Submitted for publication in the Astrophysical Journal A Preprint typeset using LTEX style emulateapj v. 11/10/09 IMPROVED ORBITAL PARAMETERS AND TRANSIT MONITORING FOR HD 156846b Stephen R. Kane1, Andrew W. Howard2,3, Genady Pilyavsky4, Suvrath Mahadevan4,5, Gregory W. Henry6, Kaspar von Braun1, David R. Ciardi1, Diana Dragomir1,7, Debra A. Fischer8, Eric Jensen9, Gregory Laughlin10, Solange V. Ramirez1, Jason T. Wright4,5 Submitted for publication in the Astrophysical Journal ABSTRACT HD 156846b is a Jovian planet in a highly eccentric orbit (e =0.85) with a period of 359.55 days. The pericenter passage at a distance of 0.16 AU is nearly aligned to our line of sight, offering an enhanced transit probability of 5.4% and a potentially rich probe of the dynamics of a cool planetary atmosphere impulsively heated during close approach to a bright star (V =6.5). We present new radial velocity (RV) and photometric measurements of this star as part of the Transit Ephemeris Refinement and Monitoring Survey (TERMS). The RV measurements from Keck-HIRES reduce the predicted transit time uncertainty to 20 minutes, an order of magnitude improvement over the ephemeris from the discovery paper. We photometrically monitored a predicted transit window under relatively poor photometric conditions, from which our non-detection does not rule out a transiting geometry. We also present photometry that demonstrates stability at the millimag level over its rotational timescale. Subject headings: planetary systems – techniques: photometric – techniques: radial velocities – stars: individual (HD 156846) 1. INTRODUCTION Several planets discovered with the radial velocity tech- The discovery of exoplanets using the transit technique nique have subsequently been found to transit, the first is becoming increasingly dominant amongst the various of which was HD 209458b (Charbonneau et al. 2000; detection methods. Examples of major contributors to Henry et al. 2000). The brightness of their host stars has the ground-based discovery of transiting exoplanets are facilitated further characterization of their atmospheres, the Hungarian Automated Telescope Network (HATNet) such as the cases of HD 189733b and HD 149026b (Bakos et al. 2004) and SuperWASP (Pollacco et al. (e.g., Knutson et al. (2009a,b), see also review arti- 2006). From the vantage-point of space, the major con- cle by Seager & Deming (2010)). The Neptune-mass tributors are the Kepler mission (Borucki et al. 2010) planet orbiting GJ 436 became the first known transiting and the CoRoT mission (Barge et al. 2008). The discov- planet around an M-dwarf primary (Gillon et al. 2007). eries provided by these surveys are producing insights The detection of transits for the planets HD 17156b into the exoplanet mass-radius relationship, extending (Barbieri et al. 2007) and HD 80606b (Laughlin et al. down towards super-Earth planets (Seager et al. 2007). 2009; Moutou et al. 2009), enabled by their high eccen- Although these space-based surveys are expected to ex- tricities (Kane & von Braun 2008, 2009), provided the tend the period sensitivity to longer periods, such as the first insights into the structures of longer-period plan- case of CoRoT-9b (Deeg et al. 2010), the picture is in- ets. Many of the known planets with orbital peri- complete since the surveys are strongly biased towards ods larger than a few days have yet to be photomet- short-period planets around relatively faint host stars. rically monitored at predicted transit times, hampered mostly by insufficient orbital parameter precision to ac- curately predict when the planet might transit. Fur- [email protected] 1 NASA Exoplanet Science Institute, Caltech, MS 100-22, 770 ther discoveries of long-period planetary transits around South Wilson Avenue, Pasadena, CA 91125 bright stars are vital to understanding the dependence arXiv:1103.4127v1 [astro-ph.EP] 21 Mar 2011 2 Department of Astronomy, University of California, Berke- of planetary structure and atmospheric dynamics on the ley, CA 94720 periastron distance of the planet (Fortney et al. 2010; 3 Space Sciences Laboratory, University of California, Berke- ley, CA 94720 Kane & Gelino 2010; Langton & Laughlin 2008). Pro- 4 Department of Astronomy and Astrophysics, Pennsylvania vided the orbital parameters can be determined with suf- State University, 525 Davey Laboratory, University Park, PA ficient precision, monitoring planets detected via the ra- 16802 dial velocity technique at predicted transit times provides 5 Center for Exoplanets & Habitable Worlds, Pennsylvania State University, 525 Davey Laboratory, University Park, PA a means to increase the sample of long-period transit- 16802 ing planets (Kane et al. 2009, 2010). There exist efforts 6 Center of Excellence in Information Systems, Tennessee to detect transits of the known radial velocity planets, State University, 3500 John A. Merritt Blvd., Box 9501, such as the Spitzer search for transits of low-mass planets Nashville, TN 37209 7 Department of Physics & Astronomy, University of British (Gillon et al. 2010). The Transit Ephemeris Refinement Columbia, Vancouver, BC V6T1Z1, Canada and Monitoring Survey (TERMS) is a program which is 8 Department of Astronomy, Yale University, New Haven, CT capable of monitoring long-period as well as short-period 06511 9 Dept of Physics & Astronomy, Swarthmore College, Swarth- planets by refining the orbital parameters of the system. more, PA 19081 Here we present a detailed analysis of one such system. 10 UCO/Lick Observatory, University of California, Santa The massive planet orbiting the star HD 156846 was dis- Cruz, CA 95064 covered by Tamuz et al. (2008) using the CORALIE in- 2 Stephen R. Kane et al. strument. The planet is in a highly-eccentric orbit with if the tidal quality factor, Q, is of order the Jovian value a period of slightly less than a year. The periastron ar- or higher (Barnes & O’Brien 2002). Given the mass ra- gument of the orbit is such that the transit probability tios observed for the Jovian planets and their satellites is significantly enhanced compared to an equivalent cir- in our own solar system, one might reasonably expect cular orbit (5.4% compared to 0.9%). Our combined fit a ∼ 0.5M⊕ satellite, which would be readily detectable to new Keck data along with the discovery CORALIE using transit timing techniques (Kipping 2009), and per- data greatly improves the orbital parameters for the sys- haps even directly via space-based photometry. Given tem, allowing an accurate prediction of possible transit the inflated transit probability for this planet, and its times. We also find no evidence for additional compan- potentially interesting dynamical history, this becomes a ions in the system through high-precision radial velocity prime candidate in this regard. (RV) data acquired during periastron passage over mul- tiple orbits. The long-term photometry presented here 3. KECK MEASUREMENTS AND REVISED ORBITAL establishes the photometric stability of the host star. We PARAMETERS present photometry acquired during a predicted transit 3.1. Observations window which places an upper-limit on a transit for this We observed HD 156846 with the HIRES echelle spec- planet. Finally, we discuss additional constraints on the trometer (Vogt et al. 1994) on the 10-m Keck I telescope mass and orbit of the planet from a potential transit with the goal of improving the accuracy of the predicted null-result. transit time to guide and temporarily anchor a photomet- 2. SCIENCE MOTIVATION ric monitoring campaign. Our Keck observations post- date the CORALIE measurements (Tamuz et al. 2008) Here we describe why the planet orbiting HD 156846 is and span 2009 May to 2010 October. The 41 Keck a particularly interesting target and the potential gains RV measurements were made from observations with an which may be achieved through further studies. iodine cell mounted directly in front of the spectrom- HD 156846 is an extraordinarily bright star (V =6.5), ∼ eter entrance slit. The dense set of molecular absorp- brighter indeed by a factor of 2.9 than either of the tion lines imprinted on the stellar spectra provide a ro- planet hosting stars HD 209458 and HD 189733. The op- bust wavelength fiducial against which Doppler shifts are portunities for follow-up studies of a fundamentally new measured, as well as strong constraints on the shape of type of planetary atmosphere would therefore be close the spectrometer instrumental profile at the time of each to optimal. Note that massive, relatively cold planets observation (Marcy & Butler 1992; Valenti et al. 1995). such as this one have intrinsically difficult atmospheres to We measured the Doppler shift of each star-times-iodine study via transmission spectroscopy. Their atmospheric spectrum using a modelling procedure descended from scale heights are of order a factor of 20 smaller than Butler et al. (1996) as described in Howard et al. (2009). typical hot Jupiters (see for example Vidal-Madjar et al. The times of observation (in barycentric Julian days), (2011)), such that a bright host star is needed to achieve relative RVs, and associated errors (excluding jitter) are adequate signal-to-noise. listed in Table 1. In cases when we observed the star Given the properties of this star (see Section 3.2), the 3–5 times in quick succession, we report the mean RV received flux of the planet at apastron will be nearly iden- and appropriately reduced uncertainty. We also observed tical to the flux received by the Earth from the Sun. It HD 156846 with the iodine cell removed to construct a is therefore not unreasonable to expect that the planet stellar template spectrum for Doppler modelling and to during this phase of the orbit will be sheathed in re- derive stellar properties. flective white water clouds. At some point prior to pe- riastron, when the received flux increases briefly to a 3.2. Stellar Properties value nearly 150 times that at apastron, the received flux should be sufficient to flash the water clouds to steam We used Spectroscopy Made Easy (Valenti & Piskunov (Sudarsky et al.