1:
2: \documentclass[12pt,preprint]{aastex}
3:
4:
5:
6:
7: \newcommand{\kepler}{{\it Kepler}}
8: \newcommand{\tpf}{{\it TPF}}
9: \newcommand{\hst}{{\it HST}}
10: \newcommand{\cmpss}{cm~s$^{-2}$}
11: \newcommand{\gmcc}{gm~cm$^{-3}$}
12: \newcommand{\mps}{m~s$^{-1}$}
13: \newcommand{\kps}{km~s$^{-1}$}
14: \newcommand{\Mjup}{\mbox{$M_{\rm J}$}}
15: \newcommand{\Rjup}{\mbox{$R_{\rm J}$}}
16: \newcommand{\Mplan}{\mbox{$M_{\rm p}$}}
17: \newcommand{\Rplan}{\mbox{$R_{\rm p}$}}
18: \newcommand{\Mstar}{\mbox{$M_{\rm *}$}}
19: \newcommand{\Rstar}{\mbox{$R_{\rm *}$}}
20: \newcommand{\Pstar}{\mbox{$P_{\rm *}$}}
21: \newcommand{\Msun}{\mbox{$M_\odot$}}
22: \newcommand{\Rsun}{\mbox{$R_\odot$}}
23: \newcommand{\Teff}{\mbox{$T_{\rm eff}$}}
24: \newcommand{\Logg}{\mbox{$\log g$}}
25: \newcommand{\Loggplanet}{\mbox{$\log g_p$}}
26: \newcommand{\MonH}{\mbox{[M/H]}}
27: \newcommand{\NaH}{\mbox{[Na/H]}}
28: \newcommand{\SiH}{\mbox{[Si/H]}}
29: \newcommand{\TiH}{\mbox{[Ti/H]}}
30: \newcommand{\FeH}{\mbox{[Fe/H]}}
31: \newcommand{\NiH}{\mbox{[Ni/H]}}
32: \newcommand{\Vsini}{\mbox{$v\sin i$}}
33: \newcommand{\kms}{\mbox{km s$^{-1}$}}
34: \newcommand{\rchisq}{\mbox{$\chi_{\rm r}^2$}}
35: \newcommand{\Mgb}{\mbox{\ion{Mg}{1} b}}
36: \newcommand{\Ytwo}{Y$^2$}
37:
38: \newcommand{\xonb}{XO-4b}
39: \newcommand{\xon}{XO-4}
40: \newcommand{\vMs}{1.32}
41: \newcommand{\eMs}{0.02}
42: \newcommand{\vRs}{1.55}
43: \newcommand{\eRs}{0.05}
44: \newcommand{\vaoRs}{7.7}
45: \newcommand{\eaoRs}{0.2}
46: \newcommand{\vRpoRs}{0.089}
47: \newcommand{\eRpoRs}{0.001}
48: \newcommand{\vAge}{2.1}
49: \newcommand{\eAge}{0.6}
50: \newcommand{\sptype}{F5V}
51: \newcommand{\vTeff}{5700}
52: \newcommand{\eTeff}{70}
53: \newcommand{\vfeoh}{-0.04}
54: \newcommand{\efeoh}{0.03}
55: \newcommand{\vlogg}{4.18}
56: \newcommand{\elogg}{0.07}
57: \newcommand{\vVsini}{8.8}
58: \newcommand{\eVsini}{0.5}
59: \newcommand{\vrvK}{163}
60: \newcommand{\ervK}{16}
61: \newcommand{\vDs}{293}
62: \newcommand{\eDs}{19}
63: \newcommand{\vjd}{2454485.9322}
64: \newcommand{\ejd}{0.0004}
65: \newcommand{\vap}{0.0555}
66: \newcommand{\eap}{0.0011}
67: \newcommand{\vbp}{0.18}
68: \newcommand{\ebp}{0.14}
69: \newcommand{\vperiod}{4.12502}
70: \newcommand{\eperiod}{0.00002}
71: \newcommand{\vMp}{1.72}
72: \newcommand{\eMp}{0.20}
73: \newcommand{\vRp}{1.34}
74: \newcommand{\eRp}{0.048}
75: \newcommand{\vincl}{88.7}
76: \newcommand{\eincl}{1.1}
77:
78: \newcommand{\valq}{0.0XXX}
79: \newcommand{\errq}{0.0XXX}
80: \newcommand{\vsect}{1.2XX}
81: \newcommand{\esect}{0.1XX}
82:
83: \newcommand{\gsc}{GSC 03793-01994}
84:
85: \slugcomment{Submitted for publication in the Astrophysical Journal}
86:
87: \received{2008 May 01}
88: \begin{document}
89:
90: \title{XO-4b: An Extrasolar Planet Transiting an F5V Star}
91:
92: \author{
93: P.~R.~McCullough\altaffilmark{1},
94: Christopher~J.~Burke\altaffilmark{1},
95: Jeff~A.~Valenti\altaffilmark{1},
96: Doug~Long\altaffilmark{1},
97: Christopher~M.~Johns-Krull\altaffilmark{2,3},
98: P.~Machalek\altaffilmark{1,3,4},
99: K.~A.~Janes\altaffilmark{5},
100: B.~Taylor\altaffilmark{6},
101: J.~Gregorio\altaffilmark{7},
102: C.~N.~Foote\altaffilmark{8},
103: Bruce~L.~Gary\altaffilmark{9},
104: M.~Fleenor\altaffilmark{10},
105: Enrique~Garc\'{i}a-Melendo\altaffilmark{11},
106: T.~Vanmunster\altaffilmark{12}
107: }
108:
109:
110: \email{pmcc@stsci.edu}
111:
112: \altaffiltext{1}{Space Telescope Science Institute, 3700 San Martin Dr., Baltimore MD 21218}
113: \altaffiltext{2}{Dept. of Physics and Astronomy, Rice University, 6100 Main Street, MS-108, Houston, TX 77005}
114: \altaffiltext{3}{Visiting Astronomer, McDonald Observatory, which is operated by the University of Texas at Austin.}
115: \altaffiltext{4}{Department of Physics and Astronomy, Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218}
116: \altaffiltext{5}{Boston University, Astronomy Dept., 725 Commonwealth Ave., Boston, MA 02215}
117: \altaffiltext{6}{Lowell Observatory, 1400 W. Mars Hill Rd., Flagstaff, AZ 86001}
118: \altaffiltext{7}{Obs. Atalaia, Alcabideche, Portugal}
119: \altaffiltext{8}{Vermillion Cliffs Observatory, Kanab, UT}
120: \altaffiltext{9}{Hereford Arizona Observatory, 5320 E. Calle Manzana, Hereford, AZ 85615}
121: \altaffiltext{10}{Volunteer Observatory, Knoxville, TN}
122: \altaffiltext{11}{Esteve Duran Observatory Foundation, Montseny 46, 08553 Seva, Spain}
123: \altaffiltext{12}{CBA Belgium Observatory, Walhostraat 1A, B-3401 Landen, Belgium}
124:
125: \begin{abstract}
126: We report the discovery of the planet \xonb, which transits the star \xon\ (\gsc, V=10.7, \sptype).
127: Transits are 1.0\% deep and 4.4 hours in duration.
128: The star \xon\ has
129: a mass of $\vMs\pm\eMs$ \Msun,
130: a radius of $\vRs\pm\eRs$ \Rsun,
131: an age of $\vAge\pm\eAge$ Gyr,
132: a distance of $\vDs\pm\eDs$ pc,
133: an effective temperature of $\vTeff\pm\eTeff$ K,
134: a logarithmic iron abundance of $\vfeoh\pm\efeoh$ relative to solar,
135: a logarithmic surface gravity in cgs units of $\vlogg\pm\elogg$,
136: and
137: a projected rotational velocity of $\vVsini\pm\eVsini$ \kps.
138: The star \xon\ has periodic radial velocity variations with a semi-amplitude of $\vrvK\pm\ervK$ \mps, due to the planet \xonb.
139: The planet \xonb\ has
140: a mass of $\vMp\pm\eMp$ \Mjup,
141: a radius of $\vRp\pm\eRp$ \Rjup,
142: an orbital period of $\vperiod\pm\eperiod$ days,
143: and
144: a heliocentric Julian date at mid-transit of $\vjd\pm\ejd$.
145: We analyze scintillation-limited differential R-band photometry of \xonb\ in
146: transit made with a 1.8-m telescope under photometric conditions,
147: yielding photometric precision of 0.6 to 2.0 millimag per one-minute interval.
148: The declination of \xon\ places it within the continuous viewing zone of
149: the Hubble Space Telescope (\hst), which permits observation without
150: interruption caused by occultation by the Earth.
151: Because the stellar rotation periods of the three hottest stars
152: orbited by transiting gas-giant planets are 2.0, 1.1, and 2.0 times
153: the planetary orbital periods, we note the possibility of resonant
154: interaction.
155: \end{abstract}
156:
157: \keywords{binaries: eclipsing -- planetary systems -- stars: individual
158: (\gsc) -- techniques: photometric -- techniques: radial velocities}
159:
160: \section{Introduction}
161:
162: During 2007, the number of reported planets that transit stars brighter than
163: $V=13$ increased markedly from 9 to 23
164: planets\footnote{Extrasolar Planets Encyclopaedia, http://exoplanet.eu}.
165: The number of transiting planets reported in the first half of 2008 has
166: already exceeded the total number reported in 2007.
167: This rapid increase in discovery rate is due in part to recently matured search
168: techniques and analysis software applied to thousands of observations
169: of hundreds of thousands of stars by multiple research groups.
170: The planet \xonb\ reported here is another in a sequence of discoveries
171: by the XO project (McCullough et al.\ 2005), facilitated by a collaboration
172: between professional and amateur astronomers.\footnote{This paper
173: includes data taken at the Haleakala summit (maintained by the University
174: of Hawaii), Lowell Observatory, McDonald Observatory (operated by the
175: University of Texas at Austin), and numerous backyard observatories.}
176:
177: The star \xon\ is bright enough to appear in a few all-sky surveys with
178: the following identifiers:
179: \gsc\ (Morrison et al.\ 2001),
180: 2MASS J07213317+5816051 (Cutri et al.\ 2003),
181: and
182: TYC 3793--01994--1 (H{\o}g et al.\ 2000).
183: For \xon\ the Tycho Catalog reports
184: right ascension
185: $\alpha = 7^{\rm h}21^{\rm m}33^{\rm s}.159\pm0.026\arcsec$,
186: declination
187: $\delta = +58\arcdeg16\arcmin4\farcs98 \pm 0.028\arcsec$,
188: and proper motions
189: $\mu_\alpha = -18.1 \pm 2.9$ mas yr$^{-1}$
190: and
191: $\mu_\delta = -4.0 \pm 2.9$ mas yr$^{-1}$,
192: all for equinox and epoch J2000.
193:
194:
195: Several recently discovered transiting planets expanded the parameter space
196: of \textit{transiting} planets in one way or another at the time of their discovery.
197: For example, GJ 426b (Gillon et al.\ 2007) is the smallest and least massive
198: transiting planet, and also the only one known to transit an M dwarf.
199: XO-3b (Johns--Krull et al.\ 2008) is the most massive ($\Mplan=12\Mjup$).
200: HAT--P--2b (Bakos et al.\ 2007) had the largest eccentricity ($e=0.52$), until
201: HD 17156b (Barbieri et al.\ 2007) was discovered
202: with a larger eccentricity ($e=0.67$) and the longest known period
203: ($P=21$ days), demonstrating the ephemeral nature of such records.
204: As Harwit (1981) discusses, the limits of the extent of any parameter space
205: begin to show themselves not first in the outliers themselves, but in the
206: ``repetitions'' of discoveries. Simply put, if nearly every new discovery is
207: unique in some significant way, then clearly we have not yet begun to bound
208: the parameter space.
209: \xonb\ is not particularly unusual with respect to other known transiting
210: planets, except perhaps that it orbits a relatively warm star. Only
211: HAT--P--6b (Noyes et al.\ 2008) and XO-3b orbit slightly warmer stars.
212: In fact, the HAT--P--6 and \xon\ systems are quite similar.
213:
214: To stimulate additional observations and analysis of the \xon\ system,
215: we report here the characteristics of both the star and the planet.
216: In Section \ref{sec:obs} we describe the observations, which include time-series
217: photometry, out-of-transit
218: ${\rm BVR_CI_C}$ photometry, and spectroscopy with and without
219: iodine absorption lines superposed.
220: In Section \ref{sec:analysis} we analyze the data and derive the ephemeris
221: of the transits, measure time-dependent radial velocities, and the
222: physical characteristics of the star \xon, using both its spectrum
223: and its light curve.
224: Section \ref{sec:lightcurve}
225: applies the numerical models commonly used for transits
226: to scintillation-limited photometry.
227: Section \ref{sec:discussion} discusses the results and summarizes the conclusions.
228:
229: \section{Observations\label{sec:obs}}
230:
231: \subsection{Survey and Extended-Team Photometry}
232:
233: In many ways \xon\ presented an excellent signature of a transiting
234: planet in the XO survey photometry. Figure \ref{fig:finder} shows the
235: star is relatively isolated: our automated scripts had estimated that
236: 86\% of the flux within the XO cameras' 75\arcsec-radius photometric
237: aperture was due to \xon. The pixel centroid shifts of \xon\ are
238: consistent with the transit signal being intrinsic to \xon\ rather
239: than the result of a fainter eclipsing stellar binary within the
240: photometric aperture. Making use of the TYCHO-2 and 2MASS catalogs,
241: the 19 mas yr$^{-1}$ proper motion combined with V$_{T}$=10.85 and
242: (V$_{T}$-J)=1.19 places \xon\ on the reduced proper motion diagram of
243: Gould \& Morgan (2003) within the $R_{\star}<1.25$ \Rsun\ cutoff,
244: although by the end of our analysis, we would derive a larger radius.
245: Transits were $\sim$1\% deep, flat-bottomed, and steep-edged, all
246: signs of a good candidate for a transiting planet (Figure \ref{fig:xolc};
247: Table \ref{table:photometry}). Initially, the 4-hour
248: transit duration from the XO survey data was too long
249: given the initial G8V spectral type estimate for \xon\ from catalog broad band
250: photometry. However, more accurate photometry and
251: spectroscopy yields an earlier, \sptype\ spectral type and
252: larger $R_{\star}$, which is consistent with the
253: 4-hour transit duration.
254:
255: The time line for XO observations covers $\sim$ 100 cycles of \xonb,
256: which provided an ephemeris sufficiently accurate for follow up photometric
257: observations to recover the transit. The XO survey observed ingresses
258: of five separate transits and egresses of six transits, with only
259: one transit observed in its entirety from ingress to egress.
260: With multiple transits
261: observed and with at least one pair separated by only one cycle, the
262: chance of misidentifying a longer-period orbit with a shorter-period
263: harmonic was small.
264:
265: Extended Team\footnote{The XO Extended Team consists of volunteers who
266: provide photometric follow up of exoplanet candidates as described
267: by McCullough \& Burke (2007).}
268: time-series photometry in 2007 November confirmed
269: \xon\ is the variable star, verified and improved the transit's ephemeris
270: and shape, i.e. depth, duration, and abruptness of ingress and egress.
271: The Extended Team recorded time-series photometry of seven transits;
272: the data are in Table \ref{table:photometry}.
273: A planetary transit has a nearly achromatic shape; many triple stars
274: do not.
275: The Extended Team observed
276: transits of \xonb\ once each in B, V, and I spectral bands and
277: multiple times in R band. The observed
278: shape of the transit is the same in each band, and
279: each observed depth equals the average depth (1.0\%) in R band, within
280: the typical uncertainty of 0.1\%.
281:
282:
283: All-sky photometric measurements were made on three dates (2008
284: January 10, 19 and 20) using a 0.35-meter telescope located in
285: Arizona in B, V, $\rm
286: R_c$, and $\rm I_c$ bands. Eight Landolt star fields and the \xon\
287: star field were observed, between airmasses of 1.18 to 1.3.
288: For the B and V bands a total of 54 Landolt stars were observed
289: multiple times per night; for the $\rm R_c$ and $\rm I_c$ bands 33
290: Landolt stars were observed multiple times. The Landolt field
291: observations provide solutions to the color and airmass terms for each
292: night. Applying the solution to the \xon\ field yields the standard
293: photometry as given in Table~\ref{table:allsky}. The r.m.s. residuals of
294: the Landolt standard stars around the solution provide the
295: uncertainties in photometry.
296:
297: \subsection{Scintillation-limited Photometry with a 1.8-m Telescope\label{sec:bestlc}}
298:
299: With the Perkins 1.8-m telescope at Lowell Observatory, we obtained R band
300: time series photometry on 2008 January 20 (UT). We
301: obtained five-second CCD exposures at 8-second cadence for six hours
302: of \xon\ and a single comparison star, for a total of 2448 exposures
303: (Figures \ref{fig:bestlc} and \ref{fig:inout}).
304: Incidentally, these observations with the PRISM instrument of
305: the \xonb\ transit followed those made by the same telescope and instrument earlier
306: the same night of a transit of XO-5b (Burke et al.\ 2008). The PRISM sensor
307: is a cryogenically-cooled Fairchild CCD with 2048 pixels by 2048 pixels,
308: of which we used a 617 pixel by 144 pixel subarray, or 4\arcmin.01 by
309: 0\arcmin.94 at 0\arcsec.39 per pixel, in order to reduce readout times
310: and thereby to increase the fraction of elapsed time collecting photons.
311: The system gain was 2.66 e$^-$ ADU$^{-1}$; the read noise was 7.8 e$^-$
312: r.m.s., and the detector has been demonstrated to be linear to within
313: $\pm$1\% for fewer than 1.0$\times 10^5$ electrons pixel$^{-1}$. The
314: shutter and filters are located near a re-imaged pupil, so any variations
315: from ideal performance will be identical across the focal plane and
316: removed by the differential photometry. The telescope was intentionally
317: de-focused to produce annulus-shaped stellar images
318: in order to reduce the peak
319: irradiance on the CCD detector from the brightest star to
320: $\la 0.8 \times 10^{5}$ electrons pixel$^{-1}$ per 5-s exposure. The de-focused
321: stellar images gradually increased in diameter from 3\arcsec\ to 6\arcsec\
322: as the telescope
323: tracked the star from airmass 1.1 to 2.1. The telescope has an offset
324: guider which maintained the image position to $\sim 1$ pixel r.m.s. in
325: each axis, throughout the 6 hours of
326: observation, except for 10 minutes during which the images of the stars
327: drifted by 30 pixels (3 diameters of the de-focused stars) and re-stabilized
328: at a new position.
329: The 10-minutes of drifting, from 0.063 to 0.070 days after mid-transit,
330: are indicated by a hash mark at bottom of Figure \ref{fig:bestlc}.
331:
332:
333: A single comparison star, GSC 03793-01990 (labeled ``1" in
334: Figure \ref{fig:finder}),
335: was selected based on its proximity to \xon\ in location, brightness,
336: and color. With respect to \xon, the comparison star is 3\arcmin\ west,
337: 0.21 mag brighter, and 0.04 mag redder in V-R$_C$ color. The photometry
338: appears to be limited by scintillation (Figure \ref{fig:scint}).
339: In a 5-second exposure at low airmass, the camera collected 3.2 and
340: 4.0 million photons from \xon\ and the comparison star, respectively,
341: implying a Poisson contribution to \xon's differential photometric
342: error of 0.9 mmag per exposure. The noise increases with airmass,
343: indicating the Earth's atmosphere limits the photometric precision.
344: The increase in noise with airmass is predominately due to scintillation
345: ($\propto$ airmass$^{1.75}$) and marginally due to fewer photons due
346: to the measured extinction of 0.12 mag
347: airmass$^{-1}$. If we average every seven consecutive exposures into
348: samples with 1-min cadence, the r.m.s. of those samples is 0.6 mmag for
349: exposures near zenith, commensurate with the 1.8 mmag r.m.s. per exposure.
350: In Figure \ref{fig:scint} the observed r.m.s. noise is fit well by the
351: solid line, which corresponds to the quadrature sum of Poisson noise
352: from each star and uncorrelated scintillation of each star that is 25\%
353: larger than the nominal value given by Dravins et al.\ (1998, Eq. 10).
354: In the photometric noise model, Poisson and scintillation noise are both
355: $\sim\sqrt{2}$ larger than the noise contribution from a single source
356: since the differential light curve is calculated from the magnitude
357: difference between \xon\ and the comparison star.
358:
359:
360:
361:
362: \subsection{Spectroscopy\label{sec:spectroscopy}}
363:
364:
365: On 2007 November 20-21, we obtained cross-dispersed echelle spectra of
366: \xon\ with the 2dcoude spectrometer (Tull et al.\ 1995) on the 2.7-m Harlan
367: J.\ Smith (HJS) telescope at McDonald Observatory. Thorium-argon lamp
368: spectra immediately preceding and following each R=60,000 stellar spectrum
369: yielded 0.1 \kms\ radial velocity precision. The \xon\ observations
370: were at similar zenith distance (34-35 degrees) both nights to improve
371: radial velocity precision. These observations eliminated
372: a stellar or brown dwarf companion in an isolated binary
373: as the cause of the transits, leaving more complex systems (e.g.,
374: Mandushev et al.\ 2005) or a planetary companion as the only plausible
375: explanations. Section \ref{sec:sme} describes our analysis of the HJS spectra to
376: obtain stellar parameters.
377:
378: Between 2007 December 21 and 2008 January 15 we obtained cross-dispersed
379: echelle spectra of \xon\ with the high-resolution spectrometer (Tull
380: 1998) on the 11-m Hobby-Eberly Telescope (HET), also located at McDonald
381: Observatory. The R=63,000 spectra were obtained through an iodine gas
382: absorption cell to provide an accurate wavelength reference for each
383: observed spectrum. We obtained one spectrum per night for the first two
384: epochs to confirm the planetary signal, and then two spectra per night for
385: the remaining 7 epochs to improve radial velocity precision by $\sqrt 2$.
386: We processed each echelle spectrum from both telescopes
387: using the optimal extraction procedure
388: described in Hinkle et al.\ (2000). Section \ref{sec:rv} describes our analysis
389: of the HET spectra to measure radial velocity variations.
390:
391: \section{Analysis\label{sec:analysis}}
392:
393: \subsection{Ephemeris}
394:
395: The heliocentric Julian date of minimum light (also mid-transit) is
396: \begin{equation}
397: t_{m.l.} = t_c + P \times E,
398: \label{eq:ephem}
399: \end{equation}
400: where E is an integer, $t_c = $\vjd$\pm$\ejd\ (HJD), and the orbital
401: period, $P =$\vperiod $\pm$\eperiod\ days. We determine the epoch $t_c$
402: and its uncertainty using a Monte Carlo method (Section \ref{sec:lightcurve})
403: of modeling the
404: high-fidelity observations of the transit observed on Jan 20, 2008. We
405: determined the period from $t_c$ and the XO survey observations of
406: two ingresses and two egresses observed 2 years earlier. Due to the
407: 10-minute sampling and $\sim$1\% precision of the survey photometry, each
408: ingress or egress has a timing uncertainty of $\sim$10 minutes. With
409: four of them observed $\sim$180 cycles from $t_c$, the uncertainty
410: of the period is $\sim 10 {\rm min} / \sqrt{4} / 180$ = \eperiod\
411: days.
412: Although the formal estimate of the period's uncertainty is far
413: smaller than required for any of the analysis reported here, future
414: planning of a time-critical observation, such as a sequence of spectra to
415: measure the Rossiter effect, one or more years from $t_c$ may benefit from
416: first verifying the ephemeris with additional precision photometry.
417:
418: \subsection{Radial Velocities\label{sec:rv}}
419:
420: We measured \xon\ radial velocities by modeling each HET spectrum
421: (obtained with an I$_2$ gas absorption cell in the light path) as the product
422: of a shifted stellar template spectrum multiplied by the known absorption
423: spectrum of the I$_2$ cell and then convolved with the line spread function
424: of the spectrograph.
425: To construct the stellar template, we scaled a high resolution
426: ($\lambda/\Delta\lambda \approx 10^6$) solar spectrum (Wallace,
427: Hinkle, \& Livingston 1998) to match the observed line depths in
428: each wavelength interval and then convolved with a rotational
429: broadening kernel that includes the effects of stellar limb-darkening.
430: We obtained an FTS spectrum of the HET I$_2$ cell (Cochran 2000)
431: from the National Solar Observatory online archive.
432:
433: For each wavelength interval, the model parameters are a continuum
434: scale factor, an exponent that scales line depth, the iodine wavelength
435: shift, the stellar radial velocity, and slight deviations from a nominal
436: Gaussian line spread function.
437: Using downhill-simplex $\chi^2$ minimization, we adjusted these free
438: parameters to fit $\sim$32 separate 1.5 nm wavelength intervals with
439: significant I$_2$ absorption (521--570 nm).
440: For each observed spectrum, we compute the mean radial velocity and
441: adopt the standard deviation divided by $\sqrt{32}$ as the uncertainty
442: in the mean.
443: The radial velocities, transformed to the barycentric frame of the solar
444: system, are in Table \ref{table:rv} and Figure \ref{fig:rviodine}.
445: We phased the radial velocities to the ephemeris of the transits,
446: assumed a circular orbit, and determined the maximum likelihood radial
447: velocity semi-amplitude K = \vrvK$\pm$\ervK\ \mps.
448:
449: Precise line bisector measurements (e.g., Torres et al.\ 2005; Johns--Krull
450: et al.\ 2008) can be used to detect triple star systems that produce shallow
451: transits that may be misinterpreted as planetary transits.
452: For each of our HET spectra, we measured the mean bisector span of
453: stellar absorption lines free of I$_2$ and telluric absorption.
454: We find no significant correlation with measured radial velocities, but the
455: significance of this result is limited by the relatively low signal noise ratio
456: of our spectra.
457:
458:
459:
460: \subsection{Stellar Properties from Spectroscopy and Isochrone Analysis\label{sec:sme}}
461:
462: We used the SME package (Valenti \& Piskunov 1996) to fit each observed
463: HJS spectrum (Section \ref{sec:spectroscopy}) with a synthetic spectrum,
464: adopting the same wavelength intervals (5150-5200 and 6000-6200 \AA),
465: line data, atmospheres, and post facto parameter adjustments as Valenti
466: \& Fischer (2005).
467: Table \ref{table:spec} lists the resulting stellar parameters for \xon:
468: effective temperature (\Teff),
469: logarithm of the gravity (\Logg),
470: metallicity (\MonH),
471: projected rotational velocity (\Vsini),
472: and
473: logarithm of the abundances of Na, Si, Ti, Fe, and Ni relative to solar
474: (\NaH, \SiH, \TiH, \FeH, and \NiH).
475: Our \MonH\ parameter is an abundance scale factor for elements other than
476: Na, Si, Ti, Fe, and Ni, so it is not equivalent to standard metallicity.
477: The 1$\sigma$ uncertainty for each parameter is listed in the ``Uncer''
478: column.
479:
480: We ran the spectroscopic analysis four times, initially allowing \Logg\
481: to be a free parameter (``Run 1'') and then fixing it at three specific
482: values.
483: Valenti \& Fischer (2005) gravities for stars cooler than \xon\ are
484: constrained almost entirely by the collisional damping wings of the \Mgb\
485: triplet lines.
486: For stars as warm as \xon, these damping wings become relatively weak,
487: providing less of a constraint on stellar gravity.
488: Fortunately, the shape of a precise transit light curve provides a strong
489: independent constraint on stellar gravity (e.g., Winn et al.\ 2008).
490: In Table \ref{table:spec}, ``Run 3'' (in bold) is our preferred solution,
491: as it assumes the value of \Logg\ favored by the transit light curve
492: analysis (Section \ref{sec:lightcurve}).
493: Tabulated results for the other three runs illustrate in a useful format
494: the covariance of stellar parameters, subject to our spectroscopic constraint.
495: Reduced $\chi^2$ for each spectroscopic fit are listed in the \rchisq\ row.
496: SME users should note that Run 1 did not converge to the absolute minimum
497: value \rchisq\ because the Levenberg-Marquardt algorithm used in SME can
498: have some difficulty following shallow valleys in the $\chi^2$ surface
499: that are not aligned with parameter axes.
500:
501: To obtain stellar mass (\Mstar), radius (\Rstar), and age, we interpreted our
502: spectroscopic and photometric results using Yonsei--Yale (\Ytwo) isochrones
503: (Demarque et al.\ 2004).
504: We calculated a bolometric correction and interpolated the isochrones
505: using the procedure described in Valenti \& Fischer (2005), except that
506: for \xon\ we were forced to assume a sequence of possible distances
507: (180 to 430 pc in steps of 10 pc).
508: Setting the isochrone gravity equal to the gravity used in the SME analysis
509: selects a preferred distance and hence preferred values for stellar mass,
510: radius and age.
511: Figure \ref{fig:iso} shows credible parameter intervals for Run 3, which
512: yielded a preferred radius ($1.56\Rsun$) nearly identical to the radius
513: favored by the light curve analysis ($\vRs\Rsun$).
514: Using the parameter relationships in Table \ref{table:spec}, we can
515: translate the $\eRs\Rsun$ uncertainty in \Rstar\ (from the light curve
516: analysis) into corresponding uncertainties in distance, stellar gravity,
517: stellar mass, and age. In Table \ref{table:spec} we adopt uncertainties
518: three times these nominal values to account crudely for possible systematic
519: errors in our analysis.
520:
521:
522: \subsection{Light Curve Modeling\label{sec:lightcurve}}
523:
524: In order to determine the physical parameters of the star and the
525: transiting planet (Table \ref{table:planet}),
526: we modeled the high-fidelity transit light curve (Figure \ref{fig:bestlc})
527: using the transit model of Mandel \& Agol (2002) and the Markov Chain Monte Carlo
528: (MCMC) methodology. Because we used
529: the same MCMC procedure as Burke et al.\ (2007; 2008), in this paper
530: we describe the assumptions and results but do not describe the MCMC procedure itself.
531: For simplicity, we assume a circular orbit;
532: Section \ref{sec:discussion} discusses implications
533: of the circular-orbit assumption being invalid.
534:
535: To fully determine the system parameters from the transit light curve,
536: we adopt an informative prior for $M_{\star}=1.34\pm 0.08$ \Msun.
537: This initial estimate for $M_{\star}$ comes from the SME isochrone analysis
538: (\S~\ref{sec:sme}). The uncertainty in the prior for $M_{\star}$
539: conservatively agrees with the typical uncertainties in $M_{\star}$
540: for the homogeneous analysis of other known transiting planets of
541: Torres et al.\ (2008). Analysis of the transit light curve provides a
542: more precise estimate of log$g$ than the spectroscopic determination.
543: One iteration of the SME analysis with log$g$ fixed to this more
544: precise estimate, the resulting variation to $M_{\star}$ was not
545: significant for our given prior on $M_{\star}$. The prior on each of the
546: other parameters, in particular $R_{\star}$ and $R_{\rm p}$, is uniform.
547: The noise in the light curve averages down as expected on one minute
548: times scales thus we assume Gaussian indendent noise for the
549: Likelihood function. For the uncertainties associated with each photometric measure of the light curve,
550: we use the analytic model of Poisson noise and scintillation noise (Section
551: \ref{sec:bestlc}; Figure \ref{fig:scint});
552: we note that $\sigma_i = 1.9$ mmag at ingress and $\sigma_e = 4.1$ mmag at egress.
553: The free parameters in the MCMC fit are $M_{\star}$, $R_{\star}$,
554: $\rho=R_{\rm p}/R_{\star}$, $\tau$, $t_{o}$, $u_{1}$, $u_{2}$, and
555: $zpt$, where $\tau$ is the total transit duration from 1$^{\rm st}$ to
556: 4$^{\rm th}$ contact, $t_{o}$ is the transit timing midpoint offset
557: from an initial ephemeris, $u_{1}$ and $u_{2}$ are the limb darkening
558: coefficients for a quadratic law, and $zpt$ is the flux ratio zero
559: point for the differential light curve.
560:
561: Table~\ref{table:spec} and Table~\ref{table:planet} show the resulting
562: parameters for \xon\ and \xonb, respectively. The median of the MCMC
563: samples provides a robust (hereafter ``best'') estimate of each parameter
564: and the uncertainties are the
565: symmetrical confidence interval containing 68\% of the samples. The
566: maximum-likelihood model, i.e. in a $\chi^{2}$ sense, is
567: the solid line in Figure~\ref{fig:bestlc}, which also
568: shows the data residuals around the maximum-likelihood model.
569:
570: The best estimate for the limb darkening coefficients ($u_{1}=0.61\pm
571: 0.1$ and $u_{2}=0.10\pm 0.23$ are significantly different than the
572: theoretical R-band limb darkening coefficients from Claret et
573: al. (2000) for a star with the properties of \xon\ ($u_{1}=0.24$ and
574: $u_{2}=0.38$). The theoretical B-band limb darkening coefficients
575: ($u_{1}=0.49$ and $u_{2}=0.29$) are closer in agreement to what is
576: measured. To investigate this difference we perform a $\chi^{2}$ fit
577: to the Perkins transit light curve fixing the limb darkening
578: parameters at their theoretically expected value in the R-band
579: ($M_{\star}=$\vMs\ also held fixed). The dashed line in
580: Figure~\ref{fig:bestlc} shows the resulting transit model. The
581: theoretical limb darkening coefficients result in steeper ingress and
582: egress with a slightly shallower mid transit flux level.
583: In Figure~\ref{fig:bestlc}, the plot
584: of residuals is with respect to the model with the
585: limb darkening coefficients treated as free parameters
586: in the MCMC analysis, thus the solid line is zero by definition and
587: the dashed line equals the difference due to fixing the limb darkening
588: coefficients at their theoretical values.
589:
590: Formally, under the assumption of Gaussian independent noise, the
591: theoretical limb darkening coefficients result in $\Delta\chi^{2}=48$
592: worse fit at $>$ 6-$\sigma$ for 2 degrees of freedom. To investigate
593: the reliability of these models, we zoom in on the egress portion of
594: the light curve in Figure~\ref{fig:inout}, where the largest amplitude
595: difference between the models occurs and also the largest photometric
596: noise. The smooth solid and dashed lines show the best-fit model with
597: varying and fixed limb darkening coefficients, respectively. The
598: jagged solid line shows binned data to improve the
599: visibility of the light curve. In our data the two most significant
600: excursions from either model are labeled with
601: arrows in Figure~\ref{fig:inout}. We consider two possibilities: either the
602: noise model is correct and we are measuring limb darkening
603: coefficients significantly in conflict with the theoretical ones, or
604: the noise model overestimates the significance of these two excursions
605: of the light curve, causing the limb darkening
606: coefficients to compensate for systematic errors in the light curve. We
607: consider the latter is the case for the excursion near
608: third contact, because during that time, the stars' positions were
609: shifting on the detector (see \S~\ref{sec:bestlc}). On the other hand, other
610: systematic deviations do not correlate with variations in external parameters;
611: an example prior to third contact is indicated also in Figure \ref{fig:bestlc}.
612: From analysis of ground-based and space-based light curves, Southworth (2008)
613: concludes that if limb-darkening coefficients are not included as fitted
614: parameters, uncertainties in other parameters may be underestimated, because the
615: highest-quality light curves (e.g. spectrophotometric, space-based
616: observations of HD 209458b)
617: show significant differences between theoretically-predicted and
618: observationally-derived limb-darkening coefficients.
619: We concur with Southworth (2008) that multiple observations may
620: be useful to discern limitations of a single light curve.
621: The choice of limb darkening profile
622: does not significantly affect $R_{\star}$ or the planetary
623: properties in the case of \xonb. The difference in parameters,
624: $\Delta R_{\star}=0.07\pm 0.05$ \Rsun,
625: $\Delta \rho=-0.0017\pm 0.0013$,
626: $\Delta R_{\rm p}=0.03\pm 0.05$ \Rjup,
627: $\Delta t_{o}=0.4\pm 0.6$ mn, are in the sense of MCMC analysis minus fixed limb darkening coefficients and the uncertainty is the uncertainty in the parameter from the MCMC analysis only.
628:
629:
630:
631:
632:
633:
634: \section{Discussion and Conclusion\label{sec:discussion}}
635:
636:
637: Table \ref{table:compare} compares physical characteristics of
638: \xon\ and two similar transiting systems,
639: HAT--P--4 (Kovacs et al. 2007) and HAT--P--6 (Noyes et al. 2008).
640: The HAT systems were selected for comparison because their
641: stars are similar to \xon.
642: The relatively large duty cycles of the transits of all three
643: planets imply low mean densities for
644: the host stars, a fact confirmed by the SME analysis of their spectra.
645: The three planets radii are similar but their masses differ by
646: as much as a factor of 2.5.
647:
648: The mass of \xonb\ (\vMp\ \Mjup) places it at the
649: the margin of the bulk of the distribution of planetary masses
650: for planets with orbital periods between 3 and 5 days.
651: For such planets, the distribution declines rapidly with mass for $M_p \ga $1 \Mjup,
652: and the upper $\sim$10\% of the distribution of gas giants
653: (4 of $\sim$40 reported) is spread broadly over the range from 1 to 12 \Mjup.
654: The latter approximate description is true for both the set of planets
655: that transit, for which the mass is known, and the set of
656: planets that do not, for which a minimum mass is known.
657: The two sets are similar in number, $\sim 20$ planets each.
658: That the distribution has a sharp decline at $\sim$1 \Mjup\ (or between 1 and 2 \Mjup)
659: is convincing, but the shape of the tail of the distribution for $M_p \ga $1 \Mjup\
660: is poorly determined for this period range, due to the small number of planets.
661: \xonb\ increases that small number by one.
662:
663: Analysis of transiting systems yields orbital inclinations and stellar radii.
664: These quantities may be combined with \Vsini\ to estimate stellar rotation
665: periods,
666: \begin{equation}
667: \Pstar={2\pi\phi\Rstar\over\Vsini},
668: \end{equation}
669: where $\phi$ is a factor of order unity that accounts for differential
670: rotation and any systematic errors in \Vsini.
671: SME yielded $\Vsini=1.7$ \kps\ for the Sun (Valenti \& Piskunov 1995),
672: which rotates every 24 days at the equator and every 30 days at the pole.
673: Adopting 27 days as the characteristic rotation period of the Sun implies
674: $\phi=0.91$.
675: With this value of $\phi$, the rotation periods of the three warmest stars
676: known to host transiting planets are 8.1, 3.4, 7.7 days for XO--4, XO--3,
677: and HAT--P--6, respectively.
678: Adopting $\Rstar=1.38\Rsun$ from Winn et al.\ (2008) for XO--3, these stellar
679: rotation periods are 2.0, 1.1, and 2.0 times the planetary orbital periods.
680: XO--3b is massive enough that the star (or at least the convective envelope)
681: may have been forced into synchronous rotation.
682: \xonb\ and HAT--P--6b have orbital periods that are twice the stellar rotation
683: period, suggesting that their orbits may be affected by resonant interactions
684: with their rotating host stars.
685: On the other hand, planets (e.g., HAT--P--2b and HAT--P--7b) orbiting
686: slightly cooler stars do not have orbital periods that are small multiples
687: of the stellar rotation period.
688: Perhaps Jupiter mass planets are only able to interact effectively with
689: stars that have very shallow surface convection zones.
690: The discovery of additional planets that transit warm F dwarfs will test whether
691: resonant interactions can affect planetary orbital periods or stellar
692: rotation periods.
693: One issue to keep in mind, however, is a possible selection effect against
694: stars with large \Vsini\ because broad lines make
695: difficult measuring radial velocity variations and the process
696: of discriminating planets from triple star systems.
697:
698: We emphasize that this analysis assumed the orbit of \xonb\ is circular,
699: i.e. $e=0$.
700: If that assumption is false, i.e.
701: $e > 0$, logically the conclusions are invalid and all the
702: derived physical parameters will change accordingly.
703: The planetary mass is proportional to $K \sqrt{(1-e^2)}$
704: (Hilditch 2001; Equation 2.53), so only an eccentricity
705: $\ga 0.5$ would change the planetary mass estimate significantly
706: compared to its fractional uncertainty ($\sim$10\%).
707: From the measured depth of the transit, the planetary radius is proportional
708: to the stellar radius, and the latter is proportional to the transverse
709: velocity of the planet at transit. The latter velocity depends on the argument
710: of periastron but is bound by its values at periastron and apasteron, i.e.
711: $1\pm e$ times its value for a circular orbit of the same period.
712: Hence an eccentricity
713: $e \ga 0.035$ could change the planetary radius estimate by $\ga 1$-$\sigma$.
714: Interestingly, the {\it amplitudes} of the radial velocity
715: curve and the light curve readily reveal the {\it ratios} of
716: planetary-to-stellar masses and radii, but the detailed {\it shapes}
717: of \underline{both} curves are required to measure precisely the radii
718: distinctly.
719:
720:
721: In addition to assuming $e = 0$, one might also assume that the
722: impact parameter $b = 0$, i.e. that the planet's path crosses
723: the center of the star. Although there is no physical
724: justification for such an assumption, it can be helpful in bounding the limits
725: of the derived physical parameters.
726: If both $e = 0$ and $b = 0$, the density of the star $\rho_*(e=0;b=0)$ is
727: determined by the orbital period and the
728: duration of the transit. In that case, the derived radius of the star
729: $R_*(e=0;b=0)$ is simply proportional to $M_*^{-1/3}$, where the mass
730: of the star, $M_*$ is estimated from the spectra and the isochrone analysis.
731: The density $\rho_*(e=0;b=0)$ is an upper limit,
732: i.e. $\rho_*(e=0;b=0) \ge \rho_*(e=0;b\ge 0)$.
733:
734: The scintillation-limited, high-cadence photometry of \xon\ presented here
735: has a small r.m.s. per unit time, 0.6 mmag min$^{-1}$, at small airmass and
736: during planetary ingress, but which increases with airmass to
737: $\ga 2$ mmag min$^{-1}$ at airmass $\ga 2$, during planetary egress.
738: To improve confidence in the parameters derived for the \xon\ system,
739: additional high-fidelity time-series photometry would be beneficial.
740: Observations above the Earth's atmosphere are not limited by scintillation
741: and can more nearly achieve Poisson-limited results, which for \xon\
742: with a comparable spectral band and optical throughput,
743: would be $\sim 0.2$ mmag min$^{-1}$ if operational overheads can be made negligible.
744: The declination of \xon\ places it within the continuous viewing zone of
745: the Hubble Space Telescope (\hst), which permits observation without
746: interruption caused by occultation by the Earth. The latter circumstance
747: may enhance \xon's potential for precision spectrophotometry, because
748: those gaps and issues associated with them potentially may
749: cause systematic errors in precision time series obtained with \hst.
750:
751:
752:
753: \acknowledgments
754:
755: We thank the staff of the University of HawaiÔi, Institute for Astronomy, Haleakala
756: Observatories. We especially thank J. Heasley, M. Maberry, and R. Ratkowski
757: for assistance in operations on Maui. We thank the staffs of McDonald Observatory
758: and Lowell Observatory.
759:
760: R. Bissinger, P. Howell, F. Mallia, G. Masi, K. Richardson, J. G., C. N. F., B. L. G.,
761: M. F., E. G.-M., and T. M. observe for the XO Extended Team.
762: Radek Poleski assisted with extrasolar planet candidate identification.
763: Lisa Prato and
764: Naved Mahmud assisted with observing at McDonald Observatory.
765: D. Bell transmitted HET data to us.
766:
767:
768: This research made use of the Royal Beowulf cluster at STScI;
769: the SIMBAD database, operated at CDS, Strasbourg, France;
770: data products from the Two Micron All Sky Survey (2MASS),
771: the Digitized Sky Survey (DSS),
772: and The Amateur Sky Survey (TASS);
773: source code for transit light-curves (Mandel \& Agol 2002);
774: and community access to the HET.
775: XO is funded primarily by NASA Origins of Solar Systems grant NNG06GG92G,
776: with additional funding from the STScI Director's Discretionary Fund.
777:
778: \begin{thebibliography}
779:
780:
781: \bibitem[Bakos et al.(2007)]{2007ApJ...670..826B} Bakos, G.~{\'A}., et al.\
782: 2007, \apj, 670, 826
783:
784: \bibitem[Barbieri et
785: al.(2007)]{2007A&A...476L..13B} Barbieri, M., et al.\ 2007, \aap, 476, L13
786:
787:
788: \bibitem[Burke et al.(2007)]{xo2b} Burke, Christopher J., et al.\ 2007, \apj, 671, 2115
789:
790: \bibitem[Burke et al.(2008)]{xo5b} Burke, Christopher J., et al.\ 2008, \apj, submitted
791:
792:
793: \bibitem[Claret(2000)]{2000A&A...363.1081C} Claret, A.\ 2000, \aap, 363,
794: 1081
795:
796: \bibitem{Cochran(2000)} Cochran, W.\ 2000,
797: FTS spectrum of I2 Cell HRS3 at 69.9 C.,
798: ftp://nsokp.nso.edu/FTS\_cdrom/FTS50/001023R0.004
799:
800: \bibitem[Cutri et al.(2003)]{twomass} Cutri, R.~M., et al.\
801: 2003, The IRSA 2MASS All-Sky Point Source Catalog, NASA/IPAC Infrared
802: Science Archive
803:
804: \bibitem[Demarque et al.(2004)]{Dem04}
805: Demarque, P., Woo, J-H, Kim, Y-C, \& Yi, S. K. 2004, \apjs, 155, 667
806:
807:
808: \bibitem[Dravins et al. 1998]{drav98} Dravins, D., Lindegren, L., Mezey, E. \& Young, A. T. 1998, PASP, 110, 610
809:
810:
811: \bibitem[Fischer \& Valenti(2005)]{2005ApJ...622.1102F} Fischer, D.~A., \&
812: Valenti, J.\ 2005, \apj, 622, 1102
813:
814: \bibitem[Gillon et
815: al.(2007)]{2007A&A...472L..13G} Gillon, M., et al.\ 2007, \aap, 472, L13
816:
817:
818: \bibitem[Harwit(1981)]{1981cdss.book.....H}
819: Harwit, M.\ 1981, Cosmic discovery. The search, scope, and heritage of astronomy,
820: Brighton: Harvester Press
821:
822:
823: \bibitem[H{\o}g et al.(2000)]{2000A&A...355L..27H} H{\o}g, E., et al.\ 2000, \aap, 355, L27
824:
825:
826: \bibitem[Hinkle et al.(2000)]{echelle_reduction} Hinkle, K., Wallace, L., Valenti, J.,\& Harmer, D.\ 2000,
827: Visible and Near Infrared Atlas of the Arcturus Spectrum 3727-9300 A, ed.~Kenneth Hinkle,
828: Lloyd Wallace, Jeff Valenti, and Dianne Harmer.~(San Francisco: ASP) ISBN: 1-58381-037-4, 2000
829:
830: %
831:
832: \bibitem[Johns-Krull et al.(2008)]{xo3b} Johns-Krull, C.~M.,
833: et al.\ 2008, \apj, 677, 657
834:
835:
836: \bibitem[Kov{\'a}cs et al.(2007)]{hat4b} Kov{\'a}cs, G., et al.\ 2007, \apjl, 670, L41
837:
838:
839: %
840: \bibitem[Mandel \& Agol(2002)]{2002ApJ...580L.171M} Mandel, K., \& Agol,
841: E.\ 2002, \apjl, 580, L171
842:
843: \bibitem[Mandushev et al.(2005)]{2005ApJ...621.1061M} Mandushev, G., et
844: al.\ 2005, \apj, 621, 1061
845:
846: \bibitem[McCullough, P. R., \& Burke, C. J.(2007)]{2007ASPC..366..70M}
847: McCullough, P. R. \& Burke, C. J.\ 2007, Astronomical Society of the Pacific Conference Series, 366, 70
848:
849: \bibitem[McCullough et al.(2005)]{2005PASP..117..783M} McCullough, P.~R.,
850: Stys, J.~E., Valenti, J.~A., Fleming, S.~W., Janes, K.~A., \& Heasley,
851: J.~N.\ 2005, \pasp, 117, 783
852:
853:
854: \bibitem[Gould \& Morgan(2003)]{2003ApJ...585.1056G}
855: Gould, A., \& Morgan, C.~W.\ 2003, \apj, 585, 1056
856:
857: \bibitem[Morrison et al.(2001)]{gsc12}
858: Morrison, J.~E., R{\"o}ser, S., McLean, B., Bucciarelli, B., \& Lasker, B.\ 2001, \aj, 121, 1752
859:
860: \bibitem[Southworth(2008)]{2008MNRAS.tmp..444S} Southworth, J.\ 2008, \mnras, 444
861:
862:
863: \bibitem{2005ApJ...619..558T}
864: Torres, G., Konacki, M., Sasselov, D.~D., \& Jha, S.\ 2005, \apj, 619, 558
865:
866: \bibitem[Torres et al.(2008)]{2008ApJ...677.1324T} Torres, G., Winn, J.~N., \& Holman, M.~J.\ 2008, \apj, 677, 1324
867:
868:
869: \bibitem[Tull(1998)]{1998SPIE.3355..387T} Tull, R.~G.\ 1998, \procspie,
870: 3355, 387
871:
872: \bibitem[Tull et al.(1995)]{1995PASP..107..251T} Tull, R.~G., MacQueen,
873: P.~J., Sneden, C., \& Lambert, D.~L.\ 1995, \pasp, 107, 251
874:
875: \bibitem[Valenti \& Fischer(2005)]{2005ApJS..159..141V} Valenti, J.~A., \&
876: Fischer, D.~A.\ 2005, \apjs, 159, 141
877:
878: \bibitem[Valenti \& Piskunov(1996)]{1996A&AS..118..595V} Valenti, J.~A., \&
879: Piskunov, N.\ 1996, \aaps, 118, 595
880:
881:
882: \bibitem[Wallace et al.(1998)]{solar_atlas}
883: Wallace, L., Hinkle, K., \& Livingston, W.\ 1998,
884: An atlas of the spectrum of the solar photosphere from 13,500 to 28,000 cm-1 (3570 to 7405 \AA),
885: Publisher: Tucson, AZ: National Optical Astronomy Observatories
886:
887: \bibitem[Winn et al.(2008)]{2008arXiv0804.4475W} Winn, J.~N., et al.\ 2008,
888: ArXiv e-prints, 804, arXiv:0804.4475
889:
890: %
891:
892: \end{thebibliography}
893:
894: \clearpage
895: \begin{deluxetable}{cccccc}
896: \tabletypesize{\small}
897: \tablewidth{0pt}
898: \tablecaption{{\rm Photometry}}
899: \startdata
900: \hline
901: \hline
902: Star${\rm ^a}$ &B &V &${\rm R_C}$ &${\rm I_C}$ &V-${\rm R_C}$\\
903: \hline
904: \xon\ &11.240 &10.674 &10.324 &10.057 & 0.350 \\
905: 1 &11.217 &10.503 &10.109 & 9.758 & 0.394 \\
906: 2 &11.853 &10.329 & 9.482 & 8.748 & 0.847 \\
907: 3 &12.030 &11.613 &11.358 &11.084 & 0.255 \\
908: 4 &13.185 &12.526 &12.184 &11.798 & 0.342 \\
909: 5 &13.018 &12.314 &11.938 &11.560 & 0.376 \\
910: 6 &12.555 &11.475 &10.904 &10.390 & 0.571 \\
911: 7 &13.828 &13.121 &12.663 &12.297 & 0.458 \\
912: 8 &14.358 &13.787 &13.408 &13.073 & 0.378 \\
913: 9 &14.833 &13.930 &13.362 &12.834 & 0.568 \\
914: \enddata
915: \tablenotetext{a}{
916: Stars are identified in Figure \ref{fig:finder}.
917: The 1-$\sigma$ uncertainties are 0.029, 0.019, 0.008, and 0.023 mag
918: for B, V, ${\rm R_C}$, and ${\rm I_C}$ respectively.
919: The 2MASS magnitudes for \xon\ are 9.667, 9.476, 9.406
920: for J, H, and K$_s$ respectively (Skrutskie et al.\ 2006).}
921: \label{table:allsky}
922: \end{deluxetable}
923:
924: \begin{deluxetable}{cccccc}
925: \tabletypesize{\small}
926: \tablewidth{0pt}
927: \tablecaption{{\rm Time-Series Photometry$^a$}}
928: \startdata
929: \hline
930: \hline
931: HJD & Brightness & Uncertainty &Filter$^b$ & N$^c$ & Observer$^d$ \\
932: & [mag] & (1-$\sigma$) [mag] & & & \\
933: \hline
934: 2453691.10864 & 0.0005 &0.0096 & W &1 & XO \\
935: 2453691.10889 &-0.0052 &0.0088 & W &1 & XO \\
936: 2453691.11645 & 0.0087 &0.0095 & W &1 & XO \\
937: 2453691.11597 &-0.0043 &0.0088 & W &1 & XO \\
938: 2453691.12256 &-0.0032 &0.0094 & W &1 & XO \\
939: \enddata
940: \\
941: \tablenotetext{a}{The entire table is in the electronic edition.
942: The printed edition contains only a sample to establish the format.}
943: \tablenotetext{b}{Standard filters, except W = wide, 400-700 nm.}
944: \tablenotetext{c}{Average of N measurements.}
945: \tablenotetext{d}{Observer initials, except XO is the XO cameras.}
946: \label{table:photometry}
947: \end{deluxetable}
948:
949:
950:
951:
952: \begin{deluxetable}{ccc}
953: \tabletypesize{\small}
954: \tablewidth{0pt}
955: \tablecaption{{\rm Radial Velocity Shifts}}
956: \startdata
957: \hline
958: \hline
959: Julian Date & Radial Velocity & Uncertainty \\
960: & Shift [\mps] & (1 $\sigma$) [\mps] \\
961: \hline
962: 2454455.7622 & 257 & 57 \\
963: 2454457.7597 & -127 & 40 \\
964: 2454469.9042 & -88 & 43 \\
965: 2454474.6989 & -166 & 31 \\
966: 2454476.8855 & 140 & 39 \\
967: 2454477.8677 & -53 & 39 \\
968: 2454479.8429 & -34 & 46 \\
969: 2454480.6910 & 172 & 47 \\
970: 2454480.8856 & 141 & 39 \\
971: \enddata
972: \label{table:rv}
973: \end{deluxetable}
974:
975: \begin{deluxetable}{lccccc}
976: \tablecaption{Stellar Properties of \xon\label{table:spec}}
977: \tablewidth{0 pt}
978: \tablehead{
979: \colhead{Parameter} &
980: \colhead{Run 1} &
981: \colhead{Run 2} &
982: \colhead{\textbf{Run 3}} &
983: \colhead{Run 4} &
984: \colhead{Uncer}
985: }
986: \startdata
987: \multicolumn{6}{l}{\textit{Spectroscopic Analysis:}}\\
988: \Teff\ (K) & 6249 & 6349 & \textbf{ 6397} & 6491 & 70 \\
989: \Logg\ (cgs) & 3.98 & 4.09\tablenotemark{a}
990: & \textbf{ 4.18\tablenotemark{a}}
991: & 4.30\tablenotemark{a}
992: & \elogg\tablenotemark{b} \\
993: \MonH & -0.05 & -0.02 & \textbf{-0.02} & 0.03 & 0.05 \\
994: \Vsini\ (\kms) & 9.0 & 8.7 & \textbf{ 8.8 } & 8.5 & 0.5 \\
995: \NaH & -0.12 & -0.05 & \textbf{-0.02} & 0.07 & 0.2 \\
996: \SiH & -0.02 & 0.01 & \textbf{ 0.03} & 0.04 & 0.02 \\
997: \TiH & -0.01 & 0.00 & \textbf{-0.02} & 0.02 & 0.07 \\
998: \FeH & -0.09 & -0.08 & \textbf{-0.04} & -0.03 & 0.03 \\
999: \NiH & -0.21 & -0.04 & \textbf{-0.02} & 0.05 & 0.05 \\
1000: \rchisq & 1.82 & 1.77 & \textbf{ 1.77} & 1.81 & \nodata \\[8pt]
1001: \multicolumn{6}{l}{\textit{Isochrone Analysis:}}\\
1002: $d$ (pc) & 343 & 312 & \textbf{ 293 } & 257 & \eDs\tablenotemark{b} \\
1003: $\Mstar/\Msun$ & 1.34 & 1.33 & \textbf{ 1.32} & 1.29 & \eMs\tablenotemark{b} \\
1004: $\Rstar/\Rsun$ & 1.96 & 1.76 & \textbf{ 1.56} & 1.33 & 0.05\tablenotemark{c} \\
1005: Age (Gyr) & 3.5 & 2.9 & \textbf{ 2.1 } & 1.3 & \eAge\tablenotemark{b} \\
1006: \enddata
1007: \tablenotetext{a}{Value of \Logg\ was fixed during the SME analysis.}
1008: \tablenotetext{b}{Three times the uncertainty obtained by propagating $0.05\Rsun$
1009: uncertainty in \Rstar.}
1010: \tablenotetext{c}{Uncertainty from light curve analysis.}
1011: \end{deluxetable}
1012:
1013:
1014: \begin{deluxetable}{lcl}
1015: \tabletypesize{\small}
1016: \tablecaption{{\rm The Planet \xonb}}
1017: \tablewidth{0pt}
1018: \tablehead{
1019: \colhead{Parameter} &
1020: \colhead{Value$^a$} &
1021: \colhead{Notes}
1022: }
1023: \startdata
1024: $P $\dotfill & \vperiod$\pm$\eperiod\ d & Period \\
1025: $t_c$\dotfill & \vjd$\pm$\ejd\ HJD & Transit midpoint\\
1026: $e$(assumed)\dotfill & 0 & Eccentricity \\
1027: $K $\dotfill & \vrvK$\pm\ervK$\ \mps & \\
1028: $M_p $\dotfill & \vMp$\pm$\eMp\ \Mjup & Mass\\
1029: $R_p $\dotfill & \vRp$\pm$\eRp\ \Rjup & Radius\\
1030: $a $\dotfill & \vap$\pm$\eap\ A.U. & Semi-major axis\\
1031: $a/R_* $\dotfill & \vaoRs$\pm$\eaoRs\ & \\
1032: $R_p/R_* $\dotfill & \vRpoRs$\pm$\eRpoRs\ & \\
1033: $i $\dotfill & \vincl$\pm$\eincl\ deg & Inclination\\
1034: $b $\dotfill & \vbp$\pm$\ebp\ & Impact parameter\\
1035: \enddata
1036: \tablenotetext{a}{\Rjup\ = 71492 km; \Mjup\ = 1.8988e27 kg}
1037: \label{table:planet}
1038: \end{deluxetable}
1039:
1040:
1041:
1042: \begin{deluxetable}{lcccc}
1043: \tablecaption{Comparison of Three Transiting Systems\tablenotemark{a}\label{table:compare}}
1044: \tablewidth{0 pt}
1045: \tablehead{
1046: \colhead{Parameter} &
1047: \colhead{HAT--P--4} &
1048: \colhead{HAT--P--6} &
1049: \colhead{\xon} &
1050: \colhead{Uncer\tablenotemark{b}}
1051: }
1052: \startdata
1053: \multicolumn{5}{l}{\textit{Stars:}}\\
1054: \Teff\ (K) & 5860 & 6570 & 6397 & 70 \\
1055: \Logg\ (cgs) & 4.14 & 4.22 & 4.18 & \elogg \\
1056: \Vsini\ (\kms) & 5.5 & 8.7 & 8.8 & 0.5 \\
1057: \FeH & +0.24 & -0.13 & -0.04 & 0.03 \\
1058: Distance, (pc) & 310 & 260 & 293 & \eDs \\
1059: Mass, (\Msun) & 1.26 & 1.29 & 1.32 & \eMs \\
1060: Radius, (\Rsun) & 1.59 & 1.46 & 1.56 & 0.05 \\
1061: Age (Gyr) & 4.2 & 2.3 & 2.1 & \eAge \\[8pt]
1062: \multicolumn{5}{l}{\textit{Planets:}}\\
1063: Period (d) & 3.06\tablenotemark{c} & 3.85\tablenotemark{c} & 4.13\tablenotemark{c} & n/a\tablenotemark{c} \\
1064: Mass (\Mjup) & 0.68 & 1.06\tablenotemark{c} & \vMp & \eMp \\
1065: Radius (\Rjup) & 1.27 & 1.33 & \vRp & \eRp \\
1066: \enddata
1067: \tablenotetext{a}{HAT--P--4 (Kovacs et al.\ 2007); HAT--P--6 (Noyes et al. 2008); \xon\ (this work).}
1068: \tablenotetext{b}{Uncertainty is for \xon; those for HAT--P--4 and HAT--P--6 are similar or smaller.}
1069: \tablenotetext{c}{Tabulated value has been rounded.}
1070: \end{deluxetable}
1071:
1072:
1073: \clearpage
1074:
1075: \begin{figure}
1076: \plotone{f1.ps}
1077: \caption{
1078: \xon\ is inside the square.
1079: Stars from Table \ref{table:allsky} are circled and numbered.
1080: The CCD image is 16\arcmin\ by 11\arcmin\ with North up and East to the left.
1081: \label{fig:finder}}
1082: \end{figure}
1083:
1084: \begin{figure}
1085: \plotone{f2.ps}
1086: \caption{
1087: Photometry of \xon\ by the two XO cameras
1088: over two seasons from Nov 2005 to Mar 2007 are shown wrapped and phased according
1089: to the ephemeris of Equation \protect{\ref{eq:ephem}} and averaged
1090: in 30-minute bins (line).
1091: \label{fig:xolc}}
1092: \end{figure}
1093:
1094: \begin{figure}
1095: \plotone{f3.ps}
1096: \caption{{\it Top}: Time-series R-band photometry of \xon\ as observed with the Lowell 1.8-m
1097: Perkins telescope and PRISM camera during the Jan 20, 2008 transit of \xonb.
1098: The data (2448 points), the best-fitting model in a $\chi^{2}$ sense from the MCMC analysis (solid line), and the best-fitting model in a $\chi^{2}$ sense with limb darkening coefficients fixed at the R-band theoretical values of Claret (2000) (dashed line). {\it Bottom}: Residual from the best-fit model from the MCMC analysis (points) and difference between models with variable and fixed to the theoretical limb darkening coefficients (dashed line).
1099: \label{fig:bestlc}}
1100: \end{figure}
1101:
1102: \begin{figure}
1103: \epsscale{0.9}
1104: \plotone{f4a.ps}
1105: \plotone{f4b.ps}
1106: \caption{Detail of ingress (upper) and egress (lower) of the Lowell
1107: 1.8-m Perkins light curve (points) along with data in bins (jagged
1108: solid line). Best-fitting transit model in a $\chi^{2}$ sense from the
1109: MCMC analysis with limb darkening coefficients as free parameters (smooth
1110: solid line). Best-fitting transit model in a $\chi^{2}$ sense with limb
1111: darkening coefficients fixed at the theoretical values from Claret (2000)
1112: (smooth dashed line). Two epochs show the largest amplitude difference
1113: between data and transit models (arrows).
1114: \label{fig:inout}}
1115: \epsscale{1.0}
1116: \end{figure}
1117:
1118: \begin{figure}
1119: \plotone{f5.ps}
1120: \caption{Scintillation-limited photometry with the Perkins 1.8-m telescope in R band.
1121: The data (squares) are the standard deviations of $\sim 225$ samples within each of
1122: twelve 30-minute intervals of differential photometry of the star \xon\
1123: after a best-fit model to the transit of \xonb\ was subtracted. Data at the moments
1124: of ingress and egress are indicated by $+$ symbols within the squares at airmasses
1125: 1.13 an 1.84, respectively.
1126: The solid line is the quadrature sum of contributions from scintillation (dashed line)
1127: and Poisson noise (0.9 mmag per 5-sec exposure).
1128: The dashed line is proportional to the airmass raised to the power 1.75, as expected for
1129: scintillation (see text).
1130: The triangles are the same data averaged in 1-minute intervals, with 7 samples
1131: per minute. The dotted line is $\sqrt{7}$ times less than the solid line.
1132: \label{fig:scint}}
1133: \end{figure}
1134:
1135: \begin{figure}
1136: \plotone{f6.ps}
1137: \caption{a) The radial velocity of \xon\ oscillates sinusoidally with a
1138: semi-amplitude K = \vrvK$\pm$\ervK\ \mps.
1139: b) The period and phase of the radial velocities were fixed at values determined
1140: by the transits. The mean stellar radial velocity with respect to the
1141: solar system's barycenter has been subtracted.
1142: c) In this representation of
1143: the data, a circular orbit yields a straight line of slope $-$K.
1144: \label{fig:rviodine}}
1145: \end{figure}
1146:
1147: \begin{figure}
1148: \plotone{f7.eps}
1149: \caption{
1150: Credible intervals for \xon\ stellar properties as a function of
1151: assumed distance, based on the ``Run 3'' isochrone analysis with
1152: $\Logg=\vlogg$ from the light curve analysis.
1153: \label{fig:iso}}
1154: \end{figure}
1155:
1156:
1157:
1158:
1159:
1160: \end{document}
1161: