ebook img

HAT-P-67b: An Extremely Low Density Saturn Transiting an F-Subgiant Confirmed via Doppler Tomography PDF

2.7 MB·
by  G. Zhou
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview HAT-P-67b: An Extremely Low Density Saturn Transiting an F-Subgiant Confirmed via Doppler Tomography

Draftversion March8,2017 PreprinttypesetusingLATEXstyleemulateapjv.2/16/10 HAT-P-67b: AN EXTREMELY LOW DENSITY SATURN TRANSITING AN F-SUBGIANT CONFIRMED VIA DOPPLER TOMOGRAPHY † G. Zhou1, G. A´. Bakos2,⋆, J. D. Hartman2, D. W. Latham1, G. Torres1, W. Bhatti2, K. Penev2, L. Buchhave3, G. Kova´cs4, A. Bieryla1, S. Quinn1, H. Isaacson5, B. J. Fulton6, E. Falco1, Z. Csubry2, M. Everett7, T. Szklenar8, G. Esquerdo1, P. Berlind1, M. L. Calkins1, B. B´eky9, R. P. Knox10, P. Hinz10, E. P. Horch11, L. Hirsch5, S. B. Howell12, R. W. Noyes1, G. Marcy5, M. de Val-Borro2, J. La´za´r8, I. Papp8, P. Sa´ri8 Draft version March 8, 2017 7 1 ABSTRACT 0 WereportthediscoveryofHAT-P-67b,ahot-SaturntransitingarapidlyrotatingF-subgiant. HAT- 2 r hPo-6st7bstahrasinaara∼di4u.s81o-fdaRypp=eri2o.d08o5r+−b00i..t00.9761WReJp,loarcbeitainngupapMer∗li=mi1t.o6n42t+−h00e..10m5752aMss⊙o,fRth∗e=pla2n.5e4t6v+−i00a..00r98a94dRia⊙l a M velocity measurements to be Mp < 0.59MJ, and lower limit of > 0.056MJ by limitations on Roche lobe overflow. Despite being a subgiant, the host star still exhibits relatively rapid rotation, with a projected rotational velocity of vsinI =35.8±1.1kms−1, making it difficult to precisely determine 6 ⋆ the massofthe planet using radialvelocities. We validatedHAT-P-67bvia two Doppler tomographic ] detections of the planetary transit, which eliminated potential eclipsing binary blend scenarios. The P Doppler tomographic observations also confirmed that HAT-P-67b has an orbit that is aligned to E within 12◦, in projection, with the spin of its host star. HAT-P-67b receives strong UV irradiation, . and is amongst the one of the lowest density planets known, making it a good candidate for future h UV transit observations to search for an extended hydrogen exosphere. p Subject headings: planetary systems — stars: individual (HAT-P-67, 03084-00533)techniques: spec- - o troscopic, photometric r t s a 1. INTRODUCTION govern planet formation and evolution. Planets orbit- [ ing high mass stars are likely born in high mass proto- Finding well-characterized planets in a variety of en- planetary disks (e.g. Muzerolle et al. 2003; Natta et al. 2 vironments is key to understanding the processes that 2006), environments that may yield higher planet oc- v currence rates (e.g. Johnson et al. 2010; Bowler et al. 6 1Harvard-Smithsonian Center for Astrophysics, Cambridge, 2010)andhigher-massplanets(e.g.Lovis & Mayor2007; 0 MA02138, USA 1 2DepartmentofAstrophysicalSciences,PrincetonUniversity, Jones et al. 2014) than around solar type stars. Planets 0 Princeton,NJ08544, USA around early type stars also receive higher incident flux 0 ⋆3CPaecnktarerdfoFreSlltoawrandPlanetFormation,NaturalHistoryMu- over their lifetimes, which in turn make them anchor- . seum of Denmark, University of Copenhagen, DK-1350 Copen- points in the planet mass-radius-equilibrium tempera- 2 hagen, Denmark ture relationships (e.g. B´eky et al. 2011; Enoch et al. 0 4KonkolyObservatoryoftheHungarianAcademyofSciences, 2012). 7 Budapest, Hungary However, only 1% of known transiting planets or- 1 5Department of Astronomy, University of California, Berke- bit stars more massive than 1.5M . Early type stars : ley,CA,USA ⊙ v 6InstituteforAstronomy,UniversityofHawaii,Honolulu,HI have larger radii, resulting in shallower transit depths i 96822, USA for any planets; they are also more likely to have rota- X 7NationalOpticalAstronomyObservatory,Tucson,AZ,USA tionally blended spectral lines due to the lack of mag- 8HungarianAstronomicalAssociation,Budapest, Hungary ar 9GoogleInc. netic braking over the main-sequence lifetime, making 10StewardObservatory,UniversityofArizona,933N.Cherry traditional radial-velocity confirmation techniques more Ave.,Tucson,AZ85721, USA difficult. One successful strategy is to conduct radial- 11Department of Physics, Southern Connecticut State Uni- velocity surveys of ‘retired A-stars’ – stars that have versity,501CrescentStreet,NewHaven,CT06515,USA 12NASA Ames Research Center, Moffett Field, CA 94035, evolved off the main-sequence and spun down enough USA to exhibit sharp spectroscopic lines that enable precise † Based on observations obtained with the Hungarian-made radial-velocity measurements. These surveys have been Automated Telescope Network. Based in part on observa- extremely successful, yielding 122 planetary systems tions made with the Keck-I telescope at Mauna Kea Observa- tory,HI(KecktimeawardedthroughNASAprogramsN029Hr, to date15 (e.g. Johnson et al. 2007; Wittenmyer et al. N108Hr, N154Hr and N130Hr and NOAO programs A289Hr, 2011; Jones et al. 2014). Recently, transit surveys have and A284Hr). Based inpart onobservations obtained withthe also been successful in discovering planets around high Tillinghast Reflector 1.5 m telescope and the 1.2 m telescope, mass stars. These include planets around subgiants both operated by the Smithsonian Astrophysical Observatory at the Fred Lawrence Whipple Observatory in Arizona. This and giants whose shallow transits were identified by workmakesuseoftheSmithsonianInstitutionHighPerformance Kepler (e.g. Kepler-56b,c Huber et al. 2013, Kepler- Cluster(SI/HPC).Basedinpartonobservationsmadewiththe 96bLillo-Box et al.2014,Kepler-432bQuinn et al.2015; Nordic Optical Telescope, operated on the island of La Palma jointly by Denmark, Finland, Iceland, Norway, and Sweden, in the Spanish Observatorio del Roque de los Muchachos of the 15 Choosinghost stars with logg <4.0, fromNASA Exoplanet Instituto deAstrof´ısicadeCanarias. Archive,July2016 2 Zhou et al. Ciceri et al. 2015, KOI-206b, KOI-680b Almenara et al. shown in Figure 2, and the data presented in Table 1. 2015, and K2-39b Van Eylen et al. 2016), and hot- Jupiters around main-sequence A-stars confirmed via 1.015 Doppler tomography(WASP-33bCollier Cameron et al. 2010b,Kepler-13bSzabo´ et al.2011;Shporer et al.2011; 1.010 Johnson et al. 2014, HAT-P-57b Hartman et al. 2015, 1.005 and KELT-17b Zhou et al. 2016b). In this paper, we present the discoveryof HAT-P-67b, x1.000 u a Saturn-mass planet found to transit an F-subgiant by Fl0.995 the HATNet survey (Bakos et al. 2004). Despite the evolvedstatusofHAT-P-67,the hoststarstillexhibits a 0.990 rapid rotation rate of vsinI = 35.8±1.1kms−1, mak- ⋆ 0.985 ing precise radial velocities difficult to obtain. Eventual confirmationwasachievedviaadetectionoftheDoppler 0.980 −0.4 −0.2 0.0 0.2 0.4 tomographic shadow of the planet during transit. When Phase a planet transits a rapidly rotating star, it successively Fig.1.— HATNetdiscoverylightcurvesshowingthetransitof blockspartsoftherotatingstellardisk,causinganasym- HAT-P-67b. The light curve is phase folded to a period of P = metry in the observed spectral line profiles. At low ro- 4.8101050days,aspertheanalysisinSection3. Greypointsshow tational velocities, the asymmetry can be measured by therawlightcurve,whilebluepointsshowthedatabinnedat0.01 theHolt-Rossiter-McLaughlineffect(Holt1893;Rossiter in phase. Solid blue line shows the best fit transit model from Section3.4. 1924; McLaughlin1924). At higher rotationalvelocities, theshadowoftheplanetcanberesolvedinthebroadened stellar spectroscopic lines (e.g. Collier Cameron et al. 2010a,b). A detection of the Doppler tomographic sig- nal, at a depth and width that are in agreement with thephotometrictransit,eliminateseclipsingbinaryblend 1.00 2011 Apr 15 i scenariosthat may mimic transiting planet signals. Fur- ther radial velocity measurements can then provide an 0.98 2011 May 19 i upper-limit mass constraint of the orbiting companion. x Ifthemasscanbeconstrainedtolessthanthatofbrown Flu0.96 2011 Jun 07 i dwarfs,the transitingobjectis confirmedto be a planet. ve ati0.94 2012 May 28 i el 2. OBSERVATIONS R0.92 2013 Apr 25 i 2.1. Photometry 0.90 2013 May 24 z ThetransitsofHAT-P-67bwerefirstdetectedwiththe HATNet survey (Bakos et al. 2004). HATNet employs a 0.88 networkofsmall,widefieldtelescopeslocatedattheFred 0.00 LawrenceWhippleObservatory(FLWO)inArizona,and the Mauna Kea Observatory(MKO) in Hawaii, USA, to photometricallymonitor selected 8×8◦ fields of the sky. −0.02 Atotalof4050I bandobservationsweretakenbyHAT- dual−0.04 5 and HAT-8 over 2005 Jan – July, and an additional si e 4518 observations were obtained in the Cousins R band R using HAT-5, HAT-7, and HAT-8 telescopes between del −0.06 o 2008Feb–Aug. ThedatareductionfollowsBakos et al. M−0.08 (2010). Lightcurveswereproducedviaaperturephotom- etry, and detrended with ExternalParameterDecorrela- −0.10 tion(EPD,Bakos et al.2007) andTrendFiltering Algo- rithm(TFA,Kov´acs et al.2005). TheBox-FittingLeast −0.12 −0.04 −0.02 0.00 0.02 0.04 Squares (BLS, Kov´acs et al. 2002) analysis revealed the Orbital Phase periodic transits of the planet candidate. The discovery light curve of HAT-P-67b is shown in Figure 1, and the photometry presented in Table 1. Fig.2.— Follow-uptransitlightcurvesofHAT-P-67bobtained byKeplerCamontheFLWO1.2mtelescope. Theindividualtran- Tobettercharacterizetheplanetaryproperties,follow- sitsare labelled, andarbitrarilyoffset along the y axis for clarity. up photometry of the transits were obtained using Ke- Therawlightcurvesareplottedingrey,andphasebinnedat0.005 plerCam on the FWLO 1.2m telescope. KeplerCamis a intervals in blue. The best fit models are plotted in blue. The 4K×4KCCDcamerawith apixelscaleof0′.′672pixel−1 residualsareshownonthebottom panel. at 2×2 pixel binning. The photometry was reduced as per Bakos et al. (2010). A full transit was observed in 2.2. Spectroscopy the Sloan-i band on 2012 May 28, and five partial tran- sits observed on 2011 Apr 15, 2011 May 19, 2011 Jun Spectroscopic observations of HAT-P-67 were carried 07, 2013 Apr 25 in the Sloan-i, and 2013 May 24 in the outusingtheFIber-fedEchelleSpectrograph(FIES),the Sloan-z band. The light curves and best fit models are Tillinghast Reflector Echelle Spectrograph (TRES), and HAT-P-67b 3 TABLE 1 Differentialphotometry of HAT-P-67 BJD Mag(Raw) a Mag(EPD) Mag(TFA) σ Mag Instrument Filter 2454521.99042 9.39859 9.72414 9.72271 0.00303 HATNet R 2454521.99445 9.39572 9.72048 9.73067 0.00325 HATNet R 2454521.99854 9.39262 9.72005 9.71695 0.00276 HATNet R 2454522.00264 9.38941 9.7199 9.72327 0.00344 HATNet R 2454522.00673 9.41475 9.74011 9.73327 0.00328 HATNet R a Thistableisavailableinamachine-readableformintheonlinejournal. Aportionisshownhere forguidanceregardingitsformandcontent. Raw, EPD, and TFA magnitudes are presented for HATNet light curves. The detrending and potentialblendingmaycausetheHATNettransittobeshallowerthanthetruetransitintheEPD and TFA light curves. This is accounted for in the global modellingby the inclusion of a third lightfactor. Follow-uplightcurveshavebeentreatedwithEPDsimultaneoustothetransitfitting. Pre-EPDmagnitudesarepresentedforthefollow-uplightcurves. TABLE 2 Summaryof photometric observations Facility Date(s) NumberofImages a Cadence(s) b Filter HATNet 2005Jan–2005Jul 4050 328 I HATNet 2008Feb–2008Aug 4518 246 CousinsR FLWO1.2mKeplerCam 2011Apr15 730 24 Sloan-i FLWO1.2mKeplerCam 2011May19 509 44 Sloan-i FLWO1.2mKeplerCam 2011Jun07 801 29 Sloan-i FLWO1.2mKeplerCam 2012May28 730 34 Sloan-i FLWO1.2mKeplerCam 2013Apr25 960 24 Sloan-i FLWO1.2mKeplerCam 2013May24 361 24 Sloan-z a Outlyingexposureshavebeendiscarded. b Median time difference between points in the light curve. Uniform sampling was not possible due to visibility,weather,pauses. theHighResolutionEchelleSpectrometer(HIRES).The vations showed that any companion orbiting HAT-P-67 observations are summarized in Table 4 and described must be sub-brown dwarf in mass. below. In addition, we observed two partial spectroscopic Initial spectroscopic characterization of HAT-P-67b transitsofHAT-P-67b,on2016Apr17and2016May16, was obtained with the FIES instrument (Telting et al. with TRES to detect the Doppler tomographic shadow 2014) on the 2.5m Nordic Optical Telescope (NOT). of the planet. These observationswere performedas per FIES is a fiber fed high resolution echelle spectrograph the strategy described in Zhou et al. (2016a). A set of witharesolutionofλ/∆λ≡R=67000andspectralcov- time series spectra, at 900s cadence, where collected on erage of 3700−7300˚A. Four FIES radialvelocities were bothnights. TheDopplertomographicanalysisforthese obtained over the 2009 Aug – Oct period. The obser- two transit sets are described in Section 3.1. vations were obtained and reduced as per the procedure To constrain the mass of the companion, we obtained fromBuchhave et al.(2010). Noradialvelocityvariation spectroscopic observations from HIRES on the 10m was detected with the FIES observations,with a scatter KECK telescope (Vogt et al. 1994) at MKO over the of 200ms−1 over the four observations. 2009 Jul – 2012 Mar period. A total of 19 observa- Additionalobservationswereobtainedwith the TRES tionswereobtainedthroughtheI2 celltoprovideprecise instrument (Fu˝r´esz 2008) on the FLWO 1.5m telescope. radial velocities. An additional I2-free observation was TRES is a fiber fed echelle with a spectral resolution of obtained to provide a template for the radial-velocity R = 44000, over the spectral region of 3850−9100˚A. measurements. The instrument was set up to use the Radial velocities and spectral classifications are mea- C2 decker, which provides a 14′′×0′.′861 slit, yielding a suredfromeachspectrumasperBuchhave et al.(2012). spectral resolution of R = 48000. The radial velocities EachTRESobservationconsistsofthreeexposurescom- weremeasuredasperButler et al.(1996),andthebisec- bined together for cosmic-ray removal, and wavelength- torspanscalculatedasperTorres et al.(2007). Thehigh calibrated by Th-Ar lamp exposures that bracket each signal-to-noiseHIRESobservationsprovidethebestcon- setofthreeexposures. TwoTRESobservationsatphase straints on the radial velocities of HAT-P-67, and were quadrature were taken in 2011 Apr 17 and 2011 Apr 20, used in the global analysis in Section 3.4. The radial with signal-to-noise at the Mg b lines of ∼100 per reso- velocities from HIRES are plotted in Figure 3 and pre- lution element. The velocity difference between the two sented in Table 3. observations was 80ms−1, with a per-point uncertainty of 100ms−1. As such, the the FIES and TRES obser- 3. ANALYSIS 4 Zhou et al. TABLE 3 KECK-HIRES relativeradialvelocities and bisector span measurementsof HAT-P-67 BJD RV a σ RV BS σ BS (UTC) (ms−1) (ms−1) (ms−1) (ms−1) 2455696.8366 -105 28 24 11 2455696.88382 -151 31 67 17 2455697.833 31 25 -58 9 2455698.92918 93 23 -53 8 2455699.83162 63 25 46 10 2455700.88206 1 22 30 13 2455704.84352 18 23 -2 13 2455705.86007 100 22 -34 8 2455706.83882 8 22 14 11 2455707.85238 35 22 -17 13 2455853.70871 -6 28 ... ... 2455945.15236 -89 24 ... ... 2455997.02884 -17 38 ... ... 2455017.0082 58 22 ... ... 2455042.88956 -65 26 ... ... 2455043.9989 -25 29 ... ... 2455044.94822 -63 26 ... ... 2455048.86097 -112 30 ... ... 2455107.71733 241 40 -3 28 a Internal errors excluding the component of astrophys- ical/instrumental jitter considered in Section 3. Bisector spans(BS)aregivenwhereavailable. TABLE 4 Summaryof spectroscopic observations Telescope/Instrument DateRange NumberofObservations Resolution ObservingMode NOT2.5m/FIES 2009Aug4–2009Oct10 5 67000 RECONRV FLWO1.5m/TRES 2011Apr17–2011Apr20 2 44000 RECONRV KECK10m/HIRES 2009Jul04–2012Mar10 19 55000 RVa FLWO1.5m/TRES 2016Apr17 14 44000 Transitb FLWO1.5m/TRES 2016May16 16 44000 Transitb a Highresolutionspectratoobtainstellaratmosphericparametersandhighprecisionradialvelocities b Highresolutionin-transitspectratodetecttheDopplertomographicsignaloftheplanet 3.1. Doppler tomographic detection of the planetary 3.2. Stellar parameters transit Stellar atmospheric parameters of HAT-P-67 were de- The significant rotational broadening of HAT-P-67 rivedfromthe32TRESspectrausingtheStellarParam- allows us to detect the spectroscopic transit of the eter Classificationpipeline (SPC, Buchhave et al. 2012). planet via Doppler tomography (Collier Cameron et al. We first run SPC to retrieve an initial estimate of the 2010a,b). Two sets of transit spectroscopy were ob- stellar atmospheric parameters. These are then incorpo- tained for HAT-P-67b with TRES. The TRES spec- rated in a first run of the globalmodeling and isochrone tra were processed as per the procedure laid out in retrievalanalysisdescribedlaterinSection3.4. We then Zhou et al. (2016a): the broadening profiles were de- re-run SPC with the stellar surface gravity logg fixed rived via a least-squares deconvolution (LSD) of the ob- to that measuredfrom the transit durationin the global served spectra against a non-rotating stellar template analysis(Section3.4,withlogg =3.854+0.014)toprovide −0.023 (as per Donati et al. 1997). Synthetic template spectra updated, and better constrained T and [Fe/H] values. eff weregeneratedusingtheSPECTRUM(Gray & Corbally WefindthatHAT-P-67isconsistentwithanF-subgiant, 1994) spectral synthesis program, using the ATLAS9 of effective temperature T = 6406±62K, metallicity eff model atmospheres (Castelli & Kurucz 2004). The syn- [m/H]=−0.08±0.05, and projected rotational velocity thetic templates were generated at the same Teff, logg vsinI = 36.5±0.3kms−1. Similarly, running SPC on and [Fe/H] as HAT-P-67, with no line broadening im- the fo⋆ur FIES spectra yield T = 6380±50K, logg = posed. A broadening profile was derived for each spec- eff 3.91±0.10, [m/H] = −0.05±0.08, vsinI = 38kms−1, trumandsubtractedfromtheaverageout-of-transitpro- ⋆ consistentwiththeinterpretationthatHAT-P-67isanF file,revealingtheplanetarytransitsignal(Figure4). We subgiant. Since anaccuratevsinI measurementisvital model the rotationalprofiles and the planetary signal as ⋆ tocorrectlymodelingtheDopplertomographicsignal,we part of our global analysis, described in Section 3.4. also use the set of time-series TRES spectra to measure the vsinI of HAT-P-67. Following Zhou et al. (2016a), ⋆ HAT-P-67b 5 R /R , and transit inclination i. Individual quadratic p ⋆ 300 limb darkening parameters are assigned to each light 200 curve (interpolated from Claret & Bloemen 2011), and ) 1− 100 fixed throughout the fitting. Separate dilution factors ms 0 are allowed for the HATNet I and RC band light curves V ( −100 to account for any distortions to the light curve shape R −200 from the TFA detrending process. The follow-up light −300 curves are simultaneously detrended against instrumen- talparametersdescribingtheX,Y pixelcentroidsofthe target star, background flux, and target airmass. The 300 ) 1− 200 radial velocities are described by an arbitrary offset γ ms 100 and orbital semi-amplitude K. The orbital eccentricity al ( 0 parametersecosω andesinω arealsoincludedwhenec- du −100 centricity is allowed to vary. The Doppler tomographic si −200 signal is modeled as per Zhou et al. (2016b), via a 2D e R −300 integration of the stellar surface covered by the planet. The free parametersdescribing the Doppler tomography 0.00 0.25 0.50 0.75 1.00 effect include the projected spin-orbit angle λ, and the Phase projected rotational broadening velocity vsinI . Note ⋆ that we do not account for the broadening of the plane- Fig.3.— Radial velocities from KECK-HIRES for HAT-P-67. tary shadow due to the motion of the planet during an The observations are marked by the open circles. The best fit exposure; the blurring of the planetary shadow during circularorbitmodelisshownbythesolidredline;thedashedlines an exposure (2kms−1) is smaller than the width of the encompassthe2σsetofmodelsallowedbythedata. Theresiduals areplottedonthebottom panel. shadow (7.2kms−1), but is not an insignificant effect. the broadeningkernelforeachspectrumis modeledby a We also allow the effective temperature Teff, metallic- rotational kernel, with width of vsinI , and a Gaussian ity [M/H], and the apparent K-band magnitude to be ⋆ kernel to account for macroturbulence and instrumen- iterated, though heavily constrained about their spec- tal broadening, finding vsinI = 30.9±2.0kms−1, and troscopic and photometric values. At each step, we de- ⋆ macroturbulence of 9.22 ± 0.5kms−1. The uncertain- rive a stellar density ρ⋆ from the transit duration as per Seager & Mall´en-Ornelas (2003); Sozzetti et al. (2007), ties are estimated from the standard deviation scatter andquerythestellarisochronestoderiveadistancemod- between exposures. The difference between the vsinI ⋆ ulus. Isochrone interpolation is performed at each step measured via SPC and that from the rotational profile using the gradient boosting regression algorithm imple- can be partially attributed to the inclusion of macrotur- mented in scikit-learn. This distance modulus is com- bulence. paredtotheactualdistanceasmeasuredfromtheGAIA 3.3. GAIA parallax parallax,with the difference applied as a penalty on the likelihood function. HAT-P-67 is included in the Tycho-GAIA- The rapid rotation rate of HAT-P-67 can introduce Astrometric-Catalogue in the first data release (DR1) a bias in the isochrone-derived parameters for the sys- of GAIA (Lindegren et al. 2016), which measured a tem. For stars with radiative envelopes, the convective parallax of 2.60 ± 0.23mas. Several literature investi- core overshoot and mixing length parameters are dif- gations have pointed out a systematic under-estimation ferent to that of non-rotating stellar models, with the in the DR1 parallaxes, as per separate studies via overall effect of lengthening the main-sequence lifetime eclipsing binaries (Stassun & Torres 2016), close-by (e.g.Meynet & Maeder2000). We adoptthe Geneva 2D Cepheids (Casertano et al. 2016), asteroseismic dis- stellarevolutionmodels(Ekstr¨om et al.2012),whichac- tances (Silva Aguirre et al. 2016), and comparison count for the effects of rotation, for our analysis. For with existing parallaxes of solar neighborhood stars the isochrone fitting, we introduce the added dimension (Jao et al. 2016). We adopt the correction offered in of equatorial velocity v into our interpolation. The Stassun & Torres (2016) of −0.325 ± 0.062mas to the eq v distribution is calculated from the measured vsinI DR1 parallax of HAT-P-67, arriving at an adopted eq ⋆ value, scaled by a uniform distribution of orientations parallax value of 2.92 ± 0.23mas, and corresponding sampled in cosI . astrometric distance measurement of 342±27pc. This ⋆ To compare our Geneva isochrone results to fittings parallax measurement is used to co-constrain the stellar withmore traditional1D isochrones,we alsopresentthe parameters during the global modeling in Section 3.4. results from analyses using the Dartmouth isochrones (Dotter et al. 2008). 3.4. Global fitting and derived planet parameters The parameter space is explored with a Markov chain We perform a global analysis of the HATNet discov- MonteCarlo(MCMC)analysis,usingtheaffine-invariant ery light curves, follow-up transit light curves, KECK- ensemble sampler emcee (Foreman-Mackey et al. 2013). HIRES I2 radial velocities, and the TRES Doppler to- The observations are fitted for twice, with the per-point mographic signal, co-constrained by stellar isochrones uncertainties for each dataset inflated such that the re- and the GAIA distance measurement. The transits duced χ2 is at unity for the second run. A cosi prior is are modelled according to Mandel & Agol (2002), with imposedonthetransitinclination,whileaGaussianprior the transit shape defined by the transit centroid time is imposed on T = 6406±64K, [M/H] = 0.08±0.05, eff T , star-planet distance a/R , planet-star radius ratio 0 ⋆ 6 Zhou et al. Fractional Variation Fractional Variation -0.1 0.0 0.1 -0.1 0.0 0.1 0.04 0.03 0.03 se 0.02 se a a h h 0.02 P P 0.01 Data 0.01 Data 0.04 0.03 Egress 0.03 se 0.02 se Egress a a Ph 0.01 Model vsini− vsini Ph 00..0012 Model vsini sini − v 0.04 0.03 0.03 se 0.02 se a a h h 0.02 P P 0.01 Residual Ingress 0.01 Residual Ingress −150 −100 −50 0 50 100 150 −150 −100 −50 0 50 100 150 Velocity(kms 1) Velocity(kms 1) − − 1.005 1.000 1.000 x x u u Fl Fl0.995 0.995 0.990 0.990 0.01 0.02 0.03 0.01 0.02 0.03 0.04 Phase Phase Fig.4.— Doppler tomographic signals for the spectroscopic transits of HAT-P-67b on 2016 Apr 17 (left) and 2016 May 16 (right). Thetoppanelsshowtheresidualbetweenthebroadeningkernelfromeachobservationandthatoftheaveragedout-of-transitbroadening kernel. Thetransitcanbe seenas the darkstreakrunningdiagonallyfrombottom left(mid-transit)totop right(post-egress). The best fitmodels areplotted below, asarethe residual aftersubtraction ofthe modeled planetary tomographic signal. The bottom panels show thereconstructedlightcurvesfromtheDopplertomographicobservation. TheseareconstructedbysummingthesignalundertheDoppler tomographic‘shadow’oftheplanet. Theredlineshowstheexpected signalfromthephotometrictransit,agreeingwiththetransitdepth modelledviaDopplertomography, eliminatingpotential blendscenariosforthesystem. andvsinI =30.9±2.0kms−1basedonthespectroscopic linkofthe MCMCchain,wecalculatethe corresponding ⋆ values outlined in Section 3.2. We note that the derived Roche lobe radius using equation A5 of Hartman et al. posterior vsinI (35.8 ± 1.1kms−1) is offset with the (2011). Links with R /a overflowing the Roche lobe are ⋆ p priorby∼2σ. Resettingthepriorto35.8±1.1kms−1did eliminated. For the circular orbit fit, the Roche lobe not change the system parameters significantly, with a provides a weak lower limit on the mass of the planet derived λ 1σ upper limit of < 11◦ (compared to < 14◦ of 0.056MJ. The posterior distribution for planet mass from our adopted results). The K-band magnitude is is plotted in Figure 5. The final mass measurement we also constrained by a Gaussian prior about its 2MASS report is the 68% confidence interval for the Roche lobe value (Skrutskie et al. 2006). The GAIA parallax is constrained posterior distribution. also heavily constrained by a Gaussian prior about our We present four sets of solutions in Tables 5 and 6 for adopted value of 2.92 ± 0.23mas as described in Sec- thecircularandeccentricorbitscenariosfromtheGeneva tion 3.2. A β distribution prioris imposed on the eccen- andDartmouthisochronefits. ThecircularorbitGeneva tricity,followingtheprescriptionforshortperiodplanets isochrone fit solution is preferred, favored over the ec- set out in Kipping (2013). Uniform priors are imposed centric solution with a Bayesian Information Criterion on all other parameters. ∆BIC of 212. That is, the increased degrees of freedom Due to the large radius of HAT-P-67b, potential solu- in an eccentric orbit fit do not justify the improvements tions in the MCMC chain lead to the planet overflowing inthe goodness offit overthat ofa circular orbitmodel. its Roche lobe (e.g. Lecavelier des Etangs et al. 2004). The evolutionary stage of HAT-P-67 is shown in Fig- We canuse this to placea lowerlimit onthe massofthe ure 6 on the Hertzsprung-Russell diagram, along with planet by assuming no Roche lobe overflow. For each evolutionary tracks of various stellar masses and rota- HAT-P-67b 7 tion rates marked for context. The derived stellar and 3.7. Imaging Constraints on Resolved Neighbors planetary parameters are presented in Tables 5 and 6, Inordertodetectpossibleneighboringstarswhichmay respectively. bedilutingthephotometrictransits,weobtainedoptical and near infrared imaging of HAT-P-67 using the Clio2 near-IR imager (Freed et al. 2004) on the MMT 6.5m Constrained by r. Roche Lobe Overflow telescope on Mt. Hopkins, in AZ, together with the Dif- y Dist RRoecghieo nlo ebxec louvdeerdfl obwy f2e0r1e1n;tiaHloSrpchecektleaSl.ur2v0e1y2,In2s0tr1u1m) eanntd(DtShSeI;WHIoYwNellHetigahl-. t RV only Resolution Infrared Camera (WHIRC), both on the bili WIYN 3.5m telescope16 at Kitt Peak National Obser- a b vatory in Arizona. o r The Clio2 images were obtained on the night of UT P 2011 June 22. Observations in H-band and L′-band 0.0 0.2 0.4 0.6 0.8 1.0 were made using the adaptive optics (AO) system. A Planet Mass (M ) possible neighbor was detected 4′.′9 to the southeast J of HAT-P-67 with a relative magnitude difference of Fig. 5.— Theposterior distributionforthemassoftheplanet. ∆H =7.4±0.5mag,butno closerobjectsareseen. The Thegreylineshowstheposteriordistributionconstrainedonlyby neighbor was blended with HAT-P-67 in the HATNet theradialvelocities,fromwhichanupperlimitof0.59MJ canbe derived. Alowerlimitof0.056MJ canalsobeappliedifweassume survey observations, but was fully resolved by all sub- the planet is not undergoing Roche lobe overflow. The resulting sequent follow-up observations. Figure 9 shows the H- mass distribution is marked by the red line, while the solutions band magnitude contrast curve for HAT-P-67 based on excludedarefilledinblack. these observations. This curve was calculated using the methodandsoftwaredescribedbyEspinoza et al.(2016). 3.5. Eccentricity constraint The band shown in this image represents the variation in the contrast limit depending on the position angle of We can constrain the eccentricity of the system via the putative neighbor. We can rule out other neighbors the photometric light curves despite a lack of detection with a magnitude difference of ∆H < 2mag, down to oftheradialvelocityorbit,sincetheGAIAparallaxpro- a separation of 0′.′3, and ∆H < 6mag, down to a sep- vides a good constraint on the stellar radius and transit aration of 0′.′8. The L′ observations suffered from high duration (Kipping 2008; Dawson & Johnson 2012). The thermal background, and the 4′.′9 neighbor was not de- eccentricityposterior,asconstrainedprimarilyfromthis tected. Meaningful constraints could not be placed on ‘photo-eccentric’effect, is showninFigure7. The eccen- closer neighbors in L′ based on these observations. tricity 2σ upper limit is 0.43, with a posterior median and 64% confidence region of ecc=0.24±0.12. J-band snapshot images of HAT-P-67 were obtained with WHIRC on the night of 2016 April 24, with a see- The parallax we choose to adopt has an effect on our ing of ∼ 0′.′9. The images were collected at four nod best-fitsolutions. IfwechoosetoadopttheGAIAparal- laxof 2.60±0.23mas (385±34pc)fromLindegren et al. positions, and were calibrated, background-subtracted, registered and median-combined using the same tools (2016) without the systematic correction offered by that we used for reducing the KeplerCam images. The Stassun & Torres (2016), we would have a modest 1.3σ 4′.′9 neighbor was not detected, and we concluded that tension between the best fit isochrone distance and the parallax distance. Adopting a distance of 385 ± 34pc it must have ∆J > 7mag. The closest neighbor de- tectedintheseobservationswasataseparationof9′.′3to yields an eccentric orbit of e = 0.356+0.072. The tidal −0.077 thenorthwest,andhasarelativemagnitudedifferenceof circularization time scale for the system is < 500Myr ∆J =4.96±0.01mag compared to HAT-P-67. Figure 9 (Dobbs-Dixon et al. 2004), so the likelihood of such an showstheJ-bandmagnitudecontrastcurvecomputedin eccentric orbit is low for the system. a similar manner to the H-band contrast curve. 3.6. Transit timing variations and additional The DSSI observations were gathered between the companions nights of UT 26 September 2015 and UT 3 October 2015. A dichroic beamsplitter was used to obtain si- To check for potential transit timing variations that multaneous imaging through 692nm and 880nm filters. may be indicative of additional orbiting companions, we Each observation consisted of a sequence of 1000 40ms re-fit the follow-up transit observations, allowing for in- exposures read-out on 128×128 pixel (2′.′8×2′.′8) sub- dividual transit centroids for each epoch. The timing frames, which were reduced to reconstructed images fol- residuals are shown in Figure 8. The transit geometry lowing Horch et al. (2011). These images were searched parameters a/R , R /R , and inclination, are heavily ⋆ p ⋆ for companions, with none detected. Based on this, constrained by Gaussian priors about their best fit val- the 5σ lower limits on the differential magnitude be- uesfromtheglobalanalysis(adoptedasthecircularorbit tween a putative companion and the primary star were fitinTable6). Wefindnoconvincingevidencefortransit timing variations, but also note that the ∼ 7hr transit determined as a function of angular separation as de- scribed in Horch et al. (2011). Based on these observa- duration makes it difficult for us to capture full transits tions we exclude neighbors with ∆m < 2.56 at 692nm, via ground-based follow-up, and partial transits provide or ∆m < 2.80 at 880nm, down to a limiting separation poorertransittimingmeasurements. Inaddition,wefind no evidence for long term radial velocity trend, with the 16 The WIYN Observatory is a joint facility of the University quadratic and linear fits to the radial velocity data con- of Wisconsin-Madison, Indiana University, the National Optical sistent with flat slopes. AstronomyObservatoryandtheUniversityofMissouri. 8 Zhou et al. TABLE 5 Stellar parametersforHAT-P-67 Parameter Circular Fit Geneva EccentricFitGeneva CircularFitDartmouth EccentricFitDartmouth Catalogue Information Tycho-2 ................ 3084-533-1 GSC ................... 03084-00533 2MASS ................ J17062656+4446371 GAIA .................. 1358614978835493120 GAIARA(J2015) ..... 17:06:26.574 GAIADEC(J2015) .... +44:46:36.794 GAIAµα (masyr−1) .. 9.32±0.88 GAIAµδ (masyr−1) ... 18.5±1.2 GAIAParallaxa (mas) . 2.92±0.23 Spectroscopic propertiesb c Teff⋆ (K)............... 6406+−6651 6408+−6635 6406+−5683 6414+−6599 [Fe/H].................. −0.08±0.05 −0.08±0.05 −0.07+0.04 −0.08±0.05 −0.05 vsinI⋆ (kms−1)........ 35.8±1.1 35.8±1.1 33.2+−11..52 33.9+−11..23 Photometricproperties GALEXFUV(ABmag) 19.759±0.137 GALEXNUV(ABmag) 14.251±0.007 GAIAg (mag).......... 9.94 APASSB (mag)........ 10.682±0.010 APASSg′ (mag)........ 10.351 APASSV (mag)........ 10.069±0.016 APASSr′ (mag)........ 10.010 TASSI (mag).......... 9.518±0.048 2MASSJ (mag)........ 9.145±0.021 2MASSH (mag)....... 8.961±0.019 2MASSKs (mag)...... 8.900±0.019 Derivedpropertiesb M⋆ (M⊙)............... 1.642+−00..105752 1.73+−00..2113 1.43±0.05 1.38+−00..0055 R⋆ (R⊙)............... 2.546+−00..009894 2.71+−00..4389 2.389+−00..004308 2.13+−00..1174 logg⋆ (cgs)............. 3.854+−00..001243 3.800+−00..100860 3.837+−00..000191 3.932+−00..003650 L⋆ (L⊙)................ 8.68+−10..5806 8.3+−41..09 8.62+−00..5570 6.8+−10..29 MV (mag).............. 2.50+−00..1233 2.57+−00..2397 2.403+−00..008633 2.67+−00..1157 MK (mag,ESO)........ 1.26+−00..1354 1.36+−00..2359 1.304+−00..004465 1.56+−00..1148 AV (mag).............. <0.051(1σ) <0.061(1σ) <0.13(1σ) <0.11(1σ) Age(Gyr).............. 1.24+0.27 1.00+0.21 2.83+0.22 3.04+0.31 −0.22 −0.41 −0.19 −0.27 Distance(pc) .......... 320+48 322+35 335+7 297+26 −14 −19 −7 −18 a Acorrectionof−0.325±0.062mashasbeenappliedtotheGAIADR1parallax,asperStassun&Torres(2016). b DerivedfromtheglobalmodellingdescribedinSection3.4,co-constrainedbyspectroscopicstellarparametersandtheGAIAparallax. c ThesestellarparametersareheavilyconstrainedbyGaussainpriorsabouttheirderivedvaluesfromtheKeck-HIRESiodine-freespectrum usingtheStellarParameterClassification(SPC)pipeline(Buchhaveetal.2012). of 0′.′2 (Figure 10). depth, confirming the lack of any significant dilution by background stars. The elimination of blend scenarios 3.8. Blend analysis andthe massupperlimitdeterminedfromHIRESradial Blend scenarios are eliminated by the detection of the velocities validates HAT-P-67b as a planet. We can also planetary Doppler tomographic transit signal. In the placestrictupperlimitsonanythirdlightcontamination cases where an eclipsing binary blended with a fore- from backgroundstars by modelling the line broadening groundstaristhecauseofthetransitsignal,theDoppler profiles from the LSD analysis. A high signal-to-noise tomographicshadowwillbesignificantlydilutedwithre- broadeningprofilewasderivedbyaveragingthe32TRES spect to the photometric transit signal. spectraobtainedforHAT-P-67. Bymodelingthisprofile The flux under the shadow of the planet, as a fraction as two stars, we place an upper limit on the flux ratio of the total flux under the rotational broadening kernel, of any potential companion to be < 0.004, or within 6 describes the area blocked by the planet. This directly magnitudesoftheprimarystar,withthecaveatthatany correspondsto a‘transitlightcurve’overthebroadband potential blended companion exhibits no radial velocity oftheTRESspectrum(followingZhou et al.2016a). We variation. plot this Doppler tomographic light curve in Figure 4 (bottom). We also plot the model transit light curve as 4. DISCUSSION pertheglobalbestfitsolution. Thespectroscopictransit We presented the discovery of HAT-P-67b, a hot- depth is consistent with that of the photometric transit Saturn transiting an F-subgiant. HAT-P-67b has a ra- HAT-P-67b 9 TABLE 6 Orbitalandplanetaryparameters Parameter Circular Fit Geneva EccentricFitGeneva CircularFitDartmouth EccentricFitDartmouth Lightcurveparameters P (days) .............. 4.8101025+0.00000043 4.8101038+0.00000054 4.8101017+0.00000034 4.8101082+0.00000052 −0.00000033 −0.00000037 −0.00000030 −0.00000051 Tc (BJD)a ............ 2455961.38467+−00..0000007664 2455961.38472+−00..0000009802 2455961.38465+−00..0000007645 2455961.3852+−00..00000180 T14 (days)a ........... 0.2912±0.0019 0.308+−00..002391 0.2910±0.0015 0.257+−00..001190 T12=T34 (days)a .... 0.0229±0.0010 0.0246±0.0027 0.02330+−00..0000003505 0.0213+−00..00001154 a/R⋆ .................. 5.691+−00..015274 5.34+−00..6416 5.659+−00..006661 6.34+−00..3402 Rp/R⋆ ................ 0.0834±0.0017 0.084+−00..00001290 0.0821+−00..00001039 0.0846+−00..00001168 b≡acosi/R⋆ ......... 0.12+−00..1028 0.12+−00..1028 0.214+−00..002435 0.20+−00..1112 i(deg) ................ 88.8+1.1 88.9±1.6 88.37+0.61 88.2+1.3 −1.3 −0.57 −1.1 |λ|(deg) .............. 2.9+6.4(<141σ) 2.5+5.8 (<121σ) −1.6+3.9 (<41σ) 2.3+6.6 (<131σ) −4.9 −4.6 −4.6 −6.4 Limb-darkeningcoefficients b ar (linearterm) ....... 0.2497 br (quadraticterm) .... 0.3765 aI ..................... 0.1701 bI ..................... 0.3744 ai ..................... 0.1897 bi ..................... 0.3747 az ..................... 0.1397 bz ..................... 0.3661 RVparameters K (ms−1) ............. <36(1σ) <52(1σ) <38(1σ) <37(1σ) ecosω ................. −0.21+0.15 −0.03+0.20 −0.14 −0.22 esinω ................. 0.027+0.10 −0.150+0.075 −0.11 −0.055 e ...................... 0.24±0.12 0.22+0.12 −0.08 ω ..................... 172+31 105+46 −43 −66 RVjitter(ms−1)c ..... 59 58 59 59 SystemicRV(kms−1)d −1.4±0.5 Planetaryparameters Mp (MJ)e ............. 0.34+−00..2159 0.49+−00..2272 0.33+−00..2127 0.29+−00..2149 Rp (RJ) ............... 2.085+−00..009761 2.25+−00..2203 1.975+−00..004358 1.78+−00..1140 ρp (gcm−3) ........... 0.052+−00..003298 0.058+−00..003295 0.058+−00..003390 0.065+−00..006424 loggp (cgs) ............ 2.32+−00..2344 2.41+−00..2205 3.837+−00..000191 2.36+−00..2487 a(AU) ................ 0.06505+0.00273 0.0663+0.0016 0.062844+0.00053 0.061994+0.00068 −0.00079 −0.0014 −0.00049 −0.00072 Teq (K) ............... 1903±25 1963+−8959 1903+−1291 1803+−6423 Θf..................... 0.0138+0.0099 0.0178+0.0098 0.015+0.025 0.015+0.013 −0.0075 −0.0077 −0.015 −0.010 hFi(109ergs−1cm−2)g 2.74+0.19 2.57+0.45 2.98+0.14 2.41+0.33 −0.17 −0.46 −0.13 −0.25 a Tc: Referenceepochofmidtransitthatminimizesthecorrelationwiththeorbitalperiod. BJDiscalculatedfromUTC.T14: totaltransit duration,timebetweenfirsttolastcontact;T12=T34: ingress/egresstime,timebetweenfirstandsecond,orthirdandfourthcontact. b Valuesfor aquadraticlaw givenseparatelyfor eachofthefilterswith whichphotometricobservationswereobtained. These valueswere adopted from the tabulations by Claret&Bloemen (2011) according to the spectroscopic (SPC) parameters listed in Table 5. The limb darkeningcoefficientsareheldfixedduringtheglobalmodelling. c ThisjitterwasaddedlinearlytotheRVuncertaintiesforeachinstrumentsuchthatχ2/dof=1fortheobservationsfromthatinstrument. d ThesystemicRVforthesystemasmeasuredrelativetothetelluriclines e Themassmeasurementisquotedasthemedianoftheposterior,withtheuncertaintiesdefinedasthe68percentileregion. f TheSafronovnumberisgivenbyΘ= 12(Vesc/Vorb)2=(a/Rp)(Mp/M⋆)(seeHansen&Barman2007). g Incomingfluxperunitsurfacearea,averagedovertheorbit. dius of 2.085+0.096R , and a mass constrained by radial flated gas giants have been discovered around subgiants −0.071 J velocity measurementsto be M <0.59M at 1σ. Con- (KOI-680b Almenara et al. 2015, EPIC 206247743b p J firmationoftheplanetarynatureofHAT-P-67binvolved Van Eylen et al. 2016, KELT-8b Fulton et al. 2015, numerous high precision follow-up transit light curves, KELT-11b Pepper et al. 2016, HAT-P-65b and HAT- radial-velocity constraints on its mass, and two Doppler P-66b Hartman et al. 2016) and giants (e.g. EPIC tomographic transits that eliminated potential blended 211351816b Grunblatt et al. 2016). One hypothesis is eclipsing binary scenarios. that these gas giants are re-inflated by the evolved host The mass, radius, and densities of HAT-P-67b are star (Lopez & Fortney 2016). In this scenario, as the plotted in Figure 11, along with selected parts of host star evolves off the main sequence, ‘warm Jupiters’ the gas giant population. HAT-P-67b is one of the are subjected to higher incident flux and stronger tidal largest, and one of the lowest density planets known heating (assuming a non-zero initial eccentricity). The (ρ = 0.052+0.039gcm−3). A number of other in- heating reaches deep enough into the planetary inte- p −0.028 10 Zhou et al. Fig. 6.— Modelevolutionarytracksofeffectivetemperature–luminosity(left)andeffectivetemperature–stellardensity(right)from theGenevaisochrones(Ekstr¨ometal.2012)areplottedforsolarmetallicitystarsofvariousmassesandrotationrates. Redtracksdenote starsof1.3M⊙,bluefor1.5M⊙,blackfor1.7M⊙. Theshadesofthelinesillustratetheinfluenceofrotationonevolution,withdarkestfor norotation,andlightestforΩ/Ωcritical=0.5,at0.1intervals. The1,2,and3σcontoursfortheposteriorprobabilitydistributionofHAT- P-67areplotted. Notethattheeffectivetemperature–stellardensitydistribution(right)ismodelindependent,witheffectivetemperature measured from spectra, and stellar density derived from the transit duration. The effective temperature – luminosity distribution (left) requiresisochroneinterpolationofluminosity,andisthereforemodeldependent. 1.0 potentially inducing an inflationofthe planetary radius. Figure 12 also plots the incident flux received by Pr.0.5 the hot-Jupiter distribution against their planet masses. 0.0 There is a paucity of low mass planets that receive 0.8 2σ upper limit highincident flux – a sharpenvelope that likely resulted 1σ upper limit from the evaporation of Saturn and Neptune mass plan- 0.6 ets in close-in orbits (e.g. Lecavelier Des Etangs 2007; Ehrenreich & D´esert 2011; Owen & Wu 2013). HAT-P- cc0.4 67b lies on the edge of the envelope – unlike planets e of similar masses that receive high incident irradiation, HAT-P-67bdidnot‘boil-off’,butsurvivedtothepresent 0.2 day. Thehighincidentfluxmayalsohavehaltedcontrac- tion early on, leading to its current radius. Since there 0.0 is a lack of inflated Saturn-massplanets in high incident −180 −90 0 90 1800.0 0.5 1.0 ω Pr. flux environments, HAT-P-67b is an important point in the mass-radius-flux relationship. Fig.7.— TheeccentricityofHAT-P-67bislargelydeterminedby The low density and high irradiation of HAT-P-67b the transit duration and the stellar radius derived from the light also results in a bloated atmosphere, with a large scale curves and GAIA distance. The eccentricity ecc and argument height of ∼500km (assuming an H atmosphere), mak- of periastron ω posteriors are plotted. The 64 and 95 percentile 2 contours areplottedingreyandblack,respectively. ing the planet a good candidate for transmission spec- troscopy follow-up studies. In addition, X-ray and EUV-driven hydrodynamic 1200 escape play an especially important role in low den- 1000 sity, low mass planets (e.g. Lecavelier Des Etangs 2007; 800 Murray-Clayet al. 2009; Ehrenreich & D´esert 2011; 600 C (s) 400 OJuwpeitner&s,JXa-crkasyonan2d01E2U;VOpwheonto&ioWniuzes20th1e3)u.ppFerorathmoot-- O- 200 sphere,causingittoheatupandexpand,resultingines- 0 caping flows. Atmospheric escape has been observed for −200 HD 209458 b (Vidal-Madjar et al. 2003, 2004) and HD −400 189733 b (Lecavelier Des Etangs et al. 2010), where the −600 −50 0 50 100 Lyman-α radii of the planets are ∼10 times larger than Epoch their optical radii, extending beyond the Roche sphere. Sincethe masslossrateis largelyproportionalto the in- Fig.8.— Transitcentroidoffsets(O−C)forthefollow-uplight cident UV and X-ray flux received by the planets (e.g. curves. Wefindnoconvincing evidence fortransitingtimingvari- ations that may be indicative of additional orbiting companions. Murray-Clayet al. 2009), we checked for existing X-ray and UV measurements of HAT-P-67. While no EUV or X-ray flux measurements exist, HAT-P-67 is identified rior to inflate the planet radius. Hartman et al. (2016)) as a source by GALEX, with flux measurements in the foundempiricalevidencethatthelevelofplanetinflation FUV (1344–1786˚A) and NUV (1771–2831˚A) bands. In is correlatedwiththe fractionalageofthe hoststar,fur- Figure 13, we compile all transiting planet systems with ther supporting the idea of re-inflation. Figure 12 shows GALEX FUV and NUV measurements (Bianchi et al. theevolutionintheincidentfluxreceivedbyHAT-P-67b 2011), as well as GAIA parallaxes and updated system over its lifetime. Currently HAT-P-67b receives ∼ 2× parameters from Stassun et al. (2016). To examine the theincident-fluxofazero-age-main-sequenceHAT-P-67, potential mass loss rate of HAT-P-67b in the context

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.