A PSF-based Approach to TESS High quality data Of Stellar clusters (PATHOS) -- III. Exploring the properties of young associations through their variables, dippers, and candidate exoplanets
aa r X i v : . [ a s t r o - ph . S R ] S e p MNRAS , 1–19 (2020) Preprint 8 September 2020 Compiled using MNRAS L A TEX style file v3.0
A PSF-based Approach to TESS High quality data OfStellar clusters (PATHOS) - III. Exploring the propertiesof young associations through their variables, dippers , andcandidate exoplanets.
D. Nardiello , ⋆ Aix Marseille Univ, CNRS, CNES, LAM, Marseille, France Istituto Nazionale di Astrofisica - Osservatorio Astronomico di Padova, Vicolo dell’Osservatorio 5, IT-35122, Padova, Italy
Accepted 2020 September 2. Received 2020 September 2; in original form 2020 July 31
ABSTRACT
Young associations in star forming regions are stellar systems that allow us to under-stand the mechanisms that characterise the stars in their early life and what happensaround them. In particular, the analysis of the disks and of the exoplanets aroundyoung stars allows us to know the key processes that prevail in their evolution andunderstand the properties of the exoplanets orbiting older stars. The
TESS mission isgiving us the opportunity to extract and analyse the light curves of association mem-bers with high accuracy, but the crowding that affects these regions makes difficultthe light curve extraction. In the PATHOS project, cutting-edge tools are used to ex-tract high-precision light curves and identify variable stars and transiting exoplanetsin open clusters and associations. In this work, I analysed the light curves of stars infive young ( . Myr) associations, searching for variables and candidate exoplanets.By using the rotational periods of the association members, I constrained the agesof the five stellar systems ( ∼ - Myr). I searched for dippers and I investigatedthe properties of the dust that forms the circumstellar disks. Finally, I searched fortransiting signals, finding 6 strong candidate exoplanets. No candidates with radius R P . . R J have been detected, in agreement with the expectations. The frequency ofgiant planets resulted to be ∼ . %);the low statistic makes this conclusion not strong, and new investigations on youngobjects are mandatory to confirm this result. Key words: techniques: image processing – techniques: photometric – Galaxy: openclusters and associations: general – stars: variables: general – planets and satellites:general
To date, more than 4000 exoplanets have been discov-ered and characterised , but their properties have notalways been those we observe today. Indeed, the ex-oplanets we observe were born with different proper-ties: in their early life, planets are subject to a se-ries of interactions with other bodies or the host star,that cause changing in their orbital and physical param-eters (migration, planetary impacts, atmospheric photo-evaporation, etc.). All these processes have been stud- ⋆ E-mail: [email protected] https://exoplanetarchive.ipac.caltech.edu/ ied in details (see, e.g., Terquem & Papaloizou 2007;Ida & Lin 2010; Hansen & Murray 2012; Lopez & Fortney2013; Owen & Wu 2013; Schlichting et al. 2015; Schlichting2018 ) and partially explain some observables, like, e.g., thegap in the radius distribution of small planets at . - . R ⊕ (Fulton et al. 2017; Fulton & Petigura 2018), the dearth ofshort-period giant planets in close-in exoplanet distribu-tion (see, e.g., Owen & Lai 2018 and references therein),and the accretion of the gaseous envelopes for giant planets(Baraffe et al. 2003; Marley et al. 2007; Spiegel & Burrows2012; Mordasini et al. 2017).In order to understand all the mechanisms that pre-vail in the life of an exoplanet, it is mandatory to searchfor and monitor stars having different ages. Unfortunately, © D. Nardiello
Table 1.
Association information
Association Name α δ r µ α cos δ µ δ π [Fe/H] ( ) N (deg.) (deg.) (deg.) (mas yr − ) (mas yr − ) (mas)ChI 166.70 − . − . ± . + . ± . . ± . − . ± . − . − . ± . − . ± . . ± . − . ± . − . − . ± . − . ± . . ± . − . ± . − . − . ± . + . ± . . ± . − . ± . − . + . ± . − . ± . . ± . − . ± . ( ) Metallicities from James et al. (2006), Spina et al. (2014a,b) stellar age is one of the most difficult parameter to mea-sure, unless the star is member of an association or of astar cluster (open or globular): in the latter cases, the ageof the star can be well constrained thanks to the use oftheoretical models. For this reason, the interest on these ob-jects has grown in recent years and many photometric andspectroscopic works have been carried out on their mem-bers until now (e.g., Quinn et al. 2012; Meibom et al. 2013;Quinn et al. 2014; David et al. 2016a; Mann et al. 2016b;Malavolta et al. 2016; Pope et al. 2016; Mann et al. 2018;Ciardi et al. 2018; Vanderburg et al. 2018; Benatti et al.2019; Newton et al. 2019; Gaidos et al. 2020).The
Kepler (Borucki et al. 2010) and K2 (Howell et al.2014) missions were a success, allowing the detection ofmany exoplanets, also around stellar cluster and associ-ation members (Meibom et al. 2013; Barros et al. 2016;Obermeier et al. 2016; Mann et al. 2016a; Nardiello et al.2016b; Libralato et al. 2016b; Pepper et al. 2017;Curtis et al. 2018; David et al. 2019a,b), but their skycoverage was limited. The Transiting Exoplanet SurveySatellite ( TESS , Ricker et al. 2015) mission is giving usthe opportunity to study stellar cluster and associationmembers with high photometric accuracy and unprece-dented sky and temporal coverage: the satellite has probedmore than 80 % of the sky in its first two years of mission,observing a large fraction of stellar clusters and associationsof the Galaxy for ∼ days or more, and on July 2020has started its extended mission. Given the low resolutionof the TESS images and the high-levels of star crowdingtypical of clusters/associations, the extraction of highprecision light curves from
TESS data needs appropriatetechniques, like the use of the difference imaging analysis(Bouma et al. 2019) or point spread function (PSF) models(Nardiello et al. 2019).The project ’A PSF-based Approach to
TESS
High Quality data Of Stellar clustersˆa ˘A´Z (PATHOS;Nardiello et al. 2019, hereafter Paper I) is aimed at find-ing and characterisation of candidate exoplanets and vari-able stars in stellar clusters and associations, by using high-precision light curves obtained with a cutting-edge toolbased on the use of empirical PSFs and neighbour subtrac-tion. This technique allows us to minimise the dilution effectsdue to neighbour contaminants, and extract high precisionphotometry even for faint stars ( T ∼ - ). The efficiency ofthe method was demonstrated in Paper I: high precision lightcurves of stars located in an extreme crowded region cen-tred on the globular cluster 47 Tuc, containing also Galacticand Small Magellanic Cloud sources, were analysed. Manyvariables and one candidate hot-Jupiter were identified. Us-ing the same technique, Nardiello et al. (2020, hereafter Pa- per II) searched for exoplanets among the light curves of ∼
163 000 stellar members of 645 open clusters observed dur-ing the first year of
TESS mission, finding 11 strong candi-dates in eight open clusters with ages between ∼ Myr and ∼ Gyr.In this third work of the series, I analysed the proper-ties of the members of five young ( . Myr) associations inas many star forming regions by using the light curves ex-tracted from the images collected during the first year of the
TESS mission. The analysed associations are: Chamaeleon I,Chamaeleon II, Lupus, γ Velorum, and Corona Australisassociations. Given their young ages ( . Myr), theseassociations host a large number of T-Tauri pre-main se-quence stars, and for this reason they are also known as
T-associations , term coined by Ambartsumian (1949) inhis study on the importance of stellar associations for theunderstanding of the stellar formation and evolution. To-day, the study of the properties of the young associationmembers allows us not only to investigate the life of thestars, but also how circumstellar disks and exoplanets areborn and evolved around them. Therefore, the analysisof the
TESS light curves of young stellar objects in starforming regions offers the unique opportunity to trace theorigin and early evolution of circumstellar disks and ex-oplanets orbiting them. In the last years, a large num-ber of studies concentrate their attention on young asso-ciations aimed to explore the metal content of their stars(e.g., James et al. 2006; Gonz´alez Hern´andez et al. 2008;Santos et al. 2008; D’Orazi et al. 2009, 2011; Biazzo et al.2011, 2012a,b; Spina et al. 2014a,b; Jeffries et al. 2017),the disks that surround their young members (e.g.,Chen et al. 2005; Carpenter et al. 2006, 2009; Chen et al.2011; Luhman & Mamajek 2012; Ansdell et al. 2016;Bodman et al. 2017; Kuruwita et al. 2018; Bohn et al. 2019;Aizawa et al. 2020; Bredall et al. 2020), and the variabil-ity of the main sequence stars (e.g., Rebull et al. 2018;Curtis et al. 2019; Rebull et al. 2020) in order to constrainttheir ages. Even if, in the last years, associations and stel-lar clusters have been the subjects of many exoplanet sur-veys, just a handful (candidate) exoplanets are known toorbit members of young ( . -150 Myr) clusters and as-sociations, and just one exoplanet orbits a (pre-)main se-quence star in a . Myr old association, K2-33b (Up-per Scorpius association, David et al. 2016b; Mann et al.2016a). Other known exoplanets and candidates in youngsystems are: the hot Jupiter HIP 67522 (Sco-Cen associa-tion, ∼ Myr, Rizzuto et al. 2020), the planetary systemaround the star V1298 Tau (Tau-Aur association, ∼ Myr,David et al. 2019b), the two candidate exoplanets PATHOS-30 and PATHOS-31 (IC 2602, ∼ Myr, Paper II), the
MNRAS , 1–19 (2020)
ATHOS III: Young Associations Neptune-size exoplanet DS Tuc Ab (Tuc-Hor association ∼ Myr, Newton et al. 2019; Benatti et al. 2019), and thesub-Neptune EPIC 247267267b (Cas-Tau group, ∼ Myr,David et al. 2018).In the present work, the PATHOS pipeline is used toextract and correct the
TESS light curves of a sample oflikely young association members (Sect. 2). Variable starshave been identified in order to constraint the associationages and analyse the dust in the circumstellar disks aroundyoung stellar objects (Sect. 3). The discovery and charac-terisation of new candidate exoplanets orbiting stars in theaforementioned associations and their frequency is reportedin Sect. 4. Section 5 is a summary and a discussion of theresults obtained in this work.
In this work, I extracted the light curves of the stars infive very young associations observed during the first yearof the
TESS mission. In particular, I used Full Frame Im-ages (FFIs) collected during Sectors 7, 8, 9, 11, 12, 13. Iproduced a total of 7150 light curves associated to 4459stars. The pipeline adopted for the light curve extractionand correction is widely described in Paper I and Paper II.The pipeline includes the use of the light curve extrac-tor
IMG2LC , developed by Nardiello et al. (2015a, 2016a)for ground-based images and improved by Libralato et al.(2016a,b) and Nardiello et al. (2016b) for
Kepler/K2 space-based data. Briefly, the routine uses empirical Point SpreadFunctions (PSFs) and an input catalogue (see Section 2.1) toextract aperture and PSF-fitting photometries of each starin the catalogue after the subtraction of all the neighboursfrom each
TESS
FFI. The raw light curves are then cor-rected for systematic effects by fitting and applying to themthe Cotrending Basis Vectors described in Paper II. Lightcurves will be released in ascii and fits format on theMikulski Archive for Space Telescopes (MAST) as a HighLevel Science Product (HLSP) under the project PATHOS (DOI: 10.17909/t9-es7m-vw14). A description of the formatof the light curves is reported in Paper I and Paper II and inthe MAST webpage of the PATHOS project. In this work, I analysed stars that have high probability tobe members of five associations: Chamaeleon I and II asso-ciations (hereafter, ChI and ChII), Lupus association (Lup),Corona Australis association (CrA), and γ Velorum associ-ation (Vel). The selection of likely association members wasperformed by using Gaia DR2 (Gaia Collaboration et al.2018) information, like proper motions and parallaxes. Foreach association, I extracted from the Gaia DR2 catalogueall the stars with G < , within circular regions (of radius r ) of the sky centred in the ( α , δ ); the values of r , α , and δ are tabulated in Table 1. For each region, I first analysedthe proper motion distributions of the stars with G < ,and I selected manually the area of the vector-point dia-gram where likely association members are located. I fitted https://archive.stsci.edu/hlsp/pathos the µ α cos δ and µ δ distributions with Gaussian functionsand I selected all the points within . σ from the meanvalues of µ α cos δ and µ δ . I fitted the parallax ( π ) distri-bution of the selected stars with a Gaussian, and I selectedall the points within . σ from the mean value of π . I it-erated the procedure 10 times, alternating the proper mo-tion and parallax selections and using only the stars thatpassed the selection criteria of the previous iteration. Anexample of likely association member selection for the CrAis illustrated in Fig. 1. The Vel and Lup associations aremore complex systems and are formed by groups of starshaving slightly different kinematical properties; in particu-lar, in the vector-point diagram association stars form dif-ferent close clumps. When possible, I fitted all the singleclumps with Gaussian functions and I selected the stars aspreviously described. The final catalogue given as input of IMG2LC contains 6629 stars. I cross-matched the final cat-alogue with the TIC v8 catalogue (Stassun et al. 2019), inorder to obtain photometric information on all the stars. Inparticular, I included in the catalogue (in addition to
TESS and Gaia magnitudes) the magnitudes in B - and V -Johnsonbands, the 2MASS J , H , and K s magnitudes (Cutri et al.2003), and the infrared WISE (Wright et al. 2010) magni-tudes W ( [ . µ m ] ), W ( [ . µ m ] ), W ( [ µ m ] ), and W ( [ µ m ] ). Reddening values ( E ( B − V ) ) were extracted foreach star in the input catalogue by using the python rou-tine mwdust implemented by Bovy et al. (2016), and the Combined19 dustmap (Drimmel et al. 2003; Marshall et al.2006; Green et al. 2019). Figure 2 shows an example G versus ( G − G RP ) colour-magnitude diagram (CMD) before(panel (a)) and after (panel (b)) the correction for reddeningfor the two close associations ChI and ChII. Following the same methodology as in Paper I and Paper II, Icalculated the following quality parameters for the cotrendedlight curves: (i) the photometric
RMS , defined as the 68.27thpercentile of the sorted residual from the 3.5 σ -clipped me-dian value of the light curve; because the simple photometric RMS is very sensitive to stellar variations, I calculated the (ii)
P2P RMS , defined as the 68.27th percentile of the sorted resid-ual from the median value of the vector δ F i = F i − F i + , where F is the flux value at a given epoch i . I fitted the RMS and
P2PRMS distributions with different polynomial functions, chang-ing the order between n = and n = , to derive the bestmean trend of each photometric method. I found that, on av-erage, the best fit was the one with n = . Figure 3 shows the RMS (top panel) and
P2P RMS (bottom panel) distributions;coloured lines are the 2nd-order polynomial fits performedfor each photometric method. As done in Paper II, I used the
P2P RMS trends to define the best photometric method foreach light curve: for not saturated stars with T . . , I usedstars extracted with the 4-pixel aperture photometry; in the . . T . . regime, 3-pixel aperture photometry gives thebest results; for stars having . . T . . the 2-pixel aper-ture photometry produces, on average, light curves with thelower P2P RMS ; PSF-fitting photometry works better thanthe aperture photometry in the range . . T . . ; in https://github.com/jobovy/mwdust MNRAS000
P2P RMS trends to define the best photometric method foreach light curve: for not saturated stars with T . . , I usedstars extracted with the 4-pixel aperture photometry; in the . . T . . regime, 3-pixel aperture photometry gives thebest results; for stars having . . T . . the 2-pixel aper-ture photometry produces, on average, light curves with thelower P2P RMS ; PSF-fitting photometry works better thanthe aperture photometry in the range . . T . . ; in https://github.com/jobovy/mwdust MNRAS000 , 1–19 (2020)
D. Nardiello
Figure 1.
Overview on the selection procedure of likely CrA association members. Panel (a) shows the vector-points diagram of propermotions for the stars in the circular region ( r = . deg.) of the sky centred on ( α , δ ) = ( . , − . ) ; panel (b) is the parallaxdistribution of the same stars as a function of the G magnitude; panel (c) and (d) show the G versus ( G − G RP ) CMD and the ( α, δ ) positions for the stars in the considered region. The red and the grey points represent the likely association members and the discardedstars in the selection procedure, respectively. In all the panel, for clarity, only 20% of the discarded stars are plotted. the faint regime, T & . , the best choice is the 1-pixelaperture photometry.After this first selection, to exclude stars contaminatedby different kind of sources (bleeding columns, bad pix-els, not-subtracted stars, blended stars), I excluded all thesources for which the mean instrumental magnitude T instr istoo different from that expected knowing the calibrated T cal .In order to select the best stars, I calculated the mean of the δ T = T instr − T cal distribution, ¯ δ T , and its standard deviation σ δ T and I excluded all the stars for which | δ T − ¯ δ T | > × σ δ T ;4088 stars passed all the selection criteria and have beenanalysed. The analysis of the stellar variability of the association mem-bers is crucial to constraint some properties of the associa-tions. In order to find periodic variable stars, I used the Gen-eralized Lomb-Scargle (GLS, Zechmeister & K¨urster 2009)routine implemented in
VARTOOLS 1.38 (Hartman & Bakos2016) to extract the periodograms of the light curves. Af-ter the identification of the period associated to the mostpowerful peak in the periodogram, the routine whitened thelight curve and extracted the periodogram of the light curveagain to find the second strongest peak period. The rea-son for this multi-period finding are: (i) some stars presentmultiple signals associated to different physical phenomena(see, e.g., Rebull et al. 2016), and the multi-period findingallows us to identify the different periods; (ii) artifacts inthe light curve or effects due to the observations (sampling,temporal gaps in the light curve, outliers, etc.) might gener-ate a peak in the periodogram stronger than that associatedto the real physical signal coming from the star; multiple- MNRAS , 1–19 (2020)
ATHOS III: Young Associations Figure 2.
Reddening correction for the associations ChI and ChII based on the dust-map released by Bovy et al. (2016). Panels (a) and(b) show the G versus ( G − G RP ) CMD before (panel (a)) and after (panel (b)) the correction. Panel (c) shows the reddening map forthe two associations.
Figure 3.
Photometric
RMS and
P2P RMS distributions as a function of the
TESS magnitude for the light curves extracted in this work.Coloured lines represent the 2nd-order polynomial interpolation to the
RMS distributions: black for PSF-fitting photometry, magenta,blue, green, and yellow for 1-px, 2-px, 3-px, and 4-px aperture photometries, respectively. Light blue dashed line represents the theoreticallimit. Grey dots and light green crosses are examples of
RMS distributions for PSF-fitting and 3-px aperture photometries, respectively.Red points are the saturated stars.MNRAS , 1–19 (2020)
D. Nardiello
Figure 4.
Selection of candidate variable stars. Panels (a) and (b) show the selection of candidate variables in the SNR versus Periodplane, for the first and the second peak period, respectively: orange points are the candidate variables selected on the basis of theirSNR; blue starred point represent the star TIC 69420071 whose light curve is shown in panel (c). In red is the sinusoidal model obtainedcombining the two peak periods found by GLS. Panel (d) shows the light curve of TIC 69420071 phased adopting the first peak period( ∼ . d); panel (e) shows the (whitened) light curve phased using the second peak period ( ∼ . d). period finding allows us to recover the real signal. For eachlight curve, I searched for periods between 0.08 d ≤ P ≤ T LC ,where T LC is the maximum temporal baseline of the lightcurve. I excluded the candidate variable stars blended withother stars in the catalogue having similar signals usingthe routine findblends implemented in VARTOOLS 1.38 : thisroutine compares the positions of the stars in a catalogue,their periods found by using GLS periodograms and the am-plitudes of their light curves to find blended stars. I usedthe Signal-to-Noise Ratio (SNR) parameter to isolate thecandidate variable stars following the method described inNardiello et al. (2015a) and shown in Fig. 4: I divided theSNR distribution in intervals of δ P = d, and I computed the3.5 σ -clipped mean and standard deviation of the SNR valuesinside each bin. I interpolated the points σ above the meanSNR values with a spline, and I considered as candidate vari-ables the points above the interpolated line (orange pointsin panels (a) and (b) of Fig. 4). I applied this procedureboth to the SNR distributions associated to the first peak ofthe periodograms and to the SNR associated to the secondpeak periods; I considered as candidate variables the starsselected in both the sample (2230 stars). Finally, I visuallyinspected the phased light curves to assign to each candi-date variable star the corrected period (or both the periodsif the star have multiple periods), or to discard it because false positive. Panels (c), (d), and (e) show an example oflight curve of a star characterised by multiple periods. Thefinal list of periodic variable stars contains 1260 stars, 28 ofthem have multiple periods. A list of periodic variable starsused in this work is available electronically. The descriptionof the columns are reported in Table A1. Because of their young age and of the low number of mem-bers, the estimation of the association ages based on theuse of theoretical models is not immediate. By using gy-rochronology, i.e. the method for the estimation of the agebased on the analysis of stellar rotation and magnetic brak-ing (Barnes 2003, 2007), it is possible to constraint the ageof the associations studied in this work.In this work, I combined the period-colour distributionanalysis and the CMD isochrone fitting, in order to con-straint the age of each association and use this parameter inthe characterisation of the candidate transiting exoplanets(Sect. 4). In a first step, I found a raw estimation of theages of the five associations comparing the period- ( V − K s ) dereddened colour distributions of the associations studiedin this work with that obtained by Rebull et al. (2018) forthe ρ Oph ( ∼ Myr) and Upper Sco ( ∼ Myr) associ-
MNRAS , 1–19 (2020)
ATHOS III: Young Associations Figure 5.
Age computation of the associations studied in this work based on the analysis of Period- ( V − K ) distributions and onthe fit of the isochrones. Panels (a) show the Period- ( V − K ) for one of the associations studied in this work (in black) compared tothe distributions obtained for other associations in other works ( ρ Oph, Taurus, and Upper Sco in red, green, and violet, respectively,Rebull et al. 2018, 2020). On the top of the panels (a) are reported the spectral classes as defined by Pecaut & Mamajek (2013, tableupdated to March 2019). Panels (b) show the best fit isochrones to the G versus ( G − G RP ) CMD of the single associations and theparameters used to obtain the fit. In this figure are shown the results for ChI, ChII (top panels), and Lup associations (bottom panels,empty black circles indicate likely field stars). ations, and with the period-colour analysis performed byRebull et al. (2020) for the Taurus association ( ∼ Myr).I used this first guess on the age to perform a fit of theisochrones and extract the age of each association. The asso-ciations studied in this work have a slightly subsolar metal-licity ([Fe/H] ∼ − . - − . , see, e.g., James et al. 2006;Spina et al. 2014a,b; Table 1). For the isochrones fitting Iused two sets of metallicities: for ChI, ChII and Lup asso-ciations I used isochrones with [Fe/H]= − . , while for Veland CrA isochrones with [Fe/H]= − . .(i) ChI & ChII associations.
Because of the low num-ber of members and because the two associations are al-most coeval and at the same distance ( ∼ pc), I anal- ysed their period-colour distributions together. Panel (a )of Fig. 5 shows the P vs ( V − K s ) distributions for ChI &ChII (black points) compared to that of ρ Oph (red points)and Taurus (green points) associations: in the four associ-ations low-mass slow-rotator ( P ∼ – d) stars prevail. Itmeans that all the associations are almost coeval. The asso-ciations ρ Oph and Taurus are very young with ages between ∼ and ∼ Myr (Rebull et al. 2018, 2020). I used this in-formation to constraint the fit of the isochrones shown inpanel (b ). I computed the median reddening E ( B − V ) and MNRAS , 1–19 (2020)
D. Nardiello
Figure 6.
As in Fig. 5, but for Vel (top panels) and CrA (bottom panels) associations. the median distance modulus ( m − M ) of the stars that be-long to ChI and ChII associations, and I performed a χ -fitof a set of PARSEC (PAdova and TRieste Stellar EvolutionCode, Girardi et al. 2002; Bressan et al. 2012; Marigo et al.2017) isochrones with ages that run from 1 to 5 Myr, instep of 0.5 Myr, to the G versus ( G − G RP ) CMD, as done byNardiello et al. (2015b, I refer the reader to this work for adetailed description of the fit procedure). I found that thebest fit is associated to an age of . ± . Myr.(ii)
Lup association.
Even if I performed strict selectionson proper motions and parallaxes for the groups that formthe complex Lup association, some field stars are still presentin the catalogue. In the analysis of variable stars, I excludedthese likely field stars on the basis of their colours and mag- By using the Gaia DR2 parallaxes corrected for the − µ masoffset found by Lindegren et al. (2018) http://stev.oapd.inaf.it/cgi-bin/cmd nitudes. Panel (a ) of Fig. 5 shows the period-colour dis-tribution of variable stars in the Lup association comparedto that derived by Rebull et al. (2018) for Upper Sco stars(empty black circles are the likely field stars). The two distri-butions are very similar, with a scattered sequence of AFGKstars which become slower as the mass decreases, and awell populated sequence of M stars, whose periods decreasefrom early to late spectral types. The age of Upper Sco is ∼ - Myr (Pecaut et al. 2012; Rebull et al. 2018): startingfrom this constraint, I performed a fit of isochrones (withages between 5 and 15 Myr) to the Lup association CMD(panel (b ) of Fig. 5). The best fit is obtained for an age of . ± . Myr.(iii)
Vel association.
Because the Vel association is furtheraway than the other associations studied in this work, lowmass stars have luminosities over the
TESS magnitude limitand the result is that the period-colour distribution is cuton the red part, as shown in panel (a ) of Fig. 6. Even if thesequence of M dwarfs is incomplete, the sequence formed MNRAS , 1–19 (2020)
ATHOS III: Young Associations by AFGK stars (which periods increases with the colour)and part of the sequence of M-type stars are very similarto that of the Upper Sco association. As done for the Lupassociation, I used a set of isochrones with ages between 5and 15 Myr to search the age that gives the best-fit. I foundan age of . ± . Myr, as shown in panel (b ) of Fig. 6(iv) CrA association.
The period-colour distribution ofvariable stars in the CrA association is shown in panel (a )of Fig. 6, compared to the distributions of Taurus and UpperSco stars. The distribution of the periods of the M stars inthe CrA association follows that of the M stars in the UpperSco, with late type stars that are faster rotators than early Mdwarfs. Unfortunately few stars with spectral types earlierthan M populate the period-colour distribution, and a directcomparison of this part of the distribution is not possible.Using these information, I performed a fit of the CMD usingisochrones with ages between 1 and 15 Myr; I found the bestfit for an age of . ± . Myr (panel (b ) of Fig. 6). Young stellar systems, like the associations analysed in thiswork, host low mass ( . M ⊙ ) T Tauri-like “dipper” starssurrounded by circumstellar disks. Dipper stars are youngstellar objects (YSOs) that show dimming events (periodicor not) in their light curves, probably caused by the dust lo-cated in the inner regions of a circumstellar disk that “tran-sits” on the stellar disk (Bodman et al. 2017). The luminos-ity of these stars usually decreases between few percent to > magnitude, on timescales between few-hours and about1 day.Characterise dipper stars and their disks in young associa-tions with different ages is essential to understand how theyevolve and which are the cleaning timescales of disks, allow-ing us to constraint the models on the planet formation.To date, few ground-based surveys have been per-formed to study these objects (see, e.g., Cody & Hillenbrand2010; Morales-Calder´on et al. 2011); in the last years datafrom telescopes in space ( K2/Kepler and
CoRot ), gave agreat contribution to the analysis of dipper stars (see, e.g.,Stauffer et al. 2015; Ansdell et al. 2016; Rodriguez et al.2017; Cody & Hillenbrand 2018), but these missions hadvery limited sky coverage. Recently, Bredall et al. (2020)characterised 11 stars in the Lupus region, combiningground-based (ASAS-SN) and space-based (
TESS ) data. Infact,
TESS is offering a unique opportunity to study with anhigh photometric accuracy the evolution of the light comingfrom these stars, over a long time baseline ( & month).In this section I describe the procedure I followed tosearch and characterise the dippers among the associationmembers for which I extracted the TESS light curves.In order to search for dipper stars, I used three differentmetrics: (i) the
RMS , sensitive to the scatter of the light curve;(ii) the peak-to-peak variability metric ( ν ), as defined bySokolovsky et al. (2017), that is sensitive to the variabilityof the star in general; (iii) the Flux Asymmetry ( M ), definedby Cody et al. (2014) and Cody & Hillenbrand (2018), sen-sitive to fading/brightening events in the light curve. First, Idivided the RMS distribution in bin of 1.0 T -magnitude, and,within each interval, I computed the mean ¯ RMS and the stan-dard deviation of the σ ¯ RMS ; I interpolated the ¯ RMS + × σ ¯ RMS points with a cubic spline and I selected all the sources above the interpolation. I performed the same procedure using asparameter ν , and I discarded all the points that were notselected in RMS and ν selections and having M < − . . I vi-sually checked the light curves of the 652 stars that passedthe selection, identifying 71 candidate dippers ( > % as-sociated to stars of spectral type K and M).Following the procedure adopted by Bredall et al.(2020), I used the All-Sky Automated Survey for Su-perNovae (ASAS-SN, Shappee et al. 2014; Kochanek et al.2017) g -band light curves to calculate the ratio between theextinction coefficient A T in T -band and that in g -sloan band, A g . This quantity is strictly linked to the grain size of thedust that surrounds the star. Defining δ T and δ g the dim-ming of the TESS and ASAS-SN light curve, the quantity ∆ ( δ g − δ T ) represents the reddening E ( g − T ) = A g − A T causedby the dust. Therefore: ∆ ( δ g − δ T ) δ g = A g − A T A g = − A T A g (1)and the quantity A T / A g can be inferred measuring the slopeof the ∆ ( δ g − δ T ) - δ g relation. I downloaded from the ASAS-SN archive the g -band light curves for all the dippers having T < . (53 stars), with a baseline that covers the TESS ob-servational period of the first year of mission. Panels (a) ofFig. 7 show two examples of light curves of dippers observedfor two consecutive
TESS sectors by
TESS (grey points)and ASAS-SN (green points). I extracted the relationshipbetween ∆ ( δ g − δ T ) and δ g splitting the light curves in sub-sectors, each one ending with the TESS down-link of thedata (about every . days): in this way I avoid (2nd-order)systematic effects due to the variation of the photometriczero-point between the first and second part of a sector. Pan-els (b) shows ∆ ( δ g − δ T ) as a function of δ g for the two starsshowed in panels (a): I performed a linear least-squares fit tothe data of each sub-sector to obtain the slope ( − A T / A g ) i ,with i = , ... N ssec is i -th sub-sector, and, finally, I averagedall the slopes. The fits obtained with the mean slope areshown in panels (b) of Fig. 7 (red lines). The catalogue ofthe identified dippers and of the A T / A g values is released aselectronic material; Table A2 reports the description of thiscatalogue.The ratio A T / A g gives information about the size of thegrains that form the surrounding disk: if the dust is domi-nated by small grains, the quantity A T / A g will be smallerthan the case in which the grains have large size; if the sizeof the grains are larger than the wavelengths in which the TESS observations were performed ( λ central ∼ nm), theratio A T / A g → , and the reddening E ( g − T ) = A g − A T → .Panel (c ) of Fig. 7 shows the A T / A g as a function ofthe de-reddened colour ( V − K s ) : the two quantities areslightly correlated (Pearson coefficient ∼ − . ), with, onaverage, earlier type stars having disks formed by largergrains. Bredall et al. (2020) found a weak relation betweenthe grain sizes and the infrared excess measured with thecolour ( K s − [ µ m ]) , with the infrared excess that inverselydecreases with the dimension of the grains. Panel (c ) ofFig. 7 illustrates the distribution of the A T / A g measured inthis work as a function of the infrared excess ( K s − [ µ m ]) :it shows that there is not a clear correlation between the https://asas-sn.osu.edu/ MNRAS000
TESS (grey points)and ASAS-SN (green points). I extracted the relationshipbetween ∆ ( δ g − δ T ) and δ g splitting the light curves in sub-sectors, each one ending with the TESS down-link of thedata (about every . days): in this way I avoid (2nd-order)systematic effects due to the variation of the photometriczero-point between the first and second part of a sector. Pan-els (b) shows ∆ ( δ g − δ T ) as a function of δ g for the two starsshowed in panels (a): I performed a linear least-squares fit tothe data of each sub-sector to obtain the slope ( − A T / A g ) i ,with i = , ... N ssec is i -th sub-sector, and, finally, I averagedall the slopes. The fits obtained with the mean slope areshown in panels (b) of Fig. 7 (red lines). The catalogue ofthe identified dippers and of the A T / A g values is released aselectronic material; Table A2 reports the description of thiscatalogue.The ratio A T / A g gives information about the size of thegrains that form the surrounding disk: if the dust is domi-nated by small grains, the quantity A T / A g will be smallerthan the case in which the grains have large size; if the sizeof the grains are larger than the wavelengths in which the TESS observations were performed ( λ central ∼ nm), theratio A T / A g → , and the reddening E ( g − T ) = A g − A T → .Panel (c ) of Fig. 7 shows the A T / A g as a function ofthe de-reddened colour ( V − K s ) : the two quantities areslightly correlated (Pearson coefficient ∼ − . ), with, onaverage, earlier type stars having disks formed by largergrains. Bredall et al. (2020) found a weak relation betweenthe grain sizes and the infrared excess measured with thecolour ( K s − [ µ m ]) , with the infrared excess that inverselydecreases with the dimension of the grains. Panel (c ) ofFig. 7 illustrates the distribution of the A T / A g measured inthis work as a function of the infrared excess ( K s − [ µ m ]) :it shows that there is not a clear correlation between the https://asas-sn.osu.edu/ MNRAS000 , 1–19 (2020) D. Nardiello
Figure 7.
Analysis of the dippers. Panels (a) show two example of light curves of dipper stars obtained using
TESS data (grey points)and ASAS-SN g -band data (green points); the magnitude limits on the y-axis cover the same interval for T and g magnitudes, in orderto directly compare light curves obtained with different instruments. Panels (b) show the ∆ ( δ g − δ T ) versus δ g diagram: red lines are thebest least squares fit. The slope of the fitted line is correlated to the reddening caused by the circumstellar disk and is related to the size ofthe grains that form the dust. Panel (c ) is the A T / A g versus the dereddened colour index ( V − K s ) diagram: on the top are reported thespectral classes as defined by Pecaut & Mamajek (2013, table updated to March 2019). Panel (c ) shows A T / A g versus the infrared excessindicator ( K s − [ µ m ]) . Panel (d) is the ( K s − [ µ m ]) versus ( K s − [ µ m ]) colour-colour diagram as reported by Luhman & Mamajek(2012): the different evolutionary stages of the disk are reported on the right. Grey points are all the stars in the input catalogue adoptedin this work, black filled circles are the dipper stars identified in this work, magenta empty squares, azure asterisks, and orange stars arethe dippers identified by Ansdell et al. (2016), Bodman et al. (2017), and Bredall et al. (2020), respectively. two quantities (Pearson coefficient: ∼ . ), and it does notconfirm what found by Bredall et al. (2020).The presence of a disk around the dippers found in thiswork is also confirmed by the analysis of the excess emis-sion in infrared shown in panel (d) of Fig. 7. In fact, theevolutionary stage of a disk can be inferred comparing thestellar luminosity in the 2MASS K s band with its WISEinfrared magnitudes. As reported by Luhman & Mamajek(2012), in a ( K s − [ µ m ]) versus ( K s − [ µ m ]) colour-colour diagram, stars with ( K s − [ µ m ]) & . are surrounded by full / transitional disks , while evolved and debris disks arelocated in the area defined by ( K s − [ µ m ]) & . and ( K s − [ µ m ]) & . . Stars with small values of the colour-colour indexes ( ∼ ) have no disk. Panel (d) of Fig. 7 con- I refer the reader to Luhman & Mamajek (2012) for a detaileddescription of the different evolutionary stages of the disksMNRAS , 1–19 (2020)
ATHOS III: Young Associations firms that the large part of YSOs-dippers found in thiswork have full or transitional disks; about 5 objects have evolved or debris disks. For completeness, I also report theresults by Ansdell et al. (2016), Bodman et al. (2017), andBredall et al. (2020). I found that about half of dippersfound in this work are located in the ChI and ChII asso-ciations (23 and 14, respectively) and the other half in theLup, CrA, and Vel associations (19, 5, and 10, respectively).Considering the number of stars for which I studied the lightcurves, I found that in the very young associations ChI andChII ( ∼ . Myr) there is a high fraction of dippers ( ∼ %and ∼ %, respectively), while the the fraction of dippersdecreases considering the other older associations: Lup andCrA ( ∼ - Myr) associations contain ∼ − % of dipperstars, while in the Vel association ( ∼ Myr) only ∼ ∼ M ⊙ ) survive up to Myr (see,e.g., Carpenter et al. 2009; Luhman & Mamajek 2012).
I searched for candidate exoplanets among the associationmembers by using the procedure described in Paper II. In or-der to search for transits in the light curves, stellar variabil-ity must be removed from them. I modelled the variability ofeach light curve interpolating to it a 5th-order spline definedon N knots . I considered three different grids of knots (withknots every 4.0 h, 6.5 h, and 13.0 h) to better model thelight curve of short- and long-period variable stars and alsoto avoid the flattening of transits whose duration is longerthan 4 or 6.5 hours. I removed bad photometric measure-ments from the flattened light curves by clipping away allthe outliers σ above the median flux, and discarding all thepoints with DQUALITY>0 and values of the local background σ above the mean background value.Adopting the routine developed by Hippke & Heller(2019), I extracted the Transit-fitting Least squares (TLS)periodograms of the flattened and “cleaned” light curves.I searched for transits with period . d ≤ P < . × T LC ,where T LC is the maximum temporal interval covered by thelight curve. I performed a first selection of candidate tran-siting objects on the basis of four parameters extracted bythe TLS routine: (i) the signal detection efficiency (SDE);(ii) signal-to-noise ratio (SNR); (iii) the significance betweenodd and even transits ( σ odd − even ); (iv) the mean depth ofthe transits ( δ t ). I selected as candidates all the stars hav-ing: (i) SDE ≥ (panel (a ) of Fig. 8); (ii) SNR ≥ (panel(a ) of Fig. 8); (iii) σ odd − even <
3; (iv) δ t <
10 %. I visually in-spected the light curves that passed the selections to checkthe odd/even transit depths (panels (c) of Fig. 8), the pres-ence of secondary eclipses, and to exclude false positives dueto the presence of artefacts in the light curves. For each sec-tor, I applied this procedure to the light curves flattened byusing the three different grids of knots previously described.I repeated this procedure considering as a first step each sec-tor independent from the others, and then considering thestacked light curves of the stars observed in more than onesector: in this way I avoided that possible artefacts in (or TLS v.1.0.24 ; https://github.com/hippke/tls different photometric precision of) the light curves of starsobserved in more than one sector decrease the detection ef-ficiency of the TLS routine. The number of candidates thatpassed this first selection is 48.I performed a series of vetting tests on the light curvesof these 48 candidate transiting objects: (i) inspection ofthe light curves obtained with different photometric aper-tures in order to check changes in the transit depths due toa close eclipsing binary; (ii) check of the light curves phasedwith a period of 0.5 × , 1 × , and 2 × the period found by theTLS routine, in order to search for secondary eclipses; (iii)comparison between the binned even/odd folded transits,in order to check if the depth of the transits are in agree-ment within the errors; (iv) analysis of the in/out-of-transitdifference centroid to check if the transit events are associ-ated to a close contaminant. I refer the reader to Paper Iand Paper II for a detailed description of the vetting tests.Figure 8 shows an overview of the main steps of the vet-ting procedure: the not-flattened light curve and the posi-tion of odd/even transits of the candidate TIC 143777072are shown in panel (b); panels (c) illustrate the comparisonbetween odd and even transits: because the mean depthsof the transits agree within the errors, the candidate passedthis test; panel (d) shows the analysis of the in/out-of-transitdifference centroid: in both the sectors in which the star wasobserved, the mean centroid is not located on the candidatebut on a star (TIC 143777056) located at ∼ arcsec fromthe target. I checked the light curve of the contaminant inthe ASAS-SN archive. The result is reported in the inset ofthe panel (d): the contaminant is confirmed to be an eclips-ing binary (depths of the primary eclipse δ T ∼ . mag in g -band).After the vetting procedure, 9 objects of interest(PATHOS-35–43) belonging to two associations (Lup andVel) survived. Among them there are two TESS
Objectsof Interest (TOI) released by the TESS team (TOI-508=PATHOS-36, TOI-831=PATHOS-41).
In order to extract physical parameters of the transiting ob-jects from the light curves of their host, some star param-eters, like temperature, mass, and radius, are mandatory. Iextracted the information for each star that hosts a candi-date transiting exoplanet by fitting isochrones to the CMDsof the associations. Stellar parameters are derived interpo-lating the colour and the magnitude of the host star on theisochrone. For the isochrone fitting, I used the set of PAR-SEC isochrones, the distance modulus, the reddening, andthe metallicity adopted in Sect. 3.1, and the ages derivedin the same section. Stellar parameters for the host of tran-siting objects are reported in Table 2. These informationare used for the transit modelling, as described in the nextsection. https://tess.mit.edu/toi-releases/go-to-alerts/ MNRAS000
In order to extract physical parameters of the transiting ob-jects from the light curves of their host, some star param-eters, like temperature, mass, and radius, are mandatory. Iextracted the information for each star that hosts a candi-date transiting exoplanet by fitting isochrones to the CMDsof the associations. Stellar parameters are derived interpo-lating the colour and the magnitude of the host star on theisochrone. For the isochrone fitting, I used the set of PAR-SEC isochrones, the distance modulus, the reddening, andthe metallicity adopted in Sect. 3.1, and the ages derivedin the same section. Stellar parameters for the host of tran-siting objects are reported in Table 2. These informationare used for the transit modelling, as described in the nextsection. https://tess.mit.edu/toi-releases/go-to-alerts/ MNRAS000 , 1–19 (2020) D. Nardiello
Figure 8.
Overview on the selection of candidate transiting objects. Panels (a) show the SDE and SNR versus Period distributionsobtained with the TLS routine; blue star represents TIC 143777072. Panel (b) shows the light curve of TIC 143777072 observed in Sectors7 and 9; green and azure triangles indicate the even and odd transits, respectively. Panels (c) are the phased even and odd transits:green/azure points are the median flux values calculated in bins of width 0.01. Panel (d) is the analysis of in/out-of-transit differencecentroid: in (0,0) is located TIC 143777072; magenta circle has the same size of the photometric aperture adopted in the light curveanalysis; light blue and red points are the mean centroids calculated for sectors 7 and 9, respectively, and indicate that the transit eventshappen on the suspected eclipsing binary TIC 143777056, whose ASAS-SN g -band light curve is plotted in the inset panel. Table 2.
Star parameters and priors for the modelling.
TIC PATHOS Assoc. α δ
T R ⋆ M ⋆ ρ ⋆ Period T LD c LD c df (deg.) (deg.) (mag.) ( R ⊙ ) ( M ⊙ ) ( ρ ⊙ ) (d) (BTJD)0081353413 35 Vel . − . . . ± .
15 0 . ± .
05 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
01 2 . ± .
05 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
10 0 . ± .
04 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
10 1 . ± .
03 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
10 8 . ± .
17 0 . ± . U( . , . ) U( . , . ) . ± .
20 0 . ± .
20 0 . ± . . − . . . ± .
11 1 . ± .
05 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
12 1 . ± .
12 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
09 1 . ± .
04 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . . − . . . ± .
08 1 . ± .
02 0 . ± . U( . , . ) U( . , . ) . ± .
10 0 . ± .
10 0 . ± . MNRAS , 1–19 (2020)
ATHOS III: Young Associations Figure 9.
Overview on the modelling procedure adopted for deriving the physical parameters of the transiting object PATHOS-38.Panel (a) shows the light curve of PATHOS-38 (TIC 123755508), whose position in the T versus G − G RP CMD is shown with a redcircle in panel (b). Panels (c) show the local polynomial fit + transit fit for each transit of the candidate exoplanet observed by TESS :the models are colour-coded as the arrows that indicate the epoch of the transit centres in panel (a). Panel (d) shows the folded transits(black points) after removing the local polynomial fits: in red the mean model of the transits. Bottom panel shows the difference betweenthe observed and the modelled transits (see text for details).
I used the python package
PYORBIT (Malavolta et al. 2016,2018; Benatti et al. 2019), developed for the modelling ofplanetary transits and radial velocities. The routine is basedon the combined use of the package BATMAN (Kreidberg2015), the affine invariant Markov chain Monte Carlo(MCMC) sampler
EMCEE (Foreman-Mackey et al. 2013), and https://github.com/LucaMalavolta/PyORBIT the global optimization algorithm PYDE (Storn & Price1997).For the transit model, I included the central time ofthe first transit ( T ), the period ( P ), the impact parameter( b ), the planetary-to-stellar-radius ratio ( R P / R ⋆ ), the stellardensity ( ρ ⋆ ). To locally model the stellar activity, for eachtransit a 2nd-degree polynomial fit is performed to the out-of-transit part of the light curve. To take into account theimpact of the dilution ( df ) on the error estimate of R P , Iincluded in the modelling this quantity as a free parameter, https://github.com/hpparvi/PyDE MNRAS000
EMCEE (Foreman-Mackey et al. 2013), and https://github.com/LucaMalavolta/PyORBIT the global optimization algorithm PYDE (Storn & Price1997).For the transit model, I included the central time ofthe first transit ( T ), the period ( P ), the impact parameter( b ), the planetary-to-stellar-radius ratio ( R P / R ⋆ ), the stellardensity ( ρ ⋆ ). To locally model the stellar activity, for eachtransit a 2nd-degree polynomial fit is performed to the out-of-transit part of the light curve. To take into account theimpact of the dilution ( df ) on the error estimate of R P , Iincluded in the modelling this quantity as a free parameter, https://github.com/hpparvi/PyDE MNRAS000 , 1–19 (2020) D. Nardiello
Table 3.
Results of transit modelling
TIC PATHOS Assoc.
P T R p / R ⋆ b a ρ ⋆ i R p R p Note(d) (BTJD) (au) ( ρ ⊙ ) (deg) ( R J ) ( R ⊕ )0081353413 35 Vel . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . (1)0095003423 37 Lup . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . (2)0238379370 40 Vel . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . (1)0374732772 42 Lup . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . (3)0411662605 43 Lup . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . . − . + . (4) ( ) Also in the TOI catalogue ( ) Radius too large, suspected eclipsing binary ( ) Another single deeper (suspected) transit ( δ T ∼ %) is present in the light curve due to a likely stellar companion or second planet. ( ) Likely field star. with a Gaussian prior obtained by using the stars in theGaia DR2 catalogue that fall in the same pixel of the tar-get, and transforming their Gaia magnitude in
TESS magni-tudes with the equations reported by Stassun et al. (2019).I extracted information on the limb darkening (LD) coeffi-cients by using the T eff and log ( g ) values obtained with theisochrone fitting, and the grid of values published by Claret(2018); the LD parametrization adopted is that described byKipping (2013). All the priors adopted for the modelling arelisted in Table 2. For the modelling of the transits I adopteda circular orbit ( e = ). In the modelling process, the rou-tine took into account of the 30-min cadence of the TESStime-series (Kipping 2010). The routine explored all the pa-rameters in linear space, using a number of walkers N walkers equal to 10 times the number of free parameters. For eachmodel, I run the sampler for 80 000 steps, cutting away thefirst 15 000 steps as burn-in, and using a thinning factor of100. An overview on the modelling procedure is reported inFig. 9 for PATHOS-38. The results of the modelling for allthe objects of interest are reported in Table 3 and Figs. 10and 11. In this work I found and modelled 9 transiting objects ofinterest. For the analysis described in this section, I ex-cluded the objects with a radius R P & R J (PATHOS-37and PATHOS-39), because of their doubtful planet nature.I excluded also PATHOS-43 because, on the basis of its po-sition on the G versus ( G − G RP ) CMD (see Fig. 11), it hasa high probability to be not an association member.All the 6 survived candidates I have detected are Jupitersize candidate exoplanets ( R P ∼ . – . R J ). No Neptune- orEarth-size candidate planets have been detected. Given thedistance of the associations ( ∼ – pc) and the pho-tometric precision of the light curves, on the basis of theanalysis performed in Paper II (see Fig. 8), it is not possi-ble to detect (super-)Earth size planets around associationmembers studied in this work. On the basis of the same anal-ysis, it is possible to detect Neptune-size exoplanets only for stars with radii R ⋆ . . R ⊙ . I calculated the expected num-ber of exoplanets ( N planet ) as done in Paper II, by using the(modified) equation: N planet = f ⋆ × Õ r N ⋆ r × Pr transit r (2)where f ⋆ is the percentage of stars with at least one exo-planet, the sum on r indicate the intervals of stellar radiiconsidered ( r = [ . , . ] , [ . , . ] , [ . , . ] , [ . , . ] R ⊙ ), N ⋆ is the number of stars in the considered stellar radius bins, Pr transit ≃ R ⋆ / a is the transit probability, with a the semi-major axis of the orbit, calculated using the third law ofKepler and by using an average period P = d. Consider-ing: (1) the stars for which I analysed the light curves; (2) thestars with R ⋆ ≤ R ⊙ ; (3) the frequencies f ⋆ for Neptune-sizeexoplanets tabulated by Fressin et al. (2013) in the case ofexoplanets with P = . − . d ( ∼ . %) and calculated inPaper II ( ∼ . %), I expect to find N planet ( R P = R N ) = ± ,in agreement with the null detection found in this work.The Jupiter size candidates found in this work orbitstars in the Lup (2 candidates) and Vel (4 candidates) associ-ations, the oldest associations studied in this work, while nocandidates have been found around ChI, ChII, and CrA as-sociations. By using equation (2), I calculated the frequencyof candidate Jupiters in Lup and Vel associations. Becausethe periods of the candidates range between ∼ . d and ∼ d, I divided the sample of candidate exoplanets in threesub-samples and, on the basis of their periods, I calculatedthe frequency f ⋆ by using the transit probabilities associ-ated to the mean period ¯ P of the candidates that form thesub-sample:(i) for candidates with period . ≤ P ≤ . , ( ¯ P ∼ . d) Ifound f ⋆ = ( . ± . ) % and f ⋆ = ( . ± . ) %, for Lupand Vel association, respectively; considering all the starsanalysed in this work (i.e., including also the ChI, ChII,and CrA members), I found f ⋆ = ( . ± . ) %. For giantplanets orbiting field stars with periods . ≤ P ≤ . ,Fressin et al. (2013) tabulated a frequency of f ⋆ = ( . ± . ) %, that is lower than the mean values found in thiswork but in agreement within ∼ σ ;(ii) for candidates with period . ≤ P ≤
30 d ( ¯ P ∼ . d)I found f ⋆ = ( . ± . ) %, f ⋆ = ( . ± . ) %, and f ⋆ = MNRAS , 1–19 (2020)
ATHOS III: Young Associations Figure 10.
Overview on the transiting objects (PATHOS-35, 36, 39, and 40) orbiting Vel association stars. Left panel shows the G versus ( G − G RP ) CMD of the likely Vel members: red dots indicate the transiting objects’ positions. Right panels show the phased transitsof the objects of interest and the fitted transit models (red); for each object, the difference between the observations and the model isshown below its folded light curve. ( . ± . ) % if I consider only Lup members, Vel members,and all the stars, respectively. For Jupiter exoplanets orbit-ing field stars with period in the range . ≤ P ≤ . ,Fressin et al. (2013) found a frequency f ⋆ = ( . ± . ) %;also in this case the mean frequency found in this work ishigher than that tabulated by Fressin et al. (2013), even ifthey agree within σ .(iii) for candidates with period . ≤ P ≤
30 d ( ¯ P ∼ . d)the frequencies of Jupiter-size exoplanets in the Lup and Velmembers are f ⋆ = ( . ± . ) % and f ⋆ = ( . ± . ) %,respectively. Considering all the stars analysed in this work,I found a frequency f ⋆ = ( . ± . ) %. Fressin et al.(2013) found for giant planets around field stars with pe-riods . ≤ P ≤ . a frequency f ⋆ = ( . ± . ) %,also in this case lower than the value found in this work, butin agreement within σ .I want to emphasise that, for the statistical analysis per-formed in this work, the completeness of the detectionmethod was not taken into account, and therefore the cal-culated frequencies might be considered as lower limits.By using the results previously obtained, I calculatedhow many transiting Jupiters are expected to be found inthe others three associations: even considering the maximummean frequency found in the previous analysis ( ∼ . %), the expected number of giants in the three associations is N P < ,in agreement with the null detection obtained in this work. In the present work, the third of the PATHOS project, Iperformed a detailed analysis of the light curves of stars infive young ( T- )associations associated to star forming re-gions: Chamaeleon I and II, Lupus, Corona Australis, and γ Velorum association. These associations have been chosenbecause of their young age ( . Myr): indeed, searching andcharacterising exoplanets orbiting very young stars allow usto constraint theoretical models on the formation of themand to understand the mechanisms that prevail in their dy-namical and physical evolution (migration, atmospheric loss,etc.).For this work, I extracted and corrected 7150 lightcurves of 4459 association members from
TESS
FFIs by us-ing the PSF-based approach pipeline already adopted withsuccess in previous works. Light curves will be publiclyavailable as HLSP on the PATHOS project webpage (DOI:10.17909/t9-es7m-vw14) of the MAST archive. https://archive.stsci.edu/hlsp/pathos MNRAS000
FFIs by us-ing the PSF-based approach pipeline already adopted withsuccess in previous works. Light curves will be publiclyavailable as HLSP on the PATHOS project webpage (DOI:10.17909/t9-es7m-vw14) of the MAST archive. https://archive.stsci.edu/hlsp/pathos MNRAS000 , 1–19 (2020) D. Nardiello
Figure 11.
As in Fig. 10, but for objects of interest that are likely members of Lup association (PATHOS-37, 41, 42, 43).
By performing an analysis of the GLS periodograms ofthe light curves, I identified 1260 periodic variable stars.Combining the gyrochronological analysis of periodic vari-able stars and the isochrone fitting of the CMD, I con-strained the ages of the associations, obtaining that thefive associations have ages that span between ∼ Myr and ∼ Myr. Because of the young age, the analysed associa-tions host a large number of YSOs surrounded by circum-stellar disk. By the analysis of the light curves, I identi-fied 71 dipper stars, i.e., stars that present in their lightcurves important drops of the flux on timescales of . day.These drops of the luminosity are due to the dust that formthe inner regions of the circumstellar disks; comparing thesimultaneous drops observed in TESS and g -band ASAS-SN light curves, I calculated the ratio between the absorp-tions in T and g bands ( A T / A g ), that gives us informationon the size of the grains that form the disk. In particular,when A T / A g → the grains have sizes comparable with thewavelengths in which TESS observes; lower values of the ra-tio A T / A g are associated to smaller grains. I found a weakanti-correlation between A T / A g and the dereddened colour ( V − K ) , with grain sizes that decrease with the mass ofthe hosting star. This work can not confirm the correlationbetween the infrared excess ( K s − [ µ m ]) and A T / A g foundby Bredall et al. (2020). Finally, I found that the highestfrequency of dippers are associated to the low mass starsof the youngest associations (ChI and ChII, ∼ . Myr, fre- quency ∼ %), and that the frequency of dippers is anti-correlated with the age of the associations, confirming thatthe timescales for the disk cleaning around low-mass stars is < Myr (Luhman & Mamajek 2012).I searched for transit signals among the light curves ofthe association members, and after the vetting tests (anal-ysis of the odd/even transits, of the in-/out-of transit cen-troid offset, etc.), 9 objects of interest passed the selections.In order to derive the physical parameters of the transit-ing objects, I modelled their transits by using their lightcurves and the stellar parameters derived through isochronefits. Excluding two objects of interest, because their radiusis too large ( R P > R J ), and another object of interest be-cause hosted by a likely field star, I detected 6 Jupiter sizecandidates: 2 in the Lup association and 4 in the Vel associ-ation. No Earth, super-Earth, and Neptune size candidateshave been detected; anyway, given the distance of the as-sociations, the number of members, and the frequency ofthese kind of exoplanets tabulated by Fressin et al. (2013)and in Paper II, the null detection is agreement with theexpectations. The mean frequency of giant planets in asso-ciations derived considering different period intervals rangesbetween ∼ % and ∼ %, higher than the values reportedby Fressin et al. (2013) for giants orbiting field stars ( . %)and in Paper II for Jupiters orbiting open cluster members( ∼ . %). Anyway, given the low number of candidates, theerrors on the calculated frequencies are too large, and the MNRAS , 1–19 (2020)
ATHOS III: Young Associations obtained results must be considered provisional. I also veri-fied if the null detection of giant planets around ChI, ChIIand CrA members is expected: even considering a frequency f ⋆ ∼ %, the number of Jupiter size exoplanet expected is N P << , in agreement with the null detection of this work.The analysis of the light curves of older associa-tion members ( ∼ APPENDIX A: ELECTRONIC MATERIAL
The catalogues of the periodic variable stars and of the dip-pers analysed in this work are available electronically as sup-porting material to this paper. Both the catalogues are in ascii and fits format. A description of the columns for thetwo catalogues are reported in Tables A1 and A2.Light curves extracted and analysed in this work areavailable in the MAST archive as HLSP under the projectPATHOS (DOI: 10.17909/t9-es7m-vw14). The updatedlist of candidate exoplanets is reported on the PATHOS web-page of the MAST archive. ACKNOWLEDGEMENTS
I acknowledge the support from the French Centre Na-tional d’Etudes Spatiales (CNES). I acknowledge thepartial support from PLATO ASI-INAF agreements n.2015-019-R0-2015 and 2015-019-R.1-2018. I warmly thankthe referee for carefully reading the manuscript. I thankM. Deleuil and G. Piotto for the useful suggestions onthis work, and L. Malavolta for his support in the use of
PYORBIT . This paper includes data collected by the
TESS mission. Funding for the
TESS mission is provided bythe NASA Explorer Program. This work has made useof data from the European Space Agency (ESA) mission
Gaia ( ), processed bythe Gaia
Data Processing and Analysis Consortium (DPAC, ).Funding for the DPAC has been provided by national in-stitutions, in particular the institutions participatingin the
Gaia
Multilateral Agreement. Some tasks of thedata analysis have been carried out using
VARTOOLS v 1.38 (Hartman & Bakos 2016) and
TLS python routine(Hippke & Heller 2019).
DATA AVAILABILITY
The data underlying this article are available inMAST at doi:10.17909/t9-es7m-vw14 and at https://archive.stsci.edu/hlsp/pathos .The data underlying this article are available in the articleand in its online supplementary material. https://archive.stsci.edu/hlsp/pathos REFERENCES
Aizawa M., Suto Y., Oya Y., Ikeda S., Nakazato T., 2020, arXive-prints, p. arXiv:2007.03393Ambartsumian V. A., 1949, Azh, 26, 3Ansdell M., et al., 2016, ApJ, 816, 69Baraffe I., Chabrier G., Barman T. S., Allard F., Hauschildt P. H.,2003, A&A, 402, 701Barnes S. A., 2003, ApJ, 586, 464Barnes S. A., 2007, ApJ, 669, 1167Barros S. C. C., Demangeon O., Deleuil M., 2016, A&A,594, A100Benatti S., et al., 2019, A&A, 630, A81Biazzo K., Randich S., Palla F., Brice˜no C., 2011, A&A, 530, A19Biazzo K., D’Orazi V., Desidera S., Covino E., Alcal´a J. M., ZusiM., 2012a, MNRAS, 427, 2905Biazzo K., Alcal´a J. M., Covino E., Frasca A., Getman F., SpezziL., 2012b, A&A, 547, A104Bodman E. H. L., et al., 2017, MNRAS, 470, 202Bohn A. J., et al., 2019, A&A, 624, A87Borucki W. J., et al., 2010, Science, 327, 977Bouma L. G., Hartman J. D., Bhatti W., Winn J. N., Bakos G. ´A.,2019, ApJS, 245, 13Bovy J., Rix H.-W., Green G. M., Schlafly E. F., Finkbeiner D. P.,2016, ApJ, 818, 130Bredall J. W., et al., 2020, MNRAS, 496, 3257Bressan A., Marigo P., Girardi L., Salasnich B., Dal Cero C.,Rubele S., Nanni A., 2012, MNRAS, 427, 127Carpenter J. M., Mamajek E. E., Hillenbrand L. A., Meyer M. R.,2006, ApJ, 651, L49Carpenter J. M., Mamajek E. E., Hillenbrand L. A., Meyer M. R.,2009, ApJ, 705, 1646Chen C. H., Jura M., Gordon K. D., Blaylock M., 2005, ApJ,623, 493Chen C. H., Mamajek E. E., Bitner M. A., Pecaut M., Su K.Y. L., Weinberger A. J., 2011, ApJ, 738, 122Ciardi D. R., et al., 2018, AJ, 155, 10Claret A., 2018, A&A, 618, A20Cody A. M., Hillenbrand L. A., 2010, ApJS, 191, 389Cody A. M., Hillenbrand L. A., 2018, AJ, 156, 71Cody A. M., et al., 2014, AJ, 147, 82Curtis J. L., et al., 2018, AJ, 155, 173Curtis J. L., Ag¨ueros M. A., Mamajek E. E., Wright J. T., Cum-mings J. D., 2019, AJ, 158, 77Cutri R. M., et al., 2003, VizieR Online Data Catalog, 2246D’Orazi V., Randich S., Flaccomio E., Palla F., Sacco G. G.,Pallavicini R., 2009, A&A, 501, 973D’Orazi V., Biazzo K., Randich S., 2011, A&A, 526, A103David T. J., et al., 2016a, AJ, 151, 112David T. J., et al., 2016b, Nature, 534, 658David T. J., et al., 2018, AJ, 156, 302David T. J., et al., 2019a, AJ, 158, 79David T. J., Petigura E. A., Luger R., Foreman-Mackey D., Liv-ingston J. H., Mamajek E. E., Hillenbrand L. A., 2019b, ApJ,885, L12Drimmel R., Cabrera-Lavers A., L´opez-Corredoira M., 2003,A&A, 409, 205Foreman-Mackey D., Hogg D. W., Lang D., Goodman J., 2013,PASP, 125, 306Fressin F., et al., 2013, ApJ, 766, 81Fulton B. J., Petigura E. A., 2018, AJ, 156, 264Fulton B. J., et al., 2017, AJ, 154, 109Gaia Collaboration et al., 2018, A&A, 616, A1Gaidos E., et al., 2020, MNRAS, 495, 650Girardi L., Bertelli G., Bressan A., Chiosi C., GroenewegenM. A. T., Marigo P., Salasnich B., Weiss A., 2002, A&A,391, 195Gonz´alez Hern´andez J. I., Caballero J. A., Rebolo R., B´ejarMNRAS000
Aizawa M., Suto Y., Oya Y., Ikeda S., Nakazato T., 2020, arXive-prints, p. arXiv:2007.03393Ambartsumian V. A., 1949, Azh, 26, 3Ansdell M., et al., 2016, ApJ, 816, 69Baraffe I., Chabrier G., Barman T. S., Allard F., Hauschildt P. H.,2003, A&A, 402, 701Barnes S. A., 2003, ApJ, 586, 464Barnes S. A., 2007, ApJ, 669, 1167Barros S. C. C., Demangeon O., Deleuil M., 2016, A&A,594, A100Benatti S., et al., 2019, A&A, 630, A81Biazzo K., Randich S., Palla F., Brice˜no C., 2011, A&A, 530, A19Biazzo K., D’Orazi V., Desidera S., Covino E., Alcal´a J. M., ZusiM., 2012a, MNRAS, 427, 2905Biazzo K., Alcal´a J. M., Covino E., Frasca A., Getman F., SpezziL., 2012b, A&A, 547, A104Bodman E. H. L., et al., 2017, MNRAS, 470, 202Bohn A. J., et al., 2019, A&A, 624, A87Borucki W. J., et al., 2010, Science, 327, 977Bouma L. G., Hartman J. D., Bhatti W., Winn J. N., Bakos G. ´A.,2019, ApJS, 245, 13Bovy J., Rix H.-W., Green G. M., Schlafly E. F., Finkbeiner D. P.,2016, ApJ, 818, 130Bredall J. W., et al., 2020, MNRAS, 496, 3257Bressan A., Marigo P., Girardi L., Salasnich B., Dal Cero C.,Rubele S., Nanni A., 2012, MNRAS, 427, 127Carpenter J. M., Mamajek E. E., Hillenbrand L. A., Meyer M. R.,2006, ApJ, 651, L49Carpenter J. M., Mamajek E. E., Hillenbrand L. A., Meyer M. R.,2009, ApJ, 705, 1646Chen C. H., Jura M., Gordon K. D., Blaylock M., 2005, ApJ,623, 493Chen C. H., Mamajek E. E., Bitner M. A., Pecaut M., Su K.Y. L., Weinberger A. J., 2011, ApJ, 738, 122Ciardi D. R., et al., 2018, AJ, 155, 10Claret A., 2018, A&A, 618, A20Cody A. M., Hillenbrand L. A., 2010, ApJS, 191, 389Cody A. M., Hillenbrand L. A., 2018, AJ, 156, 71Cody A. M., et al., 2014, AJ, 147, 82Curtis J. L., et al., 2018, AJ, 155, 173Curtis J. L., Ag¨ueros M. A., Mamajek E. E., Wright J. T., Cum-mings J. D., 2019, AJ, 158, 77Cutri R. M., et al., 2003, VizieR Online Data Catalog, 2246D’Orazi V., Randich S., Flaccomio E., Palla F., Sacco G. G.,Pallavicini R., 2009, A&A, 501, 973D’Orazi V., Biazzo K., Randich S., 2011, A&A, 526, A103David T. J., et al., 2016a, AJ, 151, 112David T. J., et al., 2016b, Nature, 534, 658David T. J., et al., 2018, AJ, 156, 302David T. J., et al., 2019a, AJ, 158, 79David T. J., Petigura E. A., Luger R., Foreman-Mackey D., Liv-ingston J. H., Mamajek E. E., Hillenbrand L. A., 2019b, ApJ,885, L12Drimmel R., Cabrera-Lavers A., L´opez-Corredoira M., 2003,A&A, 409, 205Foreman-Mackey D., Hogg D. W., Lang D., Goodman J., 2013,PASP, 125, 306Fressin F., et al., 2013, ApJ, 766, 81Fulton B. J., Petigura E. A., 2018, AJ, 156, 264Fulton B. J., et al., 2017, AJ, 154, 109Gaia Collaboration et al., 2018, A&A, 616, A1Gaidos E., et al., 2020, MNRAS, 495, 650Girardi L., Bertelli G., Bressan A., Chiosi C., GroenewegenM. A. T., Marigo P., Salasnich B., Weiss A., 2002, A&A,391, 195Gonz´alez Hern´andez J. I., Caballero J. A., Rebolo R., B´ejarMNRAS000 , 1–19 (2020) D. Nardiello [t!]
Table A1.
Description of the column content of the catalogue of variable stars.
Column Name Unit Explanation01 RA [deg.] Right ascension (J2000, epoch 2015.5)02 DEC [deg.] Declination (J2000, epoch 2015.5)03
TIC
TESS
Input Catalogue ID04
GAIA_DR2
Gaia DR2 Source ID05
PERIOD [d] Period06
Gmag [mag] Gaia DR2 G magnitude07 e_Gmag [mag] Error on Gaia DR2 G magnitude08 BPmag [mag] Gaia DR2 G BP magnitude09 e_BPmag [mag] Error on Gaia DR2 G BP magnitude10 RPmag [mag] Gaia DR2 G RP magnitude11 e_RPmag [mag] Error on Gaia DR2 G RP magnitude12 Tmag [mag]
TESS T magnitude13 e_Tmag [mag] Error on TESS T magnitude14 Bmag [mag] B -Johnson magnitude15 e_Bmag [mag] Error on B -Johnson magnitude16 Vmag [mag] V -Johnson magnitude17 e_Vmag [mag] Error on V -Johnson magnitude18 Jmag [mag] 2MASS J magnitude19 e_Jmag [mag] Error on 2MASS J magnitude20 Hmag [mag] 2MASS H magnitude21 e_Hmag [mag] Error on 2MASS H magnitude22 Kmag [mag] 2MASS K s magnitude23 e_Kmag [mag] Error on 2MASS K s magnitude24 W1mag [mag] WISE W magnitude25 e_W1mag [mag] Error on WISE W magnitude26 W2mag [mag] WISE W magnitude27 e_W2mag [mag] Error on WISE W magnitude28 W3mag [mag] WISE W magnitude29 e_W3mag [mag] Error on WISE W magnitude30 W4mag [mag] WISE W magnitude31 e_W4mag [mag] Error on WISE W magnitude32 E_BV E ( B − V ) PARALLAX mas Parallax from Gaia DR234
PM_RA mas yr − Proper motion along the RA direction from Gaia DR235
PM_DEC mas yr − Proper motion along the DEC direction from Gaia DR236
ASSOCIATION
Name of the association that host the star
V. J. S., Barrado Y Navascu´es D., Mart´ın E. L., ZapateroOsorio M. R., 2008, A&A, 490, 1135Green G. M., Schlafly E., Zucker C., Speagle J. S., Finkbeiner D.,2019, ApJ, 887, 93Hansen B. M. S., Murray N., 2012, ApJ, 751, 158Hartman J. D., Bakos G. ´A., 2016, Astronomy and Computing,17, 1Hippke M., Heller R., 2019, A&A, 623, A39Howell S. B., et al., 2014, PASP, 126, 398Ida S., Lin D. N. C., 2010, ApJ, 719, 810James D. J., Melo C., Santos N. C., Bouvier J., 2006, A&A,446, 971Jeffries R. D., et al., 2017, MNRAS, 464, 1456Kipping D. M., 2010, MNRAS, 408, 1758Kipping D. M., 2013, MNRAS, 435, 2152Kochanek C. S., et al., 2017, PASP, 129, 104502Kreidberg L., 2015, PASP, 127, 1161Kuruwita R. L., Ireland M., Rizzuto A., Bento J., Federrath C.,2018, MNRAS, 480, 5099Libralato M., Bedin L. R., Nardiello D., Piotto G., 2016a,MNRAS, 456, 1137Libralato M., et al., 2016b, MNRAS, 463, 1780Lindegren L., et al., 2018, A&A, 616, A2Lopez E. D., Fortney J. J., 2013, ApJ, 776, 2Luhman K. L., Mamajek E. E., 2012, ApJ, 758, 31Malavolta L., et al., 2016, A&A, 588, A118 Malavolta L., et al., 2018, AJ, 155, 107Mann A. W., et al., 2016a, AJ, 152, 61Mann A. W., et al., 2016b, ApJ, 818, 46Mann A. W., et al., 2018, AJ, 155, 4Marigo P., et al., 2017, ApJ, 835, 77Marley M. S., Fortney J. J., Hubickyj O., Bodenheimer P., Lis-sauer J. J., 2007, ApJ, 655, 541Marshall D. J., Robin A. C., Reyl´e C., Schultheis M., Picaud S.,2006, A&A, 453, 635Meibom S., et al., 2013, Nature, 499, 55Morales-Calder´on M., et al., 2011, ApJ, 733, 50Mordasini C., Marleau G. D., Molli`ere P., 2017, A&A, 608, A72Nardiello D., et al., 2015a, MNRAS, 447, 3536Nardiello D., et al., 2015b, MNRAS, 451, 312Nardiello D., Libralato M., Bedin L. R., Piotto G., Ochner P.,Cunial A., Borsato L., Granata V., 2016a, MNRAS, 455, 2337Nardiello D., Libralato M., Bedin L. R., Piotto G., Borsato L.,Granata V., Malavolta L., Nascimbeni V., 2016b, MNRAS,463, 1831Nardiello D., et al., 2019, MNRAS, 490, 3806Nardiello D., et al., 2020, MNRAS, 495, 4924Newton E. R., et al., 2019, ApJ, 880, L17Obermeier C., et al., 2016, AJ, 152, 223Owen J. E., Lai D., 2018, MNRAS, 479, 5012Owen J. E., Wu Y., 2013, ApJ, 775, 105Pecaut M. J., Mamajek E. E., 2013, ApJS, 208, 9MNRAS , 1–19 (2020)
ATHOS III: Young Associations Table A2.
Description of the column content of the catalogue of dipper stars.
Column Name Unit Explanation01 RA [deg.] Right ascension (J2000, epoch 2015.5)02 DEC [deg.] Declination (J2000, epoch 2015.5)03
TIC
TESS
Input Catalogue ID04
GAIA_DR2
Gaia DR2 Source ID05
AT_AG A T / A g values [ − . : not available]06 e_ATAG Error on A T / A g values [ − . : not available]07 Gmag [mag] Gaia DR2 G magnitude08 e_Gmag [mag] Error on Gaia DR2 G magnitude09 BPmag [mag] Gaia DR2 G BP magnitude10 e_BPmag [mag] Error on Gaia DR2 G BP magnitude11 RPmag [mag] Gaia DR2 G RP magnitude12 e_RPmag [mag] Error on Gaia DR2 G RP magnitude13 Tmag [mag]
TESS T magnitude14 e_Tmag [mag] Error on TESS T magnitude15 Bmag [mag] B -Johnson magnitude16 e_Bmag [mag] Error on B -Johnson magnitude17 Vmag [mag] V -Johnson magnitude18 e_Vmag [mag] Error on V -Johnson magnitude19 Jmag [mag] 2MASS J magnitude20 e_Jmag [mag] Error on 2MASS J magnitude21 Hmag [mag] 2MASS H magnitude22 e_Hmag [mag] Error on 2MASS H magnitude23 Kmag [mag] 2MASS K s magnitude24 e_Kmag [mag] Error on 2MASS K s magnitude25 W1mag [mag] WISE W magnitude26 e_W1mag [mag] Error on WISE W magnitude27 W2mag [mag] WISE W magnitude28 e_W2mag [mag] Error on WISE W magnitude29 W3mag [mag] WISE W magnitude30 e_W3mag [mag] Error on WISE W magnitude31 W4mag [mag] WISE W magnitude32 e_W4mag [mag] Error on WISE W magnitude33 E_BV E ( B − V ) PARALLAX mas Parallax from Gaia DR235
PM_RA mas yr − Proper motion along the RA direction from Gaia DR236
PM_DEC mas yr − Proper motion along the DEC direction from Gaia DR237
ASSOCIATION
Name of the association that host the star
Pecaut M. J., Mamajek E. E., Bubar E. J., 2012, ApJ, 746, 154Pepper J., et al., 2017, AJ, 153, 177Pope B. J. S., Parviainen H., Aigrain S., 2016, MNRAS, 461, 3399Quinn S. N., et al., 2012, ApJ, 756, L33Quinn S. N., et al., 2014, ApJ, 787, 27Rebull L. M., et al., 2016, AJ, 152, 114Rebull L. M., Stauffer J. R., Cody A. M., Hillenbrand L. A.,David T. J., Pinsonneault M., 2018, AJ, 155, 196Rebull L. M., Stauffer J. R., Cody A. M., Hillenbrand L. A.,Bouvier J., Roggero N., David T. J., 2020, AJ, 159, 273Ricker G. R., et al., 2015, Journal of Astronomical Telescopes, Instruments, and Systems,1, 014003Rizzuto A. C., et al., 2020, AJ, 160, 33Rodriguez J. E., et al., 2017, ApJ, 848, 97Santos N. C., Melo C., James D. J., Gameiro J. F., Bouvier J.,Gomes J. I., 2008, A&A, 480, 889Schlichting H. E., 2018, Formation of Super-Earths. p. 141,doi:10.1007/978-3-319-55333-7 141Schlichting H. E., Sari R., Yalinewich A., 2015, Icarus, 247, 81Shappee B. J., et al., 2014, ApJ, 788, 48Sokolovsky K. V., et al., 2017, MNRAS, 464, 274Spiegel D. S., Burrows A., 2012, ApJ, 745, 174Spina L., et al., 2014a, A&A, 567, A55Spina L., et al., 2014b, A&A, 568, A2Stassun K. G., et al., 2019, AJ, 158, 138Stauffer J., et al., 2015, AJ, 149, 130 Storn R., Price K., 1997, Journal of Global Optimization, 11, 341Terquem C., Papaloizou J. C. B., 2007, ApJ, 654, 1110Vanderburg A., et al., 2018, AJ, 156, 46Wright E. L., et al., 2010, AJ, 140, 1868Zechmeister M., K¨urster M., 2009, A&A, 496, 577MNRAS000