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: %% Modified 2006/10/28 Version 0.58.01 by Jianghui Ji
19: %% Modified 2006/10/30 Version 0.58.01 by Shangli Ou
20: %% Modified 2006/11/09 Version 0.68.01 by Jianghui Ji
21:
22: %%\documentclass[12pt,preprint]{aastex}
23:
24: %% manuscript produces a one-column, double-spaced document:
25:
26: %%\documentclass[manuscript]{aastex}
27:
28: %% preprint2 produces a double-column, single-spaced document:
29:
30:
31: \documentclass[preprint]{aastex}
32: %%\documentclass[preprint2]{aastex}
33: %%\usepackage{natbib}
34:
35: \oddsidemargin -0.5in
36:
37: %% Sometimes a paper's abstract is too long to fit on the
38: %% title page in preprint2 mode. When that is the case,
39: %% use the longabstract style option.
40:
41: %% \documentclass[preprint2,longabstract]{aastex}
42:
43: %% If you want to create your own macros, you can do so
44: %% using \newcommand. Your macros should appear before
45: %% the \begin{document} command.
46: %%
47: %% If you are submitting to a journal that translates manuscripts
48: %% into SGML, you need to follow certain guidelines when preparing
49: %% your macros. See the AASTeX v5.x Author Guide
50: %% for information.
51:
52: \newcommand{\vdag}{(v)^\dagger}
53: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
54: \newcommand{\Ibar}{I\hspace{-0.4em}\rule[0.35em]{0.3em}{0.02em}\hspace{0.1em}}
55:
56: %% You can insert a short comment on the title page using the command below.
57:
58: %%\slugcomment{to be submitted to ApJ}
59:
60: %% If you wish, you may supply running head information, although
61: %% this information may be modified by the editorial offices.
62: %% The left head contains a list of authors,
63: %% usually a maximum of three (otherwise use et al.). The right
64: %% head is a modified title of up to roughly 44 characters.
65: %% Running heads will not print in the manuscript style.
66:
67: \shorttitle{RWI in Disk/Planet System}
68:
69: \shortauthors{Ou, Ji, Liu, Peng}
70:
71: %% This is the end of the preamble. Indicate the beginning of the
72: %% paper itself with \begin{document}.
73:
74: \begin{document}
75:
76: %% LaTeX will automatically break titles if they run longer than
77: %% one line. However, you may use \\ to force a line break if
78: %% you desire.
79:
80: \title{{Disk-Planet Interaction Simulations: (I) Baroclinic Generation of Vortensity and
81: Non-Axisymmetric Rossby-Wave-Instability}}
82:
83: %% Use \author, \affil, and the \and command to format
84: %% author and affiliation information.
85: %% Note that \email has replaced the old \authoremail command
86: %% from AASTeX v4.0. You can use \email to mark an email address
87: %% anywhere in the paper, not just in the front matter.
88: %% As in the title, use \\ to force line breaks.
89:
90:
91: \author{Shangli Ou\altaffilmark{1}, Jianghui
92: JI\altaffilmark{2,3,4}, Lin Liu\altaffilmark{5}, and Xiaomeng Peng\altaffilmark{6}}
93:
94: \altaffiltext{1}{High Performance Computing, Center for Computation and Technology /
95: Information Technology Services,
96: Louisiana State University, Baton Rouge, LA, 70803;ou@cct.lsu.edu}
97:
98: \altaffiltext{2}{Purple Mountain Observatory , Chinese Academy
99: of Sciences, Nanjing 210008,China;jijh@pmo.ac.cn}
100:
101: \altaffiltext{3}{National Astronomical Observatory, Chinese
102: Academy of Sciences, Beijing 100012, China}
103:
104: \altaffiltext{4}{Department of Terrestrial Magnetism, Carnegie
105: Institute of Washington, 5241 Broad Branch Road NW, Washington, DC
106: 20015-1305}
107:
108: \altaffiltext{5}{Department of Astronomy, Nanjing University,
109: Nanjing 210093, China}
110:
111: \altaffiltext{6}{Department of Physics \& Astronomy,
112: Louisiana State University, Baton Rouge, LA, 70803}
113:
114:
115: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
116: %% style, abstract will output a Received/Accepted line after the
117: %% title and affiliation information. No date will appear since the author
118: %% does not have this information. The dates will be filled in by the
119: %% editorial office after submission.
120:
121: \begin{abstract}
122:
123: We use a multi-dimensional hydrodynamics code to study
124: the gravitational interaction between an embedded planet and a protoplanetary disk
125: with emphasis on the generation of vortensity (potential vorticity) through a Baroclinic Instability and
126: subsequent development of Rossby-Wave-Instability (RWI).
127: It is found that the generation of potential vorticity
128: is very common and effective in non-barotropic disks
129: through the Baroclinic Instability, especially within the coorbital region.
130: Our results also complement previous studies by \citet{KLL03}
131: that non-axisymmetric RWIs are likely to develop
132: at local minima of potential vorticity distribution
133: that are generated by the interaction between a planet and a inviscid barotropic disk.
134: This second instability appears to be very common and robust, regardless of
135: the equation of state, initial density distribution, and rotational law of the disk.
136: The development of RWIs results in non-axisymmetric density blobs,
137: which exert stronger torques onto the planet when they travel
138: in the vicinity of the planet.
139: As a result of that, large amplitude oscillations are introduced
140: to the time behavior of the total torque acted on the planet by the disk.
141: In our current simulations, RWIs do not change the overall picture of inward
142: orbital migration but bring in a non-monotonic behavior to the migration speed.
143: As a side effect, RWIs also introduce interesting structures into the disk.
144: These structures may help the formation of Earth-like planets in the Habitable Zone
145: or Hot Earths interior to a close-in giant planet.
146:
147:
148: \end{abstract}
149:
150: \keywords{accretion, accretion disks --- Baroclinic Instability --- Rossby-wave instability --- hydrodynamics ---
151: numerical methods --- planetary systems: protoplanetary disks }
152:
153:
154:
155: \section{Introduction}
156: Although theorists did not realize its relevance to the origin of
157: planet when the study on the gravitational interaction between a
158: protoplanetary disk and an embedded planet just began in late
159: 1970s \citep{GT78}, it is now one of the key ingredients of
160: current standard theory of planet formation.
161: According to standard core-accretion theory \citep{Saf69,Lis93}, the
162: formation of planets in circumstellar disks around T Tauri stars
163: consists of the formation of planetesimals via
164: collisions/coalitions of dust grains in the early stage and then
165: gravitationally accretion after they accumulate enough mass. The
166: life time of this T Tauri star phase is estimated to be short
167: ($\lesssim 10^7$ years). In order for cores of protoplanets
168: with masses comparable to that of Jupiter to accumulate enough
169: mass in the T Tauri stage, it is thought that their cores have to
170: form outside the so-called ``ice line", where the distance to the
171: central star is large \citep[typically beyond $\sim 4$ AU, see][]{Ida04}
172: so that the temperature is low enough to allow the
173: condensation of gas materials to solid ice. These additional solid
174: dusts helps to increase the dust coagulation speed and shorten the
175: time needed to form cores of protoplanets.
176:
177:
178: Since the discovery of the first Jupiter-mass planet \citep{May95}
179: orbiting the solar-type star 51 Peg, more than 200 extrasolar planets have been
180: discovered around the nearby stars within 200 pc
181: \citep[] [The Extrasolar Planets Encyclopaedia]{But06}.
182: The diversity exhibited by
183: these planetary systems shows that the masses of these planets
184: range from Jupiter-mass to Neptune-mass. The most prominent
185: characteristics is that a large number of the host stars are
186: surrounded by the so-called ``hot-Jupiters" and close-in
187: super-Earths. Around $80\%$ of the extrasolar planets are in
188: orbits with semi-major axes in the range $0.01 \lesssim a \lesssim
189: 2.5$ AU, and $\sim 25\%$ of the total population are
190: short-period planets\footnote{see http://exoplanet.eu/} with $a
191: \lesssim 0.1$ AU. This has brought one of the most interesting
192: puzzles to theorists: if protoplanets had formed in a disk region
193: beyond $\sim 4$ AU from the central star, how did the observed
194: extrasolar planets end up with orbits that are so close to their
195: host stars? If the standard theory for the formation of
196: protoplanetary cores holds (Pollack et al. 1996; Ida \& Lin 2004),
197: then the giant planets (like hot-Jupiters) must have undergone
198: inward orbital migration to their current locations
199: \citep{Lin96}.
200:
201:
202: Several mechanisms have been proposed as the drive for the
203: migration. Some authors \citep{Wei96, RF96, For99}
204: suggested that three-body interaction
205: between a central star and two or more planets can lead to the
206: ejection of one of the planets, which takes away energy from the
207: system and leaves the third object on a smaller and eccentric
208: orbit, which is finally circularized through tidal interaction
209: between the planet and its host star. \citet{mur98} suggested that
210: resonance interactions between a planet and a disk of planetesimals can
211: also lead to inward orbital migration of the planet. Another plausible
212: explanation is provided by the theory of the gravitational
213: interaction between a protoplanetary disk and an embedded planet,
214: which was formulated, as mentioned in the beginning of this
215: article, more than a decade before the discovery of extrasolar
216: planets.
217:
218: In the third mechanism, a disk interacts with an embedded planet
219: through their mutual gravitational forces. The planet causes the
220: formation of spiral waves inside the disk at the Lindblad
221: resonances; as a result of the density asymmetry induced by spiral
222: waves, the inner disk exerts a positive gravitational torque onto
223: the planet and the outer disk exerts a negative gravitational
224: torque onto the planet. According to analytical analysis,
225: the overall torque is generally negative and, hence,
226: forces the planet to migrate inward.
227: \citet{War97} pointed out that two major kinds of migration exist
228: based on the mass of the embedded protoplanet.
229: In the so called Type I migration, the planet's mass is small and
230: the response of the disk is linear; the migration speed is very
231: fast so that the migration timescale is as short as $\sim
232: 10^{4}$ yr for a planetary core of $10M_{\oplus}$ in a sufficiently
233: viscous disk. In Type II migration, the protoplanet has enough
234: mass to open a gap inside the disk, and migrates on a much longer viscous
235: timescale.
236: Besides classical analytical analysis,
237: many groups have studied the nonlinear evolution of
238: a disk-planet system using numerical multi-dimensional hydrodynamics
239: \citep[][and references there in]{Kle99,NPMK00, Pap01, NW03, bate03}.
240: Recently, \citet{deval06} also carried out a collaborative multi-code comparison
241: research.
242: These numerical simulations showed that the nonlinear evolution of
243: the orbital migration of a planet inside a disk agrees with linear analysis
244: in a qualitative manner.
245:
246: However, nature is always not as simple as human beings have naively thought.
247: Unexpected phenomena often arose in the non-linear regime.
248: For example, \citet{mas01} and \citet{mas03} suggested that the corotation torque
249: originated from the coorbital region
250: may play a very different role from that of Lindblad torques;
251: this leads to a third kind of migration often
252: referred to as Type III migration or runaway migration,
253: in which the migration happens on a timescale as short as a few tens of orbits
254: and can be directed outward in some cases.
255: %%This could occur when the planet mass is sufficient to create a dip or
256: %%gap in its adjacent region in the condition that the surrounding
257: %%disk mass is more massive than the planet mass.
258: %%in cases where a planet is unable to clear most of materials
259: %%around its orbit to open a deep gap inside the disk,
260: \cite{klahr03} described a Baroclinic Instability in non-barotropic disks
261: that may contribute to vorticity and global trubulences;
262: They argued that strong vorticities may contribute to the
263: rapid formation of Jupiter-size gas planet \citep{klahr06}.
264: \citet{KLL03} and \citet{LL05} showed that the so-called ``Rossby-wave instabilities"
265: may develop at the local minima of potential vorticity (PV), or vortensity
266: (defined as the ratio between local vorticity and surface density),
267: in an inviscid disk with initially uniform vortensity distribution.
268:
269: In simulations of \citet{LL05},
270: these non-axisymmetric instabilities lead to the formation of vorticies and
271: density blobs, which exert stronger torque onto the planet when they travel
272: around its vicinity and bring large oscillations to the total torque
273: acted on the planet.
274: They further argued that this mechanism may be possible to change the direction
275: of the migration.
276: Non-axisymmetric Rossby-wave instabilities are also relevant to
277: the evolution of a single disk \citep{LLCN99, LFLC00} and
278: stellar models with strong differential rotation \citep{OT06}.
279: The time scale for RWIs to fully develop is generally on tens of dynamical time
280: of a disk ($\sim 100$ orbits), hence, they can not be ignored in inviscid or
281: low viscosity cases.
282: Therefore, it worths more labor to study their comprehensive role on planetary orbital migration
283: and the robustness of this mechanism.
284:
285: In this paper, we present results from simulations of disk-planet interaction.
286: Our purpose is two-folded:
287: first we wish to provide further comparisons to results presented in
288: the collaborative comparison work of \citet{deval06};
289: on the other hand, we focus more on the development of non-axisymmetric RWIs in
290: disk models with non-barotropic equation of state (EOS) and
291: non-uniform initial vortensity distribution,
292: as in contrast to disk models presented in \citet{KLL03} and \citet{LL05},
293: which are isothermal disks with uniform initial vortensity everywhere.
294: In particular, we find out that the PV generation
295: is more effective and common in disk models with non-barotropic EOS.
296: Our results, together with those presented by \citet{LL05},
297: suggest that the development of RWIs is very robust in disk-planet
298: systems with different mass distribution, rotational laws and EOSs.
299: We also address the impact of RWIs on planet orbital migration
300: and formation of close-in super Earths near a giant planet.
301: In section 2, basic equations, methods and initial models are described;
302: results from test runs are given in section 3;
303: section 4 concentrates on the development of RWIs under certain circumstances;
304: we close with conclusions and discussions in section 5.
305:
306:
307:
308:
309: \section{Basic Equations, Methods and Initial Setup}
310: To study the interaction between a disk and an embedded planet
311: requires coupling hydrodynamics and orbital dynamics together.
312: Here, we follow \citet{NPMK00} and many previous investigations
313: to reduce the problem to a two-dimensional (2D) one
314: since the disk is considered to be very thin.
315: Three dimensional (3D) investigations will be postponed to future.
316: The fluid motion inside the disk is described by the vertically integrated
317: continuity equation (\ref{conteq}), radial and azimuthal components of
318: the Navier-Stokes equation (\ref{NS_eq_r}) and (\ref{NS_eq_p}),
319: \begin{eqnarray}
320: \frac{\partial \Sigma} {\partial t} + \nabla \cdot (\Sigma \vec{v}) &=& 0 \label{conteq}\\
321: \frac{\partial(\Sigma v_r)} {\partial t} + \nabla \cdot (\Sigma v_r \vec{v})
322: &=& \frac{\Sigma v^2_{\phi}} {r} - \frac{\partial P}{\partial r}
323: - \Sigma \frac{\partial \Phi}{\partial r} + f_r \label{NS_eq_r}\\
324: \frac{\partial(\Sigma v_{\phi})} {\partial t} + \nabla \cdot (\Sigma v_{\phi} \vec{v})
325: &=& -\frac{\Sigma v_r v_{\phi}} {r} - \frac{1}{r} \frac{\partial P}{\partial \phi}
326: - \frac{\Sigma}{r} \frac{\partial \Phi}{\partial \phi} + f_{\phi} \label{NS_eq_p} \,,
327: \end{eqnarray}
328: where $\Sigma$ is disk surface density, $\vec{v}$ is two fluid
329: velocities, $P$ is vertically integrated pressure, $f_r$ and $f_{\phi}$ are two
330: components of viscous forces, and $\Phi$ is the gravitational
331: potential felt by fluid elements.
332: Details regarding viscous terms can be found in \citet{NPMK00}.
333: The EOS of the disk fluid is considered as locally isothermal \citep{NPMK00}
334: as given in below
335: \begin{equation}
336: P = c_s^2 \Sigma \,,
337: \end{equation}
338: where the local isothermal sound speed is $c_s = \frac{H}{r} \sqrt{GM_*/r}$
339: with disk aspect ratio $H/r=0.05$.
340: (As will be discussed in details later, this radial variation of sound speed
341: generates vorticity wherever an azimuthal density gradient exists.)
342:
343: We further simplify our study to
344: non-self-gravitational systems, in which the self-gravity of the
345: fluid is not taken into account for the fluid motion; hence, $\Phi
346: = \Phi_* + \Phi_p$, where $\Phi_*$ is the potential field of the
347: central star and $\Phi_p$ is the potential field of the planet,
348: which is given by $\Phi_p=- M_p /\sqrt{r^2+\epsilon^2}$, where
349: $M_p$ is the planet mass and
350: $\epsilon$ is taken to be 0.2 of the Roche Lobe of the planet.
351: The initial disk model has Keplerian rotational profile and uniform
352: density, which results in an initial radial vortensity profile
353: $\xi(r)$ that is proportional to $r^{-\frac{3}{2}}$ (vortensity is
354: defined as the ratio between local vorticity and surface density).
355: The value of density and viscosity are chosen to follow those
356: specified in \citet{deval06}.
357: The disk and the non-rotating coordinate system
358: are centered at the central star instead of the center of
359: mass of the system, which locates slightly off the central star.
360: To handle the hydrodynamics part, we adopted a legacy code
361: developed by the astrophysical group at Louisiana State
362: University to study star formation \citep{T80, WT88},
363: mass transferring binary stars \citep{MTF02},
364: and rotating instabilities in neutron stars\citep{OT06}.
365: The code is explicit and 2nd order in both space and time. It splits the
366: source term and advection term in a manner similar to Zeus
367: \citep{SN92}. Other features implemented include Van Leer upwind
368: scheme, artificial viscosity to handle shock, and, staggered
369: cylindrical grids. The code is originally three-dimensional,
370: but adapted to 2D in this work.
371: At the boundary of our computational grids,
372: mass is allowed to flow off the grids but no inflow is allowed.
373: We also implemented the wave-killing boundary condition
374: specified in \cite{deval06} for comparison purpose.
375:
376: The equation of motion for the planet is:
377: \begin{eqnarray}
378: \frac{d^2 \vec{r}_p} {dt^2} = -\frac{G(M_*+M_p)}{r_p^3} \vec{r}_p -
379: G \int \frac{\Sigma(\vec{r^\prime})}{(|\vec{r}_p - \vec{r^\prime} |^2 + \epsilon^2)^\frac{3}{2}} (\vec{r}_p - \vec{r^\prime} ) r^\prime dr^\prime d\phi
380: \end{eqnarray}
381: where the first term on the right is derived from the relative motion
382: of the planet to the central star,
383: the second term is the gravitational force exerted on the planet
384: by the disk.
385: For comparisons with previous investigations \citep{LL05,deval06},
386: we turned off the disk's attraction in many of our simulations
387: so that the planet stays on a fixed circular orbit at 1 $AU$;
388: for certain runs, the disk potential was turned on to allow
389: the planet to move so that we can assess RWI's role on the migration.
390: A 4th order Runge-Kutta integrator is used to integrate forward
391: the equation of motion of the planet during each Courant timestep
392: as required by the hydro part.
393: To compute the torque acted on the planet, we followed the specification in \citet{deval06},
394: \begin{equation}
395: T_z = GM_p \int \Sigma \vec{r}_p \times \frac{ (\vec{r^\prime}-\vec{r}_p )}
396: { ( |\vec{r^\prime}-\vec{r}_p |^2 + \epsilon^2)^\frac{3}{2}} r^\prime dr^\prime d\phi \,,
397: \end{equation}
398: torques from inside and outside the Roche-lobe of the planet are computed separately
399: for both inner disk and outer disk.
400: In the following sections, we show time evolution of the total torque including
401: contributions from within and outside the Roche-lobe,
402: the temporal behavior of the total torque does not change qualitatively when torques from within
403: the Roche-lobe are excluded.
404:
405: In equations of motion for both the planet and disk, we neglected the indirect terms
406: due to the acceleration to our coordinate system by the disk
407: (the planet's acceleration to the coordinate system is neglected
408: for the disk evolution as well).
409: Since the difference between the center of the mass and
410: our coordinate origin is fairly small, we expect the effect of indirect terms
411: won't be significant to change our results qualitatively;
412: hence we dropped these terms from our equations to save computing time.
413: (The evaluation of these indirect terms involves integrals over
414: the mass distribution and is therefore relatively expensive.)
415: Finally, the accretion onto the planet is neglected as well.
416:
417:
418: The units adopted are the following: the gravitational constant
419: $G=1$, length unit is 1 $AU$, and $M_*+M_p=1$. These setups help
420: us to have a direct comparison between our results and those
421: presented in \citet{deval06}.
422: %%We also adopted their wave-killing
423: %%boundary conditions and
424: A Jupiter mass is defined as $10^{-3}$ and a Neptune mass is
425: defined as $10^{-4}$. The planet's mass is turned on gradually in
426: the first 5 orbits.
427: As shown in \citet{LL05},
428: very high resolution is required to resolve RWIs.
429: This limits our ability to push simulations to much longer time scale.
430: In general, we only advance our simulations to $ \sim 200$ orbits,
431: which is sufficient for RWIs to fully develop.
432:
433:
434: \section{Test Runs for Jupiter mass planet}
435:
436: In this section, we present results for interaction between a disk
437: and a Jupiter mass planet on a fixed orbit as a calibration to our code.
438: According to previous analytical analysis and numerical nonlinear studies \citep{LP86,TBGLGB98,NPMK00},
439: a Jupiter mass planet
440: is capable of clearing most of the materials around its orbit and
441: opening a fairly wide and deep gap in a viscous/inviscid disk.
442: In order to test whether our code can simulate this correctly,
443: we have run this configuration for both viscous and inviscid disks
444: on grids with two different resolutions,
445: $128\times384$ and $256 \times 512$, to check for convergence.
446: To avoid lengthy descriptions, we only show results for a viscous disk.
447:
448: Figure \ref{fig:rho_vort_jup} displays radial profiles of surface density
449: $\Sigma$ and vortensity $\xi$ averaged over the azimuthal
450: direction at $t \approx 0, 100, 200$ orbits for the higher
451: resolution run. At the final age, a wide and deep gap has been
452: opened from the initial flat density profile, the contrast between
453: the lowest and highest density spans over around two orders of
454: magnitude. Inside the gap where the surface density is extremely
455: low, the vortensity has increased to a very high level. (Notice
456: that the vortensity is normalized to its initial value at $r=1$.)
457: Two vortensity minima formed around the inner and outer edge of
458: the gap, where RWIs are expected to grow. But since surface
459: density inside the gap is extremely low, we don't expect it to
460: have significant effect on the evolution.
461: (We note that the existence of viscosity also affects
462: the development of RWIs.)
463: To compare with plots
464: presented in \citet{deval06},
465: a logarithmic color map of surface density distribution of the disk
466: at $t \approx 200$ orbits
467: is shown in Figure \ref{fig:isoden_jup}.
468: It is observed that spiral arms originating from
469: the planet forms in both the inner and outer disks. These two
470: spiral arms are well-known Lindblad waves caused by the potential
471: of the planet and, in turn, exert a negative total torque onto the
472: planet as predicted by classical theory \citep{GT78, LP80}. The
473: structures shown in this plot agree qualitatively with those shown
474: in \citet[][see their Fig. 10]{deval06}.
475:
476: Figure \ref{fig:torq_jup} illustrates the evolution of the total torque
477: (per unit mass) exerted on the planet by the disk for both lower
478: and higher resolution runs. The data have been smoothed using a
479: moving box method over 10 orbit period \citep{NPMK00}. Their
480: behaviors agree closely with each other, which suggests that we
481: have achieved convergence on these results. As expected, the total
482: torque has a slightly negative value and will drive the planet to
483: migrate inward gradually. The magnitude of the torque falls in the
484: same range of those presented in a recent collaborative
485: comparison study \citep[][see their Table 6]{deval06}.
486: We also monitored the mass in the computational grids,
487: there are around 93\% of the initial mass left
488: at the end of the higher resolution run.
489:
490: In general, our test runs for Jupiter mass planet agree with previous investigations in a qualitative manner.
491: In the following, we will switch to the situation of a Neptune-mass planet embedded in a protoplanetary disk
492: and study the development of RWIs in such configurations.
493:
494:
495: \begin{figure}[]
496: %%\epsscale{1.0} \plotone{den.vort.5000.jupiter.ps}
497: %%\epsscale{1.0} \plotone{den_vort.multi.jupiter.ps}
498: \epsscale{1.0} \plotone{f1.eps}
499: \caption{The
500: top and bottom panels show, respectively, $\Sigma(R)$ and $\xi(R)$
501: of a viscous disk with a Jupiter mass planet averaged azimuthally at $t
502: \approx 0$ (solid line), $100$ (dotted line), and $200$ (dashed
503: line) orbits for the higher resolution run ($256 \times 512$).
504: \label{fig:rho_vort_jup} }
505: \end{figure}
506:
507: \begin{figure}[]
508: %%\epsscale{1.0} \plotone{vel5001.jupiter.ps}
509: \epsscale{1.0} \plotone{f2.eps}
510: \caption{Logarithmic
511: color map of surface density distribution for a viscous disk with
512: a Jupiter mass planet at $t\approx 200$ orbits for the higher
513: resolution run ($256 \times 512$). The horizontal axis is the
514: radial axis ranging from 0.4 to 2.5, and the vertical axis is the
515: azimuthal direction spanning over $2\pi$ range with the planet
516: shifted to around the middle.
517: A letter ``P" is labeled next to the location of the planet.
518: The color bar represents relative rather than absolute values.
519: }
520: \label{fig:isoden_jup}
521: \end{figure}
522:
523:
524: \begin{figure}[]
525: %%\epsscale{1.0} \plotone{torque.jupiter.converge.ps}
526: \epsscale{1.0} \plotone{f3.eps}
527: \caption{Time evolutions of the total torque (per unit mass) exerted on a Jupiter mass planet
528: by the disk for lower (solid line) and higher (dotted line) resolutions}
529: \label{fig:torq_jup}
530: \end{figure}
531:
532:
533:
534:
535:
536:
537: \section{High Resolution runs for RWI in protoplanetary disk with a Neptune-mass planet }
538:
539: \subsection{Results for simulations with a Neptune mass planet on a fixed orbit }
540: In this subsection, we present results from a series of
541: simulations in which a Neptune-mass planet is on a fixed circular orbit.
542: Previous investigations \citep{NPMK00,deval06} have shown that a
543: Neptune-mass planet can only creates a shallower and narrower
544: gap inside a disk due to its smaller mass. \citet{KLL03} and \citet{LL05}
545: further suggested that, in an inviscid disk with a Neptunian mass
546: planet, radial vortensity minima are important because they may
547: trigger RWIs to develop. As a result of RWI, non-axisymmetric
548: density blobs form and orbit around the
549: central star; the torque exerted on the planet by the disk can
550: undergo large amplitude oscillations as these density blobs pass
551: by the vicinity of the planet and apply gravitational
552: perturbations to it. As suggested by \citet{LL05}, this
553: mechanism may potentially slow down the inward migration of a
554: planet and, in some extremely cases, even change the sign of the
555: total torque exerted on the planet and, thus, cause it to migrate
556: outward.
557:
558: Since the disk models studied in \citet{KLL03} and \citet{LL05} have
559: an isothermal EOS and an initial
560: flat vortensity profile, we wish to investigate if the same
561: mechanism will also work in disks with non-barotropic EOS and non-uniform vortensity.
562: With this in mind, we adopted Keplerian disk models similar to
563: those of \citet{deval06}, which has a locally isothermal EOS and an initial uniform surface
564: density and, hence, an initial monotonically decreasing vortensity
565: profile as proceeding outward radially. We carried out simulations
566: of this configuration on grids with resolutions of $128\times
567: 384$, $400\times800$, and $400\times1600$;
568: a simulation with a resolution of $800 \times 3200$ was also performed for $\sim 70$ orbits
569: to check for the convergence of vortensity profile. The viscosity is turned
570: off in these simulations.
571:
572: In general, our simulations show that the evolution of the system consists of
573: two stages: the first stage is the formation of two spiral arms which exert
574: a negative torque to the planet; the second stage is the development of
575: RWIs in valleys (minima) of the radial vortensity profile
576: some time after the formation of spiral arms.
577: %%However, in the lower resolution run, RWIs are not as prominent as they are in higher
578: %%resolution runs, mainly because the interactions happened mostly around the co-orbital region,
579: However, fine structures are only resolved accurately in higher resolution runs,
580: mainly because interactions happen mostly around the co-orbital region
581: \citep{LL05}.
582: In the following, we show detailed results from a run with a resolution of $400\times1600$.
583:
584: Figure \ref{fig:rho_vort_nep} displays radial profiles of
585: $\Sigma$ and $\xi$ averaged over the azimuthal direction at $t
586: \approx 0, 60, 120$ orbits. In order to give a better view of fine
587: structures around the co-orbital region, only regions from $r=0.6$
588: to $r=1.4$ are shown in the plot. Compared to the case of a
589: Jupiter mass planet, a Neptune mass planet only opens a relatively
590: shallower and narrower density gap. The contrast between the lowest
591: and highest density is around a factor of two. Since the
592: surface density inside the gap is not quite low, the vortensity
593: there does not increase to much high level. (Notice that the
594: vortensity is normalized to its initial value at $r=1$.) However,
595: three vortensity valleys formed across this radial region, with
596: one centered around the planet's orbit. Because the density inside
597: the gap is still significant, RWIs resulted from these vortensity
598: minima will have important effect on the evolution of the system
599: \citep{LL05}.
600:
601: A linear color map of surface
602: density distribution on a polar plot \citep{LL05} at $t \approx
603: 150$ orbits is illustrated in Figure \ref{fig:isoden_nep}.
604: Focusing on the formation of high density
605: areas (red/brown regions), we observe not only red
606: Lindblad spiral arms, but also, other non-axisymmetric high
607: density structures at different locations: the inner edge of the
608: outer disk (brown arc-shaped region centered around 7 oclock) , the outer edge
609: of the inner disk (red/brown arc-shaped region around its edge);
610: interestingly, the density inside the gap is no longer
611: axisymmetric any more,
612: as suggested by the light blue region around 10 oclock.
613: The radial locations of these non-axisymmetric
614: structures match exactly with the local minima of the vortensity
615: distribution (see bottom portion of Figure \ref{fig:rho_vort_nep})
616: as predicted by linear analysis \citep{LFLC00}.
617:
618:
619: Figure \ref{fig:torq_nep} shows the time evolution of the total torque on the planet,
620: the torque from the inner and outer disk.
621: The behavior of the total torque is totally different from that in the Jupiter mass case:
622: very large amplitude oscillations present through most of the simulation.
623: As pointed out by \citet{LL05}, these oscillations are caused by
624: the gravitational pull from those orbiting density blobs;
625: each time when they pass the vicinity of the planet,
626: they exert a stronger torque onto the planet,
627: the sign of their torques depends on their relative position to the planet.
628: The existence of these oscillations complicates the classical picture of
629: the gravitational torque exerted on the planet by the disk.
630: %%which only took into account of torques from Lindblad waves.
631: %Although the overall time-averaged torque is still negative
632: %in our current much simplified simulations,
633: %We can not draw such a definite conclusion for realistic situations.
634:
635:
636: Figure \ref{fig:jn_nep} exhibits the radial profile of the specific
637: angular momentum $J$ inside the disk around the co-orbital region
638: at different time. The solid line denotes for initial Keplerian
639: profile, dotted, dashed, and dotted-dashed lines denote for
640: profiles at later times. As the evolution goes on, the inner disk
641: exerts a positive torque to the planet and, hence, is giving away
642: angular momentum to the planet; whereas, the outer disk exerts a
643: negative angular momentum to the disk, thus, is gaining
644: angular momentum from the planet; therefore, the angular momentum
645: is being transferred from the inner disk to the outer disk due to
646: the interaction between the planet and the disk. This mechanism
647: results in a dip in the inner disk's $J$ profile but a bump in the
648: outer disk's $J$ profile. As a result, an S-shaped structure
649: formed in the vicinity of the planet's fixed orbit.
650: Again, for comparison to previous studies,
651: there are around 98\% masses left in the grid
652: at the end of the highest resolution run.
653:
654: In summary, our results of the development of RWIs in
655: an inviscid disk with a Neptune-mass planet on a fixed orbit
656: are in agreement with those of \citet{LL05};
657: however, they differ in some key physical quantities.
658: In particular, the PV maxima in \citet[][see their Figure 2 and 9]{LL05} are located at radii outside
659: the coorbital region where shocks present;
660: whereas, PV maxima in our simulations reside within the coorbital region
661: around $0.95 < r< 1.05$, where shocks are not expected to occur.
662: This discrepancy raises some concerns.
663: Because based on the evolution equation of vorticity $\vec{\zeta}$:
664: \begin{equation}
665: \frac{d \vec{\zeta}}{dt} = \frac{\nabla \Sigma \times \nabla P}{\Sigma^2}
666: \end{equation}
667: the vorticity can only be generated at regions
668: where pressure gradient and density gradient do not align with each other;
669: such generation of vorticity normally happens around shocks in barotropic disks as illustrated in \citet{LL05},
670: which contributes to vortensity maxima outside the coorbital region.
671:
672: To understand the mechanism that generates vortensity (PV) maxima within the coorbital region
673: in our simulations,
674: we note that our disk models have a non-barotropic EOS
675: with a radial variation of sound speed $c_s$,
676: which also brings in misalignment of pressure gradient and density gradient
677: wherever azimuthal density gradient appears ($\nabla c_s \times \nabla \Sigma \ne 0$).
678: This misalignment acts as a source term for the generation of vorticity
679: within the coorbital region,
680: where no shock but strong azimuthal density gradient presents.
681: Therefore, the large density depression in the same region naturally
682: gives birth to vortensity increment.
683: According to \citet{LL05}, lower resolution may result in unphysical generation of vortensity.
684: To further test that if our results are resolution-dependent,
685: we carried out a run with resolution $800\times3200$ for $\sim 70$ orbits
686: (due to the limitation of computational resources).
687: Figure \ref{fig:vorten_conv} shows the vortensity profile around coorbital regions
688: for runs with different resolutions at $t \approx 70$ orbits.
689: They all exhibit vortensity maxima within the coorbital region,
690: which indicates that our results are resolution-independent.
691: As a further evidence that vortensity is generated around where azimuthal density gradient exists,
692: Figure \ref{fig:vorten_HR} and \ref{fig:drho_HR} illustrate the distribution of
693: vortensity increment and azimuthal density gradient, respectively,
694: for the run with resolution $800 \times3200$ at $ t \sim 70$ orbits.
695: For a better view, they are zoomed in the neighborhood of the planet.
696: It is observed that vortensity is generated within the Roche lobe,
697: where strong azimuthal density gradient, instead of shocks, exists.
698: Such a generating mechanism of vortensity inside a planet's Roche lobe
699: has the same origin as baroclinic instability discussed in \citet{klahr03},
700: which contributes to global turbulence within the disk.
701: On the other hand, we also note that the oscillations of the torque acted on a neptune mass planet
702: appear much earlier in our simulations ($\sim$ 50 orbits in our Figure 6)
703: than they do in \citet[][$\sim$ 200 orbits in their Figure 6]{LL05},
704: which suggests that PV (vortensity) generation
705: is even more effective and common in non-barotropic disk models.
706:
707: As show in Figure \ref{fig:drho_HR},
708: the azimuthal density gradient flips the sign across the planet;
709: thus, we expect the vortensity increment also flip the sign there.
710: However,
711: the vortensity change within the Roche-lobe shown in Figure \ref{fig:vorten_HR}
712: is always positive.
713: We don't fully understand the underlying physics of this phenomenon,
714: but suspect that this is possiblely due to vortensity mixture along
715: librating stream lines in the horse region,
716: i.e., where fluid elements make the U-turn.
717: A detailed and vigrous analysis of fluid motion and vortensity mixture around
718: a Neptune-mass planet embedded in a non-barotropic disk
719: is beyond the scope of this paper,
720: and will be studied in a separate investigation.
721:
722: %%softenning factor.
723:
724: %%It should also be pointed out that the way we chose to soften the planet potential
725: %%is different from that in \citet{LL05},
726: %%this may also change the location of vortensity maxima to some extent.
727:
728:
729: Since the generation of vortensity in non-barotropic disks turns out to be quite universal,
730: the second instability (RWIs) triggered around vortensity extrema is probably more common
731: in realistic situations than what has been thought previously;
732: it will have strong influence on the migration of
733: a neptune mass planet in an inviscid disk.
734: Next, we will discuss the case in which the planet is allowed
735: to move freely under the torque action from the disk.
736:
737: \begin{figure}[]
738: %%\epsscale{1.0} \plotone{den.vort.3300.neptune.ps}
739: %%\epsscale{1.0} \plotone{den_vort.multi.neptune.ps}
740: \epsscale{1.0} \plotone{f4.eps}
741: \caption{The top and bottom panels show, respectively, $\Sigma(R)$
742: and $\xi(R)$ of the disk with a Neptune mass planet averaged azimuthally
743: at $t \approx 0$ (solid line), $60$ (dotted line), and $120$ (dashed line) orbits.
744: \label{fig:rho_vort_nep} }
745: \end{figure}
746:
747:
748:
749: \begin{figure}[]
750: %%\epsscale{1.0} \plotone{vel4000.neptune.ps}
751: \epsscale{1.0} \plotone{f5.eps}
752: \caption{The same as Fig. \ref{fig:isoden_jup} but in linear scale
753: for an inviscid disk with a Neptune mass planet at $t\approx 150$ orbits.
754: A letter ``P" is labeled next to the location of the planet.
755: The color bar represents relative rather than absolute values.}
756: \label{fig:isoden_nep}
757: \end{figure}
758:
759: \begin{figure}[]
760: %%\epsscale{1.0} \plotone{torque.neptune.ps}
761: \epsscale{1.0} \plotone{f6.eps}
762: \caption{Time evolutions of total torque (top), torques from the
763: inner (middle) and outer (bottom) disks exerted on a Neptune mass planet
764: for the highest resolution run ($400 \times 1600$). Raw data is shown.}
765: \label{fig:torq_nep}
766: \end{figure}
767:
768:
769: \begin{figure}[]
770: %%\epsscale{1.0} \plotone{jn.neptune.ps}
771: \epsscale{1.0} \plotone{f7.eps}
772: \caption{Radial profiles of
773: specific angular momentum around the co-orbital region of the disk
774: with a Neptune mass planet at different times. The solid line
775: denotes for initial Keplerian profile, dotted, dashed, and
776: dotted-dashed lines denote for profiles at later times. Data have
777: been normalized to the initial value at $r=1$. As evolutions goes
778: on, angular momentum is being removed in the inner region but
779: added to the exterior region due to the interaction between the
780: disk and the planet.} \label{fig:jn_nep}
781: \end{figure}
782:
783:
784: \begin{figure}[]
785: %%\epsscale{1.0} \plotone{vorten.converge.eps}
786: \epsscale{1.0} \plotone{f8.eps}
787: \caption{Vortensity profiles around the coorbital regions for runs with different resolutions
788: at $t \approx 70$ orbits.
789: All of them show maxima within the coorbital region.
790: } \label{fig:vorten_conv}
791: \end{figure}
792:
793:
794: \begin{figure}[]
795: %%\epsscale{1.0} \plotone{vorten1350.eps}
796: \epsscale{1.0} \plotone{f9.eps}
797: \caption{Distribution of vortensity increment around the planet at $t \approx 70 $ orbits
798: for the run with resolution $800 \times 3200$.
799: (The initial backrougnd vortensity is subtracted.)
800: Shocks are well resolved. Vortensity is generated within the coorbital region,
801: especially in the Roche lobe, where no shock exists.
802: The azimuthal density gradient within the same region
803: acts as a source term of vortencity generation (see Figure \ref{fig:drho_HR}).
804: Note that the vortensity increment within the Roche-lobe is always positive,
805: which contradicts our expectation. We suespect this may be resulted from
806: vortensity mixture in this region.
807: } \label{fig:vorten_HR}
808: \end{figure}
809:
810: \begin{figure}[]
811: %%\epsscale{1.0} \plotone{drhodp1350.eps}
812: \epsscale{1.0} \plotone{f10.eps}
813: \caption{Distribution of azimuthal density gradient at the same time
814: as that of Figure \ref{fig:vorten_HR}. Strong density gradient is observed
815: within the Roche lobe and coorbital region.
816: } \label{fig:drho_HR}
817: \end{figure}
818:
819:
820:
821:
822: \subsection{Simulation for a freely moving Neptune-mass planet}
823:
824: In order to assess possible roles of RWIs on the migration of a planet,
825: we carried out two simulations in which a Neptune mass planet
826: is allowed to move under the influence of
827: the gravitational torque from the disk.
828: The resolutions are $400 \times 800$ and $400 \times 1600$.
829: In these two simulations, the planet is fixed on the initial circular orbit
830: for the first 50 orbits (slightly longer than the libration period for the
831: neptune mass planet) to allow the disk to respond properly,
832: it is then released to move freely under the influence of the torque from the disk.
833: The results from these two simulations are similar to each other,
834: which suggests that convergence is achieved.
835: We use the higher resolution run to present the evolution of the system.
836:
837: Figure \ref{fig:rho_vort_nep_move} illustrates radial profiles of $\Sigma$ and
838: $\xi$ averaged over the azimuthal direction at $t \approx 50, 100, 150, 200$ orbits.
839: It is observed that at $t=50$ orbits
840: the planet has created a shallow gap within the disk around $r=1$,
841: the density level is enhanced at both the inner and outer edge of the gap
842: ($r\approx 0.88$ and $r\approx 1.12$).
843: Then, the planet migrates inward when $t>50$ orbits as an effect of an
844: overall negative total torque. As it moves in, the surface density
845: gap and vortensity valleys also migrate with it. On the other
846: hand, a shallower density gap and a vortensity valley are still preserved around its
847: initial orbital location.
848:
849: %%Figure \ref{fig:isoden_nep_move} shows
850: %%the linear color map of density distribution of the disk at $t
851: %%\approx 200$ orbits. Besides the Lindblad waves that follow the
852: %%planet, non-axisymmetric density enhancements (red/brown ring
853: %%regions) form near the planet's current and original orbital locations,
854: %%where local vortensity minima are located.
855:
856: Figure \ref{fig:torq_nep_move} exhibits the time evolution of the total torque acted
857: on the planet and its radius to the central star.
858: The overall torque exerted on the planet is negative in average throughout the simulation,
859: but undergoes some changes.
860: After the whole system settles down in the first 50 orbits,
861: large amplitude oscillations with certain period show up;
862: the torque becomes increasingly negative when the planet starts to migrate inward;
863: it then reaches its peak in magnitude at $t \approx 100$ orbits
864: ($\sim 50$ orbits after releasing);
865: after that, it is weakened and then returns back to a very small negative value
866: with smaller oscillations that have a different period from that of earlier oscillations.
867: %%which suggests that these two types of oscillations are generated by
868: %%different density blobs.
869:
870: %%due to the development of RWIs.
871:
872: We carefully analyzed the evolution between $t= 80 \sim 120 $ orbits
873: and found that the sudden change of the torque's magnitude
874: is due to the development of a relatively large density blob
875: between the planet and the edge of inner disk as a result of RWIs.
876: %%with an orbital frequency slightly faster than that of the planet.
877: During this time, the density blob accumulates materials and augments very quickly
878: because the planet is penetrating the high density region
879: near $ r \sim 0.88$ (which is introduced by the planet's potential during the first
880: $50$ orbits, see the density profile at $t=50$ orbits).
881: Consequently, the density blob exerts a stronger torque onto the planet
882: causing the oscillation amplitude of the total torque to increase,
883: the averaged torque also becomes increasingly negative during $t=80 \sim 100$ orbits.
884: After the planet passed the high density region at $t \approx 100$ orbits,
885: the large density blob saturates and wanes;
886: the torque gradually returns back to a value close to zero.
887: However, smaller density blobs do survive and contribute to smaller oscillations
888: of the total torque in the late evolution.
889: Figure \ref{fig:isoden_nep_move1} shows the density colormaps of the disk at $t\approx 100$ orbits
890: when the planet is penetrating the high density region,
891: a pronounced density blob (green arc-like structure) in the edge of the inner disk is clearly visible.
892: For comparison, similar plot for $t \approx 200$ orbits is shown in Figure \ref{fig:isoden_nep_move2},
893: where previous large density blob diminishes and only small density blob survives
894: both in the inner disk and outer disk.
895: In order to have a better view of the time behavior of the density azimuthal asymmetry
896: between the planet and the outer edge of the inner disk,
897: the radially-averaged density azimuthal distributions of this ring-belt
898: at different times are ploted in Figure \ref{fig:rhophi_nep_move2}.
899: Besides some sharp spikes that signal the locations of the planet,
900: density azimuthal asymmetry with different degrees is clearly seen.
901: We note that the density blob has the highest value at $t\approx 104$ orbits,
902: which coincides with the time when the torque has the negative maximum value
903: and when the planet is penetrating the high density region at $r\approx0.88$.
904: This further supports our analysis on the correlation between the torque magnitude
905: and the augmentation of the density blob.
906:
907: As a consequence of this interesting phenomenon,
908: the planet's radius decays quickly during $t = 80 \sim 120 $ orbits,
909: but the decaying slows down after $t \approx 110$ orbits
910: (see bottom panel of Figure \ref{fig:torq_nep_move} for this non-monotonic behavior).
911: We measured the migration timescale $\tau_{mig}$,
912: which is defined as the time needed for the planet to migrate
913: from $r=1$ to the central star given a drifting speed
914: measured at the end of our simulation.
915: Our measurement yields $\tau_{mig} \approx 1300$ orbits,
916: which is almost an order of magnitude lower than
917: that of type II migration from a similar study
918: \citep[][see their Table 1.
919: Note that their planet also locates at $r=1$ initially.]{NPMK00}.
920: This timescale is consistent with type I migration timescale,
921: which ought to be one or two orders of magnitude lower than that of type II
922: \citep{War97}.
923:
924: We note that there is a trend for the averaged torque to approach zero
925: at late times, this is probably because the mass of the disk interacting
926: with the planet decreases as it moves in.
927: On the other hand, the oscillation amplitude of the torque induced by
928: RWIs remains relatively constant, it is possible that
929: as the planet moves closer to the host star
930: a turning point may eventually appear, where the RWI-induced torque
931: beats over the Lindblad torque.
932: %%This may greatly reduce the migration rate if the system is evolved for a long time.
933: If this is true, the planet may be halted at certain radial range.
934: Since the planet is very close to the inner edge of our computational grid
935: at the end of this simulation,
936: we decided to stop the simulation and
937: will address these issue in future investigations.
938:
939: %%The ``releasing time" for the planet may also plays a role in influencing
940: %%the migration of the planet in the sense that the longer the planet is ``holded"
941: %%the deeper the density gap is,
942: %%therefore, when the planet is penetrating the inner gap edge
943:
944:
945: %%A third run was performed with resolution $400 \times 800$,
946: %%in which the planet was released from the very beginning to test
947: %%if the results would depend on the ``releasing time".
948: %%It turns out that although the disk's initial response to the planet potential
949: %%in this third run is not handled properly, it does not change the results qualitatively.
950: %%This suggests
951:
952: %%An illustration of the radial profile of the specific angular momentum
953: %%at different times is presented in Figure \ref{fig:jn_nep_mov}.
954: %%The angular momentum distribution around the orbital location
955: %%of the planet is changed through its interaction with
956: %%the disk and results in an ``S" shape.
957: %%As the planet migrates inward, the S-shaped structure also migrates
958: %%with it and changes the angular momentum distribution in
959: %%a wide scope of the disk.
960:
961: In the limited time evolutions of a freely moving planet presented here,
962: RWIs do not change the overall picture of inward migration;
963: but they have significant influence on the torque exerted on the planet
964: and consequently make the migration speed of the planet non-monotonic.
965: Further investigations are needed to study the influence of other factor
966: that may potentially bring in significant changes,
967: such as the ``releasing time" of the planet, self gravity of the disk fluid, etc.
968: One important issue is how to properly evaluate the torque from materials near the planet,
969: which is embedded in the disk.
970: In this paper, we computed this bruteforcely in 2D.
971: However, many previous studies \citep{mas02,DANGELO05,LL05} used quasi-3D treatment,
972: which tends to reduce the torque.
973: It is very likely that if proper 3D treatment is adopted,
974: the total torque will be reduced and the influence of RWIs will be much stronger.
975: This could potentially cause the migration to be reversed.
976:
977:
978:
979: \begin{figure}[]
980: %%\epsscale{1.0} \plotone{den_vort.multi.nep.mov.ps}
981: \epsscale{1.0} \plotone{f11.eps}
982: \caption{The same as Fig. \ref{fig:rho_vort_nep} but for a run at
983: $t \approx 50$ (solid line), $100$ (dotted line), $150$ (dashed
984: line), and $200$ (dash-dotted line) orbits, in which a Neptune mass planet is allowed to move after 50 orbits.
985: \label{fig:rho_vort_nep_move} }
986: \end{figure}
987:
988: \begin{figure}[]
989: \epsscale{1.0} \plotone{f12.eps}
990: %%\epsscale{1.0} \plotone{torque.neptune.mov.eps}
991: \caption{Time evolutions of the total torque (top) exerted on a Neptune mass planet
992: by the disk and the radius of the planet (bottom) for the run
993: in which the planet is allowed to move. Raw data is shown.
994: A non-monotonic behavior for the migration speed is observed.
995: \label{fig:torq_nep_move} }
996: \end{figure}
997:
998:
999:
1000: \begin{figure}[]
1001: %%\epsscale{1.0} \plotone{vel1516.nep.mov.eps}
1002: \epsscale{1.0} \plotone{f13.eps}
1003: \caption{The same
1004: as Fig. \ref{fig:isoden_nep} but at $t = 103$ orbits for a
1005: run, in which a Neptune mass planet is allowed to move.
1006: The planet is penetrating the high density region around $r \sim 0.88$
1007: and a fairly large density blob forms (green arc-like structure
1008: at the edge of inner disk).
1009: \label{fig:isoden_nep_move1} }
1010: \end{figure}
1011:
1012: \begin{figure}[]
1013: %%\epsscale{1.0} \plotone{vel2010.nep.mov.eps}
1014: \epsscale{1.0} \plotone{f14.eps}
1015: \caption{The same
1016: as Fig. \ref{fig:isoden_nep_move1} but at $t = 202$ orbits.
1017: The planet has passed the original high density region around $r \sim 0.88$.
1018: Previous large density blob wanes, smaller density blobs survive
1019: at several locations.
1020: \label{fig:isoden_nep_move2} }
1021: \end{figure}
1022:
1023: \begin{figure}[]
1024: %%\epsscale{1.0} \plotone{rho_phi.neptune.move.eps}
1025: \epsscale{1.0} \plotone{f15.eps}
1026: \caption{The radially-averaged azimuthal density distribution of the ring belt
1027: between the planet and the outer edge of the inner disk from a run
1028: in which a neptune mass planet is allowed to move.
1029: Note that the density blob has the highest value at $\phi\approx 3$ and $t\approx104$ when the planet
1030: is penetrating the high density region around $r\approx 0.88$.
1031: (The spikes reveal the locations of the planet.)
1032: \label{fig:rhophi_nep_move2} }
1033: \end{figure}
1034:
1035:
1036:
1037:
1038: %%\begin{figure}[]
1039: %%\epsscale{1.0} \plotone{jn.neptune.move.ps}
1040: %%\epsscale{1.0} \plotone{f11.eps}
1041: %%\caption{Radial
1042: %%profiles of specific angular momentum around the co-orbital region
1043: %%of the disk with a Neptune mass planet at different times. The
1044: %%solid line denotes for initial Keplerian profile, dotted and
1045: %%dashed lines denote for profiles at later times. Data have been
1046: %%normalized to the initial value at $r=1$. }
1047: %%\label{fig:jn_nep_mov}
1048: %%\end{figure}
1049:
1050:
1051: \section{Conclusion and Discussion}
1052:
1053: We have carried out high resolution simulations on the interaction
1054: between a protoplanetary disk and an embedded planet
1055: with emphasis on detailed interplay between a disk and a planet
1056: under the influence of baroclinic generation of vortensity and non-axisymmetric RWIs.
1057: Our results are consistent with classical analysis on
1058: the interaction between a protoplanetary disk and an embedded planet
1059: through Lindblad torques.
1060: We confirmed previous studies by \citet{KLL03} and \cite{LL05}
1061: that non-axisymmetric RWI is likely to develop under certain circumstances
1062: and have an important influence on the migration of a planet
1063: inside an inviscid disk.
1064: We also found that the generation of vortensity (PV)
1065: is more common and effective in disks with non-barotropic EOS
1066: through the baroclinic instability,
1067: which further favors the development of RWIs.
1068: As the asymmetry of the density distribution induced by RWIs becomes prominent,
1069: the resulting density blobs exert periodical and enhanced gravitational pull onto the planet
1070: as they pass by the vicinity of the planet,
1071: which causes the total torque received by the planet undergo large amplitude oscillations.
1072: Although in our current simulations,
1073: RWIs did not change the overall picture of an inward migration,
1074: they definitely have very important and interesting effect on the migration speed.
1075: As a side effect of an inwardly migrating planet,
1076: RWIs introduce nonaxisymmetric density blobs along the way.
1077: %%(not just limited to the vicinity of the planet).
1078: These enhanced density blobs with strong vortices may help rapid formation
1079: of new planet cores within them in a way suggested in \citep{klahr06, petersen06},
1080: %%the baroclinic instability
1081: especially in inner regions of circumstellar disks
1082: where rapid precipitation and coagulation of solid materials
1083: are likely to happen \citep{sil06}.
1084: If these new-born cores could survive
1085: during the migration of a giant planet \citep{cham06,ray06},
1086: they may produce Earth-like planets in the Habitable Zones \citep{Ji06}
1087: or Hot Earths interior to a close-in giant planet \citep{ray06}.
1088:
1089: The study presented here is just a very limited step among many
1090: efforts toward understanding disk-planet interactions.
1091: Compared to realistic situations, there are so many physics
1092: missing in our much-simplified simulations,
1093: the influence of RWIs
1094: on planet migration is still not clear in cases where all the
1095: physics are correctly taken into account. For example, if the
1096: self-gravity of the disk ignored in this study is included in the
1097: dynamics of the fluid, it is likely to enhance the RWIs because
1098: the gravitational potential field is made more asymmetric by the
1099: non-axisymmetric density distribution and more masses will be
1100: trapped into potential wells of these density blobs. Enhanced
1101: RWIs will exert a stronger positive torque onto the planet and
1102: may greatly reduce the migration speed.
1103: On the other hand, because RWI develops mostly around the coorbital region
1104: where corotation torques are originated,
1105: it may have important effect on type III migration.
1106: %%Finally, things become more interesting in cases where
1107: %%multiple planets present in the disk and
1108: %%RWIs induced by different planets interact with each other.
1109: %%We hope to study these situations in near future.
1110: These issues need to be addressed in future investigations.
1111:
1112:
1113:
1114:
1115: \acknowledgments
1116:
1117: We thank Hui Li for useful discussions and suggestions during the course of this study.
1118: We also thank Joel Tohline for helpful comments and
1119: kind permission to use LSU astrophysical hydrodynamics code.
1120: We appreciate an anonymous referee for suggesting
1121: various ways to analyze our results.
1122: S.O. was partially supported by NSF grant AST-0407070 and Center for Computaion
1123: and Technology at Louisiana State University. J.H.J. is much
1124: grateful to Paul Butler for the hospitality during the visit stay
1125: at DTM and also appreciates the astronomy group therein. J.H.J.
1126: acknowledges the financial support by the National Natural Science
1127: Foundations of China (Grants 10573040, 10673006, 10203005,
1128: 10233020) and the Foundation of Minor Planets of Purple Mountain
1129: Observatory. The computations were performed on the Supermike
1130: cluster, the Pelican cluster, the Nemeaux cluster at LSU,
1131: and IBM P5 clusters of Louisiana Optical Network Initiative.
1132:
1133:
1134:
1135: %%This work was partially supported by
1136: %%National Computational Science Alliance (NCSA) under grant MCA98N043.
1137: %%The computations were performed on the LSU
1138: %%Superhelix cluster and on NCSA's Tungsten cluster.
1139:
1140:
1141: %%Most of the simulations were carried out on
1142: %%SuperMike and SuperHelix at LSU, which are facilities operated by
1143: %%the Center for Computation and Technology whose funding largely
1144: %%comes through appropriations by the Louisiana state legislature.
1145:
1146: %% The reference list follows the main body and any appendices.
1147: %% Use LaTeX's thebibliography environment to mark up your reference list.
1148: %% Note \begin{thebibliography} is followed by an empty set of
1149: %% curly braces. If you forget this, LaTeX will generate the error
1150: %% "Perhaps a missing \item?".
1151: %%
1152: %% thebibliography produces citations in the text using \bibitem-\cite
1153: %% cross-referencing. Each reference is preceded by a
1154: %% \bibitem command that defines in curly braces the KEY that corresponds
1155: %% to the KEY in the \cite commands (see the first section above).
1156: %% Make sure that you provide a unique KEY for every \bibitem or else the
1157: %% paper will not LaTeX. The square brackets should contain
1158: %% the citation text that LaTeX will insert in
1159: %% place of the \cite commands.
1160:
1161: %% We have used macros to produce journal name abbreviations.
1162: %% AASTeX provides a number of these for the more frequently-cited journals.
1163: %% See the Author Guide for a list of them.
1164:
1165: %% Note that the style of the \bibitem labels (in []) is slightly
1166: %% different from previous examples. The natbib system solves a host
1167: %% of citation expression problems, but it is necessary to clearly
1168: %% delimit the year from the author name used in the citation.
1169: %% See the natbib documentation for more details and options.
1170:
1171: \begin{thebibliography}{}
1172: %%\bibitem[Bayly(1986)]{B86} Bayly, B. J., 1986, \prl, 57, 2160
1173:
1174: \bibitem[Bate et al.(2003)]{bate03}Bate, M.R., Lubow, S.H., Ogilvie, G.I., Miller, K.A.,
1175: 2003, \mnras, 341, 213
1176:
1177: \bibitem[Butler et al.(2006)]{But06} Butler, R. P., et al. 2006, \apj, 646, 505
1178:
1179: \bibitem[Chambers(2006)]{cham06} Chambers, J.E., 2006, \apj, 652, in press, astro-ph/0610905
1180:
1181: \bibitem[D'Angelo et al.(2005)]{DANGELO05} D'Angelo, G., Bate, M.R., Lubow, S. H., 2005, \mnras, 358,316
1182:
1183: \bibitem[Ford et al.(1999)]{For99}Ford, E. B., Rasio, F. A., \& Sills, A. 1999, \apj, 514, 411
1184:
1185: \bibitem[Goldreich \& Tremaine(1978)]{GT78} Goldreich, P., Tremaine, S., 1978, \apj, 222, 850
1186:
1187: \bibitem[Ida \& Lin(2004)]{Ida04} Ida, S., \& Lin, D.N.C. 2004, \apj, 604, 388
1188:
1189: \bibitem[Ji et al.(2006)]{Ji06} Ji, J. H., Kinoshita, H., Liu, L., \& Li, G.Y., 2007, \apj, 657, 1092
1190: %% in press, astro-ph/0611008
1191:
1192: \bibitem[Klahr \& Bodenheimer(2003)]{klahr03} Klahr, H.H., Bodenheimer, P., 2003, \apj, 582, 869
1193:
1194: \bibitem[Klahr \& Bodenheimer(2006)]{klahr06} Klahr, H.H., Bodenheimer, P., 2006, \apj, 639, 432
1195:
1196: \bibitem[Kley(1999)]{Kle99} Kley, W.\ 1999, \mnras, 303, 696
1197:
1198: \bibitem[Kley et al.(2001)]{kley01}Kley, W., D'Angelo, G., Henning, T., 2001, \apj, 547, 457
1199:
1200: \bibitem[Koller et al.(2003)]{KLL03}Koller, J., Li, H., Lin, D.N.C., 2003, \apj, 596, 91
1201:
1202: \bibitem[Li et al.(2000)]{LFLC00}Li, H., Finn, J. M., Lovelace, R. V. E.,
1203: \& Colgate, S. A., 2000, \apj, 533, 1023
1204:
1205: \bibitem[Li et al.(2001)]{LCWL01}Li, H., Colgate, S. A., Wendroff, B., Liska, R., 2001, \apj, 551, 874
1206:
1207: \bibitem[Li et al.(2005)]{LL05}Li, H., Li, S., Koller, J., Wendroff, B. B., Liska, R.
1208: Orban, C. M., Liang, E. P.T., Lin, D. N.C., 2005, \apj, 624, 1003
1209:
1210:
1211: \bibitem[Lin \& Papaloizou(1980)]{LP80}Lin, D.N.C., Papaloizou, J.C.B., 1980, \mnras, 191, 37
1212:
1213: \bibitem[Lin \& Papaloizou(1986)]{LP86}Lin, D.N.C., Papaloizou, J.C.B., 1986, \apj, 309,846
1214:
1215: \bibitem[Lin et al.(1996)]{Lin96} Lin, D. N. C., Bodenheimer, P., \& Richardson, D. C. 1996, \nat,
1216: 380, 606
1217:
1218:
1219: \bibitem[Lissauer(1993)]{Lis93}
1220: Lissauer, J.J. 1993, \araa, 31, 129
1221:
1222: \bibitem[Lovelace \& Hohlfeld(1978)]{LH78}Lovelace, R. V. E., Hohlfeld, R. G., 1978, \apj, 221, 51
1223:
1224: \bibitem[Lovelace et al.(1999)]{LLCN99}Lovelace, R. V. E., Li, H., Colgate, S. A.,
1225: Nelson, A. F., 1999, \apj, 513, 805
1226:
1227: \bibitem[Masset(2001)]{mas01}Masset, F.S., 2001, \apj, 558, 453
1228:
1229: \bibitem[Masset(2002)]{mas02}Masset, F.S., 2002, A\&A, 387, 605
1230:
1231: \bibitem[Masset \& Papaloizou(2003)]{mas03}
1232: Masset, F.S., Papaloizou, J.C.B., 2003, \apj, 588, 494
1233:
1234: %%\bibitem[Masset et al.(2006)]{mas06}Masset, F.S., Morbidelli, A.M., Crida, A., Ferreira, J., 2006, \apj, 643, 478
1235:
1236: \bibitem[Mayor \& Queloz(1995)]{May95} Mayor, M., \& Queloz, D. 1995, \nat, 378, 355
1237:
1238: \bibitem[Motl, Tohline, \& Frank(2002)]{MTF02} Motl, P. M.,
1239: Tohline, J. E., \& Frank, J. 2002, \apjs, 138, 121
1240:
1241: \bibitem[Murray et al.(1998)]{mur98}Murray, M., Hansen, B., Holman, M., Tremaine, S., 1998, Science, 279, 69
1242:
1243: \bibitem[Nelson et al.(2000)]{NPMK00}Nelson, R.P., Papaloizou, J.C.B., Masset, F., Kley, W., 2000, \mnras, 318, 18
1244:
1245: \bibitem[Nelson \& Willy(2003)]{NW03}Nelson, A., Willy, B., 2003, \apj, 589, 543
1246:
1247: \bibitem[Ou \& Tohline(2006)]{OT06}Ou, S., Tohline, J.E., 2006, \apj, 651, 1068
1248:
1249: \bibitem[Papaloizou, Nelson, \& Masset(2001)]{Pap01} Papaloizou, J. C. B., Nelson, R. P., Masset, F., 2001,
1250: A\&A, 366, 263
1251:
1252: \bibitem[Petersen et al.(2006)]{petersen06}Petersen, M.R., Stewart, G.R., Julien, K., astro-ph/0611526
1253:
1254: \bibitem[Pollack et al.(1996)]{Pol96}Pollack, J.~B., Hubickyj, O., Bodenheimer, P., Lissauer, J.~J.,
1255: Podolak, M., \& Greenzweig, Y.\ 1996, Icarus, 124, 62
1256:
1257:
1258: \bibitem[Rasio \& Ford(1996)]{RF96}Rasio, F. A., Ford, E.B., 1996, Science, 274, 954
1259:
1260: \bibitem[Raymond et al.(2006)]{ray06} Raymond, S.~N.,
1261: Mandell, A.~M., \& Sigurdsson, S.\ 2006, Science, 313, 1413
1262:
1263:
1264: \bibitem[Safronov(1969)]{Saf69} Safronov,V.S. 1969, Evolution of the Protoplanetary Cloud and
1265: Formation of the Earth and the Planets (Moscow:Nauka)
1266:
1267: \bibitem[Silverstone et al.(2006)]{sil06} Silverstone, M.D., et al., 2006, \apj, 639, 1138
1268:
1269: \bibitem[Stone \& Norman(1992)]{SN92} Stone, J.M., Norman, M.L., 1992, \apjs, 80, 753
1270:
1271: \bibitem[Tohline(1980)]{T80}Tohline, J. E. 1980, \apj, 235, 866
1272:
1273: \bibitem[Trilling et al.(1998)]{TBGLGB98}Trilling, D. E., Benz, W., Guillot, T., Lunine, J.I., Hubbard W.B., Burrows, A., 1998, \apj, 500,428
1274:
1275: \bibitem[de Val-Borro et al.(2006)]{deval06} de Val-borro, M., Edgar, R. G., Artymowicz, P., Ciecielag, P., and others, 2006, \mnras, 370, 529
1276:
1277:
1278: \bibitem[Ward(1997)]{War97} Ward, W. R. 1997, Icarus, 126, 261
1279:
1280: \bibitem[Weidenschilling \& Marzari(1996)]{Wei96}Weidenschilling, S. J., \& Marzari, F. 1996, \nat, 384, 619
1281:
1282:
1283: \bibitem[Williams \& Tohline(1988)]{WT88}
1284: Williams, H. A., \& Tohline, J. E. 1988, \apj, 334, 449
1285:
1286: \end{thebibliography}
1287:
1288: %%\clearpage
1289:
1290:
1291: %% The following command ends your manuscript. LaTeX will ignore any text
1292: %% that appears after it.
1293:
1294: \end{document}
1295:
1296: %%
1297: %% End of file `sample.tex'.
1298: