1: \documentclass[preprint,12pt]{aastex}
2: %\usepackage{latexsym,graphicx,natbib}
3: %\usepackage{apjfonts,emulateapj5,psfig}
4: %\usepackage{apjfonts,emulateapj5}
5:
6: \shortauthors{Winn et al.~2006}
7: \shorttitle{Spin-Orbit Alignment in HD~189733}
8:
9: \begin{document}
10:
11: % ------------------------------------------------------------------------
12: % New commands
13: %
14: \def\ltsima{$\; \buildrel < \over \sim \;$}
15: \def\lsim{\lower.5ex\hbox{\ltsima}}
16: \def\gtsima{$\; \buildrel > \over \sim \;$}
17: \def\gsim{\lower.5ex\hbox{\gtsima}}
18: \def\lam{\lambda=-1\fdg4 \pm 1\fdg1}
19:
20: % -------------------------------------------------------------------------
21: %
22:
23: \bibliographystyle{apj}
24:
25: \title{
26: Measurement of the Spin-Orbit Alignment in\\
27: the Exoplanetary System HD~189733$^1$
28: }
29:
30: \author{
31: Joshua N.\ Winn\altaffilmark{2},
32: John Asher Johnson\altaffilmark{3},
33: Geoffrey W.\ Marcy\altaffilmark{3},
34: R. Paul Butler\altaffilmark{4},\\
35: Steven S. Vogt\altaffilmark{5},
36: Gregory W.\ Henry\altaffilmark{6},
37: Anna Roussanova\altaffilmark{2},
38: Matthew J. Holman\altaffilmark{7},\\
39: Keigo Enya\altaffilmark{8},
40: Norio Narita\altaffilmark{9},
41: Yasushi Suto\altaffilmark{9},
42: Edwin L.\ Turner\altaffilmark{10}
43: }
44:
45: \altaffiltext{1}{Data presented herein were obtained at the W.M.~Keck
46: Observatory, which is operated as a scientific partnership among the
47: California Institute of Technology, the University of California,
48: and the National Aeronautics and Space Administration, and was made
49: possible by the generous financial support of the W.~M.~Keck
50: Foundation.}
51:
52: \altaffiltext{2}{Department of Physics, and Kavli Institute for
53: Astrophysics and Space Research, Massachusetts Institute of
54: Technology, Cambridge, MA 02139, USA}
55:
56: \altaffiltext{3}{Department of Astronomy, University of California,
57: Mail Code 3411, Berkeley, CA 94720, USA}
58:
59: \altaffiltext{4}{Department of Terrestrial Magnetism, Carnegie
60: Institution of Washington, 5241 Broad Branch Road NW, Washington
61: D.C. USA 20015-1305}
62:
63: \altaffiltext{5}{UCO/Lick Observatory, University of California at
64: Santa Cruz, Santa Cruz CA USA 95064}
65:
66: \altaffiltext{6}{Center of Excellence in Information Systems,
67: Tennessee State University, 3500 John A.\ Merritt Blvd., Box 9501,
68: Nashville, TN 37209, USA}
69:
70: \altaffiltext{7}{Harvard-Smithsonian Center for Astrophysics, 60
71: Garden Street, Cambridge, MA 02138, USA}
72:
73: \altaffiltext{8}{Institute of Space and Astronautical Science, Japan
74: Aerospace Exploration Agency, 3-1-1, Yoshinodai, Sagamihara,
75: Kanagawa, 229-8510, Japan}
76:
77: \altaffiltext{9}{Department of Physics, The University of Tokyo, Tokyo
78: 113-0033, Japan}
79:
80: \altaffiltext{10}{Princeton University Observatory, Peyton Hall,
81: Princeton, NJ 08544, USA}
82:
83: \begin{abstract}
84:
85: We present spectroscopy of a transit of the exoplanet HD~189733b. By
86: modeling the Rossiter-McLaughlin effect (the anomalous Doppler shift
87: due to the partial eclipse of the rotating stellar surface), we find
88: the angle between the sky projections of the stellar spin axis and
89: orbit normal to be $\lam$. This is the third case of a ``hot
90: Jupiter'' for which $\lambda$ has been measured. In all three cases
91: $\lambda$ is small, ruling out random orientations with 99.96\%
92: confidence, and suggesting that the inward migration of hot Jupiters
93: generally preserves spin-orbit alignment.
94:
95: \end{abstract}
96:
97: \keywords{planetary systems --- planetary systems: formation ---
98: stars:~individual (HD~189733) --- stars:~rotation}
99:
100: \section{Introduction}
101:
102: A primary reason to study planets of other stars is to learn how
103: typical (or unusual) are the properties of the Solar system. For
104: example, the nearly circular orbits of Solar system planets were once
105: considered normal, but we now know that eccentric orbits of Jovian
106: planets are common (see, e.g., Halbwachs, Mayor, \& Udry 2005; or
107: Fig.~3 of Marcy et al.~2005). Likewise, gas giants were once thought
108: to inhabit only the far reaches of planetary systems, an assumption
109: that was exploded by the discovery of ``hot Jupiters'' (Mayor \&
110: Queloz 1995, Butler et al.~1997). This inspired theoretical work on
111: planetary migration mechanisms that can deliver Jovian planets to such
112: tight orbits (as recently reviewed by Thommes \& Lissauer 2005 and
113: Papaloizou \& Terquem 2006).
114:
115: Another striking pattern in the Solar system is the close alignment
116: between the planetary orbits and the Solar spin axis. The orbit
117: normals of the 8 planets are within a few degrees of one another (Cox
118: et al.~2000, p.~295), and the Earth's orbit normal is only 7~degrees
119: from the Solar spin axis (Beck \& Giles 2005, and references
120: therein). Presumably this alignment dates back 5~Gyr, when the Sun and
121: planets condensed from a single spinning disk. Whether or not this
122: degree of alignment is universal is unknown. For hot Jupiters in
123: particular, one might wonder whether migration enforces or perturbs
124: spin-orbit alignment.
125:
126: For exoplanets, the angle between the stellar spin axis and planetary
127: orbit normal (as projected on the sky) can be measured via the
128: Rossiter-McLaughlin (RM) effect: the spectral distortion observed
129: during a transit due to stellar rotation. The planet hides some of the
130: velocity components that usually contribute to line broadening,
131: resulting in an ``anomalous Doppler shift'' (for the theory, see Ohta
132: et al.~2005, Gim\'enez 2006, or Gaudi \& Winn~2006).
133:
134: Observations of the exoplanetary RM effect have been reported for
135: HD~209458 (Bundy \& Marcy 2000, Queloz et al.~2000, Winn et al.~2005,
136: Wittenmyer et al.~2005) and HD~149026 (Wolf et al.~2006). Here we
137: report observations of the RM effect for HD~189733. This system,
138: discovered by Bouchy et al.~(2005), consists of a K dwarf with a
139: transiting Jovian planet ($M_P = 1.15$~$M_{\rm Jup}$) in a 2.2~day
140: orbit. Our observations are presented in \S~2, our model in \S~3, and
141: our results in \S~4, followed by a brief summary and discussion.
142:
143: \section{Observations}
144:
145: We observed the transit of UT~2006~August~21 with the Keck~I 10m
146: telescope and the High Resolution Echelle Spectrometer (HIRES; Vogt et
147: al.~1994) following the usual protocols of the California-Carnegie
148: planet search, as summarized here. We employed the red cross-disperser
149: and placed the I$_2$ absorption cell into the light path to calibrate
150: the instrumental response and the wavelength scale. The slit width was
151: $0\farcs85$ and the typical exposure time was 3~minutes, giving a
152: resolution of 70,000 and a signal-to-noise ratio (SNR) of
153: 300~pixel$^{-1}$. We obtained 70 spectra over 7.5~hr bracketing the
154: predicted transit midpoint. To these were added 16 spectra that had
155: been obtained by the California-Carnegie group at random orbital
156: phases.
157:
158: We determined the relative Doppler shifts with the algorithm of Butler
159: et al.~(1996). We estimated the measurement uncertainties based on the
160: scatter in the solutions for each 2~\AA~section of the spectrum. For
161: the spectra obtained on 2006~Aug~21 the typical measurement error was
162: 0.8~m~s$^{-1}$, while for the other 16 spectra the error was
163: $\approx$1.3~m~s$^{-1}$. The data are given in Table~1 and plotted in
164: Fig.~1, with enlarged error bars to account for the intrinsic velocity
165: noise of the star (see \S~3).
166:
167: We also needed accurate photometry to pin down the planetary and
168: stellar radii and the orbital inclination. We observed the transit of
169: UT~2006~Jul~21 with KeplerCam on the 1.2m telescope at the Fred L.\
170: Whipple Observatory on Mt.\ Hopkins, Arizona. We used the SDSS
171: $z$~band filter and an exposure time of 30 seconds. After bias
172: subtraction and flat-field division, we performed aperture photometry
173: of HD~189733 and 14 comparison stars. The light curve of each
174: comparison star was normalized to have unit median, and the mean of
175: these normalized light curves was taken to be the comparison
176: signal. The light curve of HD~189733 was divided by the comparison
177: signal, and corrected for residual systematic effects by dividing out
178: a linear function based on the out-of-transit data. The light curve is
179: plotted in the top panel of Fig.~1.
180:
181: \begin{figure}[p]
182: \epsscale{0.65}
183: \plotone{f1.ps}
184: \caption{
185: Photometry and spectroscopy of HD~189733.
186: The top panel shows $z$ band photometry during a transit, along with the
187: best-fitting model (solid line).
188: The middle panel show the radial velocities
189: as a function of orbital phase (expressed in days), along with the model
190: (solid line).
191: The bottom panel is a close-up near the mid-transit time.
192: In all cases, the residuals
193: (observed$-$calculated) are plotted beneath the data.
194: \label{fig:1}}
195: \end{figure}
196:
197: \section{The Model}
198:
199: We fitted the fluxes and radial velocities with a parameterized model
200: based on a star and planet in a circular orbit about the center of
201: mass.\footnote{A circular orbit is a reasonable simplifying assumption
202: because of the effects of tidal circularization (see, e.g., Rasio et
203: al.~1996, Trilling et al.~2000, Dobbs-Dixon et al.~2004).} To
204: calculate the relative flux as a function of the projected separation
205: of the planet and the star, we assumed the limb darkening law to be
206: quadratic and employed the analytic formulas of Mandel \& Agol~(2002)
207: to compute the integral of the intensity over the unobscured portion
208: of the stellar disk. We fixed the limb-darkening coefficients at the
209: values $u_1=0.320$, $u_2=0.267$, based on the calculations of
210: Claret~(2004).
211:
212: To calculate the anomalous Doppler shift, we used the technique of
213: Winn et al.~(2005): we simulated RM spectra with the same data format
214: and noise characteristics as the actual data, and determined the
215: Doppler shifts using the same algorithm used on the actual data. The
216: simulations were based on a ``template'' spectrum representing the
217: emergent spectrum from a small portion of the photosphere. We scaled
218: the template spectrum in flux by $\epsilon$ and shifted it in velocity
219: by $v_p$, representing the spectrum of the occulted portion of the
220: stellar disk. We subtracted this spectrum from a
221: rotationally-broadened version of the template spectrum (broadened to
222: 3~km~s$^{-1}$ to mimic the disk-integrated spectrum of HD~189733), and
223: then ``measured'' the anomalous Doppler shift $\Delta v$. This was
224: repeated for a grid of $\{\epsilon, v_p\}$, and a polynomial function
225: was fitted to the resulting surface.
226:
227: The template spectrum should be similar to that of HD~189733 but
228: without significant rotational broadening. We tried three different
229: choices: two empirical spectra, and one theoretical spectrum. The two
230: empirical spectra were Keck/HIRES spectra (SNR~$\approx~800$,
231: $R\approx~10^5$) of HD~3561 (G3~{\sc v}, $v\sin i_S=1.2\pm
232: 0.5$~km~s$^{-1}$) and HD~3765 (K2~{\sc v}, $0.0\pm
233: 0.5$~km~s$^{-1}$). The former is 200~K hotter than HD~189733, while
234: the latter is more metal-rich. The theoretical spectrum, with a
235: resolution of 250,000, was taken from Coelho et al.~(2005) for a
236: non-rotating star with $T_{\rm eff}=5000$~K, $\log g=4.5$, and
237: [Fe/H]~$=0.0$.
238:
239: For each choice of the template spectrum, we derived the relation
240: between $\Delta v$, $\epsilon$, and $v_p$, and optimized the model as
241: described below. With one exception, the results did not depend
242: significantly on the choice of template spectrum (in the sense that
243: measurement errors caused much larger uncertainties). The single
244: exception was $v\sin i_S$, for which the results differed as much as
245: 3\%. For our final analysis, we used the relation
246: \begin{equation}
247: \Delta v = -\epsilon~v_p\left[1.252 - 0.351 \left( \frac{v_p}{{\rm 3~km~s}^{-1}} \right)^2 \right]
248: \end{equation}
249: derived from the empirical templates, but we also included an extra
250: error term of 6\% in $v\sin i_S$ as a conservative estimate of the
251: systematic error. In summary, the projected separation of the planet
252: and the star determines the transit depth $\epsilon$ and the
253: sub-planet rotation velocity\footnote{The sub-planet velocity is the
254: projected rotation velocity of the portion of the star hidden by the
255: planet, and is calculated assuming no differential rotation, an
256: assumption justified by Gaudi \& Winn (2006).} $v_p$, and then
257: Eq.~(1) is used to calculate the anomalous Doppler shift.
258:
259: The fitting statistic was
260: \begin{equation}
261: \chi^2 =
262: \sum_{j=1}^{86}
263: \left[
264: \frac{v_j({\mathrm{obs}}) - v_j({\mathrm{calc}})}{\sigma_{v,j}}
265: \right]^2
266: +
267: \sum_{j=1}^{752}
268: \left[
269: \frac{f_j({\mathrm{obs}}) - f_j({\mathrm{calc}})}{\sigma_{f,j}}
270: \right]^2
271: +
272: \left( \frac{\Delta\gamma}{{\rm 12~m~s^{-1}}} \right)^2
273: +
274: \left( \frac{M_S/M_\odot - 0.82}{0.03} \right)^2
275: ,
276: \end{equation}
277: where $f_j$(obs) is the flux observed at time $j$, $\sigma_{f,j}$ is
278: the corresponding uncertainty, and $f_j$(calc) is the calculated
279: value. A similar notation applies to the velocities. The last two
280: terms are {\it a priori} constraints explained below.
281:
282: It is important for $\sigma_{f,j}$ and $\sigma_{v,j}$ to include not
283: only measurement errors but also any unmodeled systematic errors. To
284: account for systematic errors in the photometry, we increased the
285: Poisson estimates of the errors by a factor of 1.2, at which point
286: $\chi^2/N_{\rm DOF} = 1$ when fitting only the fluxes. Determining the
287: appropriate weights for the velocities was more complex. HD~189733 is
288: chromospherically active and should exhibit velocity noise
289: (``photospheric jitter'') with an amplitude of 11~m~s$^{-1}$ according
290: to the empirical relations of Wright~(2005). However, the time scale
291: of the noise cannot be predicted as easily. Noise from spots or plages
292: would occur on the time scale of the rotation period
293: ($\approx$13~days), while noise from oscillations and flows occurs on
294: shorter time scales.
295:
296: We took the following approach. First, we fitted only the 16
297: velocities obtained sporadically prior to 2006~Aug~21, and found the
298: root-mean-squared (RMS) residual to be 12~m~s$^{-1}$, in agreement
299: with the Wright~(2005) relations. Therefore, for fitting purposes, we
300: inflated the error bars $\sigma_{v,j}$ of those 16 velocities to
301: 12~m~s$^{-1}$. Second, we fitted only the 44 {\it out-of-transit}
302: velocities measured on 2006~Aug~21, and found the RMS residual to be
303: 1.5~m~s$^{-1}$. In addition, there were correlations in the residuals
304: on a time scale of $\sim$15~minutes ($\sim$4 data points). The
305: correlations effectively reduce the number of independent data points
306: by 4, or equivalently, they double the error per point. Therefore, for
307: fitting purposes, we inflated the error bars $\sigma_{v,j}$ of all the
308: 2006~Aug~21 velocities to 3~m~s$^{-1}$. Apparently, for HD~189733,
309: most of the velocity noise occurs on a time scale longer than one
310: night.
311:
312: Our free parameters were the two bodies' masses and radii ($M_S$,
313: $M_P$, $R_S$ and $R_P$); the orbital inclination ($i$); the
314: mid-transit time ($T_c$); the line-of-sight stellar rotation velocity
315: ($v \sin i_S$); the angle between the projected stellar spin axis and
316: orbit normal ($\lambda$; see Ohta et al.~2005 or Gaudi \& Winn~2006
317: for a diagram of the coordinate system); the velocity zero point
318: ($\gamma$); a velocity offset specific to the night of 2006~Aug~21
319: ($\Delta\gamma$); and a long-term velocity gradient
320: $\dot{\gamma}$. The parameter $\Delta\gamma$ is needed because of the
321: photospheric jitter; the first {\it a priori} constraint in Eq.~(2)
322: enforces a reasonable level of noise. The gradient $\dot{\gamma}$ was
323: included to account for the long-period orbit of HD~189733 and its
324: companion star (Bakos et al.~2006a) or possible long-period
325: planets. We fixed the orbital period to be $2.218575$~days (Bouchy et
326: al.~2005, H{\'e}brard \& Lecavelier Des Etangs~2006). A well-known
327: degeneracy prevents $M_S$, $R_S$, and $R_P$ from being determined
328: independently. We broke this degeneracy with the second {\it a priori}
329: constraint in Eq.~(2), which enforces the spectroscopic determination
330: of $M_S$ by Bouchy et al.~(2005).
331:
332: We used a Markov Chain Monte Carlo algorithm to solve for the model
333: parameters and their uncertainties (see, e.g., Tegmark et al.~2004 or
334: Ford 2005). Our jump function was the addition of a Gaussian random
335: number to each parameter value. We set the perturbation sizes such
336: that $\sim$20\% of jumps are executed. We created 10 independent
337: chains, each with 500,000 points, starting from random initial
338: positions, and discarded the first 20\% of the points in each
339: chain. The Gelman \& Rubin~(1992) $R$ statistic was close to unity for
340: each parameter, a sign of good mixing and convergence. We merged the
341: chains and took the median value of each parameter to be our best
342: estimate, and the standard deviation as the 1~$\sigma$
343: uncertainty. For the special case of $v\sin i_S$, we added an
344: additional error of 6\% in quadrature, due to the systematic error
345: noted previously.
346:
347: \section{Results}
348:
349: The results are given in Table~2. Those parameters depending chiefly
350: on the photometry ($R_P$, $R_S$, $i$) are in agreement with the most
351: accurate results reported previously (Bakos et al.~2006b). Likewise,
352: our result for the planetary mass, $M_P = 1.13\pm 0.03$~$M_{\rm Jup}$,
353: agrees with the value $1.15\pm 0.04$~$M_{\rm Jup}$ measured by Bouchy
354: et al.~(2005). Our result for the projected rotation velocity is
355: $v\sin i_S = 3.0\pm 0.2$~km~s$^{-1}$. This agrees with the value
356: $3.5\pm 1.0$~km~s$^{-1}$ reported by Bouchy et al.~(2005), which was
357: based on the line broadening in their disk-integrated stellar
358: spectrum. It also agrees with the value $3.2\pm 0.7$~km~s$^{-1}$ based
359: on a similar analysis of our own Keck spectra (D.~Fischer, private
360: communication). The most interesting result is $\lam$. The sky
361: projections of the stellar spin axis and the orbit normal are aligned
362: to within a few degrees.
363:
364: \section{Summary and Discussion}
365:
366: We have monitored the apparent Doppler shift of HD~189733 during a
367: transit of its giant planet. By modeling the RM effect, we find that
368: the stellar spin axis and the orbit normal are aligned to within a few
369: degrees.
370:
371: This is the third exoplanetary system (and the third hot Jupiter) for
372: which it has been possible to measure $\lambda$. The first system was
373: HD~209458, for which $\lambda=-4\fdg4\pm 1\fdg4$ (Winn et al.~2005;
374: see also Wittenmyer et al.~2005, who modeled the RM effect but
375: required $\lambda=0$). The second system was HD~149026 (Wolf et
376: al.~2006), for which $\lambda = 11\pm 14\arcdeg$. The small observed
377: values of $\lambda$ suggest that the most common end-state of the
378: inward migration of a hot Jupiter involves a close alignment.
379:
380: With only 3 systems, we cannot yet measure the distribution of
381: $\lambda$, but we can test the hypothesis of random orientations
382: (i.e., a uniform distribution in $\lambda$). The weighted mean of the
383: measured values of $|\lambda|$ is 2.6\arcdeg. If we replace the
384: measured values by random numbers drawn from a uniform distribution,
385: the probability that the weighted mean\footnote{Here we have assumed
386: the uncertainty in $\lambda$ is independent of $\lambda$, a good
387: approximation because all 3 systems have an intermediate impact
388: parameter (Gaudi \& Winn 2006).} will be this small
389: is only 0.04\%. Hence we rule out the hypothesis of random
390: orientations with 99.96\% confidence.
391:
392: Winn et al.~(2005) argued that tides from the star would not
393: ordinarily cause alignment within the star's main-sequence
394: lifetime. There are therefore two basic possibilities: either the
395: alignment is primordial and was not disturbed by migration, or there
396: was a different mechanism to damp any initial or induced
397: misalignment. Among the various theories of hot Jupiter migration,
398: some would tend to enhance any initial misalignments, and are thereby
399: constrained by our results. Such scenarios include planet-planet
400: scattering followed by circularization (Rasio \& Ford 1996,
401: Weidenschilling \& Marzari 1996), Kozai migration (Wu \& Murray 2003;
402: Eggenberger et al.~2004), and tidal capture (Gaudi~2003).
403:
404: The agreement among the three systems studied to date is clear, but
405: should not discourage future measurements. Obviously, a sample of
406: three is only barely sufficient to draw a conclusion. And of course,
407: the discovery of even a single example of a grossly misaligned system
408: would be of great interest.
409:
410: \acknowledgments We thank Debra Fischer for running SME on our
411: spectra, and Scott Gaudi for very helpful discussions. We recognize
412: and acknowledge the very significant cultural role and reverence that
413: the summit of Mauna Kea has always had within the indigenous Hawaiian
414: community. We are most fortunate to have the opportunity to conduct
415: observations from this mountain.
416:
417: \begin{thebibliography}{}
418:
419: \bibitem[Bakos et al.(2006a)]{2006ApJ...641L..57B} Bakos, G.~{\'A}.,
420: P{\'a}l, A., Latham, D.~W., Noyes, R.~W., \& Stefanik, R.~P.\ 2006,
421: \apjl, 641, L57
422:
423: \bibitem[Bakos et al.(2006b)]{b06b} Bakos, G.~{\'A}., et al.\ 2006,
424: ApJ, submitted [astro-ph/0603291]
425:
426: \bibitem[Beck \& Giles(2005)]{2005ApJ...621L.153B} Beck, J.~G., \&
427: Giles, P.\ 2005, \apjl, 621, L153
428:
429: \bibitem[Bouchy et al.(2005)]{2005A&A...444L..15B} Bouchy, F., et al.\
430: 2005, \aap, 444, L15
431:
432: \bibitem[Bundy \& Marcy(2000)]{2000PASP..112.1421B} Bundy, K.~A., \&
433: Marcy, G.~W.\ 2000, \pasp, 112, 1421
434:
435: \bibitem[Butler et al.(1996)]{1996PASP..108..500B} Butler, R.~P.,
436: Marcy, G.~W., Williams, E., McCarthy, C., Dosanjh, P., \& Vogt,
437: S.~S.\ 1996, \pasp, 108, 500
438:
439: \bibitem[Butler et al.(1997)]{1997ApJ...474L.115B} Butler, R.~P.,
440: Marcy, G.~W., Williams, E., Hauser, H., \& Shirts, P.\ 1997, \apjl,
441: 474, L115
442:
443: \bibitem[Claret(2004)]{Claret.2004} Claret 2004, \aap, 428, 1001
444:
445: \bibitem[Coelho et al.(2005)]{2005A&A...443..735C} Coelho, P., Barbuy,
446: B., Mel{\'e}ndez, J., Schiavon, R.~P., \& Castilho, B.~V.\ 2005,
447: \aap, 443, 735
448:
449: \bibitem[Cox(2000)]{2000asqu.book.....C} Cox, A.~N.\ 2000, Allen's
450: astrophysical quantities, 4th ed.~(New York: AIP Press)
451:
452: \bibitem[Deming et al.(2006)]{2006ApJ...644..560D} Deming, D.,
453: Harrington, J., Seager, S., \& Richardson, L.~J.\ 2006, \apj, 644,
454: 560
455:
456: \bibitem[Dobbs-Dixon et al.(2004)]{2004ApJ...610..464D} Dobbs-Dixon,
457: I., Lin, D.~N.~C., \& Mardling, R.~A.\ 2004, \apj, 610, 464
458:
459: \bibitem[Ford(2005)]{2005AJ....129.1706F} Ford, E.~B.\ 2005, \aj, 129,
460: 1706
461:
462: \bibitem[Gaudi(2003)]{gaudi03} Gaudi, B.~S.\ 20003 (astro-ph/0307280)
463:
464: \bibitem[Gaudi \& Winn(2006)]{gw06} Gaudi, B.~S.\ \& Winn, J.~N.\
465: 2006, ApJ, submitted [astro-ph/0608071]
466:
467: \bibitem[Gelman \& Rubin(1992)]{Gelman.1992} Gelman, A.\ \& Rubin,
468: D.~B.\ 1992, Stat. Sci., 7, 457
469:
470: \bibitem[Gimenez(2006)]{gim06} Gim\'enez, A.\ 2006, ApJ, 650, 408
471:
472: \bibitem[Halbwachs et al.(2005)]{2005A&A...431.1129H} Halbwachs,
473: J.~L., Mayor, M., \& Udry, S.\ 2005, \aap, 431, 1129
474:
475: \bibitem[H{\'e}brard \& Lecavelier Des
476: Etangs(2006)]{2006A&A...445..341H} H{\'e}brard, G., \& Lecavelier
477: Des Etangs, A.\ 2006, \aap, 445, 341
478:
479: \bibitem[Mandel \& Agol(2002)]{2002ApJ...580L.171M} Mandel, K., \&
480: Agol, E.\ 2002, \apjl, 580, L171
481:
482: \bibitem[Marcy et al.(2005)]{2005PThPS.158...24M} Marcy, G., Butler,
483: R.~P., Fischer, D., Vogt, S., Wright, J.~T., Tinney, C.~G., \&
484: Jones, H.~R.~A.\ 2005, Progress of Theoretical Physics Supplement,
485: 158, 24
486:
487: \bibitem[Mayor \& Queloz(1995)]{1995Natur.378..355M} Mayor, M., \&
488: Queloz, D.\ 1995, \nat, 378, 355
489:
490: \bibitem[Ohta et al.(2005)]{2005ApJ...622.1118O} Ohta, Y., Taruya, A.,
491: \& Suto, Y.\ 2005, \apj, 622, 1118
492:
493: \bibitem[Papaloizou \& Terquem(2006)]{pt06} Papaloizou, J.~C.~B.\ \&
494: Terquem, C.\ 2005, Rept.\ Prog.\ Phys.~69, 119
495:
496: \bibitem[Queloz et al.(2000)]{2000A&A...359L..13Q} Queloz, D.,
497: Eggenberger, A., Mayor, M., Perrier, C., Beuzit, J.~L., Naef, D.,
498: Sivan, J.~P., \& Udry, S.\ 2000, \aap, 359, L13
499:
500: \bibitem[Rasio \& Ford(1996)]{rasio96} Rasio, F.~A., \& Ford, E.~B.\
501: 1996, Science, 274, 954
502:
503: \bibitem[Rasio et al.(1996)]{1996ApJ...470.1187R} Rasio, F.~A., Tout,
504: C.~A., Lubow, S.~H., \& Livio, M.\ 1996, \apj, 470, 1187
505:
506: \bibitem[Tegmark et al.(2004)]{2004PhRvD..69j3501T} Tegmark, M., et
507: al.\ 2004, \prd, 69, 103501
508:
509: \bibitem[Thommes \& Lissauer(2005)]{2005asli.symp...41T} Thommes,
510: E.~W., \& Lissauer, J.~J.\ 2005, Astrophysics of Life, 41
511:
512: \bibitem[Trilling(2000)]{Trilling.2000} Trilling, D.~E.\ 2000, \apjl,
513: 537, L61
514:
515: \bibitem[Vogt et al.(1994)]{1994SPIE.2198..362V} Vogt, S.~S., et al.\
516: 1994, \procspie, 2198, 362
517:
518: \bibitem[Weidenschilling \& Marzari(1996)]{1996Natur.384..619W}
519: Weidenschilling, S.~J., \& Marzari, F.\ 1996, \nat, 384, 619
520:
521: \bibitem[Winn et al.(2005)]{2005ApJ...631.1215W} Winn, J.~N., et al.\
522: 2005, \apj, 631, 1215
523:
524: \bibitem[Wittenmyer et al.(2005)]{2005ApJ...632.1157W} Wittenmyer,
525: R.~A., et al.\ 2005, \apj, 632, 1157
526:
527: \bibitem[Wolf et al.(2006)]{wolf06} Wolf, A.\ et al.\ 2006, ApJ, in
528: press
529:
530: \bibitem[Wright(2005)]{2005PASP..117..657W} Wright, J.~T.\ 2005,
531: \pasp, 117, 657
532:
533: \end{thebibliography}
534:
535: \begin{deluxetable}{lcccc}
536: \tabletypesize{\normalsize}
537: \tablecaption{Radial Velocities of HD~189733\label{tbl:rv}}
538: \tablewidth{0pt}
539:
540: \tablehead{
541: \colhead{JD} & \colhead{Radial Velocity [m~s$^{-1}$]} & \colhead{Measurement Uncertainty [m~s$^{-1}$]}
542: }
543:
544: \startdata
545: \input tab1.tex
546: \enddata
547:
548: \tablecomments{Column 1 gives the Julian Date at the time of the
549: photon-weighted mid-exposure. Column 3 gives the measurement
550: uncertainties, which do {\it not} include the estimated photospheric
551: jitter (see \S~3).}
552:
553: \end{deluxetable}
554:
555: \begin{deluxetable}{lcccc}
556: \tabletypesize{\normalsize}
557: \tablecaption{System Parameters of HD~189733\label{tbl:params}}
558: \tablewidth{0pt}
559:
560: \tablehead{
561: \colhead{Parameter} & \colhead{Value} & \colhead{Uncertainty}
562: }
563:
564: \startdata
565: $M_S/M_\odot$& $ 0.82$ & $ 0.03$ \\
566: $M_P/M_{\rm Jup}$& $ 1.13$ & $ 0.03$ \\
567: $R_S/R_\odot$& $ 0.73$ & $ 0.02$ \\
568: $R_P/R_{\rm Jup}$& $ 1.10$ & $ 0.03$ \\
569: $i$~[deg]& $ 86.1$ & $ 0.2$ \\
570: $T_c$~[HJD]& $ 2453937.7759$ & $ 0.0001$ \\
571: $v \sin i_S$~[km~s$^{-1}$]& $ 2.97$ & $ 0.22$ \\
572: $\lambda$~[deg]& $ -1.4$ & $ 1.1$ \\
573: $\gamma$~[m~s$^{-1}$]& $ 5.0$ & $ 10.1$ \\
574: $\Delta\gamma$~[m~s$^{-1}$]& $ -15.0$ & $ 4.8$ \\
575: $\dot{\gamma}$~[m~s$^{-1}$~yr$^{-1}$]& $ -1.9$ & $ 3.3$
576: \enddata
577:
578: \end{deluxetable}
579:
580: \end{document}
581: