1: %\documentclass[12pt,preprint]{aastex}

2: 

3: \documentstyle[psfig,emulateapj]{article}

4: 

5: \def\topfraction{.9}

6: \def\bottomfraction{.1}

7: \def\textfraction{.1}

8: \def\floatpagefraction{.5}

9: 

10: \def\boldr{\mbox{\boldmath $r$}}

11: \def\boldvarepsilon{\mbox{\boldmath $\varepsilon$}}

12: \def\bnabla{\mbox{\boldmath $\nabla$}}

13: \def\divxi{\bnabla\cdot{\boldxi}}

14: \def\etal{et al.\,}

15: \def\Bruntfreq{Brunt-V{\"a}is{\"a}l{\"a}\,\,\,}

16: \def\refnew#1{(\ref{#1})}

17: \def\be{\begin{equation}}

18: \def\ee{\end{equation}}

19: \def\ni{\noindent}

20: \def\vcv{v_{\rm cv}}

21: \def\tcv{t_{\rm cv}}

22: \def\cs{c_s}

23: \def\erg{\, \rm erg}

24: \def\K{\, \rm K}

25: \def\s{\, \rm s}

26: \def\km{\, \rm km}

27: \def\cm{\, \rm cm}

28: \def\dyne{\, \rm dyne}

29: \def\ci{cos{\theta_i}}

30: \def\ph{\rm ph}

31: \def\th{\rm th}

32: \def\g{\rm g}

33: %\def\pomega{\tilde \omega}

34: \def\pomega{\varpi}

35: \def\deg{^\circ}

36: 

37: \begin{document} 

38: 	

39: \title{\mbox{Tidal Evolution Of the Planetary System around HD

40: $83443$}}

41: 

42: %\lefthead{Tidal Evolution}

43: %\righthead{Planetary System HD $83443$}

44: 

45: \author{Yanqin Wu\altaffilmark{1}, Peter Goldreich\altaffilmark{2}}

46: 

47: \altaffiltext{1}{Canadian Institute of Theoretical Astrophysics,

48: 	University of Toronto, 60 St. George Street, Toronto, ON

49: 	M5S 3H8, Canada, wu@cita.utoronto.ca}

50: \altaffiltext{2}{Division of Geophysics and Planetary Sciences, 

51: Caltech, Pasadena, CA 91125, USA, pmg@gps.caltech.edu}

52: 

53: \begin{abstract}

54: Two planets with orbital period ratio approximately 10:1 have been

55: discovered around the star HD $83443$. The inner and more massive

56: planet, HD $83443b$, has the smallest semi-major axis among all

57: currently known exoplanets. Unlike other short period exoplanets, it

58: maintains a substantial orbital eccentricity, $e_1=0.079\pm 0.008$, in

59: spite of efficient tidal damping. This is a consequence of its secular

60: interactions with HD $83443c$ whose orbital eccentricity $e_2=0.42\pm

61: 0.06$. Dissipation, associated with tides the star raises in the inner

62: planet, removes energy but not angular momentum from its orbit, while

63: secular interactions transfer angular momentum but not energy from the

64: inner to the outer planet's orbit.  The outward transfer of angular

65: momentum decreases the tidal decay rate of the inner planet's orbital

66: eccentricity while increasing that of the outer planet.  The alignment

67: of the apsides of the planets' orbits is another consequence of tidal

68: and secular interactions. In this state the ratio of their orbital

69: eccentricities, $e_1/e_2$, depends upon the secular perturbations the

70: planets exert on each other and on additional perturbations that

71: enhance the inner planet's precession rate. Tidal and rotational

72: distortions of the inner planet along with general relativity provide

73: the most important of these extra precessional perturbations, each of

74: which acts to reduce $e_1/e_2$. Provided the planets' orbits are

75: coplanar, the observed eccentricity ratio uniquely relates $\sin i$

76: and $C\equiv (k_{2}/k_{2J})(R_1/R_J)^5$, where the tidal Love number,

77: $k_2$, and radius, $R_1$, of the inner planet are scaled by their

78: Jovian equivalents.

79:  

80: \end{abstract}

81: 

82: \keywords{celestial mechanics -- (stars:) binaries: spectroscopic --

83: (stars:) planetary systems}

84: 

85: \setcounter{equation}{0}

86: 

87: \section{Introduction}

88: \label{sec:introduce}

89: 

90: \begin{deluxetable}{crrrcc} 

91: \tablecolumns{6} 

92: \tablewidth{0pt} 

93: \tablecaption{Parameters for the two planets of HD $83443$.\tablenotemark{a}

94: \label{table:parameter}}

95: \tablehead{ 

96: \colhead{Planet}    & \colhead{P (d)} &   \colhead{$e$}   & 

97: \colhead{$\pomega (\deg)$} & \colhead{$a$ (AU)} & 

98: \colhead{$m\sin i$}}

99: \startdata 

100: HD$83443b$ & $2.9853\pm 0.0009$ & $0.079\pm 0.008

101: $ & $300\pm 17$ & $0.0376$  & $1.14$ \\

102: HD$83443c$ & $29.83\pm 0.18$ & $0.42\pm 0.06$ &

103: $337\pm 10 $ &  $0.175$  &

104: $0.53$ \\

105:  \enddata

106: \tablenotetext{a}{Data taken from Mayor \etal

107: (\cite{mayor83443}). Planet masses are in unit of Saturn mass $M_{\rm

108: Sat} \approx 5.68\times 10^{29}\g$. The host star $M_* = 0.79

109: M_\odot$. }

110: \end{deluxetable}

111: 

112: Parameters of the HD $83443$ planetary system as given by Mayor \etal

113: (\cite{mayor83443}) are listed in Table \ref{table:parameter}.  The

114: substantial eccentricity of the inner planet's orbit, $e_1 = 0.079 \pm

115: 0.008$, is surprising since other exoplanets with comparable

116: semi-major axes have circular orbits.  In the absence of other

117: interactions, tidal dissipation would erode $e_1$ as $d\ln e_1/dt = -

118: 1/\tau$ with (Goldreich \cite{goldreich}, Hut \cite{hut}),

119: \begin{equation}

120: \tau \approx {2\over{27}}{{ Q}\over{k_{2}}} \left({{a_1^3}\over{G

121: M_*}}\right)^{1/2} \left({{m_1}\over{M_*}}\right)

122: \left({{a_1}\over{R_1}}\right)^5.

123: \label{eq:dampingtau}

124: \end{equation}

125: Adopting values appropriate for Jupiter: radius $R_1\approx 7\times

126: 10^4\km$, mass $m_1\approx 2\times 10^{30}\g$, tidal Love number

127: $k_{2} \approx 0.5$ (Yoder \cite{yoder}), and tidal quality factor, $Q

128: \sim 3 \times 10^5$ (Goldreich \& Soter \cite{goldreichsoter}), yields

129: $\tau \sim 3\times 10^8$ yr.  This is in striking contrast to the old

130: age of the star. Isochrone fitting suggests an age of about $6.5\,$Gyr

131: (N. Murray, private communication), which is consistent with the

132: absence of chromospheric activity and the small projected rotational

133: velocity, $v \sin i < 2 \km/\s$.

134: 

135: Secular interactions with HD $83443c$ are undoubtedly responsible for

136: the persistence of HD $83443b$'s orbital eccentricity (Mayor \etal

137: \cite{mayor83443}). Results from calculations by Lasker and

138: Benz displayed in Mayor \etal (\cite{mayor83443}) show that secular

139: interactions between the inner and outer planets induce anti-correlated

140: oscillations in the two planets' orbital eccentricities. The

141: anti-correlation results because secular interactions transfer angular

142: momentum, $J$, but not energy, $E$, and $e$ is

143: related to $E$ and $J$ by

144: \begin{equation}

145: e^2=1+{2J^2E\over m^3(GM)^2}.

146: \label{eq:eJE}

147: \end{equation}

148: Similar phenomena are noted in the hierarchical triple star

149: systems HD $109648$ (Jha \etal \cite{jha}), HD $284163$ (Griffin \&

150: Gunn \cite{griffin}), and Algol (see, e.g., Kiseleva \etal

151: \cite{kiseleva}).

152: 

153: We argue that eccentricity oscillations were tidally damped a long

154: time ago, and that currently the planets maintain apse alignment (\S

155: \ref{sec:linear}). In this state the orbital eccentricity of the inner

156: planet is proportional to that of the outer planet, and both are

157: steady on the secular time-scale. On the much longer tidal time-scale,

158: the individual eccentricities decay, with the decay rate being faster

159: for the inner planet.

160: 

161: We utilize the properties of this system to constrain parameters that

162: are not directly measurable. From the ratio of the two

163: eccentricities, we place bounds on the inclination angle, $i$, of the

164: planets' orbit plane and on the radius of the inner planet. 

165: 

166: 

167: \section{Consequences Of Apse Alignment}

168: \label{sec:nonlinear}

169: 

170: \subsection{Assumptions}

171: 

172: We study secular interactions in $HD 83443$'s planetary system under

173: the assumptions that the planets share a common orbital plane and that

174: their apsides remain aligned. The former assumption seems reasonable

175: for planets formed in a gaseous disk. The latter assumption is

176: consistent with the currently observed approximate apse alignment (see

177: Table \ref{table:parameter}) and is justified by the analysis in \S

178: \ref{sec:linear}. These assumptions allow us to infer parameters of

179: the planetary system from the observed eccentricity ratio.

180: 

181: \subsection{Secular Dynamics}

182: \label{subsec:currenteratio}

183: 

184: Mutual secular perturbations of the planets' orbits are calculated

185: following a method devised by Gauss (\cite{gauss}) as outlined in the

186: appendix. Equation \refnew{eq:gauss} describes the resulting periapse

187: and eccentricity variations. However, as a result of inner planet's

188: proximity to the star, it is subject to additional precessional

189: perturbations of which the most important arise from its related tidal

190: and rotational distortions (Sterne \cite{sterne})\footnote{The symbol

191: $k_2$ also appears as the apsidal motion constant in stellar

192: literature in which case it is a factor 2 smaller than the Love

193: number.}, and from general relativity (Einstein

194: \cite{einstein}). Expressions for these precession rates are, 

195: %as follows:

196: \begin{mathletters}

197: \begin{eqnarray}

198: \hskip-0.5cm {d\pomega_1\over dt}\Bigr|_{\rm tide} & = & 

199: 7.5 n_1 k_{2}\! {{1+\frac{3}{2}e_1^2 +

200: \frac{1}{8}e_1^4}\over{(1-e_1^2)^5}} {{M_*}\over{m_1}}\!\!

201: \left({{R_1}\over{a_1}}\right)^5,

202: \label{eq:pretide}\\ 

203: \hskip-0.7cm {d\pomega_1\over dt}\Bigr|_{J_2} \,\,& = & 0.5

204: n_1 {k_{2}\over{(1-e_1^2)^2}}\!\left({{\Omega_1}\over{n_1}}\right)^2\!\!

205: {{M_*}\over{m_1}}\!\!  \left({{R_1}\over{a_1}}\right)^5,

206: \label{eq:preJ2}\\

207: \hskip-0.5cm {d\pomega_1\over dt}\Bigr|_{\rm GR} & = & 3

208: n_1\, {{G M_*}\over{a_1 c^2 (1-e_1^2)}}.\label{eq:preGR}

209: \end{eqnarray}

210: \end{mathletters}

211: \noindent Note that in equation \refnew{eq:preJ2} the planet's spin rate,

212: $\Omega_1$, is scaled by its mean motion, $n_1$, since we expect that

213: tidal dissipation has produced approximate spin-orbit synchronization.

214: We neglect smaller precessional contributions due to the rotational

215: oblateness of the star and to the tidal distortion of the star by the

216: planet. 

217: 

218: Numerically, the total rate of extra precessions amounts to,

219: \begin{eqnarray}

220: \Delta & \equiv &

221: {d\pomega_1\over dt}\Bigr|_{\rm tide} +

222: {d\pomega_1\over dt}\Bigr|_{J_2} +

223: {d\pomega_1\over dt}\Bigr|_{\rm GR} \nonumber \\

224: & = & (7.79\, C \sin i + 1.52) \times 10^{-11}\s^{-1},

225: \label{eq:deltavalue}

226: \end{eqnarray}

227: where we set $\Omega_1 = n_1$ and introduce the dimensionless number

228: $C \equiv (k_{2}/k_{2J})(R_1/R_J)^5$, with $R_J$ and $k_{2J}$ assigned

229: values appropriate for Jupiter. Observations determine $k_{2J}=0.494$,

230: close to that of an $n=1$ polytrope, $k_2\approx 0.51$, whereas Saturn

231: has a smaller Love number, $k_2 = 0.317$ (Yoder \cite{yoder}) because

232: it is more centrally condensed as a result of possessing a relatively

233: larger core of heavy elements.

234: 

235: We incorporate the additional precessional terms into the secular

236: perturbation equations (see eq. [\ref{eq:secular}]). In a state of

237: apse alignment, there is no secular exchange of angular momentum

238: between the planets. Therefore, their orbital eccentricities are

239: constant on the secular time-scale. Moreover, the eccentricity ratio,

240: $e_1/e_2$, is a monotonically decreasing function of $\Delta$. To

241: match the observationally determined value of $e_1/e_2=0.19$ requires

242: $\Delta \approx 9.1\times 10^{-11}\,\sin^{-1} i\, \s^{-1}$. This value

243: is greater than the precession rate due to general relativity and

244: provides evidence for the tidal and rotational distortion of the inner

245: planet. Because the inner planet is not observed to transit, $\sin i <

246: 0.99$, which implies $C \ge 0.9$. Nearly pole-on orbits, $\sin i <

247: 0.20$, are excluded if we accept that $R_1 < 2 R_J$, a conservative

248: upper-limit as judged from Fig. 1 of Burrows \etal

249: \cite{burrows}. This implies $m_1 < 5.45 M_{\rm sat}$. 

250: Figure \ref{fig:secular_search} shows $e_1$ as a function of $C$ for

251: different values of $\sin i$ with $e_2$ fixed at its observed

252: value.

253: 

254: 

255: \vbox{

256: \begin{center}

257: \leavevmode \hbox{ \epsfxsize=0.85\hsize 

258: \epsffile[80 180 570 700]

259: {secular_search.ps}

260: %{f1.ps} 

261: } \figcaption{Values of the inner planet eccentricity, $e_1$,

262: as a function of $C$ for different values of $\sin i$.  The ten solid

263: lines are for $\sin i$ varying in equal increments from $0.1$ to

264: $1.0$. The horizontal band is centered on the measured value of $e_1$

265: and its vertical width spans a $1-\sigma$ error bar in each direction.

266: Relevant parameters are taken from Table

267: \ref{table:parameter}.\label{fig:secular_search}}

268: \end{center}}

269: 

270: \subsection{Tidal Dynamics}

271: \label{subsec:dampnonlinear}

272: 

273: Tidal dissipation in the approximately synchronously rotating inner

274: planet decreases its orbital energy without significantly changing its

275: orbital angular momentum. It follows from equations

276: \refnew{eq:dampingtau} and \refnew{eq:eJE} that this causes $a_1$ to

277: vary as $d\ln a_1/dt \approx - 2 e_1^2/\tau$. Since secular

278: interactions transfer angular momentum but not energy between the

279: planets' orbits, $a_2$ remains constant. Conservation of total angular

280: momentum leads to,

281: \begin{eqnarray}

282: {{dJ_1}\over{dt}} & = & J_1 \left({1\over 2} {{d \ln a_1}\over{dt}} -

283: {{e_1^2}\over{1-e_1^2}} {{d\ln e_1}\over{dt}}\right) \nonumber \\ & =

284: & - {{d J_2}\over{dt}} = J_2\,{{e_2^2}\over{1-e_2^2}} {{d\ln

285: e_2}\over{dt}},

286: \label{eq:Lconserv}

287: \end{eqnarray}

288: where $J_i = \sqrt{1-e_i^2}

289: m_i n_i a_i^2$. Apse alignment implies $e_1 = f(a_1, e_2)$, from which 

290: we deduce that

291: \begin{eqnarray}

292: {{d\ln e_1}\over{dt}}& =&  -{1\over\tau}\, 

293: {{f_{e_2} (1-e_1^2) + 2 e_2^2 f_{a_1} {{J_2 (1-e_1^2)}\over{J_1

294: (1-e_2^2)}}}\over{f_{e_2} +

295: {{J_2}\over{J_1}}\,

296: ({{1-e_1^2}\over{1-e_2^2}})\,

297: ({{e_2}\over{e_1}})^2}}, \nonumber \\

298: {{d\ln e_2}\over{dt}} & = &

299: - {1\over \tau}\, 

300: {{(1 - e_1^2) - 2e_1^2\, f_{a_1}}\over{f_{e_2} +

301: {{J_2}\over{J_1}}\,

302: ({{1-e_1^2}\over{1-e_2^2}})\,

303: ({{e_2}\over{e_1}})^2}},

304: \label{eq:relatea1e2}

305: \end{eqnarray}

306: where $f_{a_1} \equiv \partial\ln f/\partial\ln a_1$ and $f_{e_2}

307: \equiv \partial\ln f/\partial \ln e_2$. Numerically, $f_{a_1}

308: \approx 8$ and $f_{e_2} \approx 1$, so currently $e_1$ and $e_2$ are damping

309: on time-scales of $8.6\tau$ and $40\tau$, respectively; $e_1$ damps

310: faster than $e_2$ because the inner planet's semi-major axis is

311: shrinking.

312: 

313: \section{Linear Approximation}

314: \label{sec:linear}

315: 

316: In \S \ref{sec:nonlinear} we assumed that tidal dissipation had

317: delivered the planets' orbits into a state of periapse alignment. Here

318: we justify this assumption by studying the secular dynamics and 

319: tidal evolution under the somewhat inaccurate approximation of small

320: orbital eccentricities. 

321: 

322: Our starting point is the expansion of the secular disturbing function

323: to second order in $e_1$ and $e_2$. Substituting the expanded

324: disturbing function into Lagrange's equations for the variations of

325: the $e$'s and the $\pomega$'s, and adding a term to represent tidal

326: damping of $e_1$, yields the following set of linear equations

327: (MD2000)

328: \begin{equation}

329: {{dI_1}\over{dt}} = i A_{11} I_1 + i A_{12} I_2 - I_1/\tau,\hskip0.5cm

330: {{dI_2}\over{dt}} = i A_{21} I_1 + i A_{22} I_2,

331: \label{eq:Iincludetau}

332: \end{equation}

333: for the complex variable $I_i \equiv e_i \exp(i\pomega_i)$. The

334: coefficients $A_{ij}$ read:

335: \begin{eqnarray}

336: A_{11} & = & {{\alpha^2}\over{4}} {{m_2}\over{M_*}} n_1 B_1 + \Delta, 

337: \hskip0.2cm 

338: A_{12}  =  - {{\alpha^2}\over{4}} {{m_2}\over{M_*}} n_1 B_2,\nonumber \\

339: A_{21} & = & - {{\alpha }\over{4}} {{m_1}\over{M_*}} n_2 B_2, \hskip0.7cm 

340: A_{22}  =   {{\alpha }\over{4}} {{m_1}\over{M_*}} n_2 B_1,

341: \label{eq:defineAjk}

342: \end{eqnarray}

343: with $\alpha = a_1/a_2$, $B_1 = b_{3/2}^{(1)}$, and $B_2 = b_{3/2}^{(2)}$;

344: $b_s^{(j)}$ is the usual Laplace coefficient.

345: %\begin{equation}

346: %b_{s}^{(j)} (\alpha) \equiv {1\over \pi}\int_0^{2\pi}

347: %{{\cos(j\phi)}\over{(1-2 \alpha \cos\phi + \alpha^2)^{s}}} d\phi.

348: %\label{eq:laplace}

349: %\end{equation}

350: The symbols $\Delta$ and $\tau$ retain their definitions from \S

351: \ref{sec:nonlinear}.

352: 

353: In the absence of tidal damping ($\tau\to \infty$), the general

354: solution of equation \refnew{eq:Iincludetau} can be expressed as a

355: super-position of two eigenvectors ${\cal I}_*$ with

356: \begin{equation}

357: {\cal I}_* = \left[e_{1*}\exp(i\,\phi_{1*}), e_{2*}

358: \exp(i\,\phi_{2*})\right]\, \exp(i\,g_* t ),

359: \label{eq:defI}

360: \end{equation}

361: where

362: \begin{mathletters}

363: \begin{eqnarray}

364: \hskip-0.6cm g_*& = & {{(A_{11}+A_{22}) \pm \sqrt{(A_{11}-A_{22})^2 +

365: 4 A_{12} A_{21}}}\over 2}\!\!, \\\hskip-0.6cm

366: {{e_{1*}}\over{e_{2*}}}\!\!&\! = &\!  \left|{{g_* -

367: A_{22}}\over{A_{21}}}\right| =\left|{A_{12}\over {A_{11}-g_*}}\right|,

368: \\\hskip-0.6cm \phi_{1*}\!\! &\! = &\!  \phi_{2*} +

369: {\pi\over 2}\left[1 + {\mbox{sgn}}(g_*-A_{22})\right].

370: \end{eqnarray} 

371: \label{eq:eigenvectors}

372: \end{mathletters}

373: We distinguish the two eigensolutions by $*=p$ and $*=m$ according to

374: the plus or minus sign that appears before the square root in the

375: expression for $g_*$.  Since $(e_{1p}/e_{2p})\, (e_{1m}/e_{2m}) =

376: A_{12}/A_{21} \approx 1.0$, one solution satisfies $e_{1} < e_{2} $

377: and the other $e_{1} > e_{2}$. If we take $\Delta = 5\times

378: 10^{-11}\s^{-1}$, the $*=m$ solution reproduces the observed

379: eccentricity ratio $e_{1m}/e_{2m} = 0.19$ and apse

380: alignment.\footnote{This value of $\Delta$ is different from that

381: required in \S \ref{subsec:currenteratio} because the linear

382: approximation is inaccurate for $e_2 = 0.42$.}  The other solution has

383: $e_{1p}/e_{2p} = 5.26$ and anti-aligned apses. If both eigenvectors

384: have non-zero lengths, the orbital eccentricities oscillate out of

385: phase on time-scales of thousands of years ($\sim 1/g_*$) as depicted

386: in Mayor \etal (\cite{mayor83443}).

387: 

388: When $\tau$ is finite, $a_1$ and therefore $A_{ij}$ vary with

389: time. Since $\tau \gg 1/g_*$, the solution for the eigenvectors given by 

390: equation \refnew{eq:eigenvectors} remains valid. However, the eigenvector 

391: components, $e_{i*}$, decay slowly with time as

392: \begin{eqnarray}

393: \gamma_{1*} & = & {1\over \tau}\, {{1+2 e_{2*}^2\, f_{a_1}

394:  {{J_2}\over{J_1}}} \over{1+{{J_2}\over{J_1}}\,({{e_{2*}}

395:  \over{e_{1*}}})^2}},\nonumber \\ \gamma_{2*} & = &

396:  {1\over \tau}\, {{1-2 e_{1*}^2\, f_{a_1}}

397:  \over{1+{{J_2}\over{J_1}}\,({{e_{2*}}

398:  \over{e_{1*}}})^2}},

399: \label{eq:ggamma}

400: \end{eqnarray}

401: where $\gamma_{i*} \equiv - d\ln e_{i*}/dt$.  In the limit of small

402: eccentricity, the expressions for the nonlinear damping rates given by

403: equation \refnew{eq:relatea1e2} reduce to those in equation

404: \refnew{eq:ggamma} provided $f_{e_2}=1$ as is appropriate for the

405: linear problem. Each of the four eigenvector components has its

406: individual damping rate. Both components of the $p$ eigenvector decay

407: more rapidly than those of the $m$ eigenvector. If the current state

408: consists mostly the $m$ eigenvector with a small admixture of the $p$

409: eigenvector, both components in the latter would decay on time-scale

410: $\approx \tau$, while the dominant components $e_{1m}$ and $e_{2m}$

411: would decay much more slowly, as $\gamma^{-1}_{1m}\approx 7\tau$ and

412: $\gamma^{-1}_{2m}\approx 32\tau$.

413: 

414: It is notable that the asymmetry in damping rates between the aligned

415: and the anti-aligned solutions only applies if the inner planet has an

416: orbital period $\leq 5$ days. For longer periods, the additional

417: precession rates (eqs. [\ref{eq:pretide}] - [\ref{eq:preGR}]) are

418: unimportant and $(e_{1m}/e_{2m}) \sim (e_{1p}/e_{2p}) \sim 1$.

419: 

420: \section{Conclusions}

421: \label{sec:conclusion}

422: 

423: Both the survival of HD $83443b$'s orbital eccentricity and the

424: locking of its apse to that of HD $83443c$ are straightforward

425: consequences of secular and tidal dynamics. The ratio $e_1/e_2$

426: provides compelling evidence for the tidal and rotational distortion

427: of HD $83443b$. It constrains unknown system parameters by relating

428: $C\equiv (k_{2}/k_{2J})(R_1/R_J)^5$ to $\sin i$.  In particular, we

429: find $C > 0.9$.

430: 

431: Apse alignment indicates that a minimum amount of tidal dissipation

432: has occurred over the life-time of the system, taken to be $N$ Gyrs.

433: By requiring that this is at least 3 times longer than $\tau$, we

434: deduce that $Q < 5 \times 10^5\, N C \sin i$.  It may also be possible

435: to infer a lower limit for $Q$ by integrating the orbits backward in

436: time.

437: 

438: At the present time $n_1/n_2 = 9.99 \pm 0.06$, suggestive of a $10:1$

439: mean motion resonance. This is probably just a coincidence. Moreover,

440: no physically important effects would be associated with this

441: resonance should it exist.

442: 

443: \begin{acknowledgements}

444: This research was initiated while Y. Wu participated in a CITA-ITP

445: postdoctoral exchange program funded by NSF Grant PHY-9907949. She

446: thanks Drs. Man-Hoi Lee, Stan Peale, and Brad Hansen for helpful

447: discussions. P. Goldreich's research is supported by NSF Grant

448: AST-0098301.

449: \end{acknowledgements}

450: 

451: \begin{thebibliography}{}

452: \bibitem[2000]{burrows} Burrows, A., Guillot, 

453: T., Hubbard, W.\ B., Marley, M.\ S., Saumon, D., Lunine, J.\ I., \& 

454: Sudarsky, D.\ 2000, \apjl, 534, L97 

455: \bibitem[1916]{einstein} Einstein, A. 1916, Annalen der Physik, Band 49, 50

456: \bibitem[1818]{gauss} Gauss, K. F. 1818, in Collected Works, Vol 3.,

457: 331 

458: \bibitem[1963]{goldreich} Goldreich, R.\ 1963, \mnras, 126, 257

459: \bibitem[1966]{goldreichsoter} Goldreich, P.~\& Soter, S.\ 1966,

460: Icarus, 5, 375

461: \bibitem[1981]{griffin} Griffin, R. F., \& Gunn, J. E. 1981, \aj,

462: 86, 588

463: \bibitem[1981]{hut} Hut, P.\ 1981, \aap, 99, 126

464: \bibitem[2000]{jha} Jha, S., Torres, G., Stefanik, R. P., Latham,

465: D. W., \& Mazeh, T. 2000, MNRAS, 317, 375

466: \bibitem[1998]{kiseleva} Kiseleva, L. G., Eggleton, P. P., \& Mikkola,

467: S. 1998, MNRAS, 300, 292

468: \bibitem[1991]{swift} Levinson, H., \& Duncan, M. 1991, Icarus, 108, 18L

469: \bibitem[2000]{mayor83443} Mayor, M., Naef, D., Pepe, F., Queloz, D.,

470: Santos, N.C., Udry, S., Burnet, M. 2000, in Penny, A.J., Artymowicz,

471: P., Lagrange, A.- M. \& Russell, S.S., eds. Planetary Systems in the

472: Universe, Observation, Formation and Evolution, ASP Conf. Series,

473: vol.??? 

474: \bibitem[2000]{MD2000} Murray, C.D., Dermott, S.F., Solar System

475: Dynamics, Cambridge University Press, 2000 (MD2000)

476: \bibitem[1939]{sterne} Sterne, T.E. 1939, \mnras, 99, 451

477: \bibitem[1995]{yoder} Yoder, C., in Ahrens, T. J. ed. A Handbook of

478: Physical Constants: Global Earth Physics, 1995, American Geophysical

479: Union

480: \end{thebibliography}{}

481: 

482: \begin{appendix}

483: 

484: \section{Secular Perturbations}

485: \label{sec:secular}

486: 

487: Secular changes of the inner planet's periapse angle, $\pomega_1$, and

488: eccentricity, $e_1$, are described by

489: \begin{equation}

490: {de_1\over dt}  =  {de_1\over dt}\Bigr|_2, \hskip1.5cm

491: {d \pomega_1\over dt}  =   {d \pomega_1\over dt}\Bigr|_2 + {d\pomega_1\over 

492: dt}\Bigr|_{\rm tide} +

493: {d\pomega_1\over dt}\Bigr|_{J_2} +

494: {d\pomega_1\over dt}\Bigr|_{\rm GR}, 

495: \label{eq:secular}

496: \end{equation}

497: where the subscript $2$ denotes effects arising from secular

498: interactions with the outer planet.

499: Similar equations apply to the evolution of $\pomega_2$ and $e_2$ but

500: the only significant perturbations are those from the inner

501: planet. Expressions for the precession rates due to the tidal and

502: rotational distortion of the inner planet and to general relativity

503: are given in the main body of the text. We obtain $d\pomega_1/dt|_2$

504: by applying a method devised by Gauss (\cite{gauss}) as outlined

505: below.

506: 

507: Consider a pair of coplanar rings formed by spreading the mass of each

508: planet along its orbit such that the line density is inversely

509: proportional to the orbital velocity. We decompose into radial, $\cal

510: R$, and tangential, $\cal T$, components the gravitational force

511: exerted on a line element of length $r_1 df_1$ by a line element of

512: length $r_2 df_2$, where $f$ denotes true anomaly and $r$ distance

513: from the star; 

514: \begin{equation} 

515: {\cal R} = {{G \rho_1 \rho_2 r_1 r_2

516: df_1 df_2}\over {|\boldr_1 - \boldr_2|^2}} \cos\theta, \hskip1.5cm

517: {\cal T} = {{G \rho_1 \rho_2 r_1 r_2 df_1 df_2}\over {|\boldr_1 -

518: \boldr_2|^2}}\sin \theta,

519: \label{eq:radialtan}

520: \end{equation}

521: with $\theta$ the angle between $\boldr_1$ and $\boldr_2 -

522: \boldr_1$. We remove $\theta$ in favor of $\phi$, the angle between

523: $\boldr_1$ and $\boldr_2$ ($\phi = f_2 + \pomega_2 - f_1 -

524: \pomega_1$), by applying the geometric relations,

525: $|\boldr_1-\boldr_2|\, \sin \theta = r_2 \sin \phi$ and $|\boldr_1 -

526: \boldr_2|\, \cos\theta = r_2 \cos \phi - r_1$. Then we integrate over

527: $df_2$ to obtain the components of the total force due to planet 2 on

528: the mass of planet 1 in the line element $r_1 df_1$. When this result

529: is substituted into Gauss' form of the perturbation equations (Gauss

530: \cite{gauss}) and integrated over $f_1$, we obtain\footnote{Our

531: derivation is slightly different from that given in MD2000, but the

532: results are identical.}

533: \begin{eqnarray}

534: {d \pomega_1\over dt}\Bigr|_2 & = & {{\sqrt{1-e_1^2}}\over{m_1 n_1 a_1

535: e_1}}\, \oint df_1 \, \left[ - \cos f_1 \oint df_2\, {\cal R} + {{

536: (2+e_1 \cos f_1)\, \sin f_1}\over{1+e_1 \cos f_1}} \oint df_2 \, {\cal

537: T} \right], \nonumber \\

538: {de_1\over dt}\Bigr|_2 & = & {{\sqrt{1-e_1^2}}\over{m_1

539: n_1 a_1}}\,\oint df_1\, \left[ \sin f_1 \oint df_2 \,{\cal R} +

540: {{(2\cos f_1 + e_1 +e_1 \cos^2 f_1)}\over{1+e_1 \cos f_1}} \oint

541: df_2\, {\cal T}\right].

542: \label{eq:gauss}

543: \end{eqnarray}

544: We have verified that when integrating these equations with parameters

545: appropriate to HD $83443$'s planetary system, we obtain results that

546: agree well with those obtained by direct integration of the orbits

547: using the SWIFT package (Levinson \& Duncan \cite{swift}).

548: 

549: \end{appendix}

550: 

551: 

552: \end{document}

553: 

554: SECULAR RESONANCE: When the frequency, or the combination of

555: frequencies, of the variations of the orbital elements of an asteroid

556: (precession rates of its perihelion and/or node) becomes nearly equal

557: to an eigenfrequency, or a combination of eigenfrequencies, of the

558: system of coupled planetary orbits (one of the corresponding

559: precession-rates that can be associated with the major planets), the

560: amplitude of these perturbations is amplified and we speak of secular

561: resonance.

562: 

563: \begin{deluxetable}{crrrcc} 

564: \tablecolumns{6} 

565: \tablewidth{0pt} 

566: \tablecaption{Parameters for the two planets of HD $83443$.\tablenotemark{a}

567: \label{table:parameter}}

568: \tablehead{ 

569: \colhead{Planet}    & \colhead{P (d)} &   \colhead{$e$}   & 

570: \colhead{$\pomega (\deg)$} & \colhead{$a$ (AU)} & 

571: \colhead{$m\sin i$}}

572: \startdata 

573: HD$83443b$ & $2.9853$ & $0.079

574: $ & $300$ & $0.0376$  & $1.14$ \\

575: % errors

576:     &   $\pm 0.0009$ & $\pm 0.008$ & $\pm 17$ & & \\

577: \\

578: HD$83443c$ & $29.83$ & $0.42$ &

579: $337 $ &  $0.175$  &

580: $0.53$ \\

581: %errors 

582:     & $\pm 0.18$ & $\pm 0.06$ & $\pm 10$ & & \\ \enddata

583: \tablenotetext{a}{Data taken from Mayor \etal

584: (\cite{mayor83443}). Planet masses are in unit of Saturn mass $M_{\rm

585: Sat} \approx 5.68\times 10^{29}\g$. The host star $M_* = 0.79

586: M_\odot$. }

587: \end{deluxetable} 

588: 

589: \begin{figure}

590: \plotone{f1.ps}

591: \caption{Values of the inner planet eccentricity, $e_1$, as a function

592: of $C$ for different values of $\sin i$.  The ten solid lines are for

593: $\sin i$ varying in equal increments from $0.1$ to $1.0$. The

594: horizontal band is centered on the measured value of $e_1$ and its

595: vertical width spans a $1-\sigma$ error bar in each direction.

596: Relevant parameters are taken from Table \ref{table:parameter}.}

597: \label{fig:secular_search}

598: \end{figure}

599: 

600: