0711.4367/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: %test
3: \def\ms{m~s$^{-1}$}
4: \def\ks{km~s$^{-1}$}
5: \def\msini{$M_P\sin{i}$}
6: \def\asini{$a\sin{i}$}
7: \def\vsini{$V_{\rm rot}\sin{i}$}
8: \def\msun{$M_{\odot}$}
9: \def\mjup{$M_{\rm Jup}$}
10: \def\rsun{$R_{\odot}$}
11: \def\lsun{$L_{\odot}$}
12: \def\chisq{$\sqrt{\chi^2_\nu}$}
13: \def\chis{$\chi^2_\nu$}
14: \def\plmn{~$\pm$~}
15: \def\feh{[Fe/H]}
16: \def\rphk{$\log R^\prime_{HK}$}
17: \def\shk{$S$}
18: 
19: \def\npl{2}
20: \def\npllet{two}
21: \def\nplcap{Two}
22: 
23: \include{defs_C}
24: \include{defs_A}
25: \def\rhkC{-5.40}
26: \def\shkC{0.10}
27: \def\rhkA{-5.34}
28: \def\shkA{0.11}
29: \def\rhkB{-5.13}
30: \def\shkB{0.14}
31: 
32: %% Updated stellar parameters supercede values in defs_?.tex files.
33: \def\starA{HD\,167042}
34: \def\teffA{5020}
35: \def\teffeA{75}
36: \def\loggA{3.52}
37: \def\loggeA{0.08}
38: \def\feA{+0.05}
39: \def\feeA{0.06}
40: \def\vsiniA{2.5}
41: 
42: \def\starC{$\kappa$\,CrB}
43: \def\teffC{4960}
44: \def\teffeC{70}
45: \def\loggC{3.45}
46: \def\loggeC{0.09}
47: \def\feC{+0.14}
48: \def\feeC{0.05}
49: \def\vsiniC{3.0}
50: 
51: \begin{document}
52: %\title{Retired A Stars and Their Companions II: A Jovian Planet in a
53: %  Wide, Circular Orbit Around \starA\altaffilmark{1}}  
54: \title{Retired A Stars and Their Companions II: Jovian planets
55:   orbiting $\kappa$~Coronae~Borealis and \starA~\altaffilmark{1}}  
56: 
57: \author{ John Asher Johnson\altaffilmark{2,3}, 
58:   Geoffrey W. Marcy\altaffilmark{3},
59:   Debra	A. Fischer\altaffilmark{4}, 
60:   Jason T. Wright\altaffilmark{3},
61:   Sabine Reffert\altaffilmark{5},
62:   Julia M. Kregenow\altaffilmark{3},
63:   Peter K. G. Williams\altaffilmark{3},
64:   Kathryn M. G. Peek\altaffilmark{3}
65: }
66: 
67: \email{johnjohn@ifa.hawaii.edu}
68: 
69: \altaffiltext{1}{Based on observations obtained at the Lick
70:   Observatory, which is operated by the University of California.}
71: \altaffiltext{2}{Department of Astronomy, University of California,
72: Mail Code 3411, Berkeley, CA 94720}
73: \altaffiltext{3}{Institute for Astronomy, University of Hawaii,
74:   Honolulu, HI 96822}
75: \altaffiltext{4}{Department of Physics \& Astronomy, San Francisco
76:   State University, San Francisco, CA 94132}
77: \altaffiltext{5}{ZAH-Landessternwarte, K\"onigstuhl 12, 69117
78:   Heidelberg, Germany}
79: 
80: \begin{abstract}
81: We report precise Doppler measurements of \npllet\ stars, obtained at
82: Lick Observatory as part of our search for planets orbiting
83: intermediate--mass subgiants. Periodic variations in the radial
84: velocities of both stars reveal the presence of substellar
85: orbital companions. These two stars are notably massive with stellar
86: masses of \mstarC~\msun\ and 
87: \mstarA~\msun, respectively, indicating that they
88: are former A--type dwarfs that have evolved off of the main sequence
89: and are now K--type subgiants. The planet
90: orbiting \starC\ has a minimum 
91: mass \msini~$=$~\msiniC~\mjup, eccentricity $e = \eC$ and a \pC~day
92: period, corresponding to a semimajor axis $a = \arelC$~AU. The planet
93: around \starA\ has a minimum
94: mass \msini~$=$~\msiniA~\mjup\ and a \pA~day orbit, corresponding to a
95: semimajor axis $a = \arelA$~AU. The eccentricity of \starA\,b
96: is consistent with circular ($e = \eA \pm \eeA$), adding to the
97: rare class of known exoplanets in long--period, circular
98: orbits similar to the Solar System gas giants. Like all of the
99: planets previously discovered around evolved A stars, \starC b and
100: \starA b orbit beyond 0.8~AU. 
101: \end{abstract}
102: 
103: \keywords{techniques: radial velocities---planetary systems:
104:   formation---stars: individual (\starC, HD\,142091, \starA)}
105: 
106: \section{Introduction}
107: 
108: Most of what is known about planets outside of our
109: Solar System comes from Doppler surveys of Sun--like stars, with
110: spectral types ranging from K0V to F8V and 
111: masses between 0.8 and 1.2~\msun\ \citep{valenti05, butler06,
112:   takeda06}. However, recent results from planet searches 
113: around low--mass M dwarfs \citep[][]{bonfils05b,endl06,butler06b} and evolved,
114: intermediate--mass stars \citep{reffert06, johnson06b, sato07, nied07,
115:   lovis07} have 
116: expanded into the frontiers at either end of the stellar mass
117: range. These surveys have begun to reveal important
118: relationships between stellar mass and the properties of exoplanets. For
119: example, planets around ``retired'' (evolved) A--type stars reside
120: preferentially in wide orbits, with semimajor axes $a \gtrsim 0.8$~AU
121: \citep{johnson07} 
122: and appear to have larger minimum masses than planets around Sun--like
123: stars \citep{lovis07}. Also, the occurrence rate of Jovian planets
124: (\msini$\geq 0.5$~\mjup, $a \leq 2.5$~AU)
125: increases with stellar mass, rising from $< 2$\% around M
126: dwarfs, to approximately 9\% around F and A stars \citep{johnson07b}.
127: 
128: The effects of stellar mass on other characteristics of
129: exoplanets, such as orbital eccentricity and 
130: multiplicity, will become evident as the sample of planets 
131: around intermediate--mass stars grows. Here, we report the
132: detection of two Jovian planets orbiting stars with masses of
133: \mstarA~\msun\ and \mstarC~\msun. These planet detections come from our
134: sample of intermediate--mass subgiants that we have been monitoring at
135: Lick and Keck Observatories \citep{johnson06b}. We describe our
136: observations in \S~\ref{observations}. In \S~\ref{stellar} we present
137: our data, and describe 
138: the characteristics of the host stars and the orbits of their
139: planets. We conclude in \S~\ref{summary} with a summary of our results
140: and a discussion of the properties of planets orbiting evolved A--type
141: stars. 
142: 
143: \section{Observations and Doppler Measurements}
144: \label{observations}
145: 
146: We have been monitoring a sample of 159 evolved stars at Lick
147: and Keck Observatories for the past 3.5 years. The stars were selected
148: from the \emph{Hipparcos} catalog based on the color and
149: magnitude criteria described by \citet{johnson06b}, namely 
150: $0.5 < M_V < 3.5$, $0.55 < B-V < 1.0$, and $V \lesssim 7.6$. We
151: exclude from our sample clump giants with $B-V > 0.8$ and $M_V < 2.0$,
152: as well as stars within 1~mag of the mean \emph{Hipparcos} main
153: sequence, as defined by \citep{wright05}. These criteria allow us to
154: take advantage of the widely spaced, nearly parallel stellar model
155: tracks of the subgiant branch to estimate precise ages and masses. 
156: 
157: We obtained Doppler measurements of the stars presented here using the
158: 3\,m Shane and 0.6\,m Coude Auxiliary Telescope (CAT) at Lick
159: Observatory. Both telescopes feed the Hamilton 
160: spectrometer \citep{vogt87}, which has a resolution of $R \approx
161: 50,000$ at $\lambda = 5500$~\AA. 
162: Doppler shifts are measured from each spectrum
163: using the iodine cell method described in detail by \citet{butler96}
164: and summarized as follows. A
165: temperature--controlled Pyrex cell containing 
166: gaseous iodine is placed at the entrance slit of the
167: spectrometer. The dense set of narrow molecular lines imprinted on
168: each stellar spectrum from 5000 to 6000~\AA\ provides a
169: robust wavelength scale for each observation, as well as information
170: about the shape of the spectrometer's instrumental response
171: \citep{marcy92b}.  
172: 
173: Traditionally, the Doppler shift of each stellar spectrum is measured
174: with respect to an observed, iodine--free template spectrum. 
175: These template observations require higher signal and resolution than 
176: normal radial velocity observations, which in turn necessitates longer
177: exposure 
178: times. Given our large target list and the small aperture of the CAT, 
179: obtaining a traditional template for each star would represent a
180: significant fraction of our allocated observing time, resulting in a
181: smaller than optimal target list. We therefore perform
182: a preliminary analysis of each star's observations using a
183: synthetic, ``morphed'' template spectrum following the method
184: described by \citet{johnson06a}. Stars showing conspicuous Doppler
185: variations are reanalyzed using a traditional template to
186: verify the signal and search for a full orbital solution. 
187: 
188: Internal uncertainties for each velocity measurement are estimated
189: from the standard deviation of the mean velocity measured from the
190: $\sim700$ segments analyzed in each echelle spectrum. A typical $V = 6$,
191: K0 star requires a 3600 second exposure on the CAT for a
192: signal-to-noise ratio (S/N) of 180 pix$^{-1}$, which produces an
193: internal precision of 
194: 3--4~\ms. The same star observed on the Shane requires 300 seconds for
195: the same S/N owing to differences in the the collecting area and plate
196: scale at the spectrometer entrance slit. The radial velocities for the
197: \npllet\ stars presented here are listed in Table~\ref{tab1}
198: and Table~\ref{tab2}.
199: 
200: \section{Stellar Properties and Orbital Solutions} 
201: \label{stellar}
202: 
203: Our methods for determining the properties of our target stars are
204: described in \citet{johnson07} and summarized briefly as follows. We
205: use the LTE spectral synthesis code SME \citep{valenti05} to estimate
206: stellar effective temperatures, surface gravities,
207: metallicities, and projected rotational velocities by fitting a
208: synthetic spectrum to each star's iodine--free template
209: spectrum. We use the Stephan-Boltzmann Law to relate each star's 
210: radius and luminosity to its effective temperature,
211: parallax--based distance and bolometric correction. To estimate
212: stellar masses and ages, we interpolate each star's \emph{Hipparcos}
213: absolute magnitude, $B-V$ color and SME--derived metallicity onto the
214: stellar interior model grids computed by \citet{girardi02}. The
215: stellar properties and uncertainties are summarized in
216: Table~\ref{stellartable}. The position of the \npllet\ stars in the
217: H--R diagram is shown in Figure~\ref{sg_hr}, along with their
218: theoretical mass tracks and zero--age main sequence.
219: 
220: \begin{figure}[!t]
221: \epsscale{1}
222: \plotone{f1.eps}
223: \caption {\footnotesize{H--R diagram illustrating the properties of the
224:   \npllet\ subgiant planet host stars (pentagrams) compared to their
225:   main--sequence progenitors (filled circles). The connecting lines
226:   represent the \citet{girardi02} theoretical mass tracks
227:   interpolated for each star's metallicity. The thick, diagonal line
228:   is the zero--age main sequence assuming [Fe/H]=0.0.} \label{sg_hr}}    
229: \end{figure}
230: 
231: \citet{johnson07} estimate uncertainties of 7\% in stellar mass
232: and 1~Gyr for ages. Uncertainties in the SME--derived
233: parameters are given by \citet{valenti05}, however these estimates are
234: based primarily on a sample of main--sequence stars and may be
235: unrepresentative of the values obtained for our subgiants. We
236: therefore estimate uncertainties in the spectroscopic parameters by
237: exploring the degeneracy of the best--fit solution using different
238: input guesses to the SME code. We first solve for the best--fit
239: solution, and then use the resulting parameters as input for two additional
240: solutions, with $\pm 100$~K perturbations on $T_{eff}$. The
241: standard deviations of the parameters from the three SME trials are
242: then adopted as the $1\sigma$ parameter uncertainties. In cases when
243: our error estimates are less than those of \citet{valenti05}, we adopt the
244: latter values.
245: 
246: We search for the best fitting Keplerian orbital solution to each
247: radial velocity time series using a
248: Levenberg--Marquardt, least--squares minimization, and we estimate the 
249: uncertainties in the orbital parameters using a bootstrap 
250: Monte Carlo method. We first subtract the best--fit Keplerian from the
251: measured velocities. The residuals are then scrambled and added back
252: to the original measurements, and a new set of orbital parameters is
253: obtained. This process is repeated for 1000 trials and the standard
254: deviations of the parameters from all trials are adopted as
255: the formal 1$\sigma$ uncertainties. 
256: 
257: In the following subsections we describe the properties of the
258: \npllet\ new host stars and planet candidates discovered from our
259: sample of subgiants. 
260: 
261: \subsection{\starC}
262: 
263: $\kappa$ Coronae Borealis (\starC, HD\,142091, HR\,5901, HIP\,77655) is
264: listed in the SIMBAD database\footnote{http://cdsweb.u-strasbg.fr/}
265: with a spectral type K0IV and in the 
266: \emph{Hipparcos} catalog as a K0III-IV star with $V =
267: \vmagC$ and $B-V = \bvC$, and a parallax--based distance 
268: of \dC~pc \citep{hipp}. The star's apparent magnitude and distance
269: yield an absolute magnitude $M_V = \mvC$, placing the star
270: 4.3~mag above the mean main sequence of stars in the Solar
271: neighborhood \citep{wright05}. The color and absolute magnitude of \starC\
272: suggest an evolved star on the subgiant branch near the base of the red
273: giant branch, in agreement with its published spectral classifications.
274: 
275: Our LTE spectral analysis suggests that
276: \starC\ is metal--rich, with 
277: [Fe/H]~$=\feC \pm \feeC$. Our spectral analysis also yields $T_{eff} =
278: \teffC \pm \teffeC$~K, 
279: \vsini~$=\vsiniC \pm 0.5$~\ks\ and surface gravity, $\log{g} = \loggC
280: \pm \loggeC$. We 
281: estimate the mass 
282: and age of \starC\ by interpolating the star's color, absolute 
283: magnitude and metallicity onto the \citet{girardi02} stellar interior
284: models. We find a stellar mass $M_* =$~\mstarC~\msun\ and an age of
285: \ageC~Gyr. We also estimate a luminosity $L_* =
286: \lstarC$~\lsun\ and radius $R_* = \rstarC$~\rsun, assuming a
287: bolometric correction -0.302. The properties of \starC\ are summarized
288: in Table~\ref{stellartable}.
289: 
290: To assess the photospheric stability of \starC, we searched the
291: \emph{Hipparcos} Epoch
292: Photometry
293: database\footnote{http://www.rssd.esa.int/index.php?project=HIPPARCOS} 
294: and found 135 observations. The measurements have a
295: mean uncertainty of 4.7~mmag and have an rms scatter of 4.7~mmag over
296: a time span of 3.1 years\footnote{After rejecting a single
297:   outlier}. Given its photometric stability, we can rule 
298: out significant pulsation modes that can contribute to intrinsic
299: radial velocity variability. Further, based on the lack of emission in
300: its CaII H\&K line core, we measure \rphk~$=\rhkC$ indicating
301: \starC\ is chromospherically inactive, similar to other evolved stars
302: \citep{wright04b}. 
303: 
304: \begin{figure}[!t]
305: \epsscale{1}
306: \plotone{f2.eps}
307: \caption {Radial velocity time series for \starC\ (HD\,142091) measured at Lick
308:   Observatory. The dashed line shows the
309:   best--fit orbital solution, which has an orbital period of
310:   \pyearsC~years and \chisq~$ = \chisC$. \label{orbitC}}  
311: \end{figure}
312: 
313: 
314: Beginning in 2004 April, we  collected \nobsC\ Doppler
315: measurements at Lick Observatory, which are listed in
316: Table~\ref{tab1} together with their date of observation
317: and internal uncertainties. The velocities are also
318: shown in Figure~\ref{orbitC}, and the error bars represent the
319: quadrature sum of the internal measurement uncertainties and 5~\ms\ of
320: jitter, which is typical for the intrinsic Doppler variability of
321: subgiants similar to \starC\ \citep{johnson07}. 
322: 
323: Also shown in Figure~\ref{orbitC} is the best fitting Keplerian
324: orbital solution with a period $P = \pC \pm \peC$~days, eccentricity $e
325: = \eC \pm \eeC$ and velocity semiamplitude $K = \kC \pm
326: \keC$~\ms. The rms scatter of the data about the fit is \rmsC~\ms\ and
327: the reduced \chisq~$=\chisC$. Based on our stellar mass estimate of
328: $M_* = \mstarC$~\msun, the 
329: orbital solution gives a minimum planet mass \msini~$=\msiniC$~\mjup\
330: and semimajor axis $a = \arelC$~AU. The orbital parameters of
331: \starC b are summarized in Table~\ref{orbittable}. 
332: 
333: \subsection{\starA}
334: 
335: \starA\ (HR\,6817, HIP\,89047) is listed in the SIMBAD 
336: database with a spectral type K1\,III, suggesting the star is on the
337: giant branch. The
338: \emph{Hipparcos} catalog gives $V = 
339: \vmagA$, $B-V = \bvA$, and a parallax--based distance 
340: of \dA~pc \citep{hipp}. Given its distance and apparent magnitude, we
341: calculate an absolute magnitude $M_V = \mvA$, which 
342: places it 4.2~mag above the average main sequence of stars in the
343: Solar neighborhood \citep{wright05}. The position of \starA\ in the
344: H--R diagram indicates the star is better classified as a K1\,IV
345: subgiant near the upturn to the red giant branch, rather than a
346: luminosity class III giant. 
347: 
348: Based on our LTE spectral analysis \citep[SME;][]{valenti05}, we find
349: that \starA\ has Solar metal abundance with \feh~$=\feA
350: \pm \feeA$, and is 
351: slowly rotating with \vsini~$ = \vsiniA \pm 0.5$~\ks. Our spectroscopic
352: analysis also yields $T_{eff} = \teffA \pm \teffeA$~K and $\log{g} =
353: \loggA \pm \loggeA$.  
354: Interpolation of the star's color, absolute magnitude and metallicity
355: onto the \citet{girardi02} stellar model grids provides a 
356: stellar mass estimate of $M_* = \mstarA$~\msun, and an age of
357: \ageA~Gyr. Consistent with its post--main--sequence
358: evolutionary status, \starA\ is chromospherically inactive with
359: \rphk~$=$~\rhkA, as measured from its CaII~H\&K
360: emission relative to the stellar continuum. We also
361: estimate a radius $R_* = \rstarA \pm \rstareA$~\rsun\
362: and luminosity $L_* = \lstarA \pm \lstareA$~\lsun.
363: 
364: The \emph{Hipparcos} catalog lists 115
365: photometric measurements of \starA\ spanning 3.25 years. The star
366: is photometrically stable over this time baseline, with an rms scatter
367: 11.2 mmag, which is slightly higher than the median measurement uncertainty
368: of 6 mmag. Since a periodogram analysis of the photometric
369: measurements shows no significant power at periods ranging from 2 to
370: 1200 days, we expect that the contribution to the star's radial
371: velocity variability from star spots and radial pulsation should be
372: small. We follow \citet{johnson07} and adopt a jitter estimate of
373: 5~\ms, based on the velocity scatter of other chromospherically quiet
374: subgiants with properties similar to \starA.  
375: 
376: \begin{figure}[!t]
377: \epsscale{1}
378: \plotone{f3.eps}
379: \caption {Radial velocity time series for \starA\ measured at Lick
380:   Observatory. The dashed line shows the
381:   best--fit orbital solution, which has \chisq~$ = \chisA$. \label{orbitA}} 
382: \end{figure}
383: 
384: We began monitoring \starA\ in 2004 June and the
385: first 10 observations, analyzed with a synthetic
386: template, showed correlated variability with $\rm
387: rms=15$~\ms. We obtained a traditional, observed template
388: to confirm the variations with higher Doppler precision. The full set
389: of velocities is listed in Table~\ref{tab2} and
390: plotted in Figure \ref{orbitA}. The error bars in Figure~\ref{orbitA}
391: have been augmented by adding 5~\ms\ of jitter in quadrature to the
392: internal measurement uncertainties.
393: 
394: The best--fit Keplerian orbital solution is shown in
395: Figure~\ref{orbitA}. The solution has a
396: $\pA \pm \peA$~day period, a semiamplitude $K = \kA \pm \keA$~\ms, and
397: an eccentricity 
398: consistent with circular, $e = \eA \pm \eeA$. The residuals to the fit
399: have
400: $\rm rms = \rmsA$~\ms\ and reduced \chisq~$=$~\chisA, indicating that
401: the scatter is adequately modeled by the 
402: internal measurement uncertainties and estimated jitter. Assuming a
403: stellar mass 
404: $M_* = \mstarA$~\msun, the best--fit solution yields a semimajor axis $a
405: = \arelA$~AU, and minimum planet mass \msini~$=\msiniA$~\mjup. The
406: orbital parameters are summarized in Table~\ref{orbittable}.
407: 
408: \section{Summary and Discussion}
409: \label{summary}
410: 
411: We report the detection of \npllet\ Jovian planets orbiting
412: intermediate--mass subgiants. These detections come from our sample of
413: evolved stars that we are monitoring at Lick and Keck
414: Observatories. \starA\ and \starC\ have masses significantly larger
415: than solar, with $M_* = 
416: \mstarA$~\msun\ and  \mstarC~\msun, respectively. Examination of the
417: stars' theoretical mass tracks reveals that these current--day
418: subgiants began life on the main sequence as A--type
419: dwarfs (Figure~\ref{sg_hr}). 
420: 
421: The relatively low--amplitude Doppler variations induced by \starC b 
422: and \starA b would not have been detectable if the stars were
423: not in their current evolved state. Younger A-- and F--type
424: dwarfs have large rotational velocities (\vsini~$\gtrsim 50$~\ks) and
425: excessive pulsation--induced velocity jitter that can mask the reflex
426: velocity signal caused by a planet--mass companion
427: \citep{galland05}. On the other hand, older stars ascending the red giant
428: branch exhibit stochastic velocity variations in excess of 20~\ms
429: \citep{hekker06}, which would make the  planet orbiting \starC\
430: particularly difficult to detect since it induces a velocity amplitude
431: of only \kC~\ms. Subgiants, with their low 
432: rotation rates (\vsini~$ \lesssim 10$~\ks) and low jitter ($< 10$~\ms), 
433: occupy an observational ``sweet spot'' in the H--R diagram allowing
434: our Doppler survey to probe to relatively low planetary masses in a
435: wide range of orbital configurations.
436: 
437: The growing sample of planets around intermediate--mass stars is
438: beginning to reveal important relationships between stellar mass and
439: the properties of exoplanets. Of the 15
440: planets detected around evolved A--type stars ($M_* > 1.5$~\msun),
441: none have been found orbiting closer than $\sim 0.8$~AU, with the
442: majority orbiting at or beyond 1~AU \citep{johnson07}. \starC b and
443: \starA b are no 
444: exception to this trend, with semimajor axes of \arelC~AU and
445: \arelA~AU, respectively. As noted by \citet{johnson07},
446: this cannot be due to a decrease in detection sensitivity 
447: since, for a given planet and stellar mass, the amplitude of the
448: induced Doppler wobble scales as $a^{-1/2}$. Thus, the 
449: observed semimajor axis distribution of planets around A stars is
450: significantly different from that of planets around lower--mass 
451: stars. 
452: 
453: \citet{johnson07} considered the possibility that the lack of
454: close--in planets around K giants and clump giants 
455: may be attributable to engulfment by the expanding atmospheres of the
456: central stars. However, stars crossing the subgiant branch do
457: not undergo significant expansion, with radii smaller than $\sim
458: 5$~\rsun\ even at the base of the red giant branch. The fact that none
459: of the 7 planets detected around massive subgiants ($M_* > 1.5$~\msun)
460: has $a < 0.9$~AU strongly suggests that the lack of
461: close--in planets around A stars is due to the effects of stellar mass
462: on planet formation and migration, rather than post--main--sequence
463: engulfment.
464: 
465: Stellar mass also plays an important role in the likelihood
466: that a star harbors a detectable giant planet. By measuring
467: the fraction of stars with planets in three widely spaced stellar mass
468: bins, \citet{johnson07b} found that the occurrence rate of 
469: planets with \msini~$>0.8$~\mjup\ and $a < 2.5$~AU is a rising
470: function of stellar mass. This analysis reveals
471: that A stars appear to be planet--enriched by a factor of 4.5  compared
472: to low--mass M dwarfs, with a measured giant planet occurrence rate of
473: 9\% for $1.3 < M_*/M_\odot \leq 1.9$~\msun. The planet fraction for
474: higher--mass stars was based in 
475: part on three strong planet candidates from our survey of subgiants,
476: one of which is announced here (\starA b; \starC b
477: orbits beyond the 2.5 AU cut--off). This result
478: indicates that stellar mass is a strong tracer of
479: planeticity, which has important implications for the target selection
480: of future planet searches. Just as stellar metallicity is exploited in
481: the search for short--period planets around Sun--like
482: stars \citep{fischer05b, fischer05a, dasilva06},  
483: stellar mass should be an important consideration in the selection of
484: targets for future photometric, astrometric, and high--contrast direct
485: imaging surveys. 
486: 
487: The rising trend in planet occurrence toward higher stellar masses also
488: informs models of planet formation. Several theoretical studies of the
489: effects of stellar mass on 
490: planet formation have predicted that M stars should harbor
491: fewer Jovian planets than Sun--like dwarfs \citep{laughlin04,
492: ida05b}. More recently \citet{kennedy07} studied planet formation at
493: higher masses by accounting for the evolving disk mid--plane
494: temperature due to stellar irradiation and viscous evolution. They
495: predict that the fraction of stars with Jupiters should rise up to a
496: peak of $\approx 20$\% near $M_* = 3$~\msun. 
497: 
498: Testing this prediction requires searching for planets around stars
499: with masses greater than 2~\msun. Unfortunately, the mass range of
500: suitable subgiants is limited to $M* \lesssim 2.2$~\msun, primarily
501: due to our absolute magnitude criterion of $M_V > 2.0$ (see
502: \S~\ref{observations}). Doppler surveys of massive K giants with $M_*
503: > 2.5$~\msun\ provide perhaps the most favorable avenue for exploring
504: the occurrence rate of planets 
505: around more massive A stars \citep{frink01, sato05b,
506:   nied07}. In particular, \citet{lovis07} are
507: searching for planets around K giants in open clusters, which allows
508: them to take advantage of the uniform ages of the cluster members to
509: derive accurate stellar masses. The preliminary results from their
510: survey indicate that stars more massive than $\sim2$~\msun\ have an
511: enhanced abundance of super--Jupiters and brown dwarfs compared to
512: lower--mass stars.
513: 
514: To enlarge the statistical ensemble of planets detected around
515: intermediate--mass stars, we have expanded our planet search by adding
516: 300 additional subgiants at Lick and Keck Observatories. If our
517: current 9\% detection rate holds, our expanded planet
518: search is expected to yield an additional 20-30 planets over the next
519: 3 years. Together with the detections from Doppler surveys of K giants
520: and clump giants, the increased number of planets orbiting
521: intermediate--mass stars will add significantly to our understanding
522: of the effects of stellar mass on planet formation and planetary
523: system architecture.  
524: 
525: \acknowledgements 
526: We extend our gratitude to the many CAT observers who have helped 
527: with this project, including Chris McCarthy, Raj Sareen, Howard
528: Isaacson, Joshua Goldston, Bernie Walp, and Shannon Patel. We also
529: gratefully acknowledge the efforts and dedication of the Lick
530: Observatory staff, and the time assignment committee of the University
531: of California for their generous allocations of observing time. JAJ is
532: an NSF Astronomy and Astrophysics Postdoctoral Fellow and acknowledges
533: support from the NSF grant AST-0702821. We appreciate funding from 
534: NASA grant NNG05GK92G (to GWM). DAF is a Cottrell 
535: Science Scholar of Research Corporation and acknowledges support from
536: NASA Grant NNG05G164G that made this work possible. PKGW is supported
537: by an NSF Graduate Research Fellowship. This research has
538: made use of the SIMBAD database operated at CDS, Strasbourg France,
539: and the NASA ADS database. 
540: 
541: %\bibliographystyle{apj}
542: %\bibliography{apj-jour,myrefs}
543: 
544: \include{bbl}
545: 
546: \clearpage
547: \include{tab1}
548: \include{tab2}
549: 
550: \begin{deluxetable}{lll}
551: \tablecaption{Stellar Parameters\label{stellartable}}
552: \tablewidth{0pt}
553: \tablehead{
554:   \colhead{Parameter} & 
555:   \colhead{\starC\tablenotemark{a}}    &
556:   \colhead{\starA}    
557: }
558: \startdata
559: V               & \vmagC        & \vmagA              \\
560: $M_V$           & \mvC          & \mvA                \\
561: B-V             & \bvC          & \bvA                \\
562: Distance (pc)   & \dC           & \dA                 \\
563: ${\rm [Fe/H]}$  & \feC~(\feeC)   & \feA~(\feeA)       \\
564: $T_{eff}$~(K)   & \teffC~(\teffeC) & \teffA~(\teffeA) \\
565: \vsini~(\ks)    & \vsiniC~(0.5) & \vsiniA~(0.5)       \\
566: $\log{g}$       & \loggC~(\loggeC) & \loggA~(\loggeA) \\
567: $M_{*}$~(\msun) & \mstarC~(0.11)& \mstarA~(0.13)      \\
568: $R_{*}$~(\rsun) & \rstarC~(0.08)& \rstarA~(0.07)      \\
569: $L_{*}$~(\lsun) & \lstarC~(0.04)& \lstarA~(0.05)      \\
570: Age~(Gyr)       & \ageC~(1.0)   & \ageA~(1.0)         \\
571: $S_{HK}$        & \shkC         & \shkA               \\
572: $\log R'_{HK}$  & \rhkC         & \rhkA               \\
573: \enddata
574: \tablenotetext{a}{HD\,142091}
575: \end{deluxetable}
576: 
577: \clearpage
578: \begin{deluxetable}{llll}
579: \tablecaption{Orbital Parameters\label{orbittable}}
580: \tablewidth{0pt}
581: \tablehead{\colhead{Parameter} &
582:   \colhead{\starC b\tablenotemark{a}}    &
583:   \colhead{\starA b}    
584: }
585: \startdata
586: P (d)         & \pC~(\peC)   & \pA~(\peA)     \\
587: T$_p$\tablenotemark{b}~(JD)    & \tpC~(\tpeC) & \tpA~(\tpeA) \\
588: e             & \eC~(\eeC)   & \eA~(\eeA)     \\
589: K~(\ms)   & \kC~(\keC)   & \kA~(\keA)         \\
590: $\omega$~(deg)& \omC~(\omeC) & \omA~(\omeA) ) \\
591: \msini~(\mjup)& \msiniC      & \msiniA        \\
592: $a$~(AU)      & \arelC       & \arelA         \\
593: Fit RMS~(\ms) & \rmsC        & \rmsA          \\
594: Jitter~(\ms)  & 5.0          & 5.0            \\
595: \chisq        & \chisC       & \chisA         \\
596: N$_{\rm obs}$ & \nobsC       & \nobsA         \\
597: \enddata
598: \tablenotetext{a}{HD\,142091\,b}
599: \tablenotetext{b}{Time of periastron passage.}
600: \end{deluxetable}
601: 
602: \end{document}