threshold energy $E_{th}$, initiated by
425: neutrino-nucleon interactions, is given by
426: \begin{eqnarray}\label{42}
427: N_{sh} &=& T N_A \int\limits_{E_{th}}^{E_{max}} dE \, \Big[
428: \sum_{i = e, \, \mu, \, \tau} \! \Phi_{\nu_i}(E) \int\limits_0^1
429: dy \,\, \frac{d\sigma_{\nu N}^{grav} (E)}{dy} \, {\cal A}(y E) \,
430: \theta(yE - E_{th})
431: \nonumber \\
432: &+& \sum_{i = e, \, \mu, \, \tau} \! \Phi_{\nu_i}(E) \,
433: \sigma_{\nu_i N}^{SM}(E) \, {\cal A}(\bar{y}_i E_{\nu_i}) \,
434: \theta(\bar{y}_i E - E_{th}) \Big],
435: \end{eqnarray}
436: where $N_A = 6.022 \cdot 10^{23} \text{ g}^{-1}$, $T$ is a time
437: interval (one year, in our case), and ${\cal A}(E)$ is a detector
438: acceptance as a function of a shower energy (in units of km$^3$
439: steradian water equivalent $= 10^{15}$ g). The quantity
440: $\Phi_{\nu_i}(E)$ in \eqref{42} is a flux of the neutrino of type
441: $i$. Both neutrino and antineutrino are assumed in the sums in
442: Eq.~\eqref{42}. The product $E \, \Phi_{\nu_i}(E)$ is in units of
443: cm$^{-2}$ yr$^{-1}$. We have taken into account that the energy of
444: the shower resulted from the gravitational interaction is equal to
445: $yE$, and that this interaction is universal for all types of
446: neutrinos.
447:
448: For the energy distribution of the neutrino in the SM processes,
449: we have used the approximation $d\sigma_{\nu_i N}^{SM}(y,E)/dy
450: \simeq \sigma (E) \, \delta(y - \bar{y_i})$. The inelasticity
451: $\bar{y}_i$ defines a mean fraction of the neutrino energy
452: deposited into the shower in a corresponding SM process. We have
453: put $\bar{y}_e =1$ for SM \emph{charged} current interactions
454: initiated by electronic neutrino, while for SM \emph{neutral}
455: interactions initiated by $\nu_e$ and for $\nu_{\mu}/
456: \nu_{\tau}$-events we have taken $\bar{y}_e = \bar{y}_{\mu} =
457: \bar{y}_{\tau} =0.2$~\cite{Sigl:98}.
458:
459:
460: The number of extensive quasi-horizontal showers induced by
461: so-called cosmogenic neutrinos which can be detected by the array
462: of the southern site of the Pierre Auger Observatory, is presented
463: in Table~\ref{tab:cosmogenic flux} for several sets of the RS
464: parameters. These values of parameters are chosen in such a way in
465: order not to violate experimental and theoretical bounds presented
466: in Fig.~\ref{fig:bounds}.%
467: \footnote{Remember that in our notations $\kappa/\bar{M}_{Pl}
468: \equiv \mu$, $m_1 \simeq 3.83 \, \mu \Lambda_{\pi}$.}
469: %
470: The cosmogenic neutrino flux is taken from
471: Refs.~\cite{Protheroe:96}, assuming $E_{max} = 3 \cdot 10^{21}$
472: eV. The acceptance of the Auger detector is taken from
473: Ref.~\cite{Capelle:98} (it is not assumed that a shower axis falls
474: certainly in the array). The threshold energy $E_{th}$ is chosen
475: to be $10^{17}$ eV.
476:
477: For comparison, a SM background is presented in the last row of
478: the Table~\ref{tab:cosmogenic flux}. This value is in agreement
479: with the number obtained recently for the same neutrino flux in
480: Ref.~\cite{Anchordoqui:04}.
481:
482: \begin{table}[h!t]
483: \begin{center}
484: \caption{\small Yearly event rates for nearly horizontal neutrino
485: induced showers with $\theta_{zenith} > 70^{\circ}$ for the
486: cosmogenic neutrino flux from Ref.~\cite{Protheroe:96} for three
487: sets of the parameters. Number of events corresponds to one side
488: of the Auger ground array.}
489: \bigskip
490: \begin{tabular}{||c||c|c|c||}
491: \hline
492: & $\Lambda_{\pi}$=2 TeV, $\mu$=0.10 & $\Lambda_{\pi}$=3 TeV, $\mu$=0.05 &
493: $\Lambda_{\pi}$=3 TeV, $\mu$=0.10
494: \\ \hline
495: SM+grav & 0.81 & 0.66 & 0.43
496: \\ \hline
497: SM & \multicolumn{3}{c||}{0.24}
498: \\
499: \hline
500: \end{tabular}
501: \label{tab:cosmogenic flux}
502: \end{center}
503: \end{table}
504:
505: The cosmogenic neutrino flux is the most reliable one, since it
506: relies only on two assumptions: (i) the observed extremely high
507: energy cosmic rays contain protons, (ii) these cosmic rays are
508: primarily extragalactic in origin. Note, however, that the
509: cosmogenic neutrino flux may be significantly depleted, if a
510: substantial fraction of the cosmic ray primaries are heavy nuclei
511: rather than protons~\cite{Hooper:04}.
512:
513: The cosmogenic neutrino flux not only represents a lower limit on
514: the flux of ultrahigh energy neutrinos, but it also can be used to
515: put an upper limit on the neutrino flux. In Ref.~\cite{Bahcall:01}
516: an upper limit (called WB bound) on a flux of neutrinos from
517: compact sources which are optically thin to $p \gamma$ and $pp$
518: interactions (such as active galactic nuclei) has been obtained.
519: The number of showers which can be registered by the Auger
520: detector for this case is shown in Table~\ref{tab:WB bound}. We
521: have chosen the same threshold energy $E_{th}$ = 10$^{17}$ eV and
522: put $E_{max} = 10^{21}$ eV.
523:
524: \begin{table}[h!t]
525: \begin{center}
526: \caption{\small The same as in Table~\ref{tab:cosmogenic flux} but
527: for the Waxman-Bahcall neutrino flux~\cite{Bahcall:01}.}
528: \bigskip
529: \begin{tabular}{||c||c|c|c||}
530: \hline
531: & $\Lambda_{\pi}$=2 TeV, $\mu$=0.10 & $\Lambda_{\pi}$=3 TeV, $\mu$=0.05 &
532: $\Lambda_{\pi}$=3 TeV, $\mu$=0.10
533: \\ \hline
534: SM+grav & 1.03 & 0.78 & 0.49
535: \\ \hline
536: SM & \multicolumn{3}{c||}{0.28 }
537: \\
538: \hline
539: \end{tabular}
540: \label{tab:WB bound}
541: \end{center}
542: \end{table}
543:
544: Note, the so-called cascade upper limit on transparent neutrino
545: sources~\cite{Mannheim:01} (MPR bound) is 43 times higher that the
546: WB bound. It exploits the EGRET data on the diffuse gamma-ray
547: background~\cite{EGRET:98}.
548:
549: The lower bound on the cosmogenic neutrino flux was also obtained
550: under assumption that the observed extremely high energy cosmic
551: rays below 10$^{20}$ eV are protons from uniformly distributed
552: extragalactic sources~\cite{Fodor:03}. It uses the fact that the
553: proton are accumulated around the energy $E_{_{GZK}} = 4 \cdot
554: 10^{19}$ eV due to the GZK mechanism~\cite{Greisen:66}. The lower
555: cosmogenic neutrino spectrum is practically cut at $E_{\nu} \simeq
556: 2 \cdot 10^{19}$ eV~\cite{Fodor:03}. Other recent estimates of the
557: cosmogenic neutrino fluxes can be found in
558: Refs.~\cite{Kalashev:02,Anchordoqui:04}.
559:
560:
561: \section{Conclusions and discussions}
562:
563: In the present paper we have calculated the contribution from the
564: massive graviton modes to the eikonal in the RS model. The results
565: were applied to the neutrino-nucleon scattering at transplanckian
566: energies. Both differential and total cross sections are estimated
567: for the different sets of the parameters of the model. By using
568: differential cross sections, we have calculated the number of
569: quasi-horizontal neutrino induced air showers which can be
570: detected at the Auger Observatory per year. The estimates were
571: obtained for two fluxes of cosmogenic neutrinos.
572:
573: The differential cross section, $d\sigma(y)/dy$, where $y$ is the
574: fraction of the neutrino energy $E_{\nu}$ deposited to the shower,
575: can reach tens of mb at $y=0$, depending on energy
576: (Fig.~\ref{fig:dsigma_10EeV}). However, the differential cross
577: section exhibits a rapid fall-off in $y$, starting at some small
578: $y$. As a result, the gravitational cross section appears to be
579: approximately one order of magnitude larger than the SM cross
580: section at the same energy. To illustrate this statement, let us
581: fix the parameters of the RS model to be $\Lambda_{\pi} = 2$ TeV,
582: $\mu = 0.1$. Then we have $(d\sigma(y)/dy)|_{y=0} \simeq 4$ mb for
583: $E_{\nu} = 10^{10}$ GeV (see Fig.~\ref{fig:dsigma_10EeV}). As one
584: can see in Fig.~\ref{fig:dsigma_10EeV}, $d\sigma(y)/dy$ begins to
585: fall rapidly at $y > 10^{-5}$. The numerical calculations show
586: that $\sigma \simeq 4 \cdot 10^{-4}$ mb for this case (dashed
587: curve in Fig.~\ref{fig:sigma_y0_2}).
588:
589: The energy of neutrino induced air shower, $E_{sh} = y E_{\nu}$,
590: is bounded from below by a threshold energy $E_{th}$. Thus, the
591: fraction $y$ should obey the inequality $y \geqslant
592: E_{th}/E_{max}$, where $E_{max}$ is a maximum energy in a neutrino
593: spectrum. For $E_{th} = 10^{8}$ GeV and $E_{max} = 10^{11(12)}$
594: GeV, we get $y \geqslant 10^{-3(4)}$. Thus, the air showers event
595: rate is defined by the region of $y$, in which neutrino-nucleon
596: cross section $d\sigma(y)/dy$ is significantly reduced in
597: comparison with its magnitude at $y=0$. Nevertheless, gravity
598: contribution to the event rate at the Auger detector is several
599: times larger than the SM background, as one can see from
600: Tables~\ref{tab:cosmogenic flux} and \ref{tab:WB bound}.
601:
602: Recently, model independent bounds on the inelastic
603: neutrino-nucleon cross section derived from the AGASA~\cite{AGASA}
604: and RICE~\cite{RICE} search results on neutrino events were
605: obtained~\cite{Anchordoqui:04}. The bounds exploit the cosmogenic
606: neutrino fluxes from Refs.~\cite{Protheroe:96,Fodor:03}. However,
607: they were derived under an assumption that the \emph{total}
608: neutrino energy goes into shower energy, that is $y=1$. As we have
609: seen, it is not a case for the gravitational interactions
610: originated from
611: $t$-channel KK gravitons, which prefer $y \ll 1$.%
612: \footnote{In processes initiated by graviton $t$-channel exchanges
613: in large extra dimensions mean energy loss is also small, as was
614: pointed out in Ref.~\cite{Emparan:02}. On the contrary, in a
615: process of black hole production, the neutrino loses most of its
616: initial energy ($y \approx 1$).}
617: %
618: Generally, in order to extract an upper limit on $\sigma_{tot}$, a
619: dependence of $d\sigma(y)/dy$ on $y$ is needed. So, we conclude
620: that the bounds from Ref.~\cite{Anchordoqui:04} can not be
621: directly apply to the neutrino-nucleon cross sections derived in
622: our scheme.
623:
624: %%%%%%%%%%%%%%%%%%%
625: % Acknowledgments %
626: %%%%%%%%%%%%%%%%%%%
627:
628: \section*{Acknowledgments}
629:
630: The author is indebted to V.A. Petrov for discussions and valuable
631: remarks.
632:
633: %%%%%%%%%%%%
634: % Appendix %
635: %%%%%%%%%%%%
636:
637: \setcounter{equation}{0}
638: \renewcommand{\theequation}{A.\arabic{equation}}
639:
640: \section*{Appendix}
641:
642: In Appendix we calculate the dependence of the eikonal~\eqref{24}
643: on variable $u$~\eqref{22}. Let us define
644: \begin{equation}\label{A02}
645: I(u) = \int\limits_0^{\infty} \, dx \, J_0(u x) \, R(x),
646: \end{equation}
647: with
648: \begin{equation}\label{A04}
649: R(x) = \frac{I_2(x)}{I_1(x)}
650: \end{equation}
651: being the ratio of two modified Bessel functions.
652:
653: It easily to see from \eqref{A02} that $I(u) \rightarrow u^{-1}$
654: at $u \rightarrow 0$, since $R(x) \rightarrow 1$ at $x \rightarrow
655: \infty$. The asymptotics of $I(u)$ at large $u$ is defined by the
656: behavior of the integrand at small $x$ which looks like
657: \begin{equation}\label{A06}
658: R(x) \simeq \frac{x}{4} - \frac{x^3}{96} + \frac{x^5}{1536} -
659: \frac{x^7}{23040} + \mathrm{O}(x^9).
660: \end{equation}
661:
662: Let us now demonstrate that at $u \rightarrow \infty$ the function
663: $I(u)$~\eqref{A02} decreases faster that any fixed power of
664: $u^{-1}$. By using well-known relation~\cite{Erdelyi:II}
665: \begin{equation}\label{A08}
666: x^{\nu - 1}J_{\nu - 1}(u x) = \frac{1}{u} \, \Big( \frac{d}{x\,dx}
667: \Big) \big[ x^{\nu} J_{\nu}(u x)\big],
668: \end{equation}
669: and integrating \eqref{A02} by parts $k$ times, we obtain:
670: \begin{equation}\label{A09}
671: I(u) = \frac{1}{u^k} \int\limits_0^{\infty} \, dx \, J_k(u x)
672: F_k(x),
673: \end{equation}
674: where $J_k(z)$ is the Bessel function, and
675: \begin{equation}\label{A10}
676: F_k(x) = (-1)^k \, x^{k+1} \, \Big( \frac{d}{x\,dx} \Big)^k \left[
677: \frac{R(x)}{x} \right].
678: \end{equation}
679:
680: The function $F_k(x)$ \eqref{A10} has the following properties: it
681: is proportional to $x^{k+1}$ at $x\rightarrow 0$, and it decreases
682: as $x^{-k}$ at $x \rightarrow \infty$. For any positive integer
683: $k$, it depends only on x and on the ratio $R(x)$~\eqref{A04} (but
684: not on $I_1(x)$ and $I_2(x)$ separately) due to the following
685: relations between modified Bessel functions~\cite{Erdelyi:II}:
686: \begin{eqnarray}\label{A11}
687: \frac{d}{dx} I_1(x) &=& I_2(x) + \frac{1}{x} \, I_1(x),
688: \nonumber \\
689: \frac{d}{dx} I_2(x) &=& I_1(x) - \frac{2}{x} \, I_2(x).
690: \end{eqnarray}
691:
692: For instance, for $k=1$ one has
693: \begin{equation}\label{A12}
694: F_1(x) = - 1 + \frac{4R(x)}{x} + \big[ R(x) \big]^2
695: \end{equation}
696: This expression has asymptotics $x^2/48$ and $x^{-1}$ at small and
697: large $x$, respectively. For $k=2$ one gets
698: \begin{equation}\label{A14}
699: F_2(x) = \frac{1}{x} \Big\{ -6 \Big[ 1 - \frac{4R(x)}{x} \Big] -
700: 2x \, R(x) + 12 \, \big[ R(x) \big]^2 + 2x \, \big[ R(x) \big]^3
701: \Big\}
702: \end{equation}
703: The asymptotics of $F_2(x)$ are $x^3/192$ and $3 \, x^{-2}$.
704:
705: Since $k$ is an arbitrary positive integer, we conclude from
706: \eqref{A09} and \eqref{A10} that $\lim\limits_{u \rightarrow
707: \infty} u^a \,I(u) = 0$ for any $a > 0$.
708:
709: The integral in \eqref{A09}, contrary to an original one
710: \eqref{A02}, converges rapidly at $x \rightarrow \infty$ for $k
711: \geqslant 2$, and could be used for numerical calculations. It
712: cannot be calculated analytically. However, there exists an
713: expression which approximates our integral with a very high
714: accuracy:
715: \begin{equation}\label{A16}
716: \bar{I}(u) = \int\limits_0^{\infty} \, dx \, J_0(u x) \,
717: \bar{R}(x),
718: \end{equation}
719: with
720: \begin{equation}\label{A18}
721: \bar{R}(x) = \frac{\sqrt{3}}{2} \, \frac{x}{\sqrt{x^2 + 12}}.
722: \end{equation}
723: The function $\bar{R}(x)$ has the following expansion at $x^2 <
724: 12$ (compare with Eq.~\eqref{A06}):
725: \begin{equation}\label{A20}
726: \bar{R}(x) \simeq \frac{x}{4} - \frac{x^3}{96} + \frac{x^5}{1536}
727: - \frac{5 \, x^7}{110592} + \mathrm{O}(x^9).
728: \end{equation}
729: The integral \eqref{A16} is a table one~\cite{Prudnikov:II}:
730: \begin{equation}\label{A22}
731: \bar{I}(u) = \frac{\sqrt{3}}{2u} \, \exp(-2 \sqrt{3} \, u).
732: \end{equation}
733:
734: We can integrate the RHS of \eqref{A16} twice by parts,
735: \begin{equation}\label{A24}
736: \bar{I}(u) = \frac{1}{u^2} \int\limits_0^{\infty} \, dx \, J_2(u
737: x) \, \bar{F}_2(x),
738: \end{equation}
739: and compare $\bar{F}_2(x) = (3\sqrt{3}/2)\,x^3/(x^2 + 12)^{5/2}$
740: with the corresponding function $F_2(x)$~\eqref{A14}. The result
741: of our calculations is presented in Fig.~\ref{fig:integrand}.
742: Formula \eqref{A22} gives practically the same dependence on
743: variable $u$ as a numerical integration of the exact expression by
744: using formula~\eqref{A09} (with $k=2$) does, see Fig.~\ref{fig:integral}.%
745: \footnote{Some disagreement between two curves at $u \gtrsim 2.5$
746: is not important, since $I(u)$ (and, consequently,
747: $\chi_{mass}(s,u)$) is strongly suppressed in this region. Note, a
748: region of very small $u$ also gives a negligible contribution to
749: the eikonal amplitude~\eqref{18}.}
750: %
751: Thus, $I(u)$ exhibits an exponential fall-off (as we expected,
752: see above), and it becomes as small as $I(u) \simeq 0.01$ already
753: at $u=1.2$.
754:
755: Taking all said above into account, we put $I(u) \rightarrow
756: \bar{I}(u)$, that results in the analytical expression for the
757: eikonal presented in the text~\eqref{25}.
758:
759: %%%%%%%%%%%%%%
760: % References %
761: %%%%%%%%%%%%%%
762:
763: \begin{thebibliography}{99}
764:
765: % RS model
766: \bibitem{Randall:99}
767: L. Randall and R. Sundrum, Phys. Rev. Lett. {\bf 83} (1999) 3370.
768: \bibitem{Randall:99*}
769: L.~Randall and R.~Sundrum, Phys. Rev. Lett. {\bf 83} (1999) 4690;
770: J. Lykken and L. Randall, JHEP {\bf 06} (2000) 014.
771: % Bounds on the ratio k/M_Pl & UED scenario in the RS model
772: \bibitem{Davoudiasl:01}
773: H. Davoudiasl, J.L. Hewett and T.G. Rizzo, Phys. Rev. D {\bf 63}
774: (2001) 075004.
775: % RS model - collider penomenology
776: \bibitem{Davoudiasl:00}
777: H. Davoudiasl, J.L. Hewett and T.G. Rizzo, Phys. Rev. Lett. {\bf
778: 84} (2000) 2080.
779: \bibitem{Allanach:00}
780: B.C. Allanach \emph{et al.}, JHEP {\bf 0009} (2000) 019.
781: % Radion in RS model (stabilization of r_c)
782: \bibitem{Goldberger:99}
783: W.D. Goldberger and M.B. Wise, Phys. Rev. Lett. {\bf 83}
784: (1999) 4922;
785: C. Cs\'{a}ki \emph{et al.}, Phys. Rev. D {\bf 62} (2000) 045015;
786: C. Cs\'{a}ki, M. Graesser and G.D. Kribs, \emph{ibid.} D {\bf 63}
787: (2001) 064020.
788: \bibitem{Watson}
789: G.N. Watson, \emph{A Treatise on the Theory of Bessel Functions}
790: (McMillan, 1922).
791: % Decays of massive modes
792: \bibitem{Han:99}
793: T. Han, J.D. Lykken and R.-J. Zhang, Phys. Rev. D {\bf 59} (1999)
794: 105006.
795: \bibitem{Dvergsnes:03}
796: E. Dvergsnes, P. Osland and N. \"{O}zt\"{u}rk, Phys. Rev. D {\bf
797: 67} (2003) 074003.
798: % Eikonalization of t-channel exchanges in ADD model
799: \bibitem{Giudice:02}
800: G.F. Giudice, R. Rattazzi and J.D. Wells, Nucl. Phys. B {\bf 630}
801: (2002) 293.
802: % t-channel (massless and massive) gravitons in ADD model
803: \bibitem{Kisselev:03}
804: A.V. Kisselev, Eur. Phys. J. C {\bf 34} (2004) 513 .
805: % 4-dimensional gravitational amplitude
806: \bibitem{Hooft:87}
807: G. 't Hooft, Phys. Lett. B {\bf 198} (1987) 61;
808: H. Verlinde and E. Verlinde, Nucl. Phys, B {\bf 371} (1992) 246;
809: M. Fabbrichesi \emph{et al.,} Nucl. Phys. B {\bf 419} (1994) 174.
810: \bibitem{Kisselev:04}
811: A.V. Kisselev and V.A. Petrov, Eur. Phys. J. C {\bf 36} (2004) 103;
812: Eur. Phys. J. C {\bf 37} (2004) 241.
813: \bibitem{Kisselev:05}
814: A.V. Kisselev and V.A. Petrov, in preparation.
815: \bibitem{Petrov:02}
816: V.A. Petrov, {\em Proceedings of the Int. Conf. Theor. Phys., TH
817: 2002} (Paris, July 2002, Eds. D. Iagolnitzer, V. Rivasseau and J.
818: Zinn-Justin, Birkh\"{a}user Verlag, 2003). Supplement (2003) 253.
819: % PDFs
820: \bibitem{Alekhin:02}
821: S.I. Alekhin, Phys. Rev. D {\bf 68}, 014002 (2003).
822: % Non-observation of neutrino-induced events:
823: % Fly's Eye results on quasi-horizontal air showers
824: \bibitem{Fly's Eye}
825: R.M. Baltrusaitis \emph{et al.}, Phys. Rev. D {\bf 31} (1985)
826: 2192.
827: % AGASA results on quasi-horizontal air showers
828: \bibitem{AGASA}
829: N. Inoue, \emph{Proc. 26th International Cosmic Ray Conference
830: (ICRC 1999)}, eds. D. Kieda, M. Salamon and B. Dingus, Salt Lake
831: city, Utah, 1999, v.~1, p.~361; S. Yoshida \emph{et al.},
832: \emph{Proc. 27th International Cosmic Ray Conference (ICRC 2001)},
833: Hamburg, Germany, 2001, v.~3, p.~1142
834: % RICE results on quasi-horizontal air showers
835: \bibitem{RICE}
836: I. Kravchenko \emph{et al.}, Astropart. Phys. {\bf 20} (2003) 195;
837: I. Kravchenko, arXiv: astro-ph/0306408.
838: \bibitem{Auger}
839: Pierre Auger Observatory, http://www.auger.org/
840: \bibitem{Sigl:98}
841: G. Sigl, Phys. Rev. D {\bf 57}, 3786 (1998).
842: % Cosmogenic netrino flux
843: \bibitem{Protheroe:96}
844: R.J. Protheroe and P.A. Johnson, Astropart. Phys. {\bf 4}, 253
845: (1996) [Erratum, {\em ibid.} {\bf 5}, 215 (1996)]; R.J. Protheroe,
846: Nucl. Proc. Suppl. {\bf 77}, 465 (1999).
847: % Auger ground array acceptance
848: \bibitem{Capelle:98}
849: K.S. Capelle, J.W. Cronin, G. Parente and E. Zas, Astrophys. Phys.
850: \bibitem{Anchordoqui:04}
851: L. Anchordoqui, Z. Fodor, S.D. Katz, A. Ringwald and H. Tu, arXiv:
852: hep-ph/0410136.
853: \bibitem{Hooper:04}
854: D. Hooper, A. Taylor and S. Sarkar, arXiv: astro-ph/0407618.
855: % Upper bounds on the neutrino flux
856: \bibitem{Bahcall:01}
857: J.N. Bahcall and E. Waxman, Phys. Rev. D {\bf 64} (2001) 023002.
858: \bibitem{Mannheim:01}
859: K. Mannheim, R.J. Protheroe and J.P. Rachen, Phys. Rev. D {\bf 63}
860: (2001) 023003.
861: \bibitem{EGRET:98}
862: P. Sreekumar \emph{et al.}, Astrophys. J. {\bf 494} (1998) 523.
863: % Lower bound on cosmogenic neutrino flux
864: \bibitem{Fodor:03}
865: Z. Fodor, S.D. Katz, A. Ringwald and H. Tu, JCAP {\bf 0311} (2003)
866: 015;
867: A. Ringwald, arXiv: hep-ph/0409151.
868: % GZK cuttoff
869: \bibitem{Greisen:66}
870: K. Greisen, Phys. Rev. Lett. {\bf 16} (1966) 748;
871: G.T. Zatsepin and V.A. Kuzmin, JETP Lett. {\bf 4} (1966) 78.
872: % Estimates of cosmogenic netrino flux
873: \bibitem{Kalashev:02}
874: O.E. Kalashev, V.A. Kuzmin, D.V. Semikoz and G. Sigl, Phys.Rev.
875: D {\bf 66}, 063004 (2002); D.V. Semikoz and G. Sigl, JCAP {\bf
876: 0404} (2004) 003.
877: \bibitem{Erdelyi:II}
878: A. Erdelyi, W. Magnus, F. Oberhettinger and F.C. Tricomi (eds.),
879: \emph{Higher Transcendental functions}, v.2 (Mc Graw-Hill Book H.
880: Company, 1955).
881: \bibitem{Prudnikov:II}
882: A.P. Prudnikov, Yu.A. Brychkov and O.I. Marichev, \emph{Integrals
883: and series}, v.2: \emph{Special functions}, Translated from
884: Russian (NY Gordon and Breach, 1986).
885: % Eneergy loss in neutrino events
886: \bibitem{Emparan:02}
887: R. Emparan, M. Masip and R. Ratazzi, Phys. Rev. D {\bf 65} (2002)
888: 064023.
889:
890: \end{thebibliography}
891:
892: %%%%%%%%%%%
893: % Figures %
894: %%%%%%%%%%%
895:
896: \clearpage
897:
898: \begin{figure}[htpb]
899: \centering
900: \epsfig{figure=bounds.eps,height=14cm,width=8cm,angle=90}
901: \caption{Experimental and theoretical constrains on the RS model
902: in the two-parameter plane $\kappa/\bar{M}_{Pl}$ --
903: $m_1$~\cite{Davoudiasl:01}. The allowed region lies in the center
904: as indicated.}
905: \label{fig:bounds}
906: \end{figure}
907:
908: \clearpage
909:
910: \begin{figure}
911: \centering
912: \epsfig{figure=dsigma_elas.eps,height=7cm,width=10cm}
913: %\epsfysize=7cm \epsffile{dsigma_elas.eps}
914: \caption{The differential cross section as a function of
915: dimensionless variable $y$ for three fixed values of the invariant
916: energy. The parameters of the RS model are chosen to be
917: $\Lambda_{\pi}=$ 2 TeV, and $\mu=0.05$.}
918: \label{fig:dsigma elas}
919: \end{figure}
920:
921: \begin{figure}
922: \centering
923: \epsfig{figure=sigma_l.eps,height=7cm,width=10cm}
924: %\epsfysize=7cm \epsffile{sigma_l.eps}
925: \caption{The differential cross section as a function of variable
926: $y$ for three values of the mass scale $\Lambda_{\pi}$ at fixed
927: energy $s=2 \cdot 10^{11}$ GeV$^2$ (with the parameter
928: $\mu=0.1$).}
929: \label{fig:Lambda dependence}
930: \end{figure}
931:
932: \clearpage
933:
934: \begin{figure}
935: \centering
936: \epsfig{figure=sigma_m.eps,height=7cm,width=10cm}
937: %\epsfysize=7cm \epsffile{sigma_m.eps}
938: \caption{The differential cross section as a function of variable
939: $y$ for three values of the RS parameter $\mu$ at fixed energy
940: $s=2 \cdot 10^{10}$ GeV$^2$ (with the scale $\Lambda_{\pi} = $ 1
941: TeV).}
942: \label{fig:mu dependence}
943: \end{figure}
944:
945: \begin{figure}
946: \centering
947: \epsfig{figure=sigma_lm_tot.eps,height=7cm,width=10cm}
948: %\epsfysize=7cm \epsffile{sigma_lm_tot.eps}
949: \caption{The reduced (dimensionless) differential cross section as
950: a function of variable $y$ for different parameter sets
951: ($\Lambda_{\pi}$, $\mu$) at invariant energy $s=2 \cdot 10^{10}$
952: GeV$^2$. The product $\mu \Lambda_{\pi}$ is taken to be 100 GeV
953: (200 GeV) for three first (two last) sets.}
954: \label{fig:lambda-mu total dependence}
955: \end{figure}
956:
957: \clearpage
958:
959: \begin{figure}
960: \centering
961: \epsfig{figure=dsigma_10EeV.eps,height=7cm,width=10cm}
962: %\epsfysize=7cm \epsffile{dsigma_10EeV.eps}
963: \caption{The differential neutrino-proton cross section as a
964: function of $y$, the fraction of the neutrino energy deposited to
965: the proton.}
966: \label{fig:dsigma_10EeV}
967: \end{figure}
968:
969: \begin{figure}
970: \centering
971: \epsfig{figure=sigma_y0_1.eps,height=7cm,width=10cm}
972: %\epsfysize=7cm \epsffile{sigma_y0_1.eps}
973: \caption{The inelastic neutrino-proton cross-section obtained by
974: integrating the differential cross section in the region $y_0
975: \leqslant y \leqslant 1$ as a function of $y_0$, the minimal
976: fraction of the neutrino energy deposited to the proton.}
977: \label{fig:sigma_y0_1}
978: \end{figure}
979:
980: \clearpage
981:
982: \begin{figure}
983: \centering
984: \epsfig{figure=sigma_y0_2.eps,height=7cm,width=10cm}
985: %\epsfysize=7cm \epsffile{sigma_y0_2.eps}
986: \caption{The same as in Fig.~\ref{fig:sigma_y0_1} but for the
987: different value of $\mu=0.1$.}
988: \label{fig:sigma_y0_2}
989: \end{figure}
990:
991: \begin{figure}
992: \centering
993: \epsfig{figure=sigma_y0_3.eps,height=7cm,width=10cm}
994: %\epsfysize=7cm \epsffile{sigma_y0_3.eps}
995: \caption{The same as in Fig.~\ref{fig:sigma_y0_1} but for the
996: different value of $\Lambda_{\pi}=$ 4 TeV.}
997: \label{fig:sigma_y0_3}
998: \end{figure}
999:
1000: \clearpage
1001:
1002: \begin{figure}
1003: \centering
1004: \epsfig{figure=integrand.eps,height=7cm,width=10cm}
1005: %\epsfysize=7cm \epsffile{integrand.eps}
1006: \caption{The exact integrand vs. approximate one (after the
1007: integration of both integrals by parts twice). See Appendix for
1008: details.}
1009: \label{fig:integrand}
1010: \end{figure}
1011:
1012: \begin{figure}
1013: \centering
1014: \epsfig{figure=integral.eps,height=7cm,width=10cm}
1015: %\epsfysize=7cm \epsffile{integral.eps}
1016: \caption{The input integral vs. approximate one as a function
1017: of the dimensionless variable $u$~\eqref{22}. See Appendix for
1018: details.}
1019: \label{fig:integral}
1020: \end{figure}
1021:
1022: \end{document}
1023: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1024: