1: \documentstyle[12pt,epsfig,epsf,cite]{article}
2: %\documentstyle[11pt,epsfig]{article}
3: %\include{dspace12}
4: \oddsidemargin=-6pt
5: \topmargin=-1.5in
6: \textwidth=6.65in
7: \textheight=9.4in
8: %\def\baselinestretch{1.3}
9: \voffset=0.75in
10: \bibliographystyle{unsrt}
11:
12: \def\gsim{\lower0.5ex\hbox{$\:\buildrel >\over\sim\:$}}
13: \def\lsim{\lower0.5ex\hbox{$\:\buildrel <\over\sim\:$}}
14: \def\barnue{\:\raisebox{-0.35ex}{$\stackrel{(-)}{\nu_e}$}\:}
15: \def\barnumu{\:\raisebox{-0.35ex}{$\stackrel{(-)}{\nu_\mu}$}\:}
16:
17: \newcommand{\be}{\begin{equation}}
18: \newcommand{\ee}{\end{equation}}
19: \newcommand{\beq}{\begin{eqnarray}}
20: \newcommand{\eeq}{\end{eqnarray}}
21: \newcommand{\barr}{\begin{array}}
22: \newcommand{\earr}{\end{array}}
23: \newcommand{\dis}{\displaystyle}
24: \newcommand{\ev}{\; {\rm eV}}
25: \newcommand{\gev}{\; {\rm GeV}}
26:
27: %%%%%%
28: \def\atil{\widetilde A}
29: \def\delm{\Delta_{3 1}}
30: \def\ue{|U_{e 3}|}
31: \def\Um{|U_{\mu 3}|}
32: \def\cP{c_{\Psi}}
33: \def\sP{s_{\Psi}}
34: \def\cOm{c_{\Omega}}
35: \def\sOm{s_{\Omega}}
36: %%%%%%%
37:
38: \begin{document}
39: %THE TEXT STARTS HERE
40: \begin{flushright}
41: {ROME1-1391-2004}\\
42: \end{flushright}
43:
44: \begin{center}
45:
46: {\Large\bf Detecting matter effects in long baseline experiments}\\[18mm]
47: {\bf Debajyoti Choudhury\footnote{On leave from Harish-Chandra Research Institute, Allahabad, India.} and
48: Anindya Datta\footnote{E-mail: debchou@physics.du.ac.in, debchou@mail.cern.ch,
49: Anindya.Datta@roma1.infn.it} } \\[2ex]
50:
51: $^1${\em Department of Physics and Astrophysics, University of Delhi,
52: Delhi 110 007, India.}
53: \\
54: $^2${\em INFN, Sezione di Roma; Dip. di
55: Fisica, Universita La Sapienza, I-00185, Rome, Italy.}
56: \end{center}
57:
58: \vskip 20pt
59: \begin{abstract}
60: Experiments strongly suggest that the flavour mixing responsible for
61: the atmospheric neutrino anomaly is very close to being maximal.
62: Thus, it is of great theoretical as well as experimental importance to
63: measure any possible deviation from maximality. In this context, we
64: reexamine the effects of matter interactions in long baseline neutrino
65: oscillation experiments. Contrary to popular belief, the muon
66: neutrino survival probability is shown to be quite sensitive to matter
67: effects. Moreover, for moderately long baselines, the difference between
68: the survival probilities for $\nu_\mu$ and $\bar\nu_\mu$ is shown
69: to be large and sensitive to the deviation of $|U_{\mu 3}|$ from
70: maximality. Performing a realistic analysis, we demonstrate that
71: a muon-storage ring $\nu$-source alongwith an iron calorimeter detector can
72: measure such deviations. (Contrary to recent claims, this is not so
73: for the NuMI--{\sc minos} experiment.) We also discuss the possible
74: correlation in measuring $U_{\mu 3}$ and $U_{e3}$ in such experiment.
75: \end{abstract}
76:
77: \section{Introduction}
78: Neutrino masses and mixings continue to intrigue us, despite
79: the continued efforts of many recent
80: experiments~\cite{superK, sno, chooz, kamland, k2k,review}. We do
81: know that the muon-neutrino ($\nu_\mu$) mixes almost maximally
82: ($\sin^2 2\theta_{\mu3} > 0.92$ at 90\% C.L) with
83: another species $\nu_3$ (which could be identified with the tau-neutrino,
84: $\nu_\tau$) and that the mass splitting
85: $|\delta m^2_{32}| \sim 2 \times 10^{-3} \ev^2$. Furthermore, the
86: electron-neutrino ($\nu_e$) mixes with a combination of
87: $\nu_\mu$ and $\nu_\tau$ with a similarly large angle $\theta_{e2} \simeq
88: 30^\circ$ but a far smaller mass splitting
89: ($\delta m^2_{21} \sim 7 \times 10^{-5} \ev^2$). And, finally, the third
90: mixing angle $\theta_{13}$ is constrained, by the {\sc chooz}
91: experiment \cite{chooz}, to be small. What we do not know though are
92: the sign of $\delta m^2_{32}$, the precise value of $\theta_{13}$ and
93: the extent to which $\theta_{\mu 3}$ may, if at all, differ from maximality.
94: Since such knowledge is obviously essential to the formulation of a
95: theory of neutrino flavours, the importance of precise determinations
96: cannot but be underemphasised. Neutrino oscillations provide
97: just the proper platform for such an exercise.
98:
99:
100:
101: As is well known, during passage through dense matter, various
102: amplitudes for neutrino oscillations may be magnified to a great
103: degree \cite{msw} thereby rendering them measurable with relative
104: ease. While propagating in a medium, all the three neutrino flavours
105: interact with matter via neutral curent (NC) interactions with equal
106: strength. Those of the electron flavour, $\barnue$, have, in addition,
107: a extra charged current interaction as well. Since it is this extra
108: interaction that provides {\em flavour-dependence}, all matter-induced
109: effects in neutrino oscillations (including those in the
110: $\nu_\mu$--$\nu_\tau$ sector) are proportional to the size of the
111: relevant mixing with $\nu_e$.
112:
113: Recent theoretical advances in the quest of producing
114: very high intensity muon sources engender
115: optimism regarding a future neutrino factory wherein an intense
116: beam of muons is to be accelerated to a not too high energy
117: and stored in a storage ring with a
118: straight section directed towards a neutrino
119: detector. Muons would decay in this straight section thereby producing high
120: intensity neutrino beams (both of electron-- and muon-types)
121: that are highly collimated in the direction of the
122: decaying muons \cite{nu_fac_tech}. The efficacy of muon storage ring
123: neutrino sources in performing precision measurements of neutrino
124: parameters has already been discussed in the
125: literature~\cite{nu_fac}. The advantages over
126: neutrino experiments
127: with conventional neutrino beams (arising from $\pi^\pm$ decay)
128: are manifold: ($i$) a precise knowledge of the $\nu_\mu$ and $\bar
129: \nu_e$ fluxes helps reduce the systematic
130: errors; ($ii$) assuming a $\mu$-beam of, say,
131: 20 GeV energy, such neutrinos have, on the average, energies higher than
132: those of conventional neutrino beams thereby increasing the $\nu$
133: cross-section at the detector; and ($iii$) the aforementioned
134: collimation.
135:
136: Previous studies of the matter effect in Ref. \cite{barger_matter} at
137: a long base line experiment were concentrated mainly on the measurement of
138: the rate of wrong sign muon events, which, in turn is proportional to
139: the transition probablity $P_{\nu_e \to \nu_\mu}$. Such a measurement
140: would enable a determination of (or constraining of) $\sin
141: \theta_{13}$ {\em provided}\ independent and precise measurements of
142: the mixing angle $\theta_{23}$ and mass-square difference $\delta m^2
143: _{31}$ exist. Possible measurements of neutrino mixing parameters from long
144: baseline experiments using conventional neutrino beams as well as from
145: a neutrino factory, and a water \v{C}erenkov detector has also been discussed
146: in Ref.\cite{maoki}.
147:
148:
149: In the present article, we demonstrate that a sizable
150: matter effect is also evinced by a measurement of the $\nu_\mu$
151: survival probability. In addition, such a measurement has the
152: potential of detecting possible deviations from maximality of the
153: mixing angle responsible for the atmospheric neutrino anomaly. In a
154: recent study \cite{raj}, the presence of large matter effect in
155: $\nu_\mu - \nu_\tau$ oscillation as well as in $\nu_\mu$ survival
156: probability has been stressed. Although there is a large matter effect
157: present in $\nu_\mu - \nu_\tau$ sector, measurement of the
158: same is difficult due to both the lower detection efficiency for
159: the tau-events as well as the need for a more sophisticated
160: detector.
161:
162:
163: Whereas much of our analysis would be source and detector independent,
164: we also choose to examine the feasibility of making such meaurements
165: in the context of a realistic experimental setup. To be concrete, we
166: choose a 50 kT Iron detector, with detection and charge discrimination
167: capability for muons provided by a magnetic field. Such a detector was
168: proposed for Gran Sasso ({\sc monolith}) \cite{mono} and, more
169: recently, for a location in India (INO) \cite{ino}. The latter is
170: contemplated primarily as a detector for atmospheric neutrinos and
171: also as an end detector for a future neutrino factory beam. A muon
172: detection threshold of 2 GeV \cite{ino} has been used in our
173: calculation. As for the baseline, we illustrate our results for two
174: particular choices. The first one coresponds to a distance from JHF to
175: INO ($\sim$ 5000 Km) and the other one corresponds to even a longer
176: baseline, from FERMILAB to INO ($\sim$ 10000 Km). We will also compare
177: results with a baseline of 732 Km, corresponding to the Fermilab-{\sc
178: minos} distance.
179:
180: The plan for the rest of the article is as follows. In section
181: \ref{sec:analytic}, we discuss neutrino oscillation in the presence of
182: matter. We also define the asymmetry of $\mu^+$ and $\mu^-$ events and
183: explain its correlation to $U_{e3}$ and $U_{\mu 3}$. We present the
184: expressions for the asymmetry for two different aproximations: when
185: matter effect is small compared to $\frac{\delta m^2 _{32}}{2 E_\nu}$
186: i.e for short base line experiments like {\sc minos}
187: \cite{choubey_roy} and also when the matter term is comparable to the
188: $\frac{\delta m^2 _{32}}{2 E_\nu}$ i.e., for long base line
189: experiments. In section \ref{sec:minos}, we briefly review and
190: reexamine the proposal, in a recent article \cite{choubey_roy}, to
191: measure matter effect and the above correlations in the context of the
192: {\sc minos} experiment. Section \ref{sec:longbaseline} deals with the
193: expectations at lomg baseline experiments, both for the $\nu_\mu$
194: survival probability as well as for the $\nu_e$ to $\nu_\mu$
195: transition probability. An analysis of the findings in the context of
196: a realistic experimental situation is effected in Section
197: \ref{sec:storage_ring}. Finally, we conclude in Section
198: \ref{sec:concl}.
199:
200:
201: \section{Neutrino Oscillation in presence of matter}
202: \label{sec:analytic}
203: We begin by reviewing the passage of neutrino through matter.
204: While analytical results are possible, the exact answers are
205: cumbersome and not very useful. However, it is useful to
206: consider approximations such as a constant density
207: profile for the earth. While such an
208: assumption is obviously not a very accurate one, especially when
209: neutrinos traverse a longer distance through the earth,
210: the approximation serves to give us some
211: physical insights to the problem we are dealing with. Furthermore,
212: results such as those we shall obtain are still applicable if
213: the matter density within the earth may be approximated to
214: be piecewise constant. It must be kept in mind though that
215: the quantitative results presented in this article are
216: obtained not by means of such an approximation, but by
217: explicitly accounting for
218: the varying density profile of the earth using the Preliminary
219: Reference Earth Model (PREM) \cite{prem}.
220:
221:
222:
223: As with the quark sector, neutrino flavour states can be expressed
224: in terms of mass eigenstates through a relation of the form
225: \begin{equation}
226: \vert \nu_\alpha \rangle \;= \sum_i U_{\alpha i}\;\vert \nu_i \rangle \ ,
227: \end{equation}
228: where $U$ is a $3 \times 3$ unitary matrix known as the
229: Pontecorvo-Maki-Nakagawa-Sakata mixing matrix. In the absence of Majorana
230: phases (assumed to be so henceforth)\footnote{Any Majorana phase in the
231: PMNS matrix
232: cannot be probed by an oscillation experiment.},
233: $U$ can be parametrized in terms
234: of three mixing angles and a $CP$-violating phase $\delta_{\rm CP}$,
235: {\em viz.}
236: \beq
237: U = \pmatrix{
238: c_{12}c_{13} & s_{12}c_{13} & s_{13}e^{-i\delta_{\rm CP}} \cr
239: -c_{23}s_{12} - s_{23}s_{13}c_{12}e^{i\delta_{\rm CP}} & c_{23}c_{12} -
240: s_{23}s_{13}s_{12}e^{i\delta_{\rm CP}}& s_{23}c_{13}\cr
241: s_{23}s_{12} - c_{23}s_{13}c_{12}e^{i\delta_{\rm CP}}& -s_{23}c_{12} -
242: c_{23}s_{13}s_{12}e^{i\delta_{\rm CP}} & c_{23}c_{13} \cr} \ ,
243: \label{mns}
244: \eeq
245: with $c_{ij} \equiv \cos \theta_{ij}$ and $s_{ij} \equiv \sin \theta_{ij}$.
246: In vacuum, neutrino oscillations are then governed by $U$ and the
247: two independent differences in the squares of the neutrino masses,
248: $\delta m^2_{ij} \equiv m_i^2 - m_j^2$. In the rest of our analysis, we will
249: make a further simplifying assumption of $\delta_{\rm CP} = 0$.
250: With the above parametrization of $U$, experimental results can
251: be used to constrain
252: the mixing angles $\theta_{ij}$ and the mass-square differences.
253: % \begin{itemize}
254: % \item
255: Measurements of the solar neutrino fluxes, augmented by the
256: KamLAND reactor data give
257: \beq
258: 0.30 < \tan^2 \theta_{12} < 0.64, \qquad
259: 5.4\times 10^{-5} \ev^2 < \delta_{21} < 9.4 \times 10^{-5} \ev^2
260: \eeq
261: with the best-fit values given by $\tan^2 \theta_{12} = 0.40 $
262: and $\delta m^2_{21} = 6.9 \times 10^{-5} \ev^2$.
263: %
264: % \item
265: Experiments with atmospheric neutrinos (as well as the K2K
266: experiment) suggest, on the other hand, that
267: \beq
268: \sin^2 2\theta_{23} > 0.86, \qquad
269: 1.4 \times 10^{-3} \ev^2 < |\delta m^2_{32}| < 5.1 \times 10^{-3} \ev^2 \ ,
270: \eeq
271: with the best-fit value being $|\delta m^2_{32}| = 2.0 \times
272: 10^{-3}~\rm{eV^2}$.
273: %
274: %\item
275: And, finally, the negative result of the {\sc chooz} experiment implies
276: that
277: \beq
278: \sin^2 \theta_{13} < 5\times 10^{-2}~ \qquad {\rm at} \; \hbox{99.73\% C.L.}
279: \eeq
280: %\end{itemize}
281: Note that,
282: \begin{itemize}
283: \item unlike in the case for $\delta m^2_{21}$, the sign of $\delta m^2_{32}$ is
284: undetermined;
285: \item since $|\delta m^2_{32}| \approx |\delta m^2_{31}| \gg |\delta m^2_{21}|$
286: \footnote{From now on we will denote this large mass-square difference by
287: $\delta m^2_{31}$},
288: the last-mentioned (as well as the solar mixing angle $\theta_{12}$)
289: plays only a subservient role in $\nu_\mu$ oscillations;
290: \item while $|U_{\mu 3}|$ is constrained to be close to $1/\sqrt{2}$,
291: it is still allowed to be non-maximal.
292: \end{itemize}
293:
294: Working in the limit of a vanishingly small $\delta m^2_{21}$,
295: all the expressions for the oscillation/survival probabilities can be
296: expressed in terms of just two of the mixing
297: angles\footnote{While the analytical expressions below
298: have been derived under this approximation,
299: all of the numerical analysis has been performed
300: using the full expression for $U$ and a non-zero value for
301: $\delta_{12}$.},
302: {\em viz.} $\theta_{23}$ and $\theta_{13}$.
303: For a $\nu_\mu$ ($\bar \nu_\mu$) of
304: energy $E_\nu$ travelling a distance $L$ in vacuum,
305: the survival probability is then given by
306: \beq
307: P^{\rm vac}_{\mu\mu} = P^{\rm vac}_{\bar \mu \bar \mu} = 1-4 |U_{\mu 3}|^2 (1 -
308: |U_{\mu 3}|^2) \sin^2 \left(\Delta_{31}L / 2\right) \, ,
309: \label{survival_vac}
310: \eeq
311: where,
312: \beq
313: U_{e 3} = s_{13} \ , \qquad U_{\mu 3} = c_{1 3} \, s_{2 3} \ , \qquad
314: \Delta_{31} \equiv \frac{\delta m^2_{31}}{2 E_\nu} \ .
315: \eeq
316: %
317: In the
318: presence of matter, the electron (anti-)neutrino sees an additional potential
319: corresponding to charged current interactions, and the effective Hamiltonian
320: is given by
321: \beq
322: H = U \pmatrix{0 & 0 & 0 \cr 0 & \Delta_{21} & 0 \cr 0 & 0 & \Delta_{31}}
323: U^\dagger + \pmatrix{A & 0 & 0 \cr 0 & 0 & 0 \cr 0 & 0 & 0} \ ,
324: \label{effec_hamilt}
325: \eeq
326: where $A \equiv \pm \sqrt{2}G_F N_e$ for $\barnue$,
327: with $N_e$ being the instantaneous electron density of the matter.
328:
329:
330: The survival probablities suffer a
331: consequent modification. While the full expressions are cumbersome,
332: it is instructive to look at a
333: simplified form obtainable for a matter density small enough that $A$
334: can be treated as a perturbation in the effective Hamiltonian. In this
335: limit, following Ref. \cite{choubey_roy},
336: \beq
337: \barr{rcl}
338: P_{\mu\mu \, (\bar \mu \bar \mu)}
339: & \simeq & \dis P^{\rm vac} \pm (\Delta P_{\mu \mu} / 2)
340: + {\cal O}(\atil^2)
341: \\[1.5ex]
342: \Delta P_{\mu\mu} &\simeq&
343: 4 \, \atil \, |U_{e3}|^2 |U_{\mu 3}|^2
344: (1-2 |U_{\mu 3}|^2) \,
345: \left[4 \sin^2 \Psi - \Psi \, \sin (2 \, \Psi) \right]
346: + {\cal O}(\atil^2)
347: \earr
348: \label{survival_matter}
349: \eeq
350: where
351: \beq \dis
352: \atil \equiv \frac{A}{\Delta_{31}}
353: \qquad {\rm and} \qquad
354: \Psi \equiv \frac{\Delta_{31} \, L}{2} \ .
355: \eeq
356: The asymmetry $\Delta P_{\mu\mu}$ is a measure of the matter effect
357: and that it should be proportional to both $\atil$ and $|U_{\mu 3}|^2$ is
358: obvious. Note also that
359: the $\nu_\mu$ has the same interaction with matter as the $\nu_\tau$
360: with which it primarily mixes and any matter effect can only seep in
361: through a mixing with the $\nu_e$. This, then, accounts for the overall
362: factor of $|U_{e3}|^2$. All of these three proportionalities quite
363: independent of the approximation, and, in fact, are exact
364: results. The factor $(1-2 |U_{\mu 3}|^2)$,
365: however, is but a consequence of the approximation of a small
366: matter term $\atil$ and is applicable only for neutrinos
367: traversing small base lines. Since this factor vanishes identically
368: for a maximal mixing in the $\nu_\mu - \nu_\tau$ sector,
369: this matter-induced asymmetry is potentially a sensitive probe
370: of the maximality of $U_{\mu 3}$. This has already been pointed
371: out\cite{choubey_roy} and examined in the context of the
372: {\sc minos} experiment.
373: In the next section we will briefly reexamine their
374: claims.
375:
376:
377: Convoluting eqn.(\ref{survival_matter})
378: with the $E_\nu$-dependent flux and the detection efficiency of a
379: given detector, one would obtain the number of events and thereby
380: the event asymmetry
381: \beq
382: \Delta N = N_{\nu_\mu} - N_{\bar \nu_\mu} \equiv N_- - N_+ \ .
383: \eeq
384: The latter could then be used to place contraints in the
385: $|U_{\mu 3}|$--$|U_{e 3}|$ plane.
386: Of course, in a realistic case,
387: the effects of a varying matter density has to be taken into account
388: as well and this we do include in our analysis. With the
389: effect being proportional to $\atil$, it will be magnified when the
390: (anti-)neutrino beam traverses regions of high density, or in other words,
391: the core of the earth. Note, however, that a large $\atil$ results in a
392: breakdown of the simplified form given in eq.(\ref{survival_matter}) and
393: that the full expression (or, at least, a different
394: approximation) needs to be used.
395:
396: The exact expression (for an arbitrarily large but constant density) for the
397: survival probability is given by
398: \beq
399: \barr{rcl}
400: P_{\mu\mu} & = & \dis
401: 1 - 4 \, s^2_{23}\, \Bigg( c^2_{23} s^2_{\theta_m}
402: \sin^2 \left[ {\Delta_{31} L \over 4} \,
403: \left( 1 + \atil - {\cal D} \right) \right]
404: \\[1ex] && \dis \hspace*{4em} +
405: s^2_{23} s^2_{\theta_m} c^2_{\theta_m}
406: \sin^2 {\Delta_{31} L {\cal D} \over 2}
407: + c^2_{23} c^2_{\theta_m} \sin^2
408: \left[ {\Delta_{31} L \over 4} \,
409: \left( 1 + \atil + {\cal D} \right) \right] \Bigg) \ ,
410: \earr
411: \eeq
412: where
413: \beq \dis
414: {\cal D} \equiv \sqrt{1 + \atil^2 - 2 \, \atil \cos 2 \theta_{13}}
415: \ , \quad {\rm and} \quad
416: \theta_m \equiv \frac{1}{2} \tan^{-1}\frac{\sin2\theta_{13}}
417: {\cos2\theta_{13}-\atil} \ .
418: \eeq
419:
420: We now perform a double
421: expansion\footnote{This is similar in spirit to, though not the same as,
422: the expansion performed in Ref.\protect\cite{expand}.}
423: in $\ue$ as well as
424: $
425: \beta \equiv \frac{1}{2} - \Um^2
426: $
427: while allowing for any value for $\atil$. Since both the expansion
428: parameters are small, we may retain terms only upto, say, the fourth order
429: without any loss of accuracy. Then, for the survival probability, we have
430: \beq
431: \barr{rcl}
432: P_{\mu \mu} & = & \dis
433: \left(1 - 4 \, \beta^2 \right) \, \cP^2
434: \\[2ex]
435: & + & \dis
436: \frac{ \ue^2 }{{\left( 1 - \atil \right) }^2}
437: \,\Bigg[
438: \Bigg\{ \sP^2 - \sOm^2 - s_\Theta^2
439: - \atil \, \left( 1 - \atil \right) \,
440: \Psi\,
441: s_{2 \Psi} \Bigg\}
442: - 4 \beta \;
443: \Bigg\{ (1 - \atil)^2 \, \sP^2
444: - \sOm^2
445: \Bigg\}
446: \\[2ex]
447: & & \dis \hspace*{4em}
448: +
449: 4\, {\beta}^2
450: \, \Bigg\{ (1 - 4 \atil + 2 \atil^2) \,
451: \sP^2
452: +
453: \sOm^2 - s_\Theta^2
454: + \frac{\atil}{2} \left( 1 - \atil \right) \,\Psi\,
455: s_{2 \Psi} \Bigg\}
456: \Bigg]
457: \\[2ex]
458: & + & \dis
459: \frac{\ue^4}{
460: {\left( 1 - \atil \right) }^4}\,\Bigg[
461: \left(1 - 6\,\atil + {\atil}^4
462: - 4\,{\atil}^3 + 5 {\atil}^2\right) \,
463: \sP^2
464: +
465: (2\,\atil + {\atil}^2) \, \sOm^2
466: \\[2ex]
467: & & \dis \hspace*{6em}
468: - \Omega^2 \, (1 - \atil)^2 \, c_{2 \Psi}
469: + ( -1 + 6\,\atil - {\atil}^2) \,
470: s_\Theta^2
471: \\[2ex]
472: & & \dis \hspace*{6em}
473: +
474: \left( 1 - {\atil}^2 \right) \, \Omega\, s_{2 \Psi}
475: +
476: \left(1 - {\atil} \right) \, \Omega\,
477: \left( s_{2 \Omega} + 2 \, s_{2 \Theta} \right)
478: \Bigg]
479: \\[2ex]
480: & + & \dis {\cal O}(\ue^4 \beta, \ue^3 \beta^2, \ue^2 \beta^3, \ue \beta^4) \ ,
481: \earr
482: \label{survival_simp}
483: \eeq
484: where $s_\alpha \equiv \sin (\alpha)$, $c_\alpha \equiv \cos (\alpha)$
485: with
486: \beq \dis
487: \Omega \equiv \frac{\Delta_{31} \, L \, \atil}{2} = \frac{A \, L}{2}
488: = \atil \, \Psi
489: \quad
490: {\rm and}
491: \quad
492: \Theta \equiv \Omega - \Psi \ .
493: \eeq
494: %
495: With the survival probability for the antineutrinos being
496: obtained by the replacement $\atil \to - \atil$ in
497: eq.(\ref{survival_simp}), we have, for the probability asymmetry
498: \beq
499: \barr{rcl}
500: \dis
501: \frac{{- \left( 1 - {\atil}^2 \right) }^3 \, \Delta P}
502: {4\, \atil \, \ue^2\, \Um^2}
503: & = & \dis
504: ( 1 - {\atil}^2) \; \Bigg[4\,
505: \cP^2 \, \sOm^2
506: + ( 1 - {\atil}^2) \,
507: \Psi\,s_{2 \Psi} \,
508: - \frac{1 + \atil^2}{2 \, \atil} \,
509: s_{2 \Psi} \, s_{2 \Omega} \Bigg]
510: \\[2ex]
511: & + & \dis
512: 2\, \beta\,( 1 - {\atil}^2) \,
513: \Bigg[
514: -4\, \sP^2 \, \sOm^2
515: + ( 1 - {\atil}^2) \,\Psi\, s_{2 \Psi}
516: + \frac{ 1 + {\atil}^2}{2 \, \atil} \,s_{2 \Psi} \, s_{2 \Omega}
517: \Bigg]
518: \\[2ex]
519: & + & \dis
520: \ue^2\,\Bigg[ \frac{4}{1 - \atil^2}
521: \,
522: \left\{( 1 + 11\,{\atil}^2 + 2 \, {\atil}^4 ) \,
523: \sP^2 \, \cOm^2
524: + \atil^2 \, ( 3 - \atil^2 ) \, \cP^2 \, \sOm^2
525: \right\}
526: \\[2ex]
527: & & \dis \hspace*{3em}
528: + \Psi\,
529: \left\{
530: 2 \, {\atil}^2 \, (1 - {\atil}^2)^2\, \Psi \, c_{2 \Psi}
531: - (1 + 6 \, \atil^2 + \atil^4) \, \sP \, \cP \right\}
532: \\[2ex]
533: & & \dis \hspace*{3em}
534: + s_{2 \Psi} \,
535: \left\{ 2\, (1 + 3 \, \atil^2) \,
536: \Psi\, c_{2 \Omega}
537: - \,
538: \frac{ 1 - 16 \, \atil^2 - \atil^4}{4\, \atil} \, s_{2 \Omega}\right\}
539: \\[2ex]
540: & & \dis \hspace*{3em}
541: - (3 + \atil^2) \, \Omega \, s_{2 \Omega} \,
542: \left(3 - 4\, \sP^2 \right)
543: \Bigg]
544: \\[2ex]
545: & + & \dis {\cal O}(\ue^2 \beta, \ue \beta^2, \beta^3) \ .
546: \earr
547: \label{asym_simp}
548: \eeq
549: Note that
550: \begin{itemize}
551: \item The asymmetry $\Delta P$ continues to be proportional to each of
552: $\atil$, $\ue^2$ and $\Um^2$, but not to $(1 - 2 \Um^2)$.
553: \item $\lim_{\atil \to 0} \sOm / \atil = \Psi$ and hence the
554: right hand side of eq.(\ref{asym_simp}) is finite in this limit.
555: \item To ${\cal O}(\atil)$, the expressions for both $P_{\mu \mu}$ and
556: $\Delta P$ coincide with those of eq.(\ref{survival_matter}).
557: \item The strength of the ${\cal O}(1)$ term on the r.h.s. of
558: eq.(\ref{asym_simp})---which vanishes as $\atil^2$ for small $\atil$---
559: gives a measure of the violation of the $(1 - 2 \Um^2)$
560: proportionality of $\Delta P$.
561: \item While we have performed the expansion upto ${\cal O}(\ue^4)$, for all
562: practical purposes it suffices to consider only upto ${\cal O}(\ue^2)$.
563: \item The apparent singularities in
564: eqs.(\ref{survival_simp} \& \ref{asym_simp}) are not
565: real ones with
566: \[
567: \barr{rcl}
568: \dis \lim_{\atil \to 1} P_{\mu \mu} & = & \dis
569: \cP^2
570: + 4 \, {\beta}^2\, \sP^2
571: \\[2ex]
572: & + & \dis
573: \ue^2\,\Bigg[\Psi\, (s_{2 \Psi} - 2 \, \Psi \, \cP^2)
574: + 4 \, \beta\,\left( \Psi^2 - \sP^2 \right)
575: %\\[2ex]
576: % & & \dis \hspace*{4em}
577: + 4 \, {\beta}^2\,\left\{ 2 \, \left( 1 - \Psi^2 \right) \, \sP^2
578: - \Psi\, s_{2 \Psi} \right\} \Bigg]
579: \\[2ex]
580: & + & \dis
581: \ue^4\,\Psi^2 \, \left[ \sP^2 +
582: \frac{\Psi^2}{3} \, (3 + 2 \, \sP^2 )
583: - \frac{4}{3} \, \Psi\,s_{2 \Psi} \right] \ ,
584: \earr
585: \]
586: and
587: \[
588: \barr{rcl}
589: %%%%%%%%%%
590: \dis
591: \lim_{\atil \to 1} \frac{- \Delta P}{\ue^2\, \Um^2}
592: & = & \dis
593: 2 \, \Bigg[ 2 \, \Psi^2\, \cP^2
594: - \Psi\,\sP\,\cP -
595: \frac{s_{2 \Psi}^2}{4} \Bigg]
596: % \\[2ex]
597: %& + & \dis
598: +
599: \beta \,\Bigg[ - \Psi\,s_{2 \Psi}
600: + (5 - 8 \, \Psi^2) \, \sP^2
601: + s_{2 \Psi}^2 \Bigg]
602: \\[2ex]
603: & + & \dis
604: \ue^2\,\Bigg[ -s_{2 \Psi}^2
605: - \Psi^2\, \left(\frac{3}{2} + \sP^2 \right) -
606: \frac{2 \, \Psi^4}{3} (3 + 2 \, \cOm^2)
607: \\[2ex]
608: & & \dis \hspace*{4em}
609: + \frac{4\,\Psi^3}{3}\, s_{2 \Psi} +
610: \Psi\,s_{2 \Psi} \,
611: \left( \frac{5}{8} - \sP^2 \right)
612: \Bigg] \ .
613: \earr
614: \]
615:
616:
617: \end{itemize}
618:
619:
620:
621: \section{Can the NuMI--MINOS Combine See This?}
622: \label{sec:minos}
623: With the Fermilab Main Injector serving as a neutrino source,
624: the {\sc minos} detector~\cite{minos} at the Soudan Mine offers a
625: baseline of 732 km and hence a possible setup for making such a
626: measurement. This was investigated by Choubey and Roy in
627: Ref.\cite{choubey_roy}. Working with a
628: $\nu_\mu$ exposure of $16\times 10^{20}$ primary protons on target, they
629: conclude that a
630: significant rate asymmetry ($\Delta N$) can be observed for moderate values
631: of $(1 - 2 |U_{\mu 3}|^2)$ and $|U_{e 3}|^2$.
632:
633: However, while their idea was an interesting one
634: and the formalism essentially correct, a reestimation of the rate
635: asymmetry using the
636: same set of inputs reveals some discrepancies. While we agree with
637: Ref.\cite{choubey_roy} on the magnitude of $\Delta N$ for all the
638: different cases of inputs parameters, we completely disagree on the
639: issue of the error bars. Even on using an optimistic indicator of the
640: background fluctuation, namely $\sqrt{N_\mu + N_{\bar \mu}}$, we find
641: that, for the abovementioned luminosity, the statistical error bars
642: are large enough for each of the rate asymetries in
643: Ref.\cite{choubey_roy} to be consistent with zero.
644:
645: %
646: \begin{figure}[htb]
647: \vspace*{-20ex}
648: \centerline{\epsfig{figure=lumin_scale.ps,height=4.0in,width=3.5in,angle=0}}
649: \vspace*{-5ex}
650: \caption{\em The minimal ratio by which the {\sc minos} $\nu_\mu$ exposure
651: ($16\times 10^{20}$ primary protons on target) needs to be scaled up
652: to obtain $1\sigma$ effects for each individual energy bin.
653: The curves correspond to the four parameter points of
654: Ref.\protect\cite{choubey_roy}. Note that only the statistical
655: fluctuations have been included in this analysis.}
656: \label{figure:flux_scaling}
657: \end{figure}
658: %
659: %A closer look to
660: %the size of their error-bars reveals that they are equal to
661: %$\sqrt{N_\mu - N_{\bar \mu}}$ in magnitude. This simply implies an event rate
662: %of $N_{\bar \mu}$ consistent with zero. This contradicts their
663: %earlier result of small $\Delta P_{\mu \mu}$. {\bf I am not sure whether
664: %we can add these last two sentences }
665:
666: Thus, to obtain a
667: result of any statistical significance would need the flux (or, equivalently,
668: the detector size) to be
669: increased manifold (see Fig.~\ref{figure:flux_scaling}).
670: Such a large scaling,
671: unfortunately, is not feasible with the present experimental facility.
672: Note, in addition, that
673: this projection altogether neglects the systematic uncertainties, and
674: consequently is already too optimistic.
675: In reality, several uncertainties are involved
676: in the translation between the visible energy observed at a detector
677: and the energy of the neutrino to which the event is ascribed. These
678: include the uncertainties in the assumed final state multiplicities, the
679: scattering or absorption of the secondary particles etc. In a
680: recent paper~\cite{minerva},
681: the MINER$\nu$A collaboration, estimates the size of the corresponding
682: systematic effects at {\sc minos}-like experiments could be
683: comparable to, or even dominate, the statistical errors.
684:
685: \section{The case for a very long baseline}
686: \label{sec:longbaseline}
687:
688: Having established that, contrary to the claims of Ref.\cite{choubey_roy},
689: {\sc minos}, in its current incarnation,
690: is not sensitive to possible deviations of $\Um$
691: from maximality even for optimistic choices of parameters, it becomes
692: interesting to investigate if any of the other proposed long-baseline
693: experiments would do the job. And, in a similar vein, if even the
694: luminosity scaling as described in the previous section would prove adequate.
695: We start by investigating the second problem.
696:
697: As has already been emphasised above,
698: it is not only $\Delta P$ that is important,
699: but also the average survival probability
700: $P_{\rm av} \equiv (P_{\mu \mu} + P_{\bar\mu \bar\mu}) / 2$ for the latter
701: controls the total number of events and hence the size of the statistical
702: error (significance
703: $\propto \Delta P_{\mu \mu} / \sqrt{2 P_{\rm av}}$). In
704: Fig.\ref{Fig:prob_732}, we present both these quantities as function of
705: neutrino energy for a baseline of 732 km. The two choices of
706: $(U_{\mu 3}|^2, |U_{e 3}|^2)$ correspond to the most (and least) optimistic
707: sets of parameters used in Ref.\cite{choubey_roy}.
708: %%%%%%%%%%%%%%%%%%%%%
709: \begin{figure}[hbt]
710: \vspace*{-0.4cm}
711: \centerline{
712: \hspace*{3em}
713: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{732_p.eps}
714: \hspace*{-7em}
715: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{732_delp.eps}
716: }
717:
718: \vspace*{-0.8cm}
719: \caption{\em {\em (a)} The neutrino survival probability
720: as a function of neutrino energy for a baseline $L = 732$ km.
721: The curves correspond to different combinations of
722: $\delta m^2_{13}$ (in $\ev^2$), $|U_{\mu 3}|^2$ and $|U_{e 3}|^2$.
723: {\em (b)} The corresponding difference in the $\nu_\mu$ and
724: $\bar\nu_\mu$ survival probablities.}
725: \label{Fig:prob_732}
726: \end{figure}
727: %%%%%%%%%%%%%%%
728:
729: It is abundantly clear that, for such a baseline, and for
730: realistic choice of parameters,
731: $\Delta P$ is very small, especially at larger neutrino energies. This
732: was to be expected as a baseline of 732 km implies that the neutrino
733: beam neither travels through a very dense matter core nor does it
734: travel through a ``dilute'' segment long enough for the matter effect
735: to build up sufficiently. Clearly, it is dangerous to claim
736: a signal based on $\Delta P \lsim 10^{-4}$, for such an act presupposes
737: a very accurate knowledge of both the (anti-)neutrino cross sections
738: as well as the density profile within the earth. Thus, one would have to
739: concentrate only on neutrinos with relatively small energies
740: ($E_\nu \lsim 2 \gev$). Apart from the fact that the
741: asymmetry is not too large even in this
742: range\footnote{In fact, it stands to be swamped by the systematic
743: errors unless further experiments are performed~\protect\cite{minerva}.},
744: such a measurement also necessitates both a very good energy
745: resolution as well as a low energy threshold.
746: Furthermore, for such small energies, neutrino cross-section
747: scales as $E_\nu^2$ and thus restricting ourselves to a small
748: window is tantamount to rejecting a very large fraction of the events.
749: %%%%%%%%%%%%%%%%%%%%%
750: \begin{figure}[hbt]
751: \vspace*{-0.3cm}
752: \centerline{
753: \hspace*{3em}
754: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{4865_p.eps}
755: \hspace*{-7em}
756: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{4865_delp.eps}
757: }
758:
759: \centerline{
760: \hspace*{3em}
761: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{10480_p.eps}
762: \hspace*{-7em}
763: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{10480_delp.eps}
764: }
765:
766: \vspace*{-0.8cm}
767: \caption{\em As in Fig.\protect\ref{Fig:prob_732}, but for baselines
768: $L = 4865$ km (upper panels) and $L = 10480$ km (lower panels) instead.
769: }
770: \label{Fig:prob_4865_10480}
771: \end{figure}
772: %%%%%%%%%%%%%%%
773:
774: The situation, vis. a vis. the size of $\Delta P$, seems to improve
775: dramatically with an increase in the baseline.
776: For definiteness, we consider two particular values, namely
777: $L = 4865$ km (the distance between the Japan Hadron Facility and the
778: proposed Indian Neutrino Observatory site at Rammam~\cite{ino}) and
779: $L = 10480$ km (the distance between Fermilab and INO).
780: Since the matter term can no longer be treated as a perturbation,
781: we desist from using the simplified form of
782: eq.(\protect\ref{survival_matter}) and use the full expression for neutrino
783: propagation in matter with the density
784: profile being given by the Preliminary Reference Earth Model\cite{prem}.
785: As Fig.\ref{Fig:prob_4865_10480} amply demonstrates, for each of these
786: values, $\Delta P$ is large over a wide range of $E_\nu$. In fact, for
787: the larger of the two baselines $|\Delta P|$ can be as large as 0.8 for
788: certain $E_\nu$ bins. Furthermore, unlike in the case of
789: the 732 km baseline where $\Delta P$ was nearly monotonic in $E_\nu$,
790: significant structure, including sign reversal, is shown. As can be easily
791: appreciated, the latter feature has the potential of serving as a key
792: experimental signature.
793:
794: A further point of interest relates to the position of the maxima
795: of the survival probabilities as well as the asymmetry. Clearly, a
796: high resolution measurement of the same could, in principle, be
797: used to determine $\delta m_{13}^2$.
798:
799: Before we interpret Fig.~\ref{Fig:prob_4865_10480} to imply that
800: the asymmetry measurement in very long baseline experiments would
801: be a sensitive probe of the deviation of $|U_{\mu 3}|$ from maximality,
802: it should be realised that the latter is not the only source
803: for a non-zero asymmetry. In fact, as has already been hinted at
804: above, the higher order terms in $\atil$ that were dropped
805: while deriving eq.(\ref{survival_matter}) are {\em not} proportional to
806: $(1 - 2 |U_{\mu 3}|^2)$. This can also be seen explicitly from
807: eqs.(\ref{survival_simp} \& \ref{asym_simp}) obtained
808: assuming a constant density profile for earth.
809: The numerical importance of such contributions
810: naturally increases with the baseline,
811: and has been explicitly displayed in
812: Fig.~\ref{Fig:comp_4865_10480}. In fact, for the larger baseline, it is
813: quite apparent that the bulk of the effect is due to such ``higher-order''
814: terms. Thus, it may be concluded
815: that while a baseline of $\approx 5000$ km may allow for a determination of
816: $(1 - 2 |U_{\mu 3}|^2)$ through measurements of the rate asymmetry,
817: for baselines much longer than this the sensitivity reduces quite sharply,
818: and a measurement of the maximality of $\Um$ is not very
819: straightforward for ultra-long baselines.
820:
821:
822: %%%%%%%%%%%%%%%%
823: \begin{figure}[hbt]
824: \vspace*{-0.3cm}
825: \centerline{
826: \hspace*{3em}
827: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{4865_comp.ps}
828: \hspace*{-7em}
829: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{10480_comp.ps}
830: }
831: \vspace*{-0.8cm}
832: \caption{\em The normalized probability asymmetry for a baseline of
833: {\em (a)} 4865 km and {\em (b)} 10480 km. In each case, the solid
834: (dashed) line refers to $|U_{\mu 3}|^2 = 0.36$ ($0.50$).
835: Note that, according to the leading expression of
836: eq.(\protect\ref{survival_matter}), the dashed curves should have
837: coincided with the abscissa.
838: }
839: \label{Fig:comp_4865_10480}
840: \end{figure}
841: %%%%%%%%%%%%%%%
842:
843: On the other hand, the asymmetry must remain
844: proportional to $\ue^2$ even on the inclusion of $\atil^n$ terms,
845: for it is
846: only through the interaction with the $\nu_e$ that matter effect truly
847: enters $\nu_\mu$--$\nu_\tau$ oscillations. Thus, a baseline of
848: 10480 km may still serve the purpose of affording a good measurement of
849: $|U_{e 3}|$. In fact, since $U_{\mu 3}$ and $U_{e 3}$ are inextricably
850: linked even for the case of the 4865 km baseline, such an independent
851: measurement of the latter element has its own importance.
852:
853:
854: \subsection{$\nu_e \to \nu_\mu$ oscillations}
855:
856: With the matter effect playing such an important role in the survival
857: probability for the muon (anti-)neutrino, it becomes interesting to
858: consider its effect on $P(\nu_e \to \nu_\mu)$. For, a source that
859: produces $\nu_\mu$ ($\bar\nu_\mu$) copiously, whether it be a
860: beam dump or a muon storage ring, would also produce electron
861: (anti-)neutrinos.
862:
863: %%%%%%%%%%%%%%%%
864: \begin{figure}[hbt]
865: \vspace*{-0.3cm}
866: \centerline{
867: \hspace*{3em}
868: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{emu_4865_p.ps}
869: \hspace*{-7em}
870: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{emu_4865_pbar.ps}
871: }
872:
873: \centerline{
874: \hspace*{3em}
875: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{emu_10480_p.ps}
876: \hspace*{-7em}
877: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{emu_10480_pbar.ps}
878: }
879: \vspace*{-0.8cm}
880: \caption{\em The energy dependence of
881: the transition probability for $\nu_e \to \nu_\mu$ (left)
882: and $\bar\nu_e \to \bar\nu_\mu$ (right) on passage through
883: matter. The upper and lower panels correspond to baselines
884: of 4865 and 10480 km respectively.
885: }
886: \label{Fig:emu_prob}
887: \end{figure}
888: %%%%%%%%%%%%%%%
889:
890: In Fig.~\ref{Fig:emu_prob}, we present the probabilities for such transitions
891: as a function of (anti-)neutrino energy. Several observations are in
892: order:
893: \begin{itemize}
894: \item The transition probability $P(\nu_e \to \nu_\mu) $ can be quite sizable
895: for parameter values that are still allowed by experimental data.
896: \item For $E \gsim 3 \gev$, we see that
897: $P(\bar\nu_e \to \bar\nu_\mu) \lsim P(\nu_e \to \nu_\mu)$ (except,
898: of course, near the node at $\sim 8 \gev$ for the 10480 km case). Coupled
899: with the smaller detection efficiency for anti-neutrinos, this indicates
900: that it would be more profitable to work with $\nu_e$ than $\bar \nu_e$.
901: \item The transition probabilities have non-trivial dependence on both
902: $U_{e 3}$ and $U_{\mu 3}$. Unlike in the case of the aforementioned
903: asymmetry, these are {\em not} in general proportional to $|U_{e 3}|^2$.
904: \item The measurement of this effect would thus lead to a constraint in the
905: $U_{e 3}$--$U_{\mu 3}$ plane independent of that drawn from the asymmetry
906: measurement.
907: \end{itemize}
908:
909: \section{Event rate calculation from a storage ring neutrino source}
910: \label{sec:storage_ring}
911: As we have already mentioned in the introductory section, the
912: neutrino flux from a muon storage ring can be calculated very precisely.
913: Starting with a $\mu^-$ beam, the number of $\mu ^-$ events in a
914: far detector can be obtained by folding this flux with the
915: survival probability $P_{\mu\mu} (L,E)$ and the charged current
916: cross-section:
917: \begin{equation}
918: N_{\mu } = N_{n} \int \sigma(\nu_\mu + N
919: \rightarrow \mu^- + X)\; \frac{dN_{\nu}}{dE_{\nu_\mu}} \;P_{\mu\mu} (L,E)
920: \; dE_{\nu_\mu}
921: \label{no_of_evts}
922: \end{equation}
923: where $N_n$ is the total number of nucleons
924: present in the fixed target. An analogous expression obtains for
925: $\mu^+$ events as well. Note though that the $\bar \nu_e$ from
926: muon decay could also oscillate into $\bar \nu_\mu$ while traversing
927: through the earth and result in muonic events. However,
928: they result in {\em wrong sign} muons and thus
929: can be easily distinguished in a magnetized detector. We shall revert
930: back to them at a later stage.
931:
932: To be specific, we shall consider a storage ring with 20 GeV
933: muons and a 50 kT Iron calorimeter detector one such as the
934: proposed {\sc monolith}~\cite{mono}
935: or the {\sc ical/ino} experiment~\cite{ino}. The projected
936: energy threshold for muon detection is about 2~GeV and the resolution
937: is expected to be better than 0.5 GeV over the entire range.
938:
939:
940: Before delving into the actual event profile, several points are in order.
941: \begin{itemize}
942: \item As has already been demonstrated in the previous section, the difference
943: in the survival probalities ($\Delta P_{\mu\mu}$) typically grows with
944: the baseline. Thus the experimental sensitivity could be expected
945: to increase with the baseline.
946: (This is modulo the fact that the sensitivity of the asymmetry
947: to $(1 - 2 \, \Um^2)$ reduces
948: for very large baselines, while the sensitivity to $\ue^2$
949: is retained.)
950:
951: \item On the other hand, for a fixed detector size,
952: the effective solid
953: angle subtended by the detector to the source decreases with
954: increasing baseline.
955: Consequently the neutrino flux goes down
956: quadratically with the baseline, thereby resulting in a smaller number
957: of events and hence larger statistical errors.
958: This, for example, can be read off from
959: Fig.~\ref{Fig:noosc}, wherein we have displayed the number of events
960: expected if the survival probability were to be unity.
961: %%%%%%%%%%%%%%%%
962: \begin{figure}[hbt]
963: \vspace*{-0.3cm}
964: \centerline{
965: \hspace*{3em}
966: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{noosc_4865.eps}
967: \hspace*{-5em}
968: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{noosc_10480.eps}
969: }
970:
971: \vspace*{-0.0cm}
972: \caption{\em The number of events expected as a function of the
973: (anti-)neutrino energy if the survival probability were to be unity.
974: The source is a 20 GeV muon storage ring with $10^{20}$ (anti-)muons
975: decaying while the detector is a 50 kT iron calorimeter with
976: a energy threshold of 2 GeV for the $\mu^\pm$~\protect\cite{ino}.
977: The left (right) panel corresponds to a baseline of 4865 (10480) km.
978: Also shown are the events generated by the electron (anti-)neutrino
979: assuming a transition probability of unity.
980: }
981: \label{Fig:noosc}
982: \end{figure}
983: %%%%%%%%%%%%%
984:
985:
986: \item One obvious way to circumvent the latter problem is to start with higher
987: luminosity. For our numerical results we then adopt the
988: `higher luminosity' option for the muon beam, namely
989: $10^{20}$ muon decays per year.
990:
991: \item An alternative way to the same end would have been to start
992: with a higher energy for the muon beam. For example, the neutrino
993: beam from a 50 GeV storage ring is collimated sufficiently enough
994: to get a similar significance with even the `low luminosity'
995: option for the storage ring ($10^{19}$ muon decays per year). However,
996: we shall desist from using this option as such a machine is likely
997: only as a second generation facility.
998:
999: \item As Fig. \ref{Fig:prob_4865_10480} amply demonstrates, for longer
1000: baselines such as the one we are considering, the survival probabilities
1001: are very sensitive functions of the (anti-)neutrino energy, especially
1002: for low energies. Thus, extracting any information from the low energy
1003: tail of the detector muon spectrum would necessitate very good energy
1004: resolution. However, with the neutrino-nucleon cross sections being
1005: small for such $E_\nu$, these neutrinos have only a relatively
1006: small contribution to make to the total number of muon events.
1007: \end{itemize}
1008:
1009:
1010: Keeping in mind that $\nu -N$ cross-section is nearly
1011: double\footnote{For the total cross section, this ratio has
1012: only a mild dependence on the incident neutrino
1013: energy, and for the energy range we are interested in has a value
1014: close to 2.18. However, on imposition of an
1015: energy threshold $E(\mu^\pm) > 2 \gev$
1016: that such a detector is expected to have, the ratio is a more
1017: sensitive function of the (anti-)neutrino energy
1018: (see Fig.~\protect\ref{Fig:noosc}).
1019: In our numerical analysis, we explicitly account for this
1020: factor and the consequent effect on the error bars.}
1021: that of $\bar \nu -N$ we argue that the detector should be exposed to
1022: a $\bar\nu_\mu$ beam for double the time that it is exposed to a
1023: $\nu_\mu$ beam. To be specific, we consider a 50 kT-year exposure
1024: to a $\nu_\mu$ beam and a 100 kT-year exposure to $\bar \nu_\mu$ beam,
1025: or in other words, a total exposure of 3 years for the detector
1026: configuration~\cite{ino} under consideration.
1027:
1028: \subsection{The rate asymmetry for $\nu_\mu (\bar \nu_\mu)$ initiated events}
1029:
1030: Armed with the above, we can now calculate the difference in
1031: $\mu^-$ and $\mu^+$ event numbers. Defining an asymmetry
1032: of the form
1033: \beq
1034: A_N \equiv \frac{N_- - N_+}{N_- + N_+} \ ,
1035: \label{rate_asym}
1036: \eeq
1037: we present, in Fig.~\ref{Fig:diff_4865}, the
1038: same for a baseline of 4865 km. The values for ($|U_{\mu 3}|^2, |U_{e 3}|^2$)
1039: correspond to the two extreme cases of Ref.\cite{choubey_roy}. Also plotted
1040: are the analogous expectations for $|U_{\mu 3}|^2 = 0.5$, a case that, to
1041: the lowest order, would have been associated with vanishing $A_N$.
1042: %%%%%%%%%%%%%%%%%%%%%
1043: \begin{figure}[htb]
1044: \vspace*{-0.2cm}
1045: \centerline{
1046: \hspace*{3em}
1047: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.002_0.36_0.07.eps}
1048: \hspace*{-7em}
1049: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.002_0.40_0.05.eps}
1050: }
1051:
1052: \vspace*{-0.4cm}
1053: \centerline{
1054: \hspace*{3em}
1055: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.0027_0.36_0.07.eps}
1056: \hspace*{-7em}
1057: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.0027_0.40_0.05.eps}
1058: }
1059:
1060: \vspace*{-0.8cm}
1061: \caption{\em The difference between $\nu_\mu$ and $\bar \nu_\mu$
1062: charged-current events
1063: for a baseline $L = 4865$ km with an integrated luminosity of $10^{20}$ muons
1064: and double the number of antimuons. The detector is a 50 kT iron calorimeter
1065: with a muon energy threshold of $2 \gev$~\protect\cite{ino}.
1066: Each plot corresponds to a different
1067: combination of ($\delta m^2_{13}, |U_{\mu 3}|^2, |U_{e 3}|^2$).
1068: The other parameters are:
1069: $\delta m^2_{12} = 7 \times 10^{-5} \ev^2$ and $\sin^2 \theta_{12} = 0.3$.
1070: Also shown are the $1\sigma$ error bars.}
1071: \label{Fig:diff_4865}
1072: \end{figure}
1073: %%%%%%%%%%%%
1074: We have binned the expected data,
1075: keeping under consideration both the expected energy
1076: resolution~\cite{ino} as well
1077: as the number of events in a particular bin. Also shown are the
1078: statistical $1 \sigma$ error bars. Several features are easily
1079: discernible:
1080: \begin{itemize}
1081: \item In each of the cases displayed, the rate asymmetry is statistically
1082: significant in more than one bin. Together, they lead to a very
1083: considerable shift in the $\chi^2$. This amply demonstrates the
1084: sensitivity of such an experiment to a departure from maximality
1085: of $U_{\mu 3}$.
1086: \item As expected, the signal is less pronounced for the
1087: case ($|U_{\mu 3}|^2, |U_{e 3}|^2) = (0.40, 0.05)$ than
1088: for $(0.36, 0.07)$. This is just a reflection of the fact---see
1089: eq.(\ref{survival_matter})---that the rate asymmetry is proportional to
1090: $ |U_{e 3}|^2$. While the proportionality to
1091: $( 1 - 2 \, |U_{\mu 3}|^2 )$ is no longer an accurate statement, it,
1092: nevertheless still encapsulates a large measure of truth.
1093:
1094: \item In the energy range of interest, the binwise asymmetry clearly
1095: shows an oscillatory behaviour, a feature that would have been absent
1096: for a similar detector had the baseline been shorter than $\sim 1000$ km.
1097:
1098: \item The functional dependence of $\Delta N(E_\nu)$ exhibits a
1099: discernible dependence on the mixing angles
1100: ($|U_{\mu 3}|^2, |U_{e 3}|^2$), and could, in principle, be used
1101: to distinguish between values for these parameters. However, to
1102: obtain a significant resolution in the parameter space, a much larger
1103: event count would be necessary.
1104:
1105: \item The dependence on the value of
1106: $\delta m^2_{13}$\ , on the other hand, is much more pronounced. The
1107: shift in $\Delta P$ that we encountered in Fig.\ref{Fig:prob_4865_10480}
1108: is quite well reflected by a shift in $A_N$, even after the convolution with
1109: the muon spectrum and the energy-dependent cross sections as well as
1110: the finite resolution effects. Thus, our previous comment about using
1111: this measurement for a determination of $\delta m^2_{13}$ is substantiated.
1112:
1113: \item The energy dependence is quite different for the two baselines
1114: considered and can be used to advantage. For example,
1115: if storage rings come up at both the JHF~\cite{jhf}
1116: and Fermilab, a detector such as {\sc ical/ino}~\cite{ino} could
1117: use beams from both to distinguish more efficiently between the
1118: possible parameter sets.
1119: \end{itemize}
1120:
1121: \subsection{Wrong sign muons}
1122:
1123: We now turn to a discussion $\nu_e$ to $\nu_\mu$ conversion (as well
1124: the conjugate process) and their detection in the iron calorimeter
1125: detector under consideration. Such events are obviously characterised
1126: by the appearance of a muon with a charge opposite to that of the
1127: decaying particle in the storage ring. Since the detector is to be
1128: magnetized one, charge measurement is relatively straightforward thus
1129: rendering these events easily distinguishable. Efficacy of `wrong-sign-muon'
1130: events in measuring the angle $\theta_{13}$ has already been advocated in
1131: Ref.\cite{barger_matter} in the context of muon-storage ring neutrino factory
1132: experiments. Here we want to emphasise that how an independent
1133: measurement of `wrong-sign-muon' events can also supplement the
1134: more accurate measuremeant of $\theta_{23}$.
1135: We have already seen, vide Fig.\ref{Fig:emu_prob}, that
1136: the conversion probability for $\nu_e$ is typically larger than that
1137: for $\bar\nu_e$, a consequence of the sign of $\delta m_{12}^2$. This
1138: fact, alongwith the smaller detection efficiency for $\bar \nu_\mu$,
1139: implies that it is wiser to concentrate on the $\mu^+ \to \nu_e \to
1140: \nu_\mu$ chain rather than attempt to measure the conjugate process
1141: as well. And since we have already advocated twice as long an
1142: exposure to $\mu^+$ decays as compared to $\mu^-$ decays, this works
1143: in our favour too.
1144:
1145:
1146: \begin{figure}[htb]
1147: \vspace*{-0.2cm}
1148: \centerline{
1149: \hspace*{3em}
1150: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.002_emu.eps}
1151: \hspace*{-7em}
1152: \epsfxsize = 10cm \epsfysize = 7cm \epsfbox{0.0027_emu.eps}
1153: }
1154: \vspace*{-0.8cm}
1155: \caption{\em The number of $\mu^-$ events in a 50 kT iron calorimeter
1156: exposed to a total of $2 \times 10^{20}$ $\mu^+$ decays in a
1157: 20 GeV storage ring at a distance of 4865 km. The histograms correspond
1158: to different values of $(|U_{\mu 3}|^2, |U_{e 3}|^2)$ with the left
1159: (right) panels referring to $\delta m_{13}^2 = 0.002 \, (0.0027) \ev^2$.
1160: }
1161: \label{Fig:e_mu_events}
1162: \end{figure}
1163:
1164: In Fig.\ref{Fig:e_mu_events}, we present the number of $\mu^-$ events in the
1165: detector when exposed to $\mu^+$ decays. As before, the detector is supposed
1166: to be a 50 kT one, with a total exposure of $2 \times 10^{20}$ decays. Clearly,
1167: the number of events grows with $|U_{e3}|^2$. While the exact dependence
1168: is not linear (as the naive expectation would be), explicit
1169: calculations with other values of $|U_{e3}|$ shows that
1170: it is not very far from being linear. This is also the case for
1171: the larger baseline of 10480 km (which we do not exhibit here).
1172: The dependence on $|U_{\mu 3}|$ is
1173: more complicated though. As can be expected, this depends more crucially
1174: on the size of the matter effect and hence on the exact baseline
1175: (see, for example, Fig.~\ref{Fig:emu_prob}). However, given the considerably
1176: large number of such events, it is quite apparent that this measurement
1177: is likely to be a sensitive probe in the ($|U_{\mu 3}|, |U_{e 3}|$) plane.
1178: Unfortunately though, the event distribution is not very sensitive to
1179: $\delta m^2_{13}$.
1180:
1181: The main motivation for the above excercise is to show that the
1182: measurement of any of the small parameters ($\ue$ or $\beta$) at long
1183: baseline experiments cannot be done independent of each
1184: other. Although experiments with relatively small baselines (such as
1185: NuMi--{\sc minos}) seem to be quite sensitive to the parameter
1186: $\beta$, this optimism is misplaced as explained in
1187: Sections~\ref{sec:minos} \& \ref{sec:longbaseline}. On the other
1188: hand, moderate baseline experiments do show increased sensitivity to
1189: $\beta$, but only at the cost of having a more complicated dependence
1190: on $\beta$ and $\ue$ (see eq.\ref{survival_simp} and
1191: Fig. \ref{Fig:comp_4865_10480}). In either case, a measurement of
1192: $\beta$ from $\nu_\mu$-survival probability asymmetry needs an accurate
1193: knowledge of $\ue$. It is here that the wrong-sign muon rates (which
1194: is more sensitive to $U_{e3}$) have the most important role to
1195: play. Of course, a precise knowledge of $\delta m^2_{13}$ is important for
1196: the extraction of $\ue$ (although the dependence on $\delta m^2_{13}$ is
1197: not as pronounced for the wrong-sign muon events as for the right-sign
1198: ones). However, it is expected that the error-bars on $\delta m^2_{13}$
1199: would be significantly reduced before the era of such neutrino factory
1200: experiments. Then, in principle, $A_N$ and wrong-sign muon rates
1201: could be used simultaneously in a statistical analysis (in a spirit
1202: similar to that followed in Ref. \cite{maoki}) to extract $\ue$ and
1203: $\beta$ to much greater accuracy than possible today.
1204:
1205: \section{Conclusions}
1206: \label{sec:concl}
1207: Analyses of matter effects in $\nu_\mu$ oscillations have largely
1208: concentrated on transition probabilities. Revisting the problem,
1209: we demonstrate that the survival probabilities too are very sensitive
1210: to matter effects and can be used profitably in
1211: determination of several crucial parameters in the neutrino sector.
1212:
1213: To start with, we derive a set of approximate analytical
1214: expressions for the $\nu_\mu$ survival probability in presence
1215: an arbitrarily large matter density. While similar results had
1216: already been reported for small matter terms in the effective
1217: Hamiltonian, our expressions have a much wider range of validity.
1218:
1219: Since we are interested in the survival probability for $\nu_\mu$,
1220: a variable of interest can be constructed simply from the number
1221: of charged current (CC) events in the detector. Starting
1222: with $\nu_\mu$ and $\bar \nu_\mu$ beams,
1223: $A_N \equiv (N_- - N_+) / (N_- + N_+)$ describes
1224: an asymmetry between the number of `same-sign'
1225: $\mu^-$ and $\mu^+$ events generated in the detector after
1226: traversing a given baseline. We show by explicit event rate
1227: calculation that this asymmetry is a good measure of the matter
1228: effects felt by the neutrinos while propagating through the earth.
1229: However, contrary to claims in the literature~\cite{choubey_roy},
1230: the size of the asymmetry for the Fermilab-{\sc minos} combine
1231: is much smaller than the experimental sensitivity.
1232:
1233: We suggest, therefore, that a future long-baseline experiment could
1234: explore this effect. As a prototype experiment, we consider a neutrino
1235: factory (a muon storage with muon beams of energy 20 GeV) and the
1236: proposed 50 kT iron calorimeter detector~\cite{mono,ino} with a
1237: capability of muon charge determination. Using realistic experimental
1238: resolutions, we demonstrate that such a combination is sufficient to
1239: establish the aforementioned asymmetry with a very large statistical
1240: significance. However, unlike in the case of relatively short-baseline
1241: Fermilab--{\sc minos} combine, the asymmetry for a long baseline is no
1242: longer proportional to the deviation of $\Um^2$ from maximality. Even
1243: so, the sensitivity to $(1 - 2 \, \Um^2)$ remains quite pronounced for
1244: baselines of upto about 6500 Km. Using the aforementioned approximate
1245: expressions for the survival probabilities, such data can thus be used
1246: to determine $(1 - 2 \, \Um^2)$. While the sensitivity to this
1247: combination decreases for much longer baselines, asymmetry data for
1248: such baselines can still be used for precise determination of $\ue$.
1249:
1250: The latter exercise is also shown to be aided by the
1251: measurement of `wrong-sign-muon' events generated through
1252: $\nu_e \to \nu_\mu$ oscillations.
1253: Estimating the rate of such events for the
1254: same experimental setup, we demonstrate that this
1255: could be used in conjunction with the asymmetry measurement
1256: to lead to a good simultaneous determination of $\Um$ and $\ue$.
1257:
1258:
1259:
1260:
1261: \section*{Acknowledgements}
1262: The authors thank the Theory Division, CERN for hospitality during the
1263: initiation phase of the project. DC thanks INFN, Sezione di Roma, for
1264: hospitality during which a major part of this work has been done. DC
1265: acknowledges financial assistance under the {\em Swarnajayanti
1266: Fellowship} grant from the Department of Science and Technology,
1267: India. AD is partially supported by the RTN European Programme
1268: MRTN-CT-2004-503369 (Quest for Unification).
1269:
1270:
1271:
1272:
1273: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1274: % %%% Define Journal macros %
1275: % \newcommand{\araa}[3]{{\em Annu. Rev. Astron. Astrophys.\/}
1276: % {\bf#1} (19#3) #2} %
1277: % \newcommand{\ptp}[3]{{\em Prog. Theoret. Phys. (Kyoto)\/}
1278: % {\if#1} (19#3) #2} %
1279: \newcommand{\plb}[3]{{Phys. Lett.} {\bf B#1}, #2 (#3)} %
1280: \newcommand{\prl}[3]{Phys. Rev. Lett. {\bf #1}, #2 (#3) } %
1281: \newcommand{\rmp}[3]{Rev. Mod. Phys. {\bf #1} #2 (#3)} %
1282: % \newcommand{\prep}[3]{Phys. Rep. {\bf #1} (#3) #2} %
1283: \newcommand{\prep}[3]{Phys. Rep. {\bf #1} #2 (#3)} %
1284: \newcommand{\rpp}[3]{Rep. Prog. Phys. {\bf #1} #2 (#3)} %
1285: \newcommand{\prd}[3]{Phys. Rev. {\bf D#1}, #2 (#3)} %
1286: % \newcommand{\prev}[3]{{\em Phys. Rev.\/} {\bf #1} (#3) #2} %
1287: \newcommand{\np}[3]{Nucl. Phys. {\bf B#1} #2 (#3)} %
1288: \newcommand{\npbps}[3]{Nucl. Phys. B (Proc. Suppl.)
1289: {\bf #1} #2 (#3)} %
1290: \newcommand{\sci}[3]{Science {\bf #1} #2 (#3)} %
1291: \newcommand{\zp}[3]{Z.~Phys. C{\bf#1} #2 (#3)}
1292: \newcommand{\epj}[3]{Eur. Phys. J. {\bf C#1} #2 (#3)}
1293: \newcommand{\mpla}[3]{Mod. Phys. Lett. {\bf A#1} #2 (#3)} %
1294: % \newcommand{\ajp}[3]{{\em Am. J. Phys.\/} {\bf #1} (#3) #2} %
1295: % \newcommand{\ap}[3]{{\em Ann. Phys. (NY)\/} {\bf #1} (#3) #2} %
1296: \newcommand{\apj}[3]{ Astrophys. J.\/ {\bf #1} #2 (#3)} %
1297: % \newcommand{\apjs}[3]{{\em Astrophys. J. Suppl.\/}
1298: % {\bf #1} (#3) #2} %
1299: \newcommand{\jhep}[2]{{Jour. High Energy Phys.\/} {\bf #1} (#2) }%
1300: \newcommand{\astropp}[3]{Astropart. Phys. {\bf #1}, #2 (#3)} %
1301: \newcommand{\ib}[3]{{ ibid.\/} {\bf #1} #2 (#3)} %
1302: % \newcommand{\iauc}[4]{{\em IAU Circular\/} #1
1303: % (\ifcase#2\or January \or February \or March \or April \or May
1304: % \or June \or July \or August \or September
1305: % \or October \or November \or December
1306: % \fi \ #3, #4)} %
1307: \newcommand{\nat}[3]{Nature (London) {\bf #1} #2 (#3)} %
1308: \newcommand{\app}[3]{{ Acta Phys. Polon. B\/}{\bf #1} #2 (#3)}%
1309: \newcommand{\nuovocim}[3]{Nuovo Cim. {\bf C#1} #2 (#3)} %
1310: \newcommand{\yadfiz}[4]{Yad. Fiz. {\bf #1} #2 (#3); %
1311: Sov. J. Nucl. Phys. {\bf #1} #3 (#4)]} %
1312: \newcommand{\jetp}[6]{{Zh. Eksp. Teor. Fiz.\/} {\bf #1} (#3) #2;
1313: {JETP } {\bf #4} (#6) #5}%
1314: % \newcommand{\jetpl}[6]{{\em ZhETF Pis'ma\/} {\bf #1} (#3) #2 %
1315: % [English translation: {\it JETP Lett.\/} {\bf #4} (#6) #5]} %
1316: % \newcommand{\uspekhi}[6]{{\em Usp. Fiz. Nauk.\/} {\bf #1} (#3) #2 %
1317: % [English translation: {\it Sov. Phys. Usp.\/} {\bf #4} (#6) #5]}%
1318: \newcommand{\philt}[3]{Phil. Trans. Roy. Soc. London A {\bf #1} #2
1319: (#3)} %
1320: \newcommand{\hepph}[1]{hep--ph/#1} %
1321: \newcommand{\hepex}[1]{hep--ex/#1} %
1322: \newcommand{\astro}[1]{(astro--ph/#1)} %
1323: % \relax %
1324: % %%% End Journal macro definitions %
1325: % %x
1326: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1327:
1328: \begin{thebibliography}{99}
1329: \bibitem{superK}
1330: Super-Kamiokande collaboration, Y. Hayato, talk given at the EPS 2003
1331: conference (Aachen, Germany, 2003),
1332: {\tt http://eps2003.physik.rwth-aachen.de};\\
1333: S. Fukuda et al., Phys. Lett. {\bf B539}, 179 (2002).
1334: \bibitem{sno}
1335: SNO collaboration, S.N. Ahmed et al., nucl-ex/0309004.
1336: \bibitem{chooz}
1337: CHOOZ collaboration, M. Appolonio et al., Eur. Phys. J {\bf C27}, 331
1338: (2003).
1339:
1340: \bibitem{kamland}
1341: KamLAND collaboration, K. Eguchi et al., Phys. Rev. Lett. {\bf 90},
1342: 021802 (2003).
1343:
1344: \bibitem{k2k}
1345: K2K collaboration, I. Kato, talk given at the 38th {\it Recontres de
1346: Moriond on Electroweak Interactions and Unified Theories} (Les Ares,
1347: France, 2003), hep-ex/0306043.
1348:
1349: \bibitem{review}For a recent review for the limits on masses and mixing, see
1350: S. Goswami, A. Bandyopadhyay and S. Choubey, hep-ph/0409224.
1351:
1352: \bibitem{msw}L. Wolfenstein, \prd{17}{2369}{1978}; \\
1353: S. Mikheyev and A. Smirnov, Sov. J. Nucl. Phys. {\bf 42}, 913 (1985).
1354:
1355: \bibitem{nu_fac_tech}For example, see S. Geer, FERMILAB-CONF-04-133-E,
1356: 2004; \\
1357: R. Raja,. FERMILAB-CONF-04-018-E (2004), hep-ex/0402022;\\
1358: A. Blondel et. al., Nucl.Instrum. Meth. {\bf A451},102 (2000).
1359:
1360: \bibitem{nu_fac}V. Barger, S. Geer, R. Raja and K. Whisnanat,
1361: \prd{62}{013004}{2000}; \\
1362: C. Albright et al., \hepex{0008064}; \\
1363: M. Freund, P. Huber and M. Lindner, \np{585}{105}{2000}; \np{615}{331}{2001};\\
1364: M. Apollonio et al., \hepph{0210192}.
1365:
1366: \bibitem{barger_matter}V. Barger, S. Geer, R. Raja and K. Whisnanat,
1367: Phys.Lett. {\bf B485} 379 (2000); \\
1368: A. Cervera et al., \np{579}{17}{2000}.
1369:
1370: \bibitem{maoki}M. Aoki et al., \prd{67}{093004}{2003}; \\
1371: M. Aoki, K, Hagiwara and N. Okamura, \hepph{0311324}.
1372:
1373: \bibitem{raj} R. Gandhi et al., hep-ph/0408361.
1374:
1375: \bibitem{mono}N.Y.~Agafonova {\it et al.} [MONOLITH Collaboration],
1376: LNGS-P26-2000, LNGS-P26-00, CERN-SPSC-2000-031, CERN-SPSC-M-657; see
1377: {\tt http://castore.mi.infn.it/$\sim$monolith/}
1378:
1379: \bibitem{ino}See
1380: {\tt http://www.imsc.res.in/$\sim$ino}; and working reports and talks
1381: therein.
1382:
1383: \bibitem{choubey_roy}
1384: S. Choubey and P. Roy, Phys. Rev. Lett. {\bf 93}, 021803 (2004).
1385:
1386: \bibitem{prem}
1387: A. M. Dziewonski and D.L. Anderson,
1388: Phys. Earth Plan. Int. {\bf 25}, 297 (1981); \\
1389: we use the
1390: parametrisation given in R. Gandhi, C. Quigg, M. Reno and I. Sarcevic,
1391: \astropp{5}{81}{1996}.
1392:
1393: \bibitem{expand}E. Akhmedov et al., \jhep{0404:078}{2004}.
1394:
1395: \bibitem{minos}
1396: R. Saakian et al ({\sc minos} collaboration),
1397: Nucl. Phys. Proc. Suppl. {\bf 111}, 169 (2002); Yad.Fiz. {\bf 67},
1398: 1112 (2004); \\
1399: M.V. Diwan et al., Nucl. Phys. Proc. Suppl. {\bf 123}, 272 (2003).
1400:
1401: %\bibitem{numi}
1402: %D. Ayres et al., hep-ex/0210005.
1403: %\bibitem{jparc}
1404: %
1405: \bibitem{minerva}MINERvA Collaboration (D.A. Harris for the collab.),
1406: FERMILAB-PUB-04-252-E, hep-ex/0410005;
1407:
1408: \bibitem{jhf}Y. Itow et al., hep-ex/0106019; also see :
1409: {\tt http://neutrino.kek.jp/jhfnu}.
1410:
1411: %\bibitem{cngs}
1412: %F.~Arneodo, talk given at the TAUP 2003 conference,\\
1413: %{\it http://mocha.phys.washington.edu/
1414: %$\sim$int\_talk/Workshops/TAUP03/Plenary/}.
1415: %http://mocha.phys.washington.edu/taup2003/; \\ K. Kodama, talk given
1416: %at the Nufact 2003 conference,\\
1417: %http://www.cap.bnl.gov/nufact03/agenda\_ug1.xhtml.
1418:
1419: \end{thebibliography}
1420: \end{document}
1421:
1422:
1423:
1424: