1: %%
2: %% Beginning of file 'sample.tex'
3: %%
4: %% Modified 2004 January 9
5: %%
6: %% This is a sample manuscript marked up using the
7: %% AASTeX v5.x LaTeX 2e macros.
8:
9: %% The first piece of markup in an AASTeX v5.x document
10: %% is the \documentclass command. LaTeX will ignore
11: %% any data that comes before this command.
12:
13: %% The command below calls the preprint style
14: %% which will produce a one-column, single-spaced document.
15: %% Examples of commands for other substyles follow. Use
16: %% whichever is most appropriate for your purposes.
17: %%
18: %%\documentclass[12pt,preprint]{aastex}
19:
20: %% manuscript produces a one-column, double-spaced document:
21:
22: \documentclass[manuscript]{aastex}
23:
24: %\documentclass[manuscript]{emulateapj}
25:
26:
27:
28: %% preprint2 produces a double-column, single-spaced document:
29:
30: %\documentclass[preprint2]{aastex}
31:
32: %% Sometimes a paper's abstract is too long to fit on the
33: %% title page in preprint2 mode. When that is the case,
34: %% use the longabstract style option.
35:
36: %% \documentclass[preprint2,longabstract]{aastex}
37:
38: %% If you want to create your own macros, you can do so
39: %% using \newcommand. Your macros should appear before
40: %% the \begin{document} command.
41: %%
42: %% If you are submitting to a journal that translates manuscripts
43: %% into SGML, you need to follow certain guidelines when preparing
44: %% your macros. See the AASTeX v5.x Author Guide
45: %% for information.
46:
47: \newcommand{\vdag}{(v)^\dagger}
48: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
49:
50: %% You can insert a short comment on the title page using the command below.
51:
52: \slugcomment{}
53:
54: %% If you wish, you may supply running head information, although
55: %% this information may be modified by the editorial offices.
56: %% The left head contains a list of authors,
57: %% usually a maximum of three (otherwise use et al.). The right
58: %% head is a modified title of up to roughly 44 characters.
59: %% Running heads will not print in the manuscript style.
60:
61: \shorttitle{MHD waves in the solar tachocline} \shortauthors{Zaqarashvili et
62: al.}
63:
64: %% This is the end of the preamble. Indicate the beginning of the
65: %% paper itself with \begin{document}.
66:
67: \begin{document}
68:
69: %% LaTeX will automatically break titles if they run longer than
70: %% one line. However, you may use \\ to force a line break if
71: %% you desire.
72:
73: \title{Global shallow water magnetohydrodynamic waves in the solar tachocline}
74:
75: %% Use \author, \affil, and the \and command to format
76: %% author and affiliation information.
77: %% Note that \email has replaced the old \authoremail command
78: %% from AASTeX v4.0. You can use \email to mark an email address
79: %% anywhere in the paper, not just in the front matter.
80: %% As in the title, use \\ to force line breaks.
81:
82: \author{T.V. Zaqarashvili}
83: \affil{Abastumani Astrophysical Observatory at I. Chavchavadze State
84: University, Kazbegi Ave. 2a, Tbilisi 0160, Georgia}
85: \email{temury@genao.org}
86:
87:
88: \author{R. Oliver and J.L. Ballester}
89: \affil{Departament de F\'{\i}sica, Universitat de les Illes Balears,
90: E-07122 Palma de Mallorca, Spain}
91: \email{[ramon.oliver;dfsjlb0]@uib.es}
92:
93:
94: %\altaffiltext{1}{Also at Abastumani Astrophysical Observatory, Kazbegi Ave. 2a, Tbilisi 0160, Georgia}
95:
96: %% Notice that each of these authors has alternate affiliations, which
97: %% are identified by the \altaffilmark after each name. Specify alternate
98: %% affiliation information with \altaffiltext, with one command per each
99: %% affiliation.
100:
101: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
102: %% style, abstract will output a Received/Accepted line after the
103: %% title and affiliation information. No date will appear since the author
104: %% does not have this information. The dates will be filled in by the
105: %% editorial office after submission.
106:
107: \begin{abstract}
108:
109: We derive analytical solutions and dispersion relations of global
110: magnetic Poincar\'e (magneto-gravity) and magnetic Rossby waves in
111: the approximation of shallow water magnetohydrodynamics. The
112: solutions are obtained in a rotating spherical coordinate system for
113: strongly and weakly stable stratification separately in the presence
114: of toroidal magnetic field. In both cases magnetic Rossby waves
115: split into fast and slow magnetic Rossby modes. In the case of
116: strongly stable stratification (valid in the radiative part of the
117: tachocline) all waves are slightly affected by the layer thickness
118: and the toroidal magnetic field, while in the case of weakly stable
119: stratification (valid in the upper overshoot layer of the
120: tachocline) magnetic Poincar\'e and magnetic Rossby waves are found
121: to be concentrated near the solar equator, leading to equatorially
122: trapped waves. The frequencies of all waves are smaller in the upper
123: weakly stable stratification region than in the lower strongly
124: stable stratification one.
125:
126: \end{abstract}
127:
128:
129: \keywords{Sun: interior -- Sun: magnetic fields -- Sun:
130: oscillations}
131:
132: \section{Introduction}
133:
134: The tachocline, which is believed to exist near the base of the solar
135: convection zone \citep{spi92}, has become a subject of rapidly growing
136: scientific interest in the last years \citep{gou07}. The existence of
137: the tachocline is important for two main reasons. First, it may
138: prevent the spread of the solar angular momentum from the convection
139: zone to the interior \citep{spi92,gou98,gar07}, thus retaining the
140: inferred solid body rotation in the radiative zone; second, the
141: processes of solar magnetic field generation and global
142: magnetohydrodynamic (MHD) instabilities, which are crucial for
143: solar activity, may take place there \citep{dik05,gil07}. MHD waves
144: and oscillations in the tachocline may play a significant role in both
145: cases: they may redistribute the angular momentum in the horizontal
146: direction and some of them may become unstable due to the
147: differential rotation, leading to magnetic flux emergence at the
148: solar surface. Therefore, to study the complete spectrum of possible
149: wave modes in this system is of vital importance. The tachocline is
150: very thin compared to the solar radius, therefore the ordinary shallow
151: water approximation modified by the presence of a horizontal
152: large-scale magnetic field can be easily applied \citep{gil00}.
153:
154: The spectrum of various shorter scale shallow water MHD waves has
155: been recently studied in Cartesian coordinates by \citet{sch01}.
156: However, global wave modes, those with a wavelength comparable to
157: the solar radius, must be considered in spherical coordinates.
158: Recently, \citet{zaq07} have studied the spherical shallow water MHD waves
159: in the simplest case, in which the high frequency branch i.e. magnetic
160: Poincar\'e waves (or magneto-gravity waves) and the influence of the
161: tachocline thickness on the wave dynamics have been ignored (that is,
162: when the surface gravity speed is much higher than the surface
163: rotation speed). This approximation has enabled us to study the dynamics
164: of global magnetic Rossby waves in the lower, strongly stable part of
165: the tachocline, but it fails in the upper, weakly stable overshoot
166: region, where a negative buoyancy due to the subadiabatic
167: stratification strongly reduces the gravity speed \citep{gil00}.
168:
169: In this letter, we present the analytical spectrum of global linear
170: shallow water MHD waves for both parts of the tachocline in the
171: absence of differential rotation.
172:
173: \section{Global shallow water MHD waves}
174:
175: We use the linearized shallow water MHD equations in the rotating
176: spherical coordinate system ($r,\theta, \phi$), where $r$ is the
177: radial coordinate, $\theta$ is the co-latitude, and $\phi$ is the
178: longitude \citep{zaq07}. We also assume the unperturbed toroidal
179: magnetic field to be $B_{\phi}=B(\theta)\sin{\theta}$, where
180: $B(\theta)$ may have an arbitrary profile with latitude, which can
181: be chosen later.
182: %This is the
183: %simplest profile since a more complex structure of the magnetic
184: %field leads to the coupling of different harmonics and consequently
185: %an analytical dispersion relation cannot be obtained.
186: Then a normal mode analysis of the form $\exp(-i\omega t + is\phi)$
187: gives the equation \citep{zaq07}
188: \begin{eqnarray}
189: (\lambda D +s \mu)\left \{{1\over {s^2 -\epsilon
190: \lambda^2(1-\mu^2)}}\left [\lambda D -s \mu- {{\alpha^2}\over
191: {\lambda^2}}s^2\lambda(D +2\mu)\right ]\right \}{\hat u}_{\theta}
192: -(\lambda^2 -\mu^2){\hat u}_{\theta} \nonumber
193: \\ +s^2\alpha^2{\hat u}_{\theta} + 2\alpha^2\mu D{\hat u}_{\theta}
194: +\mu s {\alpha^2\over \lambda}(D+2\mu){\hat u}_{\theta}+\mu
195: D(\alpha^2) {\hat u}_{\theta}=0,
196: \end{eqnarray}
197: with
198: \begin{eqnarray*}
199: \lambda={{\omega}\over {2\Omega_0}},\,
200: \epsilon={{4\Omega^2_0R^2_0}\over {g H_0}},\,\alpha^2={{v^2_A}\over
201: {4\Omega^2_0R^2_0}},\,\,v_A={{B(\mu)}\over {\sqrt {4\pi
202: \rho}}},\nonumber
203: \\ \mu=\cos \theta,\,\,D=(1-\mu^2){{\partial
204: }\over {\partial \mu}}=-\sin \theta{{\partial }\over {\partial
205: \theta}}, \,\, {\hat u}_{\theta}=\sin {\theta} u_{\theta},
206: \end{eqnarray*}
207: where $u_{\theta}$ is the poloidal velocity component, $g$ is the
208: reduced gravity in the tachocline, $\rho$ is the medium density,
209: $\Omega_0$ is the angular velocity, $R_0$ is the distance from the
210: solar center to the tachocline, and $H_0$ is the tachocline
211: thickness. This is the general equation for the linear dynamics of
212: the shallow water MHD system, which contains various kinds of waves
213: (magnetic Poincar\'e and magnetic Rossby waves). When $\epsilon=0$
214: and $B=const$, this equation transforms into the associated Legendre
215: equation and governs the dynamics of fast and slow magnetic Rossby
216: waves \citep{zaq07}. This approximation neglects magnetic Poincar\'e
217: waves and the influence of the tachocline thickness on the dynamics
218: of magnetic Rossby waves. Therefore, equation (1) must be solved
219: with $\epsilon \not=0$ in order to give the complete spectrum of
220: waves. The subadiabatic stratification in the tachocline provides
221: negative buoyancy, which leads to a reduced gravity and consequently
222: an increased value of $\epsilon$. In the lower strongly stable part
223: of the tachocline $\epsilon$ still is much smaller than unity, but
224: in upper overshoot layer $\epsilon$ becomes much greater than unity.
225: Therefore, the solution of equation (1) for the two extreme cases
226: $\epsilon \ll 1$ and $\epsilon \gg 1$ covers both parts of the
227: tachocline. Here we use a magnetic field profile typical of the
228: Sun, namely
229: \begin{equation}B({\mu})=B_0 \mu,
230: \end{equation}
231: which has opposite signs in the northern and southern hemispheres
232: \citep{gil97}. The magnetic field profile (2) leads to
233: $\alpha=\alpha_0 \mu$, where $\alpha_0=B_0/(2\Omega_0 R_0 \sqrt{4\pi
234: \rho})$. In the next subsections we will give analytical solutions
235: and dispersion relations of global shallow water MHD waves for each
236: case separately.
237:
238: \subsection{$\epsilon \ll 1$ (strongly stable
239: stratification; valid for the radiative part of tachocline)}
240:
241: Reduced gravity in the strongly stable radiative part of the
242: tachocline can be estimated as 500--1.5 $\cdot$ 10$^4$
243: cm$\cdot$s$^{-2}$ \citep{sch01}. Then, for the layer thickness
244: $H_0$=10$^9$ cm we get $\epsilon=$ 4.5 $\cdot$ 10$^{-3}$ -- 0.13
245: (where $\Omega_0$=2.6 $\cdot$ 10$^{-6}$ s$^{-1}$, $R_0$=5$\cdot$
246: 10$^{10}$ cm have been used). Therefore, $\epsilon \ll 1$ is a good
247: approximation in the radiative part of the tachocline.
248:
249: In this approximation and using the weak magnetic field limit, i.e.
250: $\alpha_0^2 \ll 1$, equation (1) leads to the spheroidal wave
251: equation (second order terms with $\epsilon$ and $\alpha_0^2$ are
252: neglected), whose typical solutions are the spheroidal wave
253: functions $S_{sn}$ \citep{abr64}, where $n$ plays the role of the
254: poloidal wavenumber. The approximate solution can be written as
255: \begin{equation}
256: {\hat u}_{\theta} \propto u_0 S_{sn}({\epsilon_1}, \mu)\sqrt {{s^2
257: -\epsilon \lambda^2(1-\mu^2)}\over {\lambda^2 - s^2\alpha^2_0
258: \mu^2}},
259: \end{equation}
260: where $u_0$ is the amplitude and $$\epsilon_1=\sqrt {\epsilon \left
261: (1 + 3{\lambda^2\over s^2} -2 {\lambda\over s}\right ) +
262: {\alpha^2_0s^2\over \lambda^2}\left (7 + {s\over {2\lambda}} \right
263: )}.$$ Note that $n$ is integer for the solutions, which vanish at
264: the poles ($\mu=\pm 1$). The dispersion relation is found to be
265: \begin{equation}\label{dr_small_eps}
266: \epsilon (s^2 +1)\left ({{\lambda}\over s}\right )^4 -n(n+1)\left
267: ({{\lambda}\over s}\right )^2 - \left ({{\lambda}\over s}\right ) +
268: \alpha^2_0 =0.
269: \end{equation}
270:
271: The dispersion relation can be split into high (magnetic Poincar\'e)
272: and low (magnetic Rossby) frequency branches. For the magnetic
273: Poincar\'e waves we have
274: \begin{equation}\label{approx_mp_small_eps}
275: {{\lambda_{mP}}\over s} \approx \pm \sqrt { {{n(n+1)}\over {\epsilon
276: (s^2 +1)}} }.
277: \end{equation}
278: This is the dispersion relation of ordinary surface gravity waves
279: and we can recover it for the high $s$ harmonics \citep{lon65}.
280:
281: For the magnetic Rossby waves we have approximately
282: \begin{equation}
283: n(n+1)\left ({{\lambda}\over s}\right )^2 + {{\lambda}\over s} -
284: \alpha^2_0=0.
285: \end{equation}
286:
287: \noindent It is clear from equation (6) that the magnetic field causes the
288: splitting of ordinary Rossby waves into fast and slow modes
289: \citep{zaq07}. For $\alpha_0 \ll 1$ the dispersion relations for the
290: fast and slow magnetic Rossby modes can be approximated as
291: \begin{equation}\label{approx_fs_small_eps}
292: {{\lambda_f}\over s}\approx -{1\over {n(n+1)}} - \alpha^2_0,\,\,\,
293: {{\lambda_s}\over s}\approx \alpha^2_0.
294: \end{equation}
295: We see that the fast magnetic Rossby mode has a dispersion
296: relation similar to that of the ordinary Rossby waves but slightly modified
297: by the magnetic field. Slow magnetic Rossby waves, however, are new
298: wave modes as their dispersion relation is different from Rossby and
299: Alfv\'en wave dispersion relations (the dispersion relation of
300: Alfv\'en waves should be $\lambda_A={\pm}s\alpha_0$). There is a
301: significant difference between the frequencies of fast and slow
302: magnetic Rossby modes for sufficiently small $\alpha_0$.
303:
304: The dispersion diagram for the $s=1$ harmonics of shallow water MHD
305: waves according to Eq. (\ref{dr_small_eps}) is shown in Fig 1. Here $\epsilon = 0.01$
306: and $\alpha_0=0.05$. The frequencies of magnetic Poincar\'e waves
307: (solid lines) are much higher than the rotational frequency,
308: $\Omega_0$, and they increase with increasing $n$. For example, the
309: frequency of the $n=3$ harmonics is $\sim 50 \, \Omega_0$, which
310: corresponds to oscillations with period $\sim$ 13 hours. On the
311: other hand, the frequencies of magnetic Rossby waves are in general much lower
312: than the rotational frequency. The absolute value of the frequency of fast magnetic Rossby
313: waves (dotted line in the zoom) decreases with increasing $n$, such
314: as happens with the HD Rossby wave.
315:
316: Fig. 2 shows the dependence of the $s=1, n=2$ and $s=1, n=3$
317: harmonics on $\epsilon$ and $\alpha_0$ (i.e. the magnetic field
318: strength). The frequency (in absolute value) of fast (dotted) and slow (dashed) magnetic
319: Rossby modes significantly increases when the magnetic field is
320: increased. We also see that the frequency difference between the $n=2,3$ harmonics of fast magnetic
321: Rossby waves is almost independent of $\alpha_0$. On the contrary, the $n=2,3$ harmonics of slow magnetic
322: Rossby waves diverge when the magnetic field is increased. It turns
323: out that the frequencies of magnetic Rossby waves have almost no
324: dependence on $\epsilon$, therefore they are not shown in the top panel. On the
325: other hand, the frequency of magnetic Poincar\'e waves depends on
326: $\epsilon$ (lower panel) and significantly decreases with increasing
327: $\epsilon$. However, these frequencies have almost no dependence on
328: $\alpha_0$, therefore magnetic Poincar\'e waves are only slightly
329: affected by the magnetic field.
330:
331: \subsection{$\epsilon \gg 1$ (weakly
332: stable stratification; valid in the upper overshoot part of
333: tachocline)}
334:
335: In the weakly stable overshoot region (or say upper tachocline)
336: reduced gravity is much smaller, being 0.05--5 cm$\cdot$s$^{-2}$
337: \citep{sch01}. Then, for the thickness $H_0$=5 $\cdot$ 10$^8$ cm we
338: get $\epsilon=$ 27 -- 2.7 $\cdot$ 10$^{3}$. Therefore, $\epsilon \gg
339: 1$ is a good approximation in the overshoot region.
340:
341: In this case, we follow the calculation of \citet{lon68} (this paper
342: considers spherical hydrodynamic shallow water waves in the Earth
343: context), and introduce a new function $\eta = \epsilon^{1/4} \mu$
344: and also consider that $\lambda=\epsilon^{-1/4}L$ and
345: $\alpha_0=\epsilon^{-1/4} L$, where $L$ is of order unity. Then
346: after some algebra and keeping only large terms equation (1) leads
347: to the Weber (parabolic cylinder) equation, whose solutions finite
348: at $\eta \rightarrow \pm \infty$ can be expressed in terms of
349: Hermite polynomials $H_{\nu}(\eta)$ of order $\nu$, where $\nu$
350: satisfies
351: \begin{equation}
352: {{{\epsilon} \lambda^2 - {{s}\over {\lambda}} + {{\alpha_0^2
353: s^2}\over {\lambda^2}}}}=(2 \nu + 1)\sqrt {\epsilon},
354: \,\,\hspace{1cm} (\nu =0,1,2,\ldots)
355: \end{equation}
356: and the solution is proportional to
357: \begin{equation}
358: {\hat u}_{\theta} \propto e^{-{1\over 2}\eta^2}H_{\nu}(\eta).
359: \end{equation}
360: Therefore the function ${\hat u}_{\theta}$ is exponentially small
361: beyond the turning points $\cos \theta= \epsilon^{-1/4} \sqrt {2 \nu
362: + 1}$. Due to the large $\epsilon$, the co-latitude $\theta$
363: satisfying this condition is close to $90^{\rm o}$, therefore the
364: solution is confined to the neighborhood of the equator.
365:
366: Equation (8) yields the wave dispersion relation as
367: \begin{equation}
368: \lambda^4 - {{2\nu +1}\over {\sqrt {\epsilon}}}\lambda^2 -{s\over
369: {\epsilon}} \lambda + {{\alpha_0^2 s^2}\over \epsilon}=0.
370: \end{equation}
371: In the non-magnetic case this equation transforms into equation
372: (8.11) of \citet{lon68}. This is a fourth order equation and it
373: describes magnetic Poincar\'e and magnetic Rossby waves. For large
374: values of $\epsilon$, the solutions can be written separately for
375: both types of waves.
376:
377: The solutions for Poincar\'e waves can be written approximately as
378: \begin{eqnarray}\label{approx_mp_large_eps}
379: \lambda_{mP} \approx \pm {{(2\nu +1)^{1/2}}\over {\epsilon^{1/4}}} +
380: {s\over {(4\nu +2)\epsilon^{1/2}}}.
381: \end{eqnarray}
382: It turns out that the magnetic field has almost no influence on the
383: dynamics of Poincar\'e waves and the dispersion relation is the same
384: as for the nonmagnetic case \citep{lon68}.
385:
386: The dispersion relation for the magnetic Rossby waves can be written
387: as
388: \begin{equation}
389: (2\nu +1)\sqrt {\epsilon}\lambda^2 +s \lambda - \alpha_0^2 s^2=0.
390: \end{equation}
391: The magnetic field causes the splitting of ordinary Rossby waves
392: into fast and slow magnetic Rossby waves as in the $\epsilon \ll 1$
393: case. If $\sqrt {\epsilon}\alpha_0^2 \ll 1$, then we can write
394: \begin{eqnarray}\label{approx_fs_large_eps}
395: {\lambda_f \over s} \approx - {1 \over {(2\nu +1)\epsilon^{1/2}}}
396: -\alpha_0^2, \,\,\, {\lambda_s \over s} \approx \alpha_0^2.
397: \end{eqnarray}
398: Fast magnetic Rossby waves are similar to hydrodynamic Rossby
399: waves modified by the magnetic field. Slow magnetic Rossby waves
400: have a similar dispersion relation as in the $\epsilon \ll 1$ case.
401:
402: The dispersion diagram for the $s=1$ harmonics of magnetic
403: Poincar\'e and magnetic Rossby waves according to Eq. (10) is shown
404: in Fig 3. Here $\epsilon = 2700$ and $\alpha_0=0.05$. The behaviour
405: of fast magnetic Rossby waves (dotted line) is similar to the small
406: $\epsilon$ case: the absolute value of their frequency significantly decreases with
407: increasing $\nu$ (which now plays the role of the poloidal wave
408: number). However, we immediately note that the frequency now is much
409: smaller than that of the $\epsilon\ll 1$ case. This is also true for
410: magnetic Poincar\'e waves (solid lines): their frequency is
411: significantly small compared to the case of small $\epsilon$, now
412: being even smaller than the rotational frequency. The frequency of
413: slow magnetic Rossby waves only slightly depends on $\nu$, but now
414: it decreases with increasing $\nu$, contrary to $\epsilon\ll 1$
415: case.
416:
417: Fig. 4 shows the dependence of the $s=1, \nu=2$ and $s=1, \nu=3$
418: harmonics on $\epsilon$ and $\alpha_0$. The frequency of magnetic
419: Rossby mode harmonics significantly depends on $\alpha_0$; the
420: frequencies of both fast and slow magnetic Rossby modes increase
421: with increasing $\alpha_0$. On the other hand, it turns out that the
422: frequencies of magnetic Poincar\'e waves have almost no dependence
423: on $\alpha_0$. However, they are significantly reduced with
424: increasing $\epsilon$, becoming smaller than the rotational
425: frequency for $\epsilon > 1000$. Fast magnetic Rossby waves also
426: depend on $\epsilon$, their frequency significantly decreasing when
427: this parameter is increased. Slow magnetic Rossby waves have almost no dependence on $\epsilon$
428: as it is suggested from Eq. (\ref{approx_fs_large_eps}). Thus increasing $\epsilon$, which
429: is equivalent to reducing $g$, generally leads to a decrease of the
430: wave frequencies.
431:
432: The numerical solution of the dispersion relation (10) shows that
433: the frequency of the magnetic Rossby wave harmonic with $\nu=0$ becomes
434: complex for sufficiently large $\alpha_0$, i.e. large magnetic
435: field. This may point to some kind of instabilities similar to the
436: polar kink instability found by Cally (2003). However, the value of
437: $\alpha_0$ for which the frequency becomes complex is significantly
438: outside the range for which Eq. (10) is valid (note that $\alpha_0$
439: should be proportional to $\epsilon^{-1/4}$). Therefore, the complex
440: solution can be spurious and may not reflect a real instability.
441:
442: %In order to study the slow magnetic Rossby waves, one should
443: %consider the very low frequency branch so that $\lambda^2 <
444: %\alpha^2s^2$. We again follow \citet{lon68}, but now consider
445: %$\lambda=\epsilon^{-1/2}L$ instead of $\lambda=\epsilon^{-1/4}L$.
446: %Then it turns out that the solutions are concentrated near the poles
447: %and are linked to the associated Laguerre polynomials as
448: %\begin{equation}
449: %{\hat u}_{\theta} \propto x^{{{1}\over 2}s}e^{{1\over
450: %2}x}L^{s}_{\nu}(x),
451: %\end{equation}
452: %where
453: %\begin{equation}
454: %x=\sqrt {A_2} \theta^2,\,\,\, L^{s}_{\nu}(x)=\sum^{{\nu}}_{k=0}
455: %{{(s+{\nu})!}\over {(s+k)!({\nu}-k)!k!}}(-x)^k,\,\,\, A_2=
456: %-{{{\epsilon} \lambda^2}\over {\lambda^2-\alpha^2s^2}} +{1\over
457: %4}{{\epsilon^2 \alpha^4 \lambda^2 s^2}\over
458: %{(\lambda^2-\alpha^2s^2)^2}},
459: %\end{equation}
460: %so that $s>-1$.
461: %
462: %The dispersion relation for slow magnetic Rossby waves is
463: %\begin{equation} {\lambda\over s}= - {1\over
464: %\epsilon}\left (1 + 3{{\epsilon \alpha^2}} \right ) +
465: %2({{2{\nu}+s+1}}){\alpha\over {\sqrt {\epsilon}}}\sqrt {1 +
466: %{{\epsilon \alpha^2}\over 4}}.
467: %\end{equation}
468:
469: \section{Conclusions}
470:
471: We have derived analytical solutions and dispersion relations of
472: global shallow water MHD waves for the solar tachocline in a
473: rotating spherical coordinate system. The solutions and dispersion
474: relations are obtained for weakly and strongly stable
475: stratifications separately. The weakly stable stratification is
476: valid in the upper overshoot part of tachocline, while the strongly
477: stable stratification is valid for the radiative part of tachocline.
478: The solutions include the Poincar\'e and Rossby waves modified by
479: the presence of the magnetic field.
480:
481: Here we use a realistic latitudinal profile of the toroidal magnetic
482: field $B_{\phi}\sim \cos{\theta} \sin{\theta}$ that changes
483: sign at the equator. Similar to the $\epsilon=0$ case \citep{zaq07}, the
484: magnetic field leads to the splitting of hydrodynamic Rossby waves
485: into fast and slow magnetic Rossby modes.
486:
487: In the case of strongly stable stratification, the dispersion
488: relation of magnetic Poincar\'e waves is similar to that of the
489: hydrodynamic case \citep{lon65}. The dispersion relation of magnetic
490: Rossby waves \citep{zaq07} is slightly modified by the thickness of
491: the layer. In the case of weakly stable stratification, both
492: magnetic Poincar\'e and magnetic Rossby waves are concentrated near
493: the equator, thus leading to equatorially trapped waves.
494:
495: Another remarkable feature is that the frequencies of all waves are
496: significantly reduced in the large $\epsilon$ case compared to the
497: $\epsilon\ll 1$ one. This means that the upper overshoot part of the
498: tachocline supports oscillations with smaller frequency than the
499: lower radiative part of the tachocline.
500:
501: The solutions have been obtained without taking into account the
502: differential rotation which is present in the tachocline. The future
503: inclusion of differential rotation probably will lead to the
504: instability of some modes, which was found by simulations in the
505: inertial frame \citep[][and references therein]{gil07}.
506:
507: %The magnetic Rossby waves may have interesting consequences in
508: %intermediate periodicity found in the activity of the Sun and other
509: %stars.
510:
511: \acknowledgments
512:
513: This work has been supported by MICINN grant AYA2006-07637 and
514: FEDER funds, by the Conselleria de Economia, Hisenda i Innovaci\'o under grant
515: PCTIB-2005GC3-03, and by the
516: Georgian National Science Foundation grant GNSF/ST06/4-098.
517:
518:
519: \begin{thebibliography}{}
520:
521: \bibitem[Abramowitz \& Stegun(1964)]{abr64}Abramowitz, M., \& Stegun, I.A. 1964, Handbook of Mathematical Functions
522: (Washington, D.C.: National Bureau of Standards)
523:
524: \bibitem[Cally(2003)]{cal03}Cally, P. S. 2003, \mnras, 339, 957
525:
526: %\bibitem[Dikpati \& Gilman(2001)]{dik}Dikpati, M. \& Gilman, P.
527: %A. 2001, \apj, 551, 536
528:
529: \bibitem[Dikpati \& Gilman(2005)]{dik05}Dikpati, M. \& Gilman, P.
530: A. 2005, \apj, 635, L193
531:
532: %\bibitem[2001]{sterck}De Sterck, H., 2001, Physics of Plasmas, 8, 3293
533:
534: \bibitem[Garaud(2007)]{gar07}Garaud, P. 2007, in The Solar Tachocline, ed. D.W.
535: Hughes, R. Rosner \& N.O. Weiss (Cambridge: Cambridge University
536: Press), 147
537:
538: %\bibitem[1982]{gill}Gill, A.E., 1982, Atmosphere-Ocean Dynamics, (San Diego: Academic Press)
539:
540: \bibitem[Gilman \& Fox(1997)]{gil97}Gilman, P.A. \& Fox, P.A. 1997, \apj, 484,
541: 439
542:
543: \bibitem[Gilman(2000)]{gil00}Gilman, P.A. 2000, \apj, 544, L79
544:
545: \bibitem[Gilman \& Cally(2007)]{gil07}Gilman, P.A., \& Cally, P.S. 2007, in The Solar Tachocline, ed. D.W.
546: Hughes, R. Rosner \& N.O. Weiss (Cambridge: Cambridge University
547: Press), 243
548:
549: \bibitem[Gough \& McIntyre(1998)]{gou98}Gough, D., \& McIntyre, M. E. 1998, Nature, 394, 755
550:
551: \bibitem[Gough(2007)]{gou07}Gough, D. 2007, in The Solar Tachocline, ed. D.W.
552: Hughes, R. Rosner \& N.O. Weiss (Cambridge: Cambridge University
553: Press), 3
554:
555: %\bibitem[2007]{leprovost}Leprovost, N. \& Kim, E., 2007, Astrophys.
556: %J. 654, 1166
557:
558: \bibitem[Longuet-Higgins(1965)]{lon65}Longuet-Higgins, M.S., 1965, Proceedings of The Royal Society Series / A, 284, 40
559:
560: \bibitem[Longuet-Higgins(1968)]{lon68}Longuet-Higgins, M.S., 1968, Philos. Trans. R. Soc. London, A, 262, 511
561:
562: %\bibitem[2005]{mie}Miesch, M.S., 2005, Living Rev. Sol. Phys., 2, 1 [Online Article]: http://www.livingreviews.org/lrsp-2005-1
563:
564: %\bibitem[Pedlosky(1987)]{ped}Pedlosky, J. 1987, Geophysical Fluid Dynamics (New York: Springer)
565:
566: \bibitem[Schecter et al.(2001)]{sch01}Schecter, D.A., Boyd, J.F., \& Gilman, P.A. 2001, \apj, 551, L185
567:
568: \bibitem[Spiegel \& Zahn(1992)]{spi92}Spiegel, E.A., \& Zahn, J.-P. 1992, A\&A, 265, 106
569:
570: \bibitem[Zaqarashvili et al.(2007)]{zaq07}Zaqarashvili, T.V.,
571: Oliver, R., Ballester, J.L. \& Shergelashvili, B. 2007, A\&A, 470,
572: 815
573:
574: \end{thebibliography}
575:
576: \clearpage
577:
578: \begin{figure}
579: \includegraphics[angle=0,scale=0.8]{f1.eps} \caption{Dispersion diagram of spherical shallow water waves in the
580: presence of a toroidal magnetic field for small $\epsilon$. The two
581: extreme upper and lower solutions (solid lines) correspond to
582: magnetic Poincar\'e waves, whereas the two low frequency modes shown
583: also in the zoom are fast (dotted) and slow (dashed) magnetic Rossby
584: waves. Symbols denote solutions obtained with the approximate
585: formulas (\ref{approx_mp_small_eps}) and (\ref{approx_fs_small_eps}). The parameters used to obtain the dispersion
586: diagram are $\epsilon =0.01$, $s=1$, and $\alpha_0=0.05$. }
587: \end{figure}
588:
589:
590: \begin{figure}
591: \includegraphics[angle=0,scale=0.8]{f2.eps} \caption{Upper panel:
592: frequencies of the $s=1, n=2$ and $s=1, n=3$ harmonics of fast
593: (dotted) and slow (dashed) magnetic Rossby modes vs. $\alpha_0$ for
594: $\epsilon=0.0045$. Magnetic Poincar\'e waves do not depend
595: significantly on the magnetic field, therefore they are not shown
596: here. Lower panel: frequencies of the $s=1, n=2$ and $s=1, n=3$
597: harmonics of magnetic Poincar\'e waves vs. $\epsilon$ for
598: $\alpha_0=0.05$. }
599: \end{figure}
600:
601:
602: \begin{figure}
603: \includegraphics[angle=0,scale=0.8]{f3.eps} \caption{Dispersion
604: diagram of spherical shallow water waves in the presence of a
605: toroidal magnetic field for large $\epsilon$. The two extreme upper
606: and lower solutions (solid lines) correspond to magnetic Poincar\'e
607: waves, whereas the low frequency modes (also shown with zoom) are
608: the fast (dotted) and slow (dashed) magnetic Rossby waves. Symbols
609: denote solutions obtained with the approximate formulas (\ref{approx_mp_large_eps}) and (\ref{approx_fs_large_eps}). The
610: parameters used to obtain the dispersion diagram are $\epsilon
611: =2700$, $s=1$, and $\alpha_0=0.05$. }
612: \end{figure}
613:
614: \begin{figure}
615: \includegraphics[angle=0,scale=0.7]{f4.eps} \caption{Upper panel:
616: frequencies of the $s=1, \nu=2$ and $s=1, \nu=3$ harmonics of fast
617: (dotted) and slow (dashed) magnetic Rossby modes vs. $\alpha_0$ for
618: $\epsilon=400$. Middle panel: frequencies of the $s=1, \nu=2$ and
619: $s=1, \nu=3$ harmonics of magnetic Poincar\'e and magnetic Rossby
620: modes vs. $\epsilon$ for $\alpha_0=0.05$. Lower panel: same as in
621: middle panel, but only for magnetic Rossby waves. }
622: \end{figure}
623:
624:
625: \end{document}
626:
627: %%
628: %% End of file `sample.tex'.
629: