1: %\documentclass[letter]{aastex}
2: \documentclass[12pt,preprint]{aastex}
3: %\documentclass{emulateapj}
4: \newcommand{\vdag}{(v)^\dagger}
5: \newcommand{\myemail}{aniedzi@astri.uni.torun.pl}
6:
7:
8: %
9: \usepackage{natbib}
10: \usepackage{amsmath,amsfonts,amssymb}
11: \usepackage{graphicx}
12: %
13: % Local macros
14: %
15: \def\v#1{{\bf #1}}
16: \def\u#1{{\mbox{[#1]}}}
17: \def\bdelta{{\pmb \delta}}
18: \def\bphi{{\pmb \phi}}
19: \def\deg{{\rm o}}
20: \def\idm#1{{\mbox{\scriptsize #1}}}
21: \def\D#1{\frac{\mbox{d}\,#1}{\mbox{d\,t}}}
22: \def\Dtau#1{\frac{\mbox{d}^2\,#1}{\mbox{d}\,t^2}}
23: \newcommand\Ym{\langle Y\rangle}
24: \newcommand\Chi{{(\chi^2_\nu)^{1/2}}}
25: \def\astrobj#1{#1\ }
26: \def\bibcode#1{(\texttt{#1})}
27: \def\url#1{\texttt{#1}}
28: \newcommand\pstar{{\astrobj{SIM}}}
29:
30:
31: \shorttitle{Planet Around HD 102272}
32: \shortauthors{Niedzielski et al.}
33: %
34: \begin{document}
35:
36: \title{A Planet in a 0.6-AU Orbit Around the K0 Giant HD 102272}
37:
38: \author{A. Niedzielski\altaffilmark{1,2}, K. Go\' zdziewski\altaffilmark{1}, A. Wolszczan\altaffilmark{2}, M. Konacki\altaffilmark{3,4}, G. Nowak\altaffilmark{1}, P. Zieli\'nski\altaffilmark{1} }
39:
40: \altaffiltext{1}{Toru\'n Center for Astronomy, Nicolaus Copernicus University, ul. Gagarina 11, 87-100 Toru\'n, Poland, Andrzej.Niedzielski@astri.uni.torun.pl, Krzysztof.Gozdziewski@astri.uni.torun.pl, Grzegorz.Nowak@astri.uni.torun.pl, Pawel.Zielinski@astri.uni.torun.pl}
41: \altaffiltext{2}{Department of Astronomy and Astrophysics, Pennsylvania State University, 525 Davey Laboratory, University Park, PA 16802, alex@astro.psu.edu}
42: \altaffiltext{3}{Nicolaus Copernicus Astronomical Center, ul. Rabia\'nska 7, 87-100 Toru\'n, Poland, maciej@ncac.torun.pl}
43: \altaffiltext{4}{Astronomical Observatory, A. Mickiewicz University, ul. Sloneczna 36, 60-286 Pozna\'n, Poland}
44:
45: \begin{abstract}
46: We report the discovery of one or more planet-mass companions to the K0-giant HD 102272
47: with the Hobby-Eberly Telescope. In the absence of any correlation of the observed
48: periodicities with the standard
49: indicators of stellar activity, the observed radial velocity variations
50: are most plausibly explained in terms
51: of a Keplerian motion of at least one planet-mass body around the star. With the estimated
52: stellar mass of 1.9M$_\odot$, the minimum mass of the confirmed planet is 5.9M$_J$.
53: The planet's orbit
54: is characterized by a small but nonzero eccentricity of $e$=0.05 and
55: the semi-major axis of 0.61 AU, which makes it the most compact one discovered so far around GK-giants. This detection
56: adds to the existing evidence that, as predicted by theory, the minimum size of
57: planetary orbits around intermediate-mass giants is affected by both planet formation processes and stellar evolution.
58: The currently available evidence for another planet around HD 102272 is insufficient to obtain an unambiguous two-orbit solution.
59:
60: \end{abstract}
61:
62: \keywords{planetary systems-stars: individual (HD 102272)}
63:
64: \section{Introduction}
65:
66: Searches for planets around giant stars offer a unique way to extend studies of planetary
67: system formation and evolution to stellar masses substantially larger than 1 M$_{\odot}$ \citep{2003ApJ...597L.157S}.
68: These evolved stars have cool atmospheres and many
69: narrow spectral lines, which can be utilized in precision radial velocity (RV) measurements ($<$10 m s$^{-1}$). Searches for planets around giant
70: stars \citep[][and references therein]{2008arXiv0802.2590S, 2006A&A...457..335H, 2008A&A...480..215H, 2007ApJ...669.1354N} are beginning to provide the statistics, which are needed to constrain
71: the efficiency of planet formation as a function of stellar mass and chemical composition. In fact, the initial analyses
72: by \cite{2007A&A...472..657L} and \cite{2007ApJ...665..785J} suggest that the frequency of occurence
73: of massive planets is correlated with stellar mass. Because more massive
74: stars probably have more massive disks, these results appear to support
75: the core accretion scenario of planet formation \citep{2008ApJ...673..502K}. Furthermore, \cite{2008arXiv0801.3336P}
76: have used the apparent lack of correlation between the frequency of planets
77: around giants and stellar metallicity to argue that this effect may imply
78: a pollution origin of the observed planet frequency - metallicity correlation
79: for main sequence stars \citep{2005ApJ...622.1102F}.
80:
81: In time, the ongoing surveys will also create an experimental basis with which to study
82: the dynamics of planetary systems orbiting evolving stars \citep[e.g.][]{1996Sci...274..954R, 1998Icar..134..303D}.
83: Sufficiently large surveys of post-MS giants should furnish enough
84: planet detections to meaningfully address the problem of a long-term survival of
85: planetary systems around stars that are off the main-sequence (MS) and on their way to the white dwarf stage. In fact, the existing
86: data suggest a deficiency or absence of orbits with radii below 0.6-0.7 AU, which still awaits a fully satisfying
87: explanation \citep{2007ApJ...665..785J, 2008arXiv0802.2590S}. In addition,
88: the recent detection of a planet around a post-red-giant phase
89: subdwarf V391 Peg \citep{2007Natur.449..189S} has already demonstrated that planets can survive both the RGB
90: and AGB phases of giant evolution.
91:
92: In this paper, we describe the detection of a planet around the K0-giant HD 102272 and discuss a possibility
93: that at least one more planet-mass body exists in this system.
94: An outline of the observing procedure and description of the basic properties of HD 102272 are given in Section 2,
95: followed by discussion of rotation and stellar activity indicators of the star in Section 3.
96: The analysis of radial velocity
97: measurements, is given in Section 4. Finally, our results are summarized and further discussed in Section 5.
98:
99: \section{Observations and properties of the star}
100:
101: Our survey, the observing procedure, and data analysis have been described in detail elsewhere \citep{2007ApJ...669.1354N, 2008IAUS..249...43N}.
102: Briefly, observations were made between 2004, July and 2008, January, with the Hobby-Eberly Telescope (HET) \citep{lwr98} equipped
103: with the High Resolution Spectrograph (HRS) \citep{tull98} in the queue scheduled mode \citep{HetQ}. The spectrograph was used in the R=60,000 resolution mode with a gas cell ($I_2$) inserted into the optical path, and it was fed with a 2 arcsec fiber.
104:
105: Radial velocities of HD 102272 were measured at 35 epochs spanning the period of about 1500 days.
106: Typically, the signal-to-noise ratio per resolution element in the spectra was 200-250 at 594 nm in
107: 9-16 minutes of integration, depending on the atmospheric conditions. Radial velocities were measured
108: using the standard I$_2$ cell calibration technique \cite{Butler+1996}. A template spectrum was
109: constructed from a high-resolution Fourier Transform Spectrometer (FTS) I$_2$ spectrum and
110: a high signal-to-noise stellar spectrum measured without the I$_2$ cell. Doppler shifts were
111: derived from least-squares fits of template spectra to stellar spectra with the imprinted I$_2$
112: absorption lines. The radial velocity for each epoch was derived as a
113: mean value of the independent measurements from the 17 usable echelle orders with
114: a typical uncertainty of 6-8 m s$^{-1}$ at 1$\sigma$-level. This RV precision made it quite sufficient to
115: use the \cite{1980A&AS...41....1S} algorithm to refer the measured RVs to the Solar System barycenter.
116:
117: HD 102272 (BD+14 2434) is a K0-giant
118: with V=8${^m}$.69, B-V=1${^m}$.02 and U-B=0${^m}$.69 \citep{1973A&AS...12..381} and $\pi$= 2.76$\pm$1.11 mas
119: \citep{hipparcos}.
120: Its atmospheric parameters were estimated using the method of \citet{2005PASJ...57...27T, 2005PASJ...57..109T} as T${_{eff}}$=4908$\pm$35 K, log(g)=3.07$\pm$0.12, and [Fe/H]=-0.26$\pm$0.08.
121: Comparing the star's position in the HR diagram with evolutionary tracks of \citet{girardi2000}, we estimate the mass and radius of HD 102272 to be M/M$_{\odot}$=1.9$\pm$0.3 and R/R${_\odot}$=10.1$\pm$4.6, respectively, from calibrations of \cite{alonso2000}.
122:
123:
124: \section{Analysis of the radial velocity data}
125:
126: Radial velocity measurements of HD 102272 derived from the HRS spectra
127: are shown in Figure 1.
128: They reveal a correlated behavior of radial velocity variations of the star
129: on a timescale of about 130 days. Moreover, observations
130: made around MJD 54140 indicate that the amplitude of the RV curve may vary in time.
131:
132: We have used both the nonlinear least-squares and the genetic (GA) algorithm
133: \citep{1995ApJS..101..309C} to model the observed RV variations with the standard, six-parameter
134: Keplerian orbits. A fixed 15 m s$^{-1}$ error was quadratically added to the
135: formal RV uncertainities to account for any random RV variations intrinsic
136: to the star, such as fluctuations of the stellar surface and possible solar-type oscillations. The anticipated value of this additional jitter was conservatively adopted to lie between 20 m s$^{-1}$, which is typical of stable K-giants \citep{2006A&A...454..943H}, and 5 m s$^{-1}$ as measured for stable dwarfs \citep{2005PASP..117..657W}. For example the solar-type oscillations alone, when extrapolated for HD 102272 as described by \cite{1995A&A...293...87K}, would account for a 6 m s$^{-1}$ RV variation.
137: The 15 m s$^{-1}$ jitter is equivalent to f=0.005 spot on the surface of HD 102272 that would produce 2 m s$^{-1}$ BVS at the same time \citep{Hatzes2002}.
138:
139: The best-fit of a single orbit and its residuals are shown
140: in Figure 1. The fit, if interpreted in terms of an orbiting body, calls for
141: a $\sim$6 M$_{Jup}$ companion in a 127.6-day, slightly eccentric orbit around
142: the star (Table 1). The 0.61 A.U. radius of the orbit is even smaller than
143: the tightest planetary orbit around a red giant identified so far \citep{2008arXiv0802.2590S}. However, not
144: unexpectedly, the single planet model leaves highly correlated post-fit
145: residuals, indicating a possible presence of additional periodicities.
146:
147: We have performed a search for the best-fitting two-companion Keplerian model
148: of the observed RV variations and identified several solutions of comparable
149: quality with very similar $\chi^2$ values of the fits. For example, the orbits
150: with the respective approximate periods of 179, 350, and 520 days and
151: eccentricities of 0.3, 0.5 and 0.7 are all almost indistinguishable in terms
152: of their goodness of fit. At the same time, the parameters of the 127.6-day
153: day orbit remain practically unchanged for all of the best two-orbit solutions.
154: An example of such a formal solution with the second orbit of 520 days is
155: shown in Figure 1.
156:
157: Because of a sparse sampling of the observed RV curve and the resulting
158: ambiguities in its modeling, it is useful to perform an additional test of
159: non-randomness of the residuals obtained from the best fit of a single
160: planet model to data. To accomplish this, we have chosen the test of
161: scrambled velocities \citep{Butler+2004} using the GA algorithm. In this test,
162: the residuals were randomly scrambled at the epochs of observations and
163: then searched for the best-fit parameters of the second orbit. A likelihood,
164: $p_{\idm{H}}$,
165: that the residuals represent a white noise can be quantified as a ratio
166: of the number of best fits, for which $\chi^2$ is less than or equal to
167: the one derived from the fit to unscrambled data, to the total number of
168: trials. The result of the test with 100,000 trials is shown in Figure 2.
169: Clearly, the fit to the real signal stands apart from those for the randomly
170: scrambled data with $p_{\idm{H}}\sim 10^{-5}$, which strongly suggests
171: the presence of a second, non-random signal in the RV data.
172:
173: In principle, the existing ambiguities in the modeling of the RV variations
174: in HD 102272 can be constrained by imposing the obvious requirement that
175: the true two-planet solution is dynamically stable. A convenient way
176: to apply this constraint is to use the GAMP algorithm \citep{2006ApJ...645..688G}, in which
177: an N-body model of the observed RV curve has the $\chi^2$ criterion modified
178: by a penalty term for unstable orbits, and the MEGNO test \citep{2000A&AS..147..205C} is
179: applied to check the stability of a given two-planet solution.
180:
181: We have performed a GAMP search for the dynamically acceptable orbits of
182: the second companion, with the MEGNO integration time set to $\sim$600
183: orbital periods of the outer companion. The results, in which the $\chi^2$
184: values relative to the best solution found by the search are mapped onto
185: the semi-major axis - eccentricity plane, are shown in Figure 3.
186:
187: The plot reveals three regions of stability, one of which is located
188: between the mean motion resonances (MMR) of 5:2 and 3:1, another one is extended
189: along the 4:1 MMR, and yet another one occupies a high-eccentricity region
190: between the 4:1 and 5:1 MMRs. Further discussion of these results will be
191: given in the Chapter 5.
192:
193:
194:
195:
196: \section{Stellar photometry, rotation, and line bisector analysis}
197:
198: We have followed the standard procedure to verify that the observed RV variations are due to orbital motion. It included an examination of the existing photometry of the star for possible variations and an analysis of the bisectors and curvatures of the selected spectral line profiles. We have also searched for stellar acivity signatures in the H$\alpha$ line.
199:
200: The most precise existing photometry of HD 102272 consists of 70 photometric observations of the star in the Hipparcos \citep{hipparcos} H$_{P}$ filter between JD 2447877.95490 and 2448961.44322. The observed scatter in H$_{P}$ amounted to only 0.015 mag, and no variability above this threshold was detectable.
201: The star was also observed by the Northern Sky Variability Survey (NSVS) \citep{2004AJ....127.2436W}. We have analyzed the 31 photometric observations made between JD=2451318.188426 and JD=2451630.300686, which were of a sufficient quality, and found no photometric variations of the star.
202: Finally, our Lomb-Scargle periodogram analysis of these data did not reveal
203: any spectral peaks above a threshold level set at 6 times
204: the mean power computed for periods shorter than one month. Of course, it should be kept in mind that the Hipparcos and NSVS data were not simultaneous with our RV measurements.
205:
206: The projected rotational velocity of HD 102272, $v sin i= $3 $\pm$ 1 km s$^{-1}$, was estimated using the cross-correlation method \citep{Benz+Mayor1984}. From this value and the adopted stellar radius we have obtained an estimate of the rotation period of P$_{rot}\sim$170 days, which is longer than the observed 127.6-day period of radial velocity variation and appears to be typical for K0 giants \citep{deMedeiros1996}. In principle, given the error estimates of \cite{alonso2000}, the rotation period of HD 102272 may range from 90 to 250 days.
207:
208:
209:
210: Most analyses of the variations of spectral lines using line bisectors are based on a cross-correlation function (CCF) representation of an ``average'' spectral line of the observed star. As the $I_2$ lines affect stellar spectra in the iodine cell method of RV determination, they have to be properly removed before the CCF computation. An efficient method to accomplish this has been proposed by \citep{mf05}.
211:
212: In our implementation of the method, we first cross-correlated the stellar spectrum with that of the flat-field $I_2$, in order to measure the wavelength offset between the iodine lines present in both spectra. The flat field flux was adjusted to the new wavelength scale, adding the previously determined offset by using a Hermite spline interpolation \citep{hill82}.
213:
214: The CCFs were computed by correlating the stellar spectrum with a numerical "zero-one", in which the non-zero points were aligned with the positions of stellar absorption lines at zero velocity. The numerical mask was constructed out of 980 lines of a very high SNR spectrum of HD 17092, the first star with a planet in our survey \citep{2007ApJ...669.1354N}, cleaned from spectral features lying within $\pm \, 30 \; km \, s^{-1}$ of the known telluric lines \cite{Han03}. Out of the 35 resulting bisector measurements, three were affected by a low SNR in the original spectra and hence removed from further analysis. The typical uncertainties of the $BVS$, calculated with the expression given by \cite{mf05} were around $\sim$ 20 m s$^{-1}$. The mean value of BVS was calculated to be 25 $\pm$ 21 m s$^{-1}$.
215:
216: As both the Ca II K emission line (393.4 nm) and the infrared Ca II triplet lines at 849.8-854.2 nm are outside the range of our spectra, we used the H$\alpha$ line (656.28 nm) as a chromospheric activity indicator. To minimize a contamination from the telluric lines, we measured the EW of the central part of the line profile defined by I/I$_{c}$$\leq$0.7. Our analysis has shown
217: that the EW exhibited a 0.69\% rms scatter around its
218: mean value of $944.4\pm 6.5 m\AA$, which was not correlated with the observed RV variability (r = 0.06).
219:
220: The analysis of the existing photometry and bisector variability span for HD 102272 is summarized in Figure 4 which shows the Lomb-Scargle periodograms computed for all the time series of interest. We conclude that the only measured parameter, which exhibits a significant periodic variability is radial velocity. Similarly, as evident in the plot of BVS against the RVs in Figure 5, no correlation between the bisector velocity span and the radial velocities exists in HD 102272.
221:
222: Using the scatter seen in the Hipparcos photometry of the star, and its rotational velocity determined above, we can estimate the amplitude of radial velocity variations and of the bisector velocity span due to a possible presence of a spot on the stellar surface \citep{Hatzes2002}. The observed radial velocity amplitude of HD 102272 is almost an order of magnitude larger than 35 m s$^{-1}$ RV amplitude predicted by \cite{Hatzes2002} for a spot with a filling factor of f=0.015, on a star rotating at $v sin i=$3 km s$^{-1}$. The expected bisector variations of 5 m s$^{-1}$ are comparable to the precision of our radial velocities measurements and cannot have a detectable effect on our results. Similar results have been derived with the NSVS photometry using the spot-induced RV variation modeling described by \citep{sd97}.
223:
224: In principle, it is possible that HD 102272, having no spots over the periods covered by both the Hipparcos data ($\sim$1083 days) and those from the NSVS ($\sim$312 days), had developed a spot later on, at the time of our RV measurements. Using the results of \cite{Hatzes2002} and \cite{sd97}, we estimate that, to generate RV variations as large as K=155 m s$^{-1}$, at a period similar to that of HD 102272 rotation, would require a large spot with the filling factor of f=0.17-0.22. Such a spot would translate into a 50-60 m s$^{-1}$ bisector velocity span that is most certainly excluded by our measurements. We also find it unlikely that such a large spot, apparently not present in the NSVS data mere $\sim$1400 days prior to our observations, would show no evolution in our measurements collected over 1500 days or 11 cycles of the observed periodicity. Therefore, we conclude that a rotating spot on the surface of HD 102272 is unlikely to be a cause of the observed periodic radial velocity variations.
225:
226: \section{Discussion}
227:
228: Our observations of the K0-giant, HD 102272, reveal that the measured radial velocities of this star undergo
229: strictly periodic variations at a period of 127.6$\pm$0.3 days. When interpreted in terms of a Keplerian motion, this periodicity indicates
230: the presence of a sub-stellar companion with a minimum mass of 5.9M$_J$, in a low
231: eccentricity, e=0.05$\pm$0.04 orbit, 0.61 A.U. away of the star (Table 1). In addition, our data indicate a possible presence of another, more
232: distant planetary-mass companion. Unfortunately, in this case, the sparse sampling of a putative second orbit does not allow an unambiguous distinction between orbital motion and intrinsic stellar effects. In any case, the inner planet has the shortest orbital period and the most compact orbit among planets around GK-giants discovered
233: so far \citep[][and references therein]{2008arXiv0802.2590S}.
234:
235: As the long period RV variations in red giants may also be related to a combination of effects including
236: stellar rotation, activity, and non-radial pulsations \citep[e.g.][]{2006A&A...457..335H}, we have analyzed the photometric data
237: and the behavior of line bisectors of the star, following the established practices \citep{Quelozetal2001}. The details of our procedure
238: have been described by \citep{2008IAUS..249...49N}. Its application to HD 102272 has not shown any significant correlation between
239: the RV variations and the photometry or line variability of the star. Consequently, the most plausible explanation of our data is the presence of at least one sub-stellar mass companions around the star.
240:
241: The orbital parameters of the confirmed planet are, within errors, independent of the choice of the orbit of a putative outer companion. The small orbital radius of the planet reinforces the existing evidence that orbits of GK-giant planets appear to be wider than
242: $\sim$0.6 A.U., as suggested by \citep{2007ApJ...665..785J, 2008arXiv0802.2590S}. These authors have considered the two obvious scenarios to create such a zone of avoidance around the GK-giants, namely the tidal capture of a nearby planet by the expanding star, and
243: a paucity of compact orbits around giants caused by peculiarities of the disk evolution around intermediate-mass stars
244: \citep{2007ApJ...660..845B}. In particular, simulations carried out by \citep{2008arXiv0802.2590S} indicate that, for giants which are over the peak
245: of the red giant branch, a tidal capture of planets at orbital radii $\leq$0.5 A.U. is possible. Clearly, much more observational evidence is needed to place these tentative conclusions on a firmer statistical footing.
246:
247: Some of the past reports on planet discoveries around GK-giants do indicate a possible
248: existence of more planet-sized companions to the stars in question \citep{2008arXiv0802.2590S} , but none of them include analyses designed to
249: constrain their possible orbital parameters. As discussed above, and illustrated in Figure 3, our analysis has revealed three regions of plausible two-planet solutions. The better two of them, located between the MMRs of 5:2 and 3:1 and close to the 4:1 MMR (Table 1), offer a significant improvement of the quality of the fit, in comparison with the one-planet model. In addition,
250: these solutions remain dynamically stable over at least 1 Gyr, as shown by our numerical integration of the corresponding orbits. On the other hand, a more in-depth analysis of these orbits (Go\' zdziewski et al., in preparation) suggests that, because of the compactness of the stability regions for the two solutions, a probability for such systems to become trapped in the respective MMRs in the course of their dynamical evolution must be very low. Evidently, a confirmation or dismissal of a possible dynamical origin of the observed RV variations will have to wait until more data become available in the near future.
251:
252:
253: \acknowledgments
254: We thank the HET resident astronomers and telescope operators for support. AN, AW and GN were
255: supported in part by the Polish Ministry of Science and Higher Education grant 1P03D 007 30.
256: AW acknowledges a partial support from the NASA Astrobiology Program.
257: M.K. is supported by the Foundation for Polish Science through a FOCUS
258: grant. K.G. was supported by the Polish Ministry of Sciences and Education grant 1P03D-021-29.
259: GN is a recipient of a graduate stipend of the Chairman of the Polish Academy of Sciences.
260: The Hobby-Eberly Telescope (HET) is a joint project of the University of Texas at Austin, the Pennsylvania State University, Stanford University, Ludwig-Maximilians-Universit\"at M\"unchen, and Georg-August-Universit\"at G\"ottingen. The HET is named in honor of its principal benefactors, William P. Hobby and Robert E. Eberly.
261:
262:
263: \begin{thebibliography}{}
264:
265: \bibitem[Alonso et al.(2000)]{alonso2000}
266: Alonso, A., Salaris, M., Arribas, S., Martinez-Roger, S., Asensio Ramos, A., 2000, \aap, 355, 1060
267:
268: \bibitem[Benz \& Mayor(1984)]{Benz+Mayor1984}
269: Benz, W., Mayor, M. 1984, \aap, 138, 183
270:
271: \bibitem[Burkert \& Ida(2007)]{2007ApJ...660..845B}
272: Burkert, A., \& Ida, S.\ 2007, \apj, 660, 845
273:
274: \bibitem[Butler et al.(1996)]{Butler+1996}
275: Butler, R. P., Marcy, G. W., Williams, E., McCarthy, C., \& Dosanjh, P. 1996, \pasp, 108, 500
276:
277: \bibitem[Butler et al.(2004)]{Butler+2004}
278: {Butler}, R.~P., {Vogt}, S.~S., {Marcy}, G.~W., {Fischer}, D.~A.,
279: {Wright}, J.~T., {Henry}, G.~W., {Laughlin}, G.,
280: {Lissauer}, J.~J., 2004, \apj, 617, 580
281:
282: %\bibitem[Cannon \& Pickering(1924)]{hd}
283: %Cannon A.J., Pickering E.C. 1924, Harv. Ann., 91-100 (1918-1924)
284:
285: \bibitem[Charbonneau(1995)]{1995ApJS..101..309C} Charbonneau, P.\ 1995,
286: \apjs, 101, 309
287:
288: \bibitem[Cincotta
289: \& Sim{\'o}(2000)]{2000A&AS..147..205C} Cincotta, P.~M., \& Sim{\'o}, C.\ 2000, \aaps, 147, 20
290:
291: \bibitem[Duncan \& Lissauer(1998)]{1998Icar..134..303D}
292: Duncan, M.J., \& Lissauer, J.J.\ 1998, Icarus , 134, 303
293:
294: \bibitem[Fischer \& Valenti(2005)]{2005ApJ...622.1102F} Fischer, D.~A., \&
295: Valenti, J.\ 2005, \apj, 622, 1102
296:
297: \bibitem[Girardi et al.(2000)]{girardi2000}
298: Girardi, L., Bressan, B., Bertelli, G., Chiosi, C. 2000, \aaps, 141, 371
299:
300: \bibitem[Go{\'z}dziewski et al. (2006)]{2006ApJ...645..688G}
301: {Go{\'z}dziewski}, K., {Konacki}, M., {Maciejewski}, A.~J,
302: 2006, \apj, 645, 688
303:
304: %bibitem[Go{\'z}dziewski
305: %\& Konacki(2006)]{2006ApJ...647..573G} Go{\'z}dziewski, K., \& Konacki, M.\ 2006, \apj, 647, 573
306:
307: \bibitem[Haggkvist \& Oja (1973)]{1973A&AS...12..381}
308: Haggkvist, L., Oja, T. 1973, \aaps, 12, 381
309:
310: \bibitem[Hanuschnik (2003)]{Han03} Hanuschnik, R.W., 2003, \aap, 407, 1157
311:
312: \bibitem[Hatzes (2002)]{Hatzes2002} Hatzes, A.P. 2002 AN323, 392
313:
314: \bibitem[Hatzes et al.(2006)]{2006A&A...457..335H}
315: Hatzes, A.~P., et al.\ 2006, \aap, 457, 335
316:
317: \bibitem[Hekker et
318: al.(2006)]{2006A&A...454..943H} Hekker, S., Reffert, S., Quirrenbach, A., Mitchell, D.~S., Fischer, D.~A., Marcy, G.~W., \& Butler, R.~P.\ 2006, \aap, 454, 943
319:
320:
321: \bibitem[Hekker et
322: al.(2008)]{2008A&A...480..215H} Hekker, S., Snellen, I.~A.~G., Aerts, C., Quirrenbach, A., Reffert, S., \& Mitchell, D.~S.\ 2008, \aap, 480, 215
323:
324: \bibitem[Hill (1982)]{hill82} Hill, G., Publications of the Dominion Astrophysical Observatory Victoria, 1982, 16, 67.
325:
326: \bibitem[Johnson et al.(2007)]{2007ApJ...665..785J} Johnson, J.~A., et al.\
327: 2007, \apj, 665, 785
328:
329: \bibitem[Johnson et al.(2007)]{2007ApJ...670..833J} Johnson, J.~A., Butler,
330: R.~P., Marcy, G.~W., Fischer, D.~A., Vogt, S.~S., Wright, J.~T.,
331: \& Peek, K.~M.~G.\ 2007, \apj, 670, 833
332:
333: \bibitem[Kennedy
334: \& Kenyon(2008)]{2008ApJ...673..502K} Kennedy, G.~M., \& Kenyon, S.~J.\ 2008, \apj, 673, 502
335:
336: \bibitem[Kjeldsen
337: \& Bedding(1995)]{1995A&A...293...87K} Kjeldsen, H., \& Bedding, T.~R.\ 1995, \aap, 293, 8
338:
339: \bibitem[Linsky \& Haisch(1979)]{1979ApJ...229L..27L} Linsky, J.~L., \&
340: Haisch, B.~M.\ 1979, \apjl, 229, L27
341:
342: \bibitem[Mart{\'{\i}}nez Fiorenzano (2005)]{mf05} Mart{\'{\i}}nez Fiorenzano , A. F., et al. 2005, \aap 442, 775.
343:
344: \bibitem[Lovis \& Mayor(2007)]{2007A&A...472..657L} Lovis, C., \& Mayor,
345: M.\ 2007, \aap, 472, 657
346:
347: \bibitem[de Medeiros et al.(1996)]{deMedeiros1996}
348: de Medeiros, J.R., da Rocha, C., Mayor, M., 1996, \aap, 314, 499
349:
350: \bibitem[Niedzielski et al.(2007)]{2007ApJ...669.1354N} Niedzielski, A., et
351: al.\ 2007, \apj, 669, 1354
352:
353: \bibitem[Niedzielski et al.(2008)]{2008IAUS..249...49N} Niedzielski, A.,
354: Nowak, G., \& Zieli{\'n}ski, P.\ 2008, IAU Symposium, 249, 49
355:
356: \bibitem[Niedzielski \& Wolszczan(2008)]{2008IAUS..249...43N} Niedzielski, A., \& Wolszczan, A.\ 2008, IAU Symposium, 249, 43
357:
358: \bibitem[Pasquini et al.(2008)]{2008arXiv0801.3336P} Pasquini, L.,
359: Doellinger, M.~P., Hatzes, A., Setiawan, J., Girardi, L., da Silva, L.,
360: \& de Medeiros, J.~R.\ 2008, ArXiv e-prints, 801, arXiv:0801.3336
361:
362: \bibitem[Queloz~et~al.(2001)]{Quelozetal2001}
363: Queloz,~D., Henry,~W., Sivan,~J.~P.,~et~al. 2001, \aap, 379, 279
364:
365: \bibitem[Ramsey et al.(1998)]{lwr98}
366: Ramsey, L.W., et al. 1998, \procspie, 3352, 34
367:
368: \bibitem[Rasio \& Ford(1996)]{1996Sci...274..954R}
369: Rasio, F.~A., \& Ford, E.~B.\ 1996, Science, 274, 954
370:
371: \bibitem[Saar \& Donahue (1997)]{sd97} Saar,S.H., Donahue, R.A. \ 1997, \apj 485, 319
372:
373: \bibitem[Sato et al.(2003)]{2003ApJ...597L.157S} Sato, B., et al.\ 2003,
374: \apjl, 597, L157
375:
376: \bibitem[Sato et al.(2008)]{2008arXiv0802.2590S} Sato, B., et al.\ 2008,
377: ArXiv e-prints, 802, arXiv:0802.2590
378:
379: \bibitem[Shetrone et al.(2007)]{HetQ} Shetrone, M., et al.\
380: 2007, \pasp, 119, 556
381:
382: \bibitem[Silvotti et al.(2007)]{2007Natur.449..189S} Silvotti, R., et al.\
383: 2007, \nat, 449, 189
384:
385: \bibitem[Stumpff(1980)]{1980A&AS...41....1S} Stumpff, P.\ 1980, \aaps, 41,
386: 1
387:
388: %\bibitem[Takeda et al.(2002)]{2002PASJ...54..451T} Takeda, Y., Ohkubo, M.,
389: %\& Sadakane, K.\ 2002, \pasj, 54, 451
390:
391: \bibitem[Takeda et al.(2005a)]{2005PASJ...57...27T} Takeda, Y., Ohkubo, M.,
392: Sato, B., Kambe, E., \& Sadakane, K.\ 2005, \pasj, 57, 27
393:
394: \bibitem[Takeda et al.(2005b)]{2005PASJ...57..109T} Takeda, Y., Sato, B.,
395: Kambe, E., Izumiura, H., Masuda, S., \& Ando, H.\ 2005, \pasj, 57, 109
396:
397: \bibitem[Tull(1998)]{tull98}
398: Tull, R.G., 1998, \procspie, 3355, 387
399:
400: \bibitem[ESA(1997)]{hipparcos}
401: The Hipparcos and Tycho Catalogues, European Space Agency SP-1200, 1997
402:
403: \bibitem[Wo{\'z}niak et al.(2004)]{2004AJ....127.2436W}
404: Wo{\'z}niak, P.~R., et al.\ 2004, \aj, 127, 2436
405:
406: \bibitem[Wright(2005)]{2005PASP..117..657W} Wright, J.~T.\ 2005, \pasp,
407: 117, 657
408:
409: \end{thebibliography}
410:
411: \clearpage
412:
413: \input{tab1.tex}
414:
415: \clearpage
416:
417:
418: %
419: % FIGURES WITH CAPTIONS
420: %
421:
422:
423: \begin{figure}
424: \includegraphics[angle=0,scale=1.0]{f1.eps}
425: \caption{(a) Radial velocity measurements of HD 102272 (filled circles) and
426: the best-fit models of a single orbit (gray line) and a two-planet system
427: close to the 4:1 MMR (solid line). (b-c) The respective best-fit residuals for
428: the above models.}
429: \end{figure}
430:
431:
432: \begin{figure}
433: \includegraphics[angle=0,scale=1.0]{f2.eps}
434: \caption{A histogram of $\Chi$ for fits of a single-orbit, Keplerian model to 100,000 sets of RV residuals
435: scrambled with a randomly generated signal of the outer planet around HD 102272.
436: Position of the best-fit solution for the real data is marked with an arrow.}
437: \end{figure}
438:
439: \begin{figure}
440: \includegraphics[angle=0,scale=1.0]{f3.eps}
441: \caption{An ensemble of Newtonian, best-fit models of a two-planet system
442: generated with the GAMP code and projected onto the ($a_c,e_c$)-plane.
443: A location of each model in the diagram is marked with a grey-shaded circle.
444: The consecutive shades of grey (from the darkest to the lightest one) correspond to models with the values of $\Chi$ located
445: within 1$\sigma$, 2$\sigma$, and 3$\sigma$ of the $\Chi$ of the best-fit solution identified in the GAMP search, respectively.
446: Vertical lines denote positions of low-order mean motion resonances
447: between planets~b and~c.
448: The double-lined curve marks the so called collision line of orbits defined by a condition:
449: $a_b (1+e_b) = a_c (1-e_c)$.
450: As a reference, the best-fit Newtonian model obtained without
451: stability constraints is marked with a crossed circle. See text for details.}
452: \end{figure}
453:
454: \begin{figure}
455: \includegraphics[angle=0,scale=1.0]{f4.ps}
456: \caption{Lomb-Scargle periodograms of (a) radial velocities, (b) bisector velocity span, (c) Hipparcos H$_p$ photometry, and (d) NSVS photometry. The position of the periodicity identified in the RV data is indicated with a vertical arrow. Two false alarm probability levels, 10$\%$ and 1$\%$ are marked with dotted lines.}
457: \end{figure}
458:
459: \begin{figure}
460: \includegraphics[angle=0,scale=1.0]{f5.ps}
461: \caption{ Radial velocity measurements of HD 102272 plotted against the bisector velocity span. Within errors, the BVS remains constant during the period covered by observations at the mean value of 25 m s$^{1}$ and rms scatter of 1$\sigma =$ 21 m s$^{-1}$. There is no correlation between RV and BVS (r=0.03).}
462: \end{figure}
463:
464:
465: \end{document}