0805.3733/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: 
3: \newcommand{\vdag}{(v)^\dagger}
4: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
5: 
6: 
7: \shorttitle{A Region void of Jovian Irregulars}
8: \shortauthors{Haghighipour and Jewitt}
9: 
10: 
11: 
12: 
13: \begin{document}
14: 
15: \title{A Region Void of Irregular Satellites Around Jupiter}
16: 
17: \author{N. Haghighipour and D. Jewitt}
18: \affil{Institute for Astronomy and NASA Astrobiology Institute,\\
19: University of Hawaii, Honolulu, HI 96822}
20: \email{nader@ifa.hawaii.edu, jewitt@ifa.hawaii.edu}
21: 
22: 
23: \begin{abstract}
24: An interesting feature of the giant planets of our solar system is
25: the existence of regions around these objects where no irregular 
26: satellites are observed. Surveys have shown that, around Jupiter,
27: such a region extends from the outermost regular satellite Callisto, 
28: to the vicinity of Themisto, the innermost irregular satellite. 
29: To understand the reason for the existence of such a 
30: {\it satellite-void} region, we have studied the dynamical evolution 
31: of Jovian irregulars by numerically integrating the orbits of several 
32: hundred test particles, distributed in a region between 30 and 
33: 80 Jupiter-radii, for different values of their semimajor axes, 
34: orbital eccentricities, and inclinations. As expected, our simulations
35: indicate that objects in or close to the influence zones of the
36: Galilean satellites become unstable because of interactions with 
37: Ganymede and Callisto. However, these perturbations 
38: cannot account for the lack of irregular satellites in the entire
39: region between Callisto and Themisto. It is suggested that at 
40: distances between 60 and 80 Jupiter-radii, Ganymede and Callisto 
41: may have long-term perturbative effects, which may require the 
42: integrations to be extended to times much longer than 10 Myr. 
43: The interactions of irregular satellites with protosatellites 
44: of Jupiter at the time of the formation of Jovian regulars 
45: may also be a destabilizing mechanism in this region. 
46: We present the results of our numerical simulations and discuss 
47: their applicability to similar satellite void-regions around other 
48: giant planets.
49: \end{abstract}
50: 
51: \keywords{planets and satellites: general, celestial mechanics,
52: solar system: general, methods: N-body simulations}
53: 
54: 
55: \section{Introduction}
56: 
57: Despite the differences in their compositions, structures and
58: mechanisms of formation, the giant planets of our solar system 
59: have one common feature. They all host irregular satellites. 
60: Thanks to wide field charge-coupled-devices (CCDs), the past 
61: few years have witnessed the discovery of a large number of 
62: these objects [see \citet{Jewitt07} for a comprehensive review]. 
63: At the time of writing of this article, 108 irregular satellites 
64: have been discovered, of which 55 belong to Jupiter, making 
65: the Jovian satellite system the largest among all planets.
66: 
67: Due to its proximity, the irregular satellites of Jupiter have
68: been the subject of extensive observational and theoretical 
69: research. Many of the dynamical characteristics of these objects, 
70: such as their orbital stability, dynamical grouping and their 
71: collision probability have long been studied 
72: \citep{Saha93,Carruba02,Nesvorny03,Nesvorny04,Beauge06,
73: Beauge07,Douskos07}. 
74: There is, however, one interesting feature in the distribution 
75: of Jovian irregulars that has not yet been fully understood. 
76: As shown by \citet{Sheppard03}, the region extending from the 
77: orbit of Callisto, the outermost Galilean satellite at 26 Jupiter-radii
78: $({R_J})$, to the periastron of Themisto $(\sim 76{R_J})$, Jupiter's
79: innermost irregular satellite, is void of irregulars. 
80: 
81: Observations suggest the presence of similar void
82: regions around all four giant planets. Table 1 and figure 1 show 
83: this in more detail.  As seen from figure 1, satellite void regions also exist 
84: {\it between} the
85: currently known irregular satellites of the giant planets. Theoretical
86: studies have indicated that there may be two possible scenarios for 
87: the existence of such void regions; 
88: ejection from the system due to mutual interactions with other
89: irregular satellites and, in the case of satellites that are
90: the remnants of collisions, clustering around their parent bodies
91: \citep{Kuiper56,Pollack79,Kessler81,Thomas91,Krivov02,
92: Nesvorny03,Nesvorny04,Beauge07}.
93: The focus of this paper is, however, on the lack of irregular 
94: satellites in the {\it boundary} between regulars and irregulars.
95: We are interested in understanding of why no irregular satellite exists
96: between the outermost Galilean satellite and Jupiter's innermost 
97: irregular one. 
98: 
99: 
100: The lack of irregular satellites in the boundary between
101: regulars and irregulars may be attributed to the 
102: distribution of the orbits of the latter bodies. Since irregular satellites
103: appear to have been captured from heliocentric orbits, it may be
104: natural to expect them to preferably have large semimajor axes,
105: and therefore not to exist in close orbits.
106: Proving this to be so would be an important contribution to the 
107: subject, but, unfortunately,
108: none of the models of capture is sufficiently specific to be 
109: used in this way. The N-body capture model of \citet{Nesvorny07}
110: does roughly match the distribution of irregular 
111: satellites of some planets, but not of Jupiter.
112: In this paper, we examine the 
113: possibility of a dynamical origin for the existence of this 
114: satellite-void boundary region.
115: 
116: The origin of irregular satellites and the mechanisms of their
117: capture remain unknown. The high values of the orbital inclinations 
118: and eccentricities of these objects imply an origin outside the
119: primordial circumplanetary disk from which the regular satellites of
120: giant planets were formed. It is believed that irregular satellites 
121: were formed elsewhere and were captured in their current orbits 
122: \citep{Kuiper56,Pollack79,Nesvorny03,Nesvorny04,Jewitt07}. 
123: 
124: The capture of irregular satellites might have occurred during 
125: and/or after the formation of the regular satellites of the giant 
126: planets. Given that the latter objects are formed through the 
127: collisional growth of small bodies in a circumplanetary disk
128: \citep{Canup02,Mosqueira03a,Mosqueira03b,Estrada06},  
129: the orbits of captured irregulars might have been altered by 
130: perturbations from these objects during their formation and after 
131: they are fully formed. In the case of Jovian irregulars, the 
132: migrations of Ganymede and Callisto 
133: \citep{Tittemore88,Tittemore89,Tittemore90,Goldreich80,Canup02}
134: have also had significant effects on the dynamics of irregular 
135: satellites. 
136: 
137: In this paper we study the dynamics and stability of irregular satellites
138: between Callisto and Themisto. We present the details of our
139: model in \S 2, and an analysis of the results in \S 3. Section 4
140: concludes this study by reviewing our study and discussing its
141: limitations.
142: 
143: 
144: \section{Numerical Simulations}
145: We numerically integrated the orbits of 
146: several hundred test particles in a region interior to the orbit 
147: of Themisto, the innermost Jovian irregular satellite. We assumed 
148: that the regular satellites of Jupiter were fully formed and studied 
149: the perturbative effects of the Galilean satellites on the dynamics 
150: of small objects in their vicinities.
151: We considered a system consisting of Jupiter, the Galilean satellites, 
152: and 500 test particles uniformly distributed between 30 and 
153: 80 Jupiter-radii. The initial orbital elements of the test 
154: particles were chosen in a systematic way as explained below.
155: 
156: \noindent
157: 1) At the beginning of each simulation, test particles were placed
158: in orbits with semimajor axes starting at $30{R_J}$ and increasing 
159: in increments of 0.1$R_J$. 
160: 
161: \noindent
162: 2) For each initial value of the semimajor axis of a test 
163: particle $(a_p)$, the initial orbital eccentricity $(e_p)$ 
164: was chosen to be 0, 0.2, 0.4, and 0.6. This choice of orbital 
165: eccentricity matches the range of the current values of the orbital
166: eccentricities of Jovian irregulars, as shown in figure 1. 
167: 
168: \noindent
169: 3) The initial orbital inclinations of test particles $(i_p)$ 
170: were varied between $0^\circ$ and $180^\circ$ in steps of 
171: $20^\circ$. As shown by figure 2, irregular satellites are 
172: absent at inclinations between $55^\circ$ and $130^\circ$ due 
173: to perturbations resulting from the Kozai resonance
174: \citep{Kozai62,Hamilton91,Carruba02,Nesvorny03}.
175: In choosing the initial orbital inclinations of these objects, 
176: we made a conservative assumption and considered the region of 
177: the influence of Kozai resonance to be between $60^\circ$ and 
178: $120^\circ$. We did not integrate the orbits of the test particles 
179: for ${i_p}={80^\circ}, {100^\circ}$ and $120^\circ$.
180: 
181: \noindent
182: 4) Since we were interested in studying the effects of the perturbations 
183: of regular satellites on the variations of the orbital eccentricities
184: and inclinations of test particles, we considered the initial values 
185: of the argument of the periastron, longitude of the ascending 
186: node, and the mean-anomaly of each test particle to be zero. 
187: This is an assumption that was made solely for the 
188: purpose of minimizing the initial-value effects.
189: 
190: We numerically integrated the orbits of the Galilean 
191: satellites\footnote{The orbital elements of the Galilean satellites 
192: were obtained from documentation on solar system dynamics 
193: published by the Jet Propulsion Laboratory 
194: (http://ssd.jpl.nasa.gov/?sat\_elem).} and the test particles 
195: of our system for different values of the test particles' 
196: orbital eccentricities and inclinations. Simulations were 
197: carried out for 10 Myr using the N-body integration package 
198: MERCURY \citep{Chambers99}. Since the objects of our interest 
199: are close to Jupiter, we neglected the perturbation of the Sun and 
200: considered Jupiter to be the central massive object of the system. 
201: This assumption is consistent with the findings of \citet{Hamilton97},
202: who have shown that around a giant planet with a Hill radius $R_H$,
203: the gravitational force of the Sun destabilizes the orbit of 
204: prograde irregular satellite at distances larger than 0.53$R_H$ 
205: and those of the retrograde irregulars at distances beyond 
206: 0.69$R_H$. For Jupiter, these values translate to 389$R_J$ 
207: for prograde irregulars and 507$R_J$ for retrograde ones. 
208: We carried out all integrations with respect to Jupiter with 
209: timesteps equal to the 1/20 of the orbital period of Io.
210: 
211: 
212: \section{Discussion and Analysis of the Results}
213: 
214: To study the relation between the orbital parameters of test 
215: particles and their stability, we determined the lifetime of 
216: each particle, considering ejection from the system and 
217: collision with other bodies. We considered a particle to be 
218: ejected when it reached a distance of 2000$R_J$ or larger from 
219: the center of Jupiter. A collision, on the other hand, occurred 
220: when the distance between a particle and a Galilean satellite 
221: became smaller than ${R_{GH}}={a_{GS}}{({M_{GS}}/3{M_J})^{1/3}}$, 
222: or the particle's closest distance to the center of Jupiter 
223: became smaller than Jupiter's radius. Here, $a_{GS}$ and $M_{GS}$ 
224: represent the semimajor axis and mass of a Galilean satellite, and
225: $M_J$ is the mass of Jupiter. Figure 3 shows the graphs of the 
226: test particles' lifetimes in terms of their initial semimajor 
227: axes for particles in two coplanar systems. The graph at the 
228: top corresponds to particles initially in circular orbits, 
229: and the one at the bottom shows the lifetimes of particles 
230: with initial eccentricities of 0.2. The positions and lifetimes 
231: of the regular satellites of Jupiter and the orbit of Themisto 
232: are also shown. As shown by the upper graph, test particles in 
233: circular orbits are mostly stable (for the duration of integrations)
234: except for a few that are close to Callisto. The region of 
235: stability, however, becomes smaller (instability progresses 
236: toward larger distances) in simulations in which the initial 
237: eccentricities of test particles are larger. This can be seen 
238: more clearly in figure 4 where from the graphs of figure 3,
239: only the regions between $30{R_J}$ and $80{R_J}$ are shown. The 
240: islands of instability, corresponding to mean-motion resonances 
241: with Callisto (indicated by the subscript C) and Ganymede 
242: (indicated by the subscript G) are also shown.
243: 
244: The migration of unstable regions to larger distances in systems
245: where test particles were initially in eccentric orbits was observed 
246: in all our simulations. Figure 5 shows another example of such a 
247: system. In this figure, the lifetimes of test particles with initial 
248: eccentricities of 0.4 and initial inclinations of 20$^\circ$ are 
249: shown. The unstable region extends to distances beyond their
250: corresponding regions in figures 3 and 4.
251: 
252: We also simulated the dynamics of test particles having
253: orbital inclinations larger than 90$^\circ$ (retrograde orbits).
254: As shown by figure 2, the number of irregular satellites is larger 
255: at these angles implying that retrograde orbits have longer lifetimes 
256: \citep{Hamilton97,Touma98,Nesvorny03}. Our simulations also show 
257: that retrograde orbits are more stable than their corresponding 
258: prograde ones. Figure 6 shows this for two sets of test particles. 
259: The particles in black correspond to a system in which ${e_p}=0.4$ 
260: and ${i_p}=40^\circ$. The particles in red correspond to a system
261: with similar orbital eccentricity, but with ${i_p}=140^\circ$.
262: As expected, the particles on retrograde orbits are more
263: stable and maintain their orbits for longer times.
264: 
265: The fact that the region of instability of test particles,
266: having a given semimajor axis,  expands by increasing the initial 
267: values of their orbital eccentricities can be attributed to the 
268: interactions of these particles with Jupiter's regular satellites. 
269: Given that the orbits of Jovian regulars are almost circular, 
270: an eccentric orbit for a test particle implies a smaller periastron 
271: distance for this object, and consequently a closer approach to 
272: the system's regular moons. Instability occurs when the perturbative 
273: effects of regular satellites disturb the motion of a test particle 
274: in its close approach. The outer boundaries of the influence 
275: zones\footnote{We define the {\it influence zone} of a Galilean 
276: satellite as the region between $({a_{GS}}-3{R_{GH}})$ and 
277: $({a_{GS}}+3{R_{GH}})$, where the dynamics of a small object is 
278: primarily affected by the gravitational force of the satellite.} 
279: of the Galilean satellites (Table 2) mark the extent of these 
280: perturbations. Particles with periastron distances beyond these 
281: boundaries, i.e., ${a_p}(1-{e_p}) > ({a_{GS}}+3{R_{GH}})$, will 
282: more likely have longer lifetimes.  
283: 
284: Figure 7 shows the boundaries of the stable and unstable test 
285: particles for all Galilean satellites. The initial positions of 
286: test particles with $({e_p},{i_p})$ equal to (0,0), (0.2,0), 
287: (0.4,40), (0.4,140), (0.6,60), and (0.6,120), are also shown. 
288: The stable particles are shown in black and unstable ones are 
289: in red. As shown here, particles with higher initial orbital 
290: eccentricities penetrate the influence zones of Ganymede and 
291: Callisto, and their orbits become unstable. Figure 7 also shows 
292: that for particles with similar semimajor axes, the boundary 
293: between the stable and unstable regions extends to larger 
294: distances by increasing the particle's eccentricity. For instance, 
295: when interacting with Callisto, in order for a particle to 
296: maintain stability, ${a_p}(1-{e_p}) > 28 {R_J}$. However, for 
297: particles in figure 5, where ${e_p}=0.2$, this implies that the 
298: region of instability extends to at least $35{R_J}$. 
299: 
300: Although for a given semimajor axis, the boundary of stable and 
301: unstable regions expands with increasing orbital eccentricities 
302: of the test particles, for a given value of this eccentricity, the 
303: destabilizing effects of the Galilean satellites reach to larger
304: semimajor axis, beyond their influence zones. At such distances, 
305: although the perturbative effects of Galileans are small they may, 
306: in the long term, disturb the motions of other objects and render 
307: their orbits unstable. An example of such instability can be seen 
308: in figure 7 for $({e_p},{i_p})=(0.2,0), (0.4,40), (0.6,60)$ and 
309: also in figure 5, where the unstable region extends to approximately 
310: $46{R_J}$. These results also imply that in simulations similar 
311: to those shown in figure 3, the region of instability may migrate 
312: outward if the integrations are continued to much larger times 
313: beyond 10 Myr.
314: 
315: Ganymede and Callisto may have undergone inward radial migrations 
316: after their formation \citep{Goldreich80,Canup02}. As noted by 
317: \citet{Canup02}, Ganymede might have started its inward migration 
318: from a distance not larger than approximately $30{R_J}$, and 
319: Callisto might have migrated inward from approximately $35{R_J}$. 
320: The perturbative effects of these satellites have, therefore, 
321: influenced a larger region beyond their current influence zones. 
322: Figure 8 shows this in more detail. In this figure, the top graph 
323: shows the boundary of stable and unstable test particles for 
324: Ganymede before and after its radial migration. The bottom graph 
325: in figure 8 shows the similar curves for both Ganymede and Callisto. 
326: The influence zones of these two satellites were initially 
327: larger, implying that many small objects might have been destabilized 
328: during the migrations phase.
329: 
330: Even when including radial migration, the influence zones of 
331: Ganymede and Callisto do not cover the entire region between Callisto 
332: and Themisto. For the 10 Myr integration time presented here, the 
333: interactions with Ganymede and Callisto do not seem to account for 
334: the lack of irregular satellites at distances beyond 60$R_J$. Since 
335: at such distances, the perturbative effects of Ganymede and Callisto 
336: are weaker, extension of integrations to longer times may reveal that 
337: this region is indeed unstable. We also speculate that the lack of 
338: irregular satellites at such distances is the result of a clearing 
339: process that has occurred during the formation of the Jovian regular 
340: moons. As shown by \citet{Canup02}, and by 
341: \citet{Mosqueira03a,Mosqueira03b}, regular satellites of giant planets 
342: might have formed through the collisional growth of smaller 
343: objects (satellitesimals) in a circumplanetary disk. Similar to 
344: the formation of terrestrial planets in our solar system, where 
345: the mutual collisions of planetesimals around the Sun resulted in 
346: the formation of many protoplanetary objects, satellitesimals might 
347: have also collided and formed a disk of protosatellite bodies
348: around giant planets. The interaction between protosatellites and
349: smaller bodies in such circumplanetary disks could have destabilized
350: the orbits of many of these objects and resulted in their collisions,
351: accretion by protosatellites, and/or ejection from the system. 
352: The final locations of surviving irregular 
353: satellites at smaller distances were then limited by the outer 
354: boundary of a region that included the influence zones of regular 
355: satellites, as well as the above-mentioned dynamically cleared area.
356: For Jovian irregulars, a conservative assumption places this limit 
357: at $76{R_J}$ on the curve of constant-periastron of Themisto. 
358: Figure 9 shows this limit in light blue. At larger distances, on 
359: the other hand, the stability of irregular satellites
360: is governed by the perturbation from the Sun. The boundaries 
361: of the Sun-perturbed regions have been shown in figure 9 as curves 
362: of constant-apastron, with the constant value equal to 0.53$R_{JH}$ 
363: for prograde irregulars and 0.69$R_{JH}$ for retrograde ones 
364: \citep{Hamilton97}. The quantity $R_{JH}$
365: is the Hill radius of Jupiter. It is important to note that the values
366: of the eccentricities and semimajor axes of the irregular satellites
367: in figure 9 were obtained from the documentation on solar system dynamics
368: published by JPL, in which the orbital parameters of a body have their 
369: mean values and, unlike the test particles in our simulations, 
370: their angular elements are non-zero. This implies
371: that although figure 9 portrays a qualitatively reliable
372: picture of the stability of the Jovian irregular satellites, a
373: more detailed mapping could be obtained by assuming zero values for the
374: initial angular elements of irregular satellites and simulating
375: their stability for 10 Myr.
376:  
377: 
378: 
379: \section{Summary and Concluding Remarks}
380: 
381: We numerically integrated the orbits of 500 test particles 
382: for different values of their orbital elements, in a region 
383: between $30{R_J}$ and $80{R_J}$. Our integrations indicated 
384: that the long-term stability of these objects is affected 
385: by the values of their initial periastron distances. For given 
386: values of their semimajor axes, the region of instability of test 
387: particles extended to larger distances as the initial values 
388: of their orbital eccentricities were increased. 
389: 
390: Our numerical simulations also showed that, except at large
391: distances from the outer boundaries of the influence zones of 
392: Ganymede and Callisto, the lack of irregular satellites between 
393: Callisto and Themisto can be attributed to the instability of 
394: test particles caused by their interactions with the two outermost 
395: Galilean satellites. At larger distances (e.g., between 
396: $\sim 40{R_J}$ to 80$R_J$ for particles in circular orbits, 
397: and between $\sim 60{R_J}$ to 80$R_J$ for particles with initial 
398: orbital eccentricities of 0.4), however, the perturbations of 
399: Galilean satellites do not seem to be able to account for the 
400: instability of small bodies. A possible explanation is that their 
401: instability is the result of interactions with Jovian satellitesimals 
402: and protosatellites during the formation of Jupiter's regular moons.
403: 
404: Because the test particles in our simulations were initially close 
405: to Jupiter, we neglected the effect of solar perturbations.
406: As shown by \citet{Hamilton97}, for Jupiter, the shortest critical 
407: distance beyond which the perturbation from the Sun cannot be neglected
408: corresponds to prograde orbits and is equal to 389$R_J$.
409: In our simulations, the outermost test particle was placed
410: well inside this region at 80$R_J$. It is important to note that,
411: although the effects of solar perturbations
412: on our test particles are small and will not cause
413: orbital instability, they may, in the long term, create noticeable changes
414: in the orbital evolution of test particles. For instance, solar
415: perturbations may enhance the perturbative effects of regular satellites
416: in increasing the orbital eccentricity of test particles and
417: result in their capture in Kozai resonance. More numerical simulations 
418: are needed to explore these effects.
419: 
420: As mentioned in \S 2, different non-zero
421: angular elements may in fact affect the stability of individual 
422: test particles. However, the analysis of
423: the stability of the system, as obtained from our numerical 
424: simulations, portrays a picture of the dynamical
425: characteristics of the test particles that, in general, is also
426: applicable to Jovian irregular satellite systems with other 
427: initial angular variables.
428:  
429: The applicability of our results and the extension of our analysis 
430: to the satellite-void boundary regions around other giant planets may be 
431: limited due to fact that their satellite systems are different from 
432: that of Jupiter. Although the above-mentioned dynamical-clearing process
433: can still account for the instability of many small objects around 
434: these planets, numerical simulations, similar to those presented here, 
435: are necessary to understand the dynamical characteristics of their 
436: small bodies in more detail.
437: 
438: 
439: \acknowledgments
440: We acknowledge the use of the computational facilities
441: at the Department of Terrestrial Magnetism at the
442: Carnegie Institution of Washington. 
443: This work has been supported by the NASA Astrobiology
444: Institute under Cooperative Agreement NNA04CC08A at the Institute 
445: for Astronomy at the University of Hawaii for NH.
446: 
447: 
448: \begin{thebibliography}{}
449: 
450: \bibitem[Beaug\'e, Nesvorn\'y and Dones (2006)]{Beauge06}
451: Beaug\'e, C., Nesvorn\'y, D., and Dones, L., 2006, \aj, 131, 2299
452: 
453: \bibitem[Beaug\'e and Nesvorn\'y (2007)]{Beauge07}
454: Beaug\'e, C., and Nesvorn\'y, D., 2007, \aj, 133, 2537
455: 
456: \bibitem[Canup and Ward (2002)]{Canup02}
457: Canup, R. M., and Ward, W. R., 2002, \aj, 124, 3404
458: 
459: \bibitem[Carruba et al. (2002)]{Carruba02}
460: Carruba, V., Burns, J. A., Nicholson, P. D., and Gladman, B., 
461: 2002, Icarus, 158, 434 
462: 
463: \bibitem[Chambers (1999)]{Chambers99}
464: Chambers, J. E., 1999, \mnras, 304, 793
465: 
466: \bibitem[Douskos, Kalantonis and Markellos (2007)]{Douskos07}
467: Douskos, C., Kalantonis, V., and Markellos, P., 2007, 
468: Astrophys. Space. Sci., 310, 245
469: 
470: \bibitem[Estrada and Mosqueira (2006)]{Estrada06}
471: Estrada, P. R., and Mosqueira, I., 2006, Icarus, 181, 486
472: 
473: \bibitem[Goldreich and Tremaine (1980)]{Goldreich80}
474: Goldreich, P., and Tremaine, S., 1980, \apj, 241, 425
475: 
476: \bibitem[Hamilton and Burn (1991)]{Hamilton91}
477: Hamilton, D. P., and Burns, J. A., 1991, Icarus, 92, 118
478: 
479: \bibitem[Hamilton and Krivov (1997)]{Hamilton97}
480: Hamilton, D. P., and Krivov, A. V., 1997, Icarus, 128, 241
481: 
482: \bibitem[Jewitt and Haghighipour (2007)]{Jewitt07}
483: Jewitt, D., and Haghighipour, N. 2007, \araa, {\bf 45}, 261
484: 
485: \bibitem[Kessler (1981)]{Kessler81}
486:  Kessler, D. J., 1981,   Icarus, 48, 39
487: 
488: \bibitem[Krivov et al. (2002)]{Krivov02}
489:  Krivov, A. V., Wardinski, I., Spahn, F., Kruger, H. 
490: and Grun, E., 2002,  Icarus 157, 436
491: 
492: \bibitem[Kozai (1962)]{Kozai62}
493: Kozai, Y., 1962, \aj, 67, 591
494: 
495: \bibitem[Kuiper (1956)]{Kuiper56}
496: Kuiper, G. P., 1956, Vistas in Astronomy, 2, 1631
497: 
498: \bibitem[Mosqueira and Estrada (2003a)]{Mosqueira03a}
499: Mosqueira, I., and Estrada, P. R., 2003, Icarus, 163, 198
500: 
501: \bibitem[Mosqueira and Estrada (2003b)]{Mosqueira03b}
502: Mosqueira, I., and Estrada, P. R., 2003, Icarus, 163, 232
503: 
504: \bibitem[Nesvorn\'y et al. (2003)]{Nesvorny03}
505: Nesvorn\'y, D., Alvarellos, J. L. A., Dones, L., and Levison, H. F., 
506: 2003, \aj, 126, 398
507: 
508: \bibitem[Nesvorn\'y, Beaug\'e, and Dones (2004)]{Nesvorny04}
509: Nesvorn\'y, D., Beaug\'e, C., and  Dones, L., 2004, \aj, 127, 1768
510: 
511: \bibitem[Nesvorn\'y, Vokrouhlick\'y, and Morbidelli(2007)]{Nesvorny07}
512: Nesvorn\'y, D., Vokrouhlick\'y, D., and  Morbidelli, A.,
513: 2007, \aj, 133, 1962
514: 
515: \bibitem[Pollack, Burns, and Tauber (1979)]{Pollack79}
516: Pollack, J. B., Burns, J. A., Tauber, M. E., 1979, Icarus, 37, 587 
517: 
518: \bibitem[Saha and Tremaine (1993)]{Saha93}
519: Saha, P., and Tremaine, S., 1993, Icarus, 106, 549
520: 
521: \bibitem[Sheppard and Jewitt (2003)]{Sheppard03}
522: Sheppard, S. S., and Jewitt, D., 2003, Nature, 423, 261
523: 
524: \bibitem[Thomas et al.(1991)]{Thomas91} 
525: Thomas, P., Veverka, J., and Helfenstein, P., 1991,  
526: \jgr, 96, 19253 
527: 
528: \bibitem[Tittemore and Wisdom (1988)]{Tittemore88}
529: Tittemore, W. C., and Wisdom, J., 1988, Icarus, 74, 172
530: 
531: \bibitem[Tittemore and Wisdom (1989)]{Tittemore89}
532: Tittemore, W. C., and Wisdom, J., 1989, Icarus, 78, 63
533: 
534: \bibitem[Tittemore and Wisdom (1990)]{Tittemore90}
535: Tittemore, W. C., and Wisdom, J., 1990, Icarus, 85, 394
536: 
537: \bibitem[Touma and Wisdom (1998)]{Touma98}
538: Touma, J., and Wisdom, J., 1998, \aj, 115, 1653
539: 
540: \end{thebibliography}
541: 
542: 
543: 
544: 
545: \clearpage
546: \begin{figure}
547: \plottwo{f1a.eps}{f1b.eps}
548: \plotone{f1c.eps}
549: \plotone{f1d.eps}
550: \vskip 0.1in
551: \caption{Satellite systems of the four giant planets. The 
552: $x-$axes show the semimajor axes of satellites in the units
553: of the radii of their host planets.
554: \label{fig1}}
555: \end{figure}
556: 
557: 
558: 
559: 
560: 
561: 
562: \clearpage
563: \begin{figure}
564: \plotone{f2.eps}
565: \vskip -4.2in
566: \caption{Distribution of irregular satellites around the 
567: four giant planets. The quantities on the axes represent 
568: the semimajor axis of a satellite, $a$, the Hill radius 
569: of its host planet, $R_H$, and the satellite's orbital 
570: inclination, $i$. The distance of each satellite from the 
571: origin of the graph is equivalent to its semimajor axis, 
572: and its radial excursion (the distance from its periastron 
573: to apastron) is given by the length of its associated line. 
574: The angle between this line and the horizontal axis is 
575: equal to satellite's orbital inclination.
576: \label{fig2}}
577: \end{figure}
578: 
579: 
580: 
581: 
582: \clearpage
583: \begin{figure}
584: \vskip 2in
585: \includegraphics[height=8cm]{f3a.eps}
586: \includegraphics[height=8cm]{f3b.eps}
587: \vskip 2.2in
588: \caption{Lifetimes of test particles in a coplanar system
589: with ${e_p}=0$ (top) and ${e_p}=0.2$ (bottom). The locations of
590: the regular satellites of Jupiter and Themisto are also shown.
591: \label{fig3}}
592: \end{figure}
593: 
594: 
595: 
596: 
597: \clearpage
598: \begin{figure}
599: \vskip 0.85in
600: \includegraphics[height=8cm]{f4a.eps}
601: \includegraphics[height=8cm]{f4b.eps}
602: \vskip 35pt
603: \caption{Lifetimes of the test particles of figure 3. Both 
604: systems are coplanar. In the upper graph ${e_p}=0$, and in 
605: the lower graph ${e_p}=0.2$. The locations of mean-motion 
606: resonances with Ganymede and Callisto are also shown.
607: \label{fig4}}
608: \end{figure}
609: 
610: 
611: 
612: 
613: \clearpage
614: \begin{figure}
615: \vskip 0.8in
616: \includegraphics[height=8cm]{f5a.eps}
617: \includegraphics[height=8cm]{f5b.eps}
618: \caption{Lifetimes of test particles in a system in which 
619: ${e_p}=0.4$ and ${i_p}=20^\circ$. The locations of
620: mean-motion resonances with Ganymede and Callisto are also shown.
621: As shown here, compared with the systems of figures 3 and 4, 
622: as the initial eccentricities of test particle increase, their
623: region of instability extends to farther distances.
624: \label{fig5}}
625: \end{figure}
626: 
627: 
628: 
629: \clearpage
630: \begin{figure}
631: \centering
632: \includegraphics[height=9cm]{f6.eps}
633: \vskip 10pt
634: \caption{Lifetimes of test particles with initial eccentricities of
635: ${e_p}=0.4$ and orbital inclinations of ${i_p}=40^\circ$ (black)
636: and ${i_p}=140^\circ$ (red). As shown here, particles in retrograde
637: orbits (red) are more stable. The locations of mean-motion resonances 
638: with Ganymede and Callisto are also shown. 
639: \label{fig6}}
640: \end{figure}
641: 
642: 
643: 
644: 
645: \clearpage
646: \begin{figure}
647: \centering
648: \includegraphics[height=10.5cm]{f7.eps}
649: \caption{Boundaries of stable and unstable regions of a test 
650: particle (i.e., curves of constant-periastron for which the 
651: constant value is equal to the distance of the outer 
652: boundary of the influence zones of the Galilean satellites). 
653: The black and red horizontal lines represent the initial semimajor
654: axes of test particles between $30{R_J}$ and $80{R_J}$ for
655: different values of their orbital eccentricities and inclinations.
656: The particles in black maintained their orbits for the duration of 
657: integration (10 Myr) whereas the particles in red became unstable.
658: \label{fig7}}
659: \end{figure}
660: 
661: 
662: 
663: 
664: 
665: \clearpage
666: \begin{figure}
667: \centering
668: \vskip -0.3in
669: \includegraphics[height=9cm]{f8a.eps}
670: \vskip 10pt
671: \includegraphics[height=9cm]{f8b.eps}
672: \vskip -10pt
673: \caption{Top: curves of the constant-periastron (boundaries 
674: of stable and unstable regions) of a test particle for which 
675: the constant values are equal to the distance of the outer 
676: boundary of the influence zone of Ganymede before its migration
677: (the curve denoted by ${\rm G}_{\rm i}$) and after it migrates to 
678: its present orbit (the curve denoted by ${\rm G}_{\rm f}$).
679: The stable (black) and unstable (red) test particles in 
680: the region between $30{R_J}$ and $80{R_J}$ are also shown. 
681: Bottom: Similar constant-periastron curves as in the above
682: for Ganymede and Callisto. Note that, because the pre-migration
683: semimajor axes of Ganymede and Callisto are close to one
684: another ($30{R_J} $ and $35{R_J}$, respectively), it seems
685: as though the two
686: curves ${\rm G}_{\rm i}$ and ${\rm C}_{\rm i}$ in the bottom graph
687: are in contact with one another. The stable
688: and unstable test particle are also similar to the top graph.
689: \label{fig8}}
690: \end{figure}
691: 
692: 
693: \clearpage
694: \begin{figure}
695: \centering
696: \vskip 1.9in
697: \includegraphics[height=12cm]{f9.eps}
698: \vskip -15pt
699: \caption{Graph of the region of the stability of Jovian 
700: irregular satellites. Prograde satellites are in purple 
701: and retrograde ones are in orange. The inner boundary of 
702: this region, shown in light blue, corresponds to the curve 
703: of constant-periastron of Themisto. Its outer boundary,
704: shown in purple for prograde irregulars and in orange for 
705: retrograde ones, is a curve for constant-apastron equal to 
706: the largest distance a Jovian irregular satellite can travel 
707: before its orbit becomes unstable by solar perturbation. The 
708: curves of constant-periastron corresponding to the influence 
709: zones of Ganymede and Callisto, as explained in figure 8, and 
710: the stable (black) and unstable (red) test particles, within 
711: the region of $30{R_J}$ to $80{R_J}$, are also shown.
712: \label{fig9}}
713: \end{figure}
714: 
715: 
716: 
717: \clearpage
718: \begin{deluxetable}{lc} 
719: \tablewidth{0pt} 
720: \tablecaption{Irregular satellite-void regions around giant planets}
721: \tablehead{ 
722: \colhead{Satellite} &
723: \colhead{Void Region (Planet Radii)}}
724: \startdata 
725: Jupiter   & 30-80         \\
726: Saturn    & 59-184         \\
727: Uranus    & 23-167         \\
728: Neptune   & 223-635           \\    
729: 
730: \enddata 
731: \end{deluxetable} 
732: 
733: 
734: 
735: 
736: 
737: \clearpage
738: \begin{deluxetable}{lccc} 
739: \tablewidth{0pt} 
740: \tablecaption{Influence zones of the Galilean Satellites.}
741: \tablehead{ 
742: \colhead{Satellite} &
743: \colhead{Semimajor Axis $(R_J)$} &
744: \colhead{Inner boundary $(R_J)$} &
745: \colhead{Outer boundary $(R_J)$}}
746: \startdata 
747: Io         & 5.8     & 5.4      & 6.3         \\
748: Europa     & 9.3     & 8.7      & 9.8          \\
749: Ganymede   & 14.8    & 13.5     & 16.1          \\
750: Callisto   & 26.0    & 24.0     & 28.0           \\    
751: 
752: \enddata 
753: \end{deluxetable} 
754: 
755: 
756: 
757: \end{document}
758: