1:
2: %%
3: %% Beginning of file 'ms.tex'
4: %%
5: %% Modified 03 Jan 01
6: %%
7: %% This is a sample manuscript marked up using the
8: %% AASTeX v5.x LaTeX 2e macros.
9:
10: %% The first piece of markup in an AASTeX v5.x document
11: %% is the \documentclass command. LaTeX will ignore
12: %% any data that comes before this command.
13:
14: %% The command below calls the preprint style
15: %% which will produce a one-column, single-spaced document.
16: %% Examples of commands for other substyles follow. Use
17: %% whichever is most appropriate for your purposes.
18:
19: %\documentclass[12pt,preprint]{aastex}
20:
21: %% manuscript produces a one-column, double-spaced document:
22:
23: %\documentclass[manuscript]{aastex}
24:
25: %% preprint2 produces a double-column, single-spaced document:
26:
27: \documentclass[preprint2]{aastex}
28:
29: %% If you want to create your own macros, you can do so
30: %% using \newcommand. Your macros should appear before
31: %% the \begin{document} command.
32: %%
33: %% If you are submitting to a journal that translates manuscripts
34: %% into SGML, you need to follow certain guidelines when preparing
35: %% your macros. See the AASTeX v5.x Author Guide
36: %% for information.
37:
38: %% You can insert a short comment on the title page using the command below.
39:
40: \slugcomment{ApJ Letters, 2004, 603, L3-L96}
41:
42: %% If you wish, you may supply running head information, although
43: %% this information may be modified by the editorial offices.
44: %% The left head contains a list of authors,
45: %% usually a maximum of three (otherwise use et al.). The right
46: %% head is a modified title of up to roughly 44 characters. Running heads
47: %% will not print in the manuscript style.
48:
49: \shorttitle{Forced oscillations and kHz QPOs}
50:
51: \shortauthors{Lee, Abramowicz \& Klu\'{z}niak}
52:
53: %% This is the end of the preamble. Indicate the beginning of the
54: %% paper itself with \begin{document}.
55: %
56: %\received{2003 December 18}
57: \begin{document}
58:
59: %% LaTeX will automatically break titles if they run longer than
60: %% one line. However, you may use \\ to force a line break if
61: %% you desire.
62:
63: \title{Resonance in Forced Oscillations of an Accretion Disk\\ and
64: Kilohertz Quasi--Periodic Oscillations}
65:
66: %% Use \author, \affil, and the \and command to format
67: %% author and affiliation information.
68: %% Note that \email has replaced the old \authoremail command
69: %% from AASTeX v4.0. You can use \email to mark an email address
70: %% anywhere in the paper, not just in the front matter.
71: %% As in the title, you can use \\ to force line breaks.
72:
73: \author{William H. Lee}
74: \affil{Instituto de Astronom\'{\i}a,
75: Universidad Nacional Aut\'{o}noma de M\'{e}xico, \\
76: Apdo. Postal 70-264, Cd. Universitaria,
77: M\'{e}xico D.F. 04510}
78:
79: \author{Marek A. Abramowicz}
80: \affil{Astrophysics Department, Chalmers University, S--41296
81: G\"{o}teborg, Sweden}
82:
83: \and
84:
85: \author{W\l odek Klu\'{z}niak}
86: \affil{Institute of Astronomy, Zielona G\'{o}ra University, Lubuska
87: 2, 65--265 Zielona G\'{o}ra, Poland
88: \\Copernicus Astronomical Center, ul. Bartycka 18, 00-716 Warszawa, Poland}
89:
90:
91: %% Notice that each of these authors has alternate affiliations, which
92: %% are identified by the \altaffilmark after each name. Specify alternate
93: %% affiliation information with \altaffiltext, with one command per each
94: %% affiliation.
95:
96: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
97: %% style, abstract will output a Received/Accepted line after the
98: %% title and affiliation information. No date will appear since the author
99: %% does not have this information. The dates will be filled in by the
100: %% editorial office after submission.
101:
102: \begin{abstract}
103: We have performed numerical simulations of a radially perturbed
104: ``accretion'' torus around a black hole or neutron star and find that
105: the torus performs radial and vertical motions at the appropriate
106: epicyclic frequencies. We find clear evidence that vertical motions
107: are excited in a nonlinear resonance when the applied perturbation is
108: periodic in time. The strongest resonant response occurs when the
109: frequency difference of the two oscillations is equal to one-half the
110: forcing frequency, precisely as recently observed in the accreting
111: pulsar, SAX J1808.4-3658, where the observed kHz QPO peak separation
112: is half the spin frequency of 401 Hz.
113: \end{abstract}
114:
115: %% Keywords should appear after the \end{abstract} command. The uncommented
116: %% example has been keyed in ApJ style. See the instructions to authors
117: %% for the journal to which you are submitting your paper to determine
118: %% what keyword punctuation is appropriate.
119:
120: \keywords{Stars: neutron --- X-rays: binaries --- accretion disks ---
121: hydrodynamics }
122:
123: %% From the front matter, we move on to the body of the paper.
124: %% In the first two sections, notice the use of the natbib \citep
125: %% and \citet commands to identify citations. The citations are
126: %% tied to the reference list via symbolic KEYs. The KEY corresponds
127: %% to the KEY in the \bibitem in the reference list below. We have
128: %% chosen the first three characters of the first author's name plus
129: %% the last two numeral of the year of publication as our KEY for
130: %% each reference.
131:
132: \section{Introduction}\label{intro}
133:
134: Millisecond oscillations in the X--ray light curves of systems known
135: to contain neutron stars or possibly black holes have been observed
136: for several years with the Rossi X--Ray Timing Explorer (see
137: \cite{vdk00} for a review). Recently, it has been pointed out that the
138: centroid frequencies of the corresponding kilohertz quasi periodic
139: oscillations (kHz QPOs) are in rational ratios of small integers, as
140: 3:2 \citep{ak01,rmmo02,Mc03,abbk03,akklr03}. This supports
141: suggestions that a resonance of some kind is responsible for the
142: observed properties \citep{ka01,ka03,t02}. In addition, for at
143: least one system in which coherent pulsations have been detected,
144: indicating the spin frequency ($\nu_{s}=401$~Hz in SAX J1808.4-3658),
145: the separation in frequency between the kHz peaks has been recently
146: reported to be consistent with $\nu_{s}/2$ \citep{wetal03}, implying
147: that the pulsar is exciting motions in the accretion disk in a
148: nonlinear fashion \citep{kakls03}.
149:
150: In this Letter we show that the kHz oscillations detected in SAX
151: J1808.4-3658 can be attributed to forcing of epicyclic motions in the
152: accretion disk by the 2.5 ms pulsar, which induces resonance
153: at selected frequencies. The coupling between the pulsar and the disk
154: could be due to the magnetic field, or to some structure on the
155: surface of the star. In other, similar systems, a frequency
156: separation equal to the stellar spin frequency is also possible.
157:
158: \section{Response of a torus to an external perturbation}\label{response}
159:
160: In order to study the response of an accretion disk to an external
161: perturbation, we make several simplifying assumptions. The first is to
162: consider not a full accretion disk but a slender torus in hydrostatic
163: equilibrium, orbiting a central body of mass $M$. By virtue of the
164: pressure effects, the rotation curve is not Keplerian, and for all the
165: cases studied here, we consider constant distributions of specific
166: angular momentum within the torus, which has low mass and is slender
167: in the sense that its mass $m \ll M$, and its extension $L \ll R$,
168: where $R$ is the distance separating it from the mass $M$. We thus
169: neglect the self--gravity of the torus, which is ellipsoidal in
170: cross section in a meridional slice. We additionally assume azimuthal
171: symmetry.
172:
173: The second assumption is to use a potential for the central mass
174: $\Phi_{KL}=M[1-\exp(r_{ms}/r)]/r_{ms}$ \citep{kl02}, which reproduces
175: two features of Einstein's gravity that are relevant in our context:
176: (1) there is a marginally stable orbit for test masses in circular
177: orbits at $r_{ms}=6M$, and (2) the vertical and radial epicyclic
178: frequencies (in the equatorial plane) are given by
179: $\zeta^{2}(r)=(1/4\pi)^{2}(M/r^{3})\exp(r_{ms}/r)$ and
180: $\kappa^{2}(r)=(1/4\pi)^{2}(M/r^{3})\exp(r_{ms}/r)(1-r_{ms}/r)$,
181: respectively, so that their ratio is
182: $\kappa(r)/\zeta(r)=(1-r_{ms}/r)^{1/2}$, coinciding with the exact
183: solution in the Schwarzschild metric in general relativity.
184:
185: In hydrostatic equilibrium, pressure and centrifugal support balance
186: gravity in the radial direction, and pressure alone balances gravity
187: vertically. An ideal equation of state for the fluid closes the
188: system of equations (we have used an adiabatic index $\gamma=4/3$
189: throughout). The center of the torus can be conveniently defined as
190: the locus of maximum density, where the rotation velocity is
191: Keplerian, and its position is denoted by $r_{0}$.
192:
193: Clearly a detailed answer to the general problem being explored here
194: requires modeling of the full accretion disk, whether it is
195: geometrically thin or thick. We will explore this in future work. For
196: the time being, analyzing the behavior of such a slender torus can be
197: considered analogous to splitting the disk into thin annuli and
198: investigating the properties of one such ring --- as in analysis of
199: radiation--induced warping \citep{petterson77,pringle96}. The slender
200: torus can be thought of as a density enhancement in the accretion
201: disk, the properties of which might be transmitted and/or amplified
202: into the overall X--ray light curve observed from the system by an
203: unspecified mechanism.
204:
205: The actual tori are constructed from the specified gravitational
206: potential and equation of state in standard fashion (see
207: e.g. \citep{ica96,zb86}), and their dynamical behavior (see below) is
208: followed using a two--dimensional smooth particle hydrodynamics (SPH)
209: code \citep{monaghan92,lrr02} that uses cylindrical coordinates
210: $(r,z)$. A cross section (in density) of an equilibrium configuration
211: thus realized is shown in Fig.~\ref{torusIC}. We now detail the
212: results obtained from using two different types of perturbations to
213: such an initial state.
214:
215: \subsection{Impulsive perturbation}\label{impulsive}
216:
217: As a first example, we consider an impulsive perturbation to the torus,
218: analogous to that considered recently by \citet{zrf03}. (We note that
219: these modes might also be relevant for kHz QPOs in black hole systems;
220: see \cite{rymz03}.) The equilibrium torus is given a perturbed
221: velocity field at $t=0$ and allowed to oscillate freely
222: thereafter. The velocity field of the perturbation is purely radial
223: (in a cylindrical sense). The kick is small, so that the induced
224: oscillations in the radial direction have an amplitude that is of the
225: same order as the extent of the torus itself. The torus performs small
226: oscillations in the radial direction, primarily at a frequency
227: consistent with the local epicyclic frequency $\kappa_{0}$, as would
228: be expected. However, small vertical motions are also induced owing to
229: pressure coupling between both modes, and these can be clearly seen in
230: the vertical oscillations of the center of the torus, occurring with
231: greatest power at a frequency that is consistent with the local
232: vertical epicyclic frequency, $\zeta_{0}$. Both the radial and
233: vertical oscillations are shown in Fig.~\ref{RZimpulsive}, along with
234: their Fourier transforms.
235:
236: It is thus clear that if radial motions of some kind are present in
237: the torus, they can induce vertical oscillations, and both of these
238: occur at the epicylic frequencies for test particles in nearly
239: circular orbits.
240:
241: \subsection{Periodic perturbation}\label{periodic}
242:
243: We now consider a situation in which there is a periodic perturbation
244: at work, which is external to the disk. The pulsar at the center of
245: the gravitational potential well is clearly an example of this, and
246: one would expect it to have an effect on the disk. The magnetic field
247: of the pulsar, or some deformation on its surface, can perturb the
248: disk at intervals given by the inverse of the spin period, $\Delta
249: T=1/\nu_{s}$ (in the case of SAX J1808.4-3658, $\nu_{s}$=401~Hz). We
250: have considered then a forcing in the radial direction, which
251: manifests itself through a small, radial acceleration, the magnitude
252: of which is shown in Fig.~\ref{forcingamp}. It is not purely
253: sinusoidal but repeats at a fixed interval $\Delta T$, as given
254: above. The adopted profile of this perturbation deserves some comment:
255: we have used it to mimic the passage of a brief (compared with the
256: repetition time), but fairly stronger than average, disturbance in the
257: accretion disk (e.g., the corresponding polar magnetic field of the
258: pulsar, sweeping around the disk). Different shapes for this pulse
259: will be explored elsewhere.
260:
261: The radial forcing obviously induces a radial oscillation of the
262: torus. Because of pressure coupling, a vertical oscillation is again
263: apparent, albeit at a reduced magnitude (compared with the radial
264: amplitude). The motion of the center of the torus (as defined above)
265: can be Fourier--analyzed to extract the relevant frequencies. We find
266: that the radial and vertical motions occur primarily at the local
267: radial and epicylic frequencies, $\kappa_{0}$ and $\zeta_{0}$
268: respectively, at $r_{0}$. For a torus orbiting a mass
269: $M=1.38$M$_{\odot}$, and center at $r_{0}=12.25M=24.8\,$km,
270: $\kappa_{0}=500$~Hz and $\zeta_{0}=700$~Hz.
271:
272: The question now is: what effect, if any, does the repetition time
273: $\Delta T$ of the perturbation have on the power of oscillatory motion
274: induced in the torus? The radial motion is always driven at a
275: relatively high amplitude, simply because the perturbation is itself
276: applied as a radial acceleration. We find, however, that the power of
277: the vertical motions varies greatly as $\Delta T$ is altered.
278:
279: We have kept the initial position of the torus fixed at $r_{0}=12.25M$
280: and performed over a dozen simulations, differing only in the value of
281: $\nu_{s}$, and covering a range $100<\nu_{s} \mbox{(Hz)}<600$. For
282: each one of these, we have computed the peak power $P_{M,z}$ in the
283: vertical oscillation at $\zeta_{0}$. Since one could equivalently
284: keep $\nu_{s}$ fixed (as is actually the case in nature) and vary
285: $r_{0}$ (which would alter the difference $\zeta_{0}-\kappa_{0}$), in
286: Fig.~\ref{vertpower} we show $P_{M,z}$ as a function of the ratio
287: $\nu_{s}/(\zeta_{0}-\kappa_{0})$. The response of the torus is clearly
288: greatest when $\nu_{s}=2 (\zeta_{0}-\kappa_{0})$, as was observed in
289: the single instance for SAX J1808.4-3658 when two QPO peaks in the kHz
290: range were seen.
291:
292: We note that there is also a strong response when $\nu_{s}$ and
293: $\zeta_{0}-\kappa_{0}$ are in a 1:1 or 3:2 correspondence. The first
294: of these would allow for the possibility of twin peaks with a
295: separation of 401~Hz in SAX J1808.4-3658, while the second would imply
296: a separation of $802/3=267$~Hz. The first option would occur at
297: $\zeta_{0}$=1054~Hz, $\kappa_{0}$=653~Hz (see also \citet{kakls03}),
298: and $r_{0}$=9.75M=19.7~km for 1.38M$_{\odot}$. The second would imply
299: $\zeta_{0}$=832~Hz and $\kappa_{0}$=565~Hz at a radius
300: $r_{0}$=11.13M=22.5~km for the same mass. Neither has been observed as
301: yet.
302:
303: \section{Discussion}\label{disc}
304:
305: The appearance of twin kHZ QPOs in the millisecond pulsar SAX
306: J1808.4-3658, with a separation consistent with half the known spin
307: frequency of the pulsar, strongly indicates that a nonlinear
308: resonance is at work, coupling the spin to vibrational modes in the
309: disk. Using a simple hydrodynamical model, we identify these modes
310: with the radial and epicyclic oscillations of fluid elements slightly
311: displaced from exact circular orbits. In this respect, the model is
312: crucially dependent on the effects of strong gravity, to break the
313: degeneracy between the orbital and epicyclic frequencies present in
314: the Newtonian regime.
315:
316: Under the unique assumption that the pulsar provides a periodic
317: driving radial force to a slender torus in orbit (which we consider as
318: a stand--in for a density enhancement in the accretion disk; see
319: \S~\ref{response}), we show that the response of the torus is greatest
320: when the spin frequency is twice the difference between the vertical
321: and radial epicyclic frequencies. We also show that there are other
322: possibilities for resonant motion, when the above numbers are in a 1:1
323: or 3:2 correspondence.
324:
325: For the case of SAX J1808.4-3658, the frequency ratio 700:500=1.4, and
326: the actual values of the frequencies observed would allow us to fix
327: the mass of the pulsar at 1.38 solar masses in the Schwarzschild
328: metric. However, the actual value will be different, as the epicyclic
329: frequencies for a neutron star rotating at 401 Hz depart from the
330: Schwarzschild values \citep{kakls03}.
331:
332: Two further points deserve comment in the context of kHZ QPOs in
333: systems with a millisecond pulsar whose spin frequency is
334: known. First, there is an apparent dichotomy in the values of the QPO
335: peak separation with respect to the spin frequency. For the ``fast''
336: rotators, like SAX J1808.4-3658, the separation is half the spin
337: frequency, $\Delta \nu=\nu_{s}/2$, while for ``slow'' rotators, like XTE
338: J1807-294 (where $\nu_{s}=190$~Hz), the separation is consistent with
339: the spin frequency, $\Delta \nu=\nu_{s}$. This could be explained
340: within the current framework, since for fast rotators, the resonant
341: point corresponding to $\Delta \nu = \nu_{s}$ is so close to the
342: neutron star that is it unlikely to be seen (one could say that there
343: is no room around the pulsar for this mode to occur). For slow
344: rotators, both resonances, at $\nu_s$ and $\nu_s/2$, could in principle
345: be observed, since they would manifest themselves at greater distances
346: from the neutron star. However, if the modulations in flux generated
347: in the innermost regions of the accretion disk have a dominant effect
348: on the overall X--ray light curve, one would then expect to see the
349: strongest signal at frequency $\nu_s$.
350:
351: Second, the twin kHz QPOs then need not necessarily occur always at
352: the same frequencies, as the torus shifts its radial position, but
353: their separation should nevertheless remain roughly constant (and
354: equal to $\nu_{s}$ or $\nu_{s}/2$, depending on the system) under this
355: excitation mechanism, assuming mode locking to occur. Further timing
356: observations of millisecond pulsars will no doubt shed light on these
357: matters.
358:
359: More generally, our simulation shows directly that a radial perturbation
360: present in the torus can excite forced oscillation at (other) eigenfrequencies.
361: This may have implications for mode coupling in black hole accretion disks.
362:
363: \acknowledgments
364:
365: Financial support for this work was provided in part by CONACyT
366: (36632E) and by the Polish Committee for Scientific Research (KBN
367: grant 2P03D01424). We thank the referee for his comments on the
368: original version of the manuscript.
369:
370: %% Appendix material should be preceded with a single \appendix command.
371: %% There should be a \section command for each appendix. Mark appendix
372: %% subsections with the same markup you use in the main body of the paper.
373:
374: %% Each Appendix (indicated with \section) will be lettered A, B, C, etc.
375: %% The equation counter will reset when it encounters the \appendix
376: %% command and will number appendix equations (A1), (A2), etc.
377:
378: %% The reference list follows the main body and any appendices.
379: %% Use LaTeX's thebibliography environment to mark up your reference list.
380: %% Note \begin{thebibliography} is followed by an empty set of
381: %% curly braces. If you forget this, LaTeX will generate the error
382: %% "Perhaps a missing \item?".
383: %%
384: %% thebibliography produces citations in the text using \bibitem-\cite
385: %% cross-referencing. Each reference is preceded by a
386: %% \bibitem command that defines in curly braces the KEY that corresponds
387: %% to the KEY in the \cite commands (see the first section above).
388: %% Make sure that you provide a unique KEY for every \bibitem or else the
389: %% paper will not LaTeX. The square brackets should contain
390: %% the citation text that LaTeX will insert in
391: %% place of the \cite commands.
392:
393: %% We have used macros to produce journal name abbreviations.
394: %% AASTeX provides a number of these for the more frequently-cited journals.
395: %% See the Author Guide for a list of them.
396:
397: %% Note that the style of the \bibitem labels (in []) is slightly
398: %% different from previous examples. The natbib system solves a host
399: %% of citation expression problems, but it is necessary to clearly
400: %% delimit the year from the author name used in the citation.
401: %% See the natbib documentation for more details and options.
402:
403: \begin{thebibliography}{}
404: \bibitem[Abramowicz et al.(2003a)]{abbk03}Abramowicz M.A., Bulik T.,
405: Bursa M., Klu\'zniak W. 2003, A\&A, 404, L21
406: \bibitem[Abramowicz et al.(2003b)]{akklr03}Abramowicz, M.A., Karas, V.,
407: Klu\'{z}niak, W., Lee, W.H., Rebusco, P. 2003, PASJ, 55, 467
408: \bibitem[Abramowicz \& Klu\'{z}niak(2001)]{ak01}Abramowicz M.A.,
409: Klu\'zniak W. 2001, A\&A, 374, L19
410: \bibitem[Igumenshchev, Chen \& Abramowicz(1996)]{ica96}Igumenshchev,
411: I. V., Chen, X. \& Abramowicz, M. A. 1996, MNRAS, 278, 236
412: \bibitem[Klu\'{z}niak \& Abramowicz(2001)]{ka01} Klu\'{z}niak, W.,
413: Abramowicz, M.A.2001 Acta Phys. Polonica, 32, 3605
414: \bibitem[Klu\'{z}niak \& Abramowicz(2003)]{ka03} Klu\'{z}niak, W.,
415: Abramowicz, M.A. 2003, preprint (astro-ph/0304345)
416: \bibitem[Klu\'{z}niak \& Lee(2002)]{kl02}Klu\'{z}niak, W. \& Lee,
417: W. H. 2002, MNRAS, 335, L29
418: \bibitem[Klu\'{z}niak et al.(2003)]{kakls03}Klu\'{z}niak, W.,
419: Abramowicz, M.A., Kato, S., Lee, W.H., Stergioulas, N., 2004, ApJ,
420: 603, L89
421: \bibitem[Lee \& Ramirez--Ruiz(2002)]{lrr02}Lee, W.H., Ramirez--Ruiz,
422: E. 2002, ApJ, 577, 893
423: \bibitem[McClintock \& Remillard(2003)]{Mc03}McClintock J.E, Remillard
424: R.A. 2004, in Compact Stellar X--Ray Sources, ed. W.~H.~G. Lewin \&
425: M. van der Klis (Cambridge: Cambridge Univ. Press), in press
426: (astro-ph0306213)
427: \bibitem[Monaghan(1992)]{monaghan92} Monaghan, J.~J. 1992, ARAA, 30,
428: 543
429: \bibitem[Petterson(1977)]{petterson77}Petterson, J.A., 1977, ApJ, 214,
430: 550
431: \bibitem[Pringle(1996)]{pringle96} Pringle, J.~E, 1996, MNRAS, 281,
432: 357
433: \bibitem[Remillard et al.(2002)]{rmmo02}Remillard, R.A., Muno, M.P.,
434: McClintock, J.E., Orosz, J.A. 2002, ApJ, 580, 1030
435: \bibitem[Rezzolla, Yoshida, Maccarone \&
436: Zanotti(2003)]{rymz03}Rezzolla, L., Yoshida, S'i., Maccarone, T.J.,
437: Zanotti, O. 2003, MNRAS, 344, L37
438: \bibitem[Titarchuk(2002)]{t02}Titarchuk, L. 2002, ApJ, 578, L71
439: \bibitem[van der Klis(2000)]{vdk00}van der Klis, M. 2000, ARAA, 38,
440: 717
441: \bibitem[Wijnands et al.(2003)]{wetal03}Wijnands, R., van der Klis,
442: M., Homan, J., Chakrabarty, D., Markwardt, C.~B., Morgan, E.~H. 2003,
443: Nature, 424, 44
444: \bibitem[Zanotti, Rezzolla \& Font(2003)]{zrf03}Zanotti, O., Rezzolla, L.,
445: Font, J.A. 2003, MNRAS, 341, 832
446: \bibitem[Zurek \& Benz(1986)]{zb86}Zurek, W.H. \& Benz, W. 1986, ApJ,
447: 308, 123
448: \end{thebibliography}
449:
450: %% Use the figure environment and \plotone or \plottwo to include
451: %% figures and captions in your electronic submission.
452:
453: %\clearpage
454:
455: \begin{figure}
456: \plotone{f1.eps}
457: \caption{Logarithmic density contours in a meridional slice of a
458: slender torus in hydrostatic equilibrium, spaced every 0.2 dex, with
459: the lowest one at $\log (\rho/\rho_{\mbox{max}})=-1$. The center of the
460: torus is at $r_{0}=12.25$~M.
461: \label{torusIC}}
462: \end{figure}
463:
464: %\clearpage
465:
466: \begin{figure}
467: \plotone{f2.eps}
468: \caption{Left: Radial (solid line) and vertical (dashed line)
469: oscillations performed by the center of the torus after an impulsive
470: radial perturbation at $t=0$. Right: Fourier transforms of the radial
471: (solid line) and vertical (dashed line) oscillations shown in the left
472: panel. The two curves have been rescaled so as to fit on the same
473: graph.
474: \label{RZimpulsive}}
475: \end{figure}
476:
477: %\clearpage
478:
479: \begin{figure}
480: \plotone{f3.eps}
481: \caption{Amplitude of the radial perturbation (normalized to the
482: maximum value) applied to the torus, with a repetition frequency of
483: $\nu_{s}=401$~Hz.
484: \label{forcingamp}}
485: \end{figure}
486:
487: %\clearpage
488:
489: \begin{figure}
490: \plotone{f4.eps}
491: \caption{Peak power in the vertical epicyclic oscillations induced in
492: the torus by the periodic perturbation of Fig.~\ref{forcingamp},
493: as a function of $\nu_{s}/(\zeta_{0}-\kappa_{0})$. Three
494: peaks at 1.0, 1.5, and 2.0 are clearly visible. For this calculation,
495: $\nu_{s}$, the pulsar perturbation frequency, varied while
496: $\zeta_{0}-\kappa_{0}$ was held fixed at 200~Hz.
497: \label{vertpower}}
498: \end{figure}
499:
500:
501: \end{document}
502:
503: %%
504:
505: