A super-Earth and a sub-Neptune orbiting the bright, quiet M3 dwarf TOI-1266

Astronomy & Astrophysics A&A 642, A49 (2020) https://doi.org/10.1051/0004-6361/202038616 © ESO 2020 A super-Earth and a sub-Neptune orbiting the bright, quiet M3 dwarf TOI-1266 B.-O. Demory1, F. J. Pozuelos2,3, Y. Gómez Maqueo Chew4, L. Sabin5, R. Petrucci4,6,7, U. Schroffenegger1, S. L. Grimm1, M. Sestovic1, M. Gillon3, J. McCormac8,9, K. Barkaoui3,10, W. Benz1, A. Bieryla11, F. Bouchy12, A. Burdanov13,14, K. A. Collins11, J. de Wit13, C. D. Dressing15, L. J. Garcia3, S. Giacalone15, P. Guerra16, J. Haldemann1, K. Heng1,8, E. Jehin2, E. Jofré4,6,7, S. R. Kane17, J. Lillo-Box18, V. Maigné1, C. Mordasini19, B. M. Morris1, P. Niraula13, D. Queloz20, B. V. Rackham13,21, A. B. Savel15,22, A. Soubkiou10, G. Srdoc23, K. G. Stassun24, A. H. M. J. Triaud25, R. Zambelli26, G. Ricker21, D. W. Latham11, S. Seager13,21,27, J. N. Winn28, J. M. Jenkins29, T. Calvario-Velásquez5, J. A. Franco Herrera5, E. Colorado5, E. O. Cadena Zepeda5, L. Figueroa5, A. M. Watson4, E. E. Lugo-Ibarra5, L. Carigi4, G. Guisa5, J. Herrera5, G. Sierra Díaz5, J. C. Suárez30,31, D. Barrado18, N. M. Batalha32, Z. Benkhaldoun10, A. Chontos33, F. Dai34, Z. Essack13,21, M. Ghachoui10, C. X. Huang21, D. Huber33, H. Isaacson15,35, J. J. Lissauer29, M. Morales-Calderón18, P. Robertson36, A. Roy34, J. D. Twicken29,37, A. Vanderburg38, and L. M. Weiss33 (Affiliations can be found after the references) Received 9 June 2020 / Accepted 20 July 2020 ABSTRACT We report the discovery and characterisation of a super-Earth and a sub-Neptune transiting the bright (K=8.8), quiet, and nearby (37 pc) M3V dwarf TOI-1266. We validate the planetary nature of TOI-1266 b and c using four sectors of TESS photometry and data from the newly-commissioned 1-m SAINT-EX telescope located in San Pedro Mártir (México). We also include additional groundbased follow-up photometry as well as high-resolution spectroscopy and high-angular imaging observations. The inner, larger planet has a radius of R=2.37+0.16 −0.12 R⊕and an orbital period of 10.9 days. The outer, smaller planet has a radius of R=1.56+0.15 −0.13 R⊕on an 18.8-day orbit. The data are found to be consistent with circular, co-planar and stable orbits that are weakly influenced by the 2:1 mean motion resonance. Our TTV analysis of the combined dataset enables model-independent constraints on the masses and eccentricities of the planets. We find planetary masses of Mp=13.5+11.0 −9.0M⊕(<36.8M⊕at 2-σ) for TOI-1266 b and 2.2+2.0 −1.5M⊕(<5.7M⊕at 2-σ) for TOI-1266 c. We find small but non-zero orbital eccentricities of 0.09+0.06 −0.05 (<0.21 at 2-σ) for TOI-1266 b and 0.04 ±0.03 (<0.10 at 2-σ) for TOI-1266 c. The equilibrium temperatures of both planets are of 413 ±20 and 344 ±16 K, respectively, assuming a null Bond albedo and uniform heat redistribution from the day-side to the night-side hemisphere. The host brightness and negligible activity combined with the planetary system architecture and favourable planet-to-star radii ratios makes TOI-1266 an exquisite system for a detailed characterisation. Key words. instrumentation: detectors – planets and satellites: detection 1. Introduction The science of exoplanets has been historically driven by dedicated astronomical observations. Currently, the Transiting Exoplanet Survey Satellite (TESS; Ricker et al. 2015) is leading the discovery of multi-planetary transiting systems with relatively small planets (i.e. sub-Neptune or smaller), orbiting around bright M-dwarf stars in the solar neighbourhood (e.g. Günther et al. 2019;Jenkins et al. 2019;Kostov et al. 2019;Cloutier et al. 2020). Their brightness allows for a detailed characterisation of small planets, and, in the near future, a glimpse into their atmospheric composition with the James Webb Space Telescope (JWST). Furthermore, when multiple transit-like signals are detected from a single star, the signals are likely to be genuine as opposed to false positives such as eclipsing binaries (Latham et al. 2011;Lissauer et al. 2012a;Morton et al. 2016). In some cases, in particular where the planets are near resonant orbits, time-series photometry alone not only allows for the measurement of the planet size, but also places dynamical constraints on the planet mass (Holman 2005;Agol et al. 2005). Measuring the planet mass and radius allows for the derivation of the bulk density, thus constraining planetary structure models (e.g. Dorn et al. 2017). There are more than 3000 transiting exoplanets known to date1, including 499 planetary systems with more than one detected transiting planet. This large sample of transiting exoplanets allows for in-depth exploration of the distinct exoplanet populations. One such study by Fulton et al. (2017) identified a bi-modal distribution for the sizes of super-Earth and subNeptune Kepler exoplanets, with a peak at ∼1.3 R⊕and another at ∼2.4 R⊕. The interval between the two peaks is called the radius valley and it is typically attributed to the stellar irradiation received by the planets, with more irradiated planets being smaller due to the loss of their gaseous envelopes. Studying more than one super-Earth or sub-Neptune planet in a single system allows for tighter constraints on formation models (e.g. Owen & Campos Estrada 2020;Kubyshkina et al. 2019) and thus the exploration of the effects of other physical 1http://exoplanet.eu/ retrieved on 10 Aug 2020. Article published by EDP Sciences A49, page 1 of 21 A&A 642, A49 (2020) Table 1. Ground-based time-series photometric observations of TOI-1266. Date (UT) Filter Facility Exp. time [s] Notes 25 Dec 2019 Ic OAA-0.4m 160 b full 29 Jan 2020 z0SAINT-EX-1m 12 c partial 29 Feb 2020 z0SAINT-EX-1m 12 b full 21 Mar 2020 z0TRAPPIST-N-0.6m 15 b full 21 Mar 2020 r0Artemis-1m 10 b partial 01 Apr 2020 VTRAPPIST-N-0.6m 60 b partial 23 Apr 2020 TESS Kotizarovci-0.3m 50 b full 23 Apr 2020 clear ZRO-0.4m 200 b full processes such as the core and envelope mass distribution (e.g. Modirrousta-Galian et al. 2020). Here we present the discovery and characterisation of the planetary system TOI-1266, which was first identified from the TESS photometry2. We confirmed the planetary nature of the transits through ground-based follow-up observations, including time-series photometry, high-angular resolution images, spectroscopy, and archival imagery. The paper is structured as follows. Section 2describes all observations. The stellar characterisation of the planet host is described in Sect. 3. The validation of the transit signals in the light curves is presented in Sect. 4. The search for transit signals and the analysis of the light curves to derive physical properties are presented in Sect. 5. We also include a stability analysis and mass constraints from a dynamical analysis in Sect. 5.2. Finally, in Sect. 6, we discuss the implications for the formation and evolution of the TOI-1266 planetary system given the measured planet radii, orbital periods, and constraints on the masses, as well as prospects for atmospheric characterisation. A summary of our results and their implications is presented in Sect. 7. 2. Observations In this section, we present all the observations of TOI-1266 obtained with TESS and ground-based facilities. A summary of all ground-based time-series photometric observations of TOI-1266 is shown in Table 1. 2.1. TESS photometry TOI-1266 is a late-type star with a measured parallax that is part of the TESS Candidate Target List (Stassun et al. 2018). It was observed by TESS with 2-min-cadence in sectors 14–15 (18 July to 11 Sep 2019) and 21–22 (21 Jan to 18 Mar 2020). TOI-1266’s astrometric and photometric properties from the literature are reported in Table 2. The time-series observations of TOI-1266 were processed with the TESS Science Processing Operations Center (SPOC) pipeline (Jenkins 2002;Jenkins et al. 2016,2017), which resulted in the detection of two periodic transit signals: TOI-1266.01 and .02, the latter being at the detection limit using sectors 14-15 data alone, thus requiring additional data to strengthen that signal. We retrieved the Presearch Data Conditioning Simple Aperture Photometry (PDC-SAP; Stumpe et al. 2012,2014;Smith et al. 2012) from the Mikulski Archive for Space Telescopes and removed all datapoints flagged as “bad quality”. We identified 819/19337 such datapoints for sector 14, 912/18757 for sector 15, 2During the review of this paper, we were made aware of a preprint by Stefansson et al. (2020) that also reports on the discovery of the TOI-1266 system. Table 2. TOI-1266 stellar astrometric and photometric properties. Parameter Value Source Target designations TIC 467 179 528 1 2MASS J13115955+6550017 2 UCAC 4 780-025091 3 Gaia DR2 1678074272650459008 4 Photometry TESS 11.040 ±0.007 1 B14.58 ±0.05 3 V12.94 ±0.05 3 Gaia 12.1222 ±0.0002 4 u16.527 ±0.007 5 g14.950 ±0.005 5 r12.584 ±0.002 5 i11.600 ±0.001 5 z11.608 ±0.005 5 J9.71 ±0.02 2 H9.07 ±0.03 2 K8.84 ±0.02 2 WISE 3.4 µm 8.72 ±0.02 6 WISE 4.6 µm 8.61 ±0.02 6 WISE 12 µm 8.50 ±0.02 6 WISE 22 µm 8.23 ±0.21 6 Astrometry RA (J2000) 13 11 59.56 4 Dec (J2000) +65 50 01.70 4 RA PM (mas yr−1)−150.652 ±0.041 4 Dec PM (mas yr−1)−25.368 ±0.039 4 Parallax (mas) 27.7397 ±0.0226 4 References. 1. Stassun et al. (2018), 2. Cutri et al. (2003), 3. Zacharias et al. (2013), 4. Brown et al. (2018), 5. Alam et al. (2015), 6. Cutri & et al. (2013). 1074/19694 for sector 21, and a larger count (5652/19579) for sector 22. Figure 1shows the TESS fields of view and apertures used for TOI-1266 over each of the four sectors with the location of nearby Gaia DR2 sources superimposed. 2.2. SAINT-EX photometry We obtained ground-based photometric time-series observations of TOI-1266 from the SAINT-EX Observatory2(Search And characterIsatioN of Transiting EXoplanets), which was 2https://www.saintex.unibe.ch/saint_ex_observatory/ A49, page 2 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 628 630 632 634 636 638 Pixel Column Number 1252 1254 1256 1258 1260 1262 Pixel Row Number E N TIC 467179528 - Sector 14 m = -2.0 m = 0.0 m = 2.0 m = 5.0 m = 8.0 1 2 3 4 5 6 0.00 0.02 0.04 0.06 0.08 0.10 0.12 Flux ×104 (e ) 554 556 558 560 562 564 Pixel Column Number 1454 1456 1458 1460 1462 1464 Pixel Row Number E N TIC 467179528 - Sector 15 m = -2.0 m = 0.0 m = 2.0 m = 5.0 m = 8.0 1 2 3 4 5 6 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 Flux ×104 (e ) 980 982 984 986 988 990 Pixel Column Number 1596 1598 1600 1602 1604 1606 Pixel Row Number E N TIC 467179528 - Sector 21 m = -2.0 m = 0.0 m = 2.0 m = 5.0 m = 8.0 1 2 3 4 5 60.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175 Flux ×104 (e ) 860 862 864 866 868 870 Pixel Column Number 1670 1672 1674 1676 1678 1680 Pixel Row Number E N TIC 467179528 - Sector 22 m = -2.0 m = 0.0 m = 2.0 m = 5.0 m = 8.0 1 2 3 4 5 6 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Flux ×104 (e ) Fig. 1. TESS target pixel files (TPFs) of the four sectors that observed TOI-1266. The plots were created with tpfplotter4(Aller et al. 2020). The apertures used to extract the photometry (Twicken et al. 2010;Morris et al. 2017) by the SPOC pipeline are shown as red shaded regions. The Gaia DR2 catalogue (Gaia Collaboration 2018) is overplotted, with all sources up to 8 magnitudes in contrast with TOI-1266 are shown as red circles. We note that the symbol size scales with the magnitude contrast. commissioned in March 2019. SAINT-EX is a 1-m F/8 RitcheyChrétien telescope built by the ASTELCO company and uses a similar design to the telescopes of the SPECULOOS Southern Observatory (Delrez et al. 2018;Jehin et al. 2018). SAINT-EX is located at the Observatorio Astronómico Nacional, in the Sierra de San Pedro Mártir in Baja California, México (31.04342 N, 115.45476 W) at 2780 m altitude. The telescope is installed on an ASTELCO equatorial NTM-1000 mount equipped with directdrive motors, which enables operations without meridian flip. The telescope is installed in a 6.25-m wide dome built by the Gambato company. In terms of mount performance, SAINT-EX typically achieves a RMS better than 300 relative to the pointing model and a tracking accuracy – without autoguiding – better than 200 over 15-min timescales. To improve this figure further, SAINT-EX uses the DONUTS autoguiding software (McCormac et al. 2013), which increases the guiding precision to 0.200 RMS or better that is less than a pixel. SAINT-EX is equipped with an Andor iKon-L camera that integrates a deepdepletion e2v 2K×2K CCD chip with a BEX2-DD coating that is optimised for the near infrared (NIR). The filter wheel includes the Sloan ugriz0broad-band filters, as well as special blue-blocking (transmittance >90% beyond 500 nm) and NIR (transmittance >95% beyond 705 nm) filters. The detector gives a field of view of 120×120with 0.3400per pixel. SAINT-EX operations are robotic and the data reduction and analysis are automated by a custom pipeline PRINCE (Photometric Reduction and In-depth Nightly Curve Exploration) that ingests the raw science and calibration frames and produces light curves using differential photometry. The PRINCE pipeline performs standard image reduction steps, applying bias, dark, and flat-field corrections. Astrometric calibration is conducted using Astrometry.net (Lang et al. 2010) to derive correct world coordinate system (WCS) information for each exposure. Photutils star detection (Bradley et al. 2019) is run on a median image of the whole exposure stack to create a pool of candidate stars in the field of view. Stars whose peak value in the largest aperture is above the background by a certain threshold, defined by an empirical factor times the median background noise of the night, are kept as reference stars for the differential photometric analysis. From the WCS information and the detected stars’ coordinates, the pipeline runs centroiding, aperture and annulus photometry on each detected star from the common pool, using LMFit (Newville et al. 2014) and Astropy (Astropy Collaboration 2013,2018), and repeats this for each exposure to obtain the measured lightcurves for a list of apertures. The measured lightcurves for each aperture are corrected for systematics using either a PCA approach (Pedregosa et al. 3https://github.com/jlillo/tpfplotter 2011) or a simple differential photometry approach that corrects a star’s lightcurve by the median lightcurve of all stars in the pool except for the target star. SAINT-EX observed one transit of each planet of the TOI-1266 system in early 2020. The observing strategy was to use the z0filter with a slightly-defocused 12-s exposure time to mitigate shutter noise and scintillation. A partial transit of TOI-1266.02 was observed on 29 January 2020 from 7:36 to 12:05 UT. A full transit of TOI-1266.01 was then observed on 29 February 2020 from 6:16 to 11:03 UT. We reduced both datasets with PRINCE using differential aperture photometry. We corrected our differential light curves for variations in both the airmass and the full width at half maximum (FWHM) along both horizontal and vertical axes on the detector. This correction is performed simultaneously to the transit fit in our MCMC framework detailed in Sect. 5.1.2. 2.3. TRAPPIST-North photometry We used the 60-cm TRAPPIST-North telescope located at Oukaimeden Observatory in Morocco (Jehin et al. 2011;Gillon et al. 2013;Barkaoui et al. 2019) to observe one full and one partial transit of TOI-1266.01. TRAPPIST-North is equipped with a thermoelectrically cooled 2K×2KAndor iKon-L BEX2-DD CCD camera with a pixel scale of 0.600 and a field of view of 200×200. We used the TESS Transit Finder tool, which is a customised version of the Tapir software package (Jensen 2013), to schedule the photometric time-series. Data reduction and photometric analysis were performed with the TRAPPHOT pipeline developed in IRAF as described in Gillon et al. (2013). The first transit was observed on 21 March 2020 in the Sloan-z0filter with an exposure time of 15 s. We took 476 raw images and performed aperture photometry in an uncontaminated aperture of 6.5 pixels (4.000) and a FWHM of 2.600. The second transit was observed on 01 April 2020 in the Johnson-Vfilter with an exposure time of 60 s. We took 338 raw images and we performed aperture photometry in an uncontaminated aperture of 8.1 pixels (5.000) and a FWHM of 3.400. During that second observation of TOI-1266, the telescope underwent a meridian flip at HJD 2 458941.5431, and we noticed the presence of thin clouds during the ingress. 2.4. Artemis photometry One partial transit of TOI-1266.01 was acquired on 21 March 2020 with the Artemis telescope, which constitutes the SPECULOOS-North facility located at the Teide Observatory (Canary Islands, Spain). The Artemis telescope is a twin of the SPECULOOS-South and SAINT-EX telescopes, which are all operated similarly and utilise Andor iKon-L cameras with A49, page 3 of 21 A&A 642, A49 (2020) e2v 2K ×2K deep-depletion CCDs. We acquired 668 images in the Sloan-r0filter with an exposure time of 10 s. Data reduction consisted of standard calibration steps and subsequent aperture photometry with the TRAPPHOT pipeline. Best comparison stars and optimum aperture size were selected on the basis of the minimisation of the out-of-transit scatter of the light curve. 2.5. Observatori Astronòmic Albanyà A full transit of TOI-1266 b was observed in ICband on 25 Dec. 2019 from Observatori Astronòmic Albanyà (OAA) in Girona, Spain. The 0.4 m telescope is equipped with a 3056 ×3056 Moravian G4-9000 camera with an image scale of 1. 0044 pixel−1, resulting in a 360×360field of view. The images were calibrated and the photometric data were extracted using the AstroImageJ (AIJ) software package (Collins et al. 2017). 2.6. Kotizarovci observatory A full transit of TOI-1266 b was observed in Baader R longpass 610 nm band on 23 Apr. 2020 from Kotizarovci Observatory near Viskovo, Croatia. The 0.3 m telescope is equipped with a765 ×510 SBIG ST7XME camera with an image scale of 1. 002 pixel−1resulting in a 15.30×10.20field of view. The images were calibrated and the photometric data were extracted using AIJ. 2.7. ZRO oobservatory A full transit of TOI-1266 b was observed without filter on 23 Apr. 2020 from the Zambelli Roberto Observatory (ZRO) in Liguria, Italy. The 0.4-m telescope is equipped with a 3072 ×2048 SBIG STXL-6303E camera with an image scale of 1. 0016 pixel−1resulting in a 490×330field of view. The images were calibrated and the photometric data were extracted using AIJ. 2.8. TRES spectroscopy We obtained two “reconnaissance” spectra with the Tillinghast Reflector Echelle Spectrograph (TRES; Furesz 2008) instrument mounted on the Fred Lawrence Whipple Observatory’s 1.5-m telescope on 25 January 2020 and 30 January 2020. The TRES wavelength coverage spans 385 to 910 nm and the resolving power is 44000. The purpose of these spectra is to discard obvious false positive scenarios and provide basic constraints on the stellar host. The two TRES spectra were extracted using the procedures outlined in Buchhave et al. (2010). Radial velocities were determined using special procedures developed by Jonathan Irwin for the analysis of spectra of M dwarfs, rather than the standard pipeline analysis, which was optimised for Solar-type stars, and uses the wavelength region near the Mg b features at 519 nm and a library of calculated templates to yield absolute velocities. Irwin’s tools use an observed spectrum of Barnard’s star as the template, and the wavelength region near 715 nm that contains TiO features rich in radial-velocity information. For M dwarfs, the S/N per resolution element is significantly stronger at 715 nm, in this case 30, compared to 12 at 519 nm. The two TRES observations were obtained at phases 16.79 and 17.25 of the TESS ephemeris for the inner planet (i.e. at opposite quadratures for a circular orbit), and yielded absolute velocities of −41.511 and −41.662 km s−1, thus ruling out a stellar companion or brown dwarf orbiting TOI-1266, as the source of the transits for the inner planet. 2.9. HIRES spectroscopy We obtained a single spectrum with the HIRES (Vogt et al. 1994) instrument mounted on the Keck-I 10-m telescope on 15 December 2019. HIRES has a wavelength coverage of 390 to 900 nm and a resolving power of 50000. The spectrum with which we conducted these analyses has a S/N per resolution element of 80. 2.10. High-angular resolution imaging We used high-angular resolution imaging to rule out falsepositive signals caused by unresolved blended stars in the timeseries photometry. This is particularly important for the TESS light curves, given that the TESS pixels are ∼2100. 2.10.1. AstraLux at Calar Alto TOI-1266 was observed on 30 October 2019 with the AstraLux instrument (Hormuth et al. 2008), a high-spatial resolution camera installed at the 2.2-m telescope of the Calar Alto Observatory (Almería, Spain). This fast-readout camera uses the lucky-imaging technique (Fried 1978) by obtaining thousands of short-exposure frames to subsequently select a small percentage of them showing the best Strehl ratio (Strehl 1902), and combining them into a final image. We observed this target using the Sloan Digital Sky Survey zfilter (SDSSz), which provides the best resolution and contrast capabilities for the instrument (Hormuth et al. 2008), and obtained 35 000 frames with 20 ms exposure time and a 600 ×600 field-of-view. The datacube was subsequently reduced using the observatory pipeline (Hormuth et al. 2008), and we used a 10% selection rate for the best frames to obtain a final high-resolution imaging with a total exposure time of 70 s. We then computed the sensitivity curve by using our own astrasens package4, following the procedure described in Lillo-Box et al. (2012,2014). The image allows us to discard stellar companions with magnitude contrasts down to ∆m=4at 0.200 (corresponding to 7.2 au at the distance of this system), and hence, establish a maximum contamination in the planet transit better than 3%. The AstraLux contrast curve is shown in Fig. 2. 2.10.2. ShARCS TOI-1266 was observed on 12 November 2019 with the adaptiveoptics-assisted ShARCS camera (McGurk et al. 2014;Gavel et al. 2014) on the Shane 3-m telescope at Lick Observatory. We collected observations in both the Ksand Jfilters with exposure times of 6 and 12 s, respectively. Observations were performed with a four-point dither pattern, with the distance between subsequent exposures being 4.0000 on a side. We reduced our data with SImMER (Savel, Hirsch et al., in prep.), an open-source, Python-based pipeline5. Prior to aligning images, the pipeline implements standard dark-subtraction and flat-fielding. To align our science images for each target, we adapted methods from Morzinski et al. (2015), performing rotations about points within a search radius and minimising the summed residuals from the original image. To determine our sensitivity to undetected stellar companions to TOI-1266, we calculated the minimum detectable companion brightness at increasing angular separations from the target. We performed this step by constructing concentric annuli centred on TOI-1266 and determining the mean and standard deviation of the flux within each annulus (e.g. Marois et al. 2006,Nielsen et al. 2008, 4https://github.com/jlillo/astrasens 5https://github.com/arjunsavel/SImMER A49, page 4 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 0.1 1 Angular separation (arcsec) 2.0 2.5 3.0 3.5 4.0 4.5 Contrast (mag) 10% cont. 5% cont. 2% cont. TOI-1266 (AstraLux - SDSSz) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Angular separation (arcsec) 0 2 4 6 8 Contrast (mag) TOI-1266 (ShARCS Ks ) TOI-1266 (ShARCS J ) ( Fig. 2. Contrast curves for TOI-1266 resulting from high-angular resolution imaging. Top: 5σcontrast curve of the AstraLux image and contamination levels (horizontal dotted lines) obtained in SDSS zfilter. Bottom: 5σcontrast curves computed from ShARCS images of TOI-1266 taken in the Ks(solid line) and J(dashed line) filters. Janson et al. 2011,Wang et al. 2015). We subsequently took our contrast curve to be 5σabove the mean. Through this method, we find that we are sensitive in the KSband to companions 4 magnitudes fainter than the host beyond a separation of 0.5100 and companions 8 magnitudes fainter than the host beyond a separation of 1.6700. The full contrast curves are shown in Fig. 2. 3. Stellar characterisation 3.1. Spectroscopic analysis 3.1.1. TRES spectroscopy We used the TRES spectra to derive initial stellar parameters employing the method described in Maldonado et al. (2015). Briefly, the effective temperature, spectral type and iron abundance were computed from the measurements of pseudo-equivalent widths of several spectral features. This analysis yields Teff=3570 ±100 K, spectral type M3, and [Fe/H] =−0.03 ±0.18 dex. We also used the derived temperature and metallicity as inputs in the empirical relations by Maldonado et al. (2015) to derive the stellar mass (M?=0.48 ± 0.10 M), the radius (R?=0.47 ±0.10 R), and the surface gravity (log g=4.78 ±0.10 dex). 3.1.2. HIRES spectroscopy We analysed the HIRES spectrum with SpecMatch-Empirical (Yee et al. 2017), which classifies stars by comparing their optical spectra to a library of spectra of well-characterised stars. With this technique, we recovered Teff=3548 ±70 K, [Fe/H] = −0.24 ±0.09 dex, and R?=0.43 ±0.10 R, confirming the dwarf nature of the host star. Using the spectrum, we also ruled out double-lined spectroscopic binaries with contamination ratios down to 1% (Kolbl et al. 2014), greatly constraining the parameter space within which false positive scenarios can exist. 3.2. SED analysis As an independent check on the derived stellar parameters, we performed an analysis of the broadband spectral energy distribution (SED) together with the Gaia DR2 parallax in order to determine an empirical measurement of the stellar radius, following the procedures described in Stassun & Torres (2016); Stassun et al. (2017,2018). We obtained the BVgri magnitudes from APASS, the JHKSmagnitudes from 2MASS, the W1– W4 magnitudes from WISE, and the GGBPGRP magnitudes from Gaia (see Table 2). In addition, we obtained the NUV flux from GALEX in order to assess the level of chromospheric activity, if any. Together, the available photometry spans the full stellar SED over the wavelength range 0.2–22 µm (see Fig. 3). We performed a fit using NextGen stellar atmosphere models (Hauschildt et al. 1999), with the fitted parameters being the effective temperature (Teff) and metallicity ([Fe/H]). We set the extinction (AV) to zero due to the star being nearby. We used the Tefffrom the TIC (Stassun et al. 2018) as an initial guess. The broadband SED is largely insensitive to the surface gravity (log g), thus we simply adopted the value from the TIC. The resulting fit is satisfactory (Fig. 3) with a reduced χ2of 1.9. The best-fit parameters are Teff=3600 ±150 K and [Fe/H] = −0.5± 0.5dex. Integrating the (unreddened) model SED gives the bolometric flux at Earth of Fbol =6.72 ±0.16 ×10−10 erg s−1cm−2. Taking the Fbol and Tefftogether with the Gaia DR2 parallax, adjusted by +0.08 mas to account for the systematic offset reported by Stassun & Torres (2018), gives the stellar radius as R∗=0.420 ±0.037 R. Finally, estimating the stellar mass from the empirical relations of Mann et al. (2019) gives M∗=0.45 ± 0.03 M, which with the empirical radius measurement gives the mean stellar density as ρ?=8.7±2.3g cm−3. In a separate global fit, we used the EXOFASTv26analysis package (Eastman et al. 2019) to derive the stellar parameters. EXOFASTv2 simultaneously utilises the SED, MIST stellar models (Dotter 2016;Choi et al. 2016), Gaia DR2 parallax and enforces an upper limit on the extinction of AV=0.04185 from the Schlafly & Finkbeiner (2011) dust maps. In addition, we set Gaussian priors on Teffand [Fe/H] from the spectral analysis, presented in Sect. 3.1. The resulting EXOFASTv2 fit provides values for Teff=3533 ±45 K, R∗=0.428 ±0.012 R, and M∗=0.447 ±0.023 M. In Table 3, we present a summary of the values of the stellar parameters obtained from the different instruments and methods previously described. We also added stellar properties derived in this work. We adopt the mass and metallicity values from the SED+Mann analysis as priors in our global analysis, because it assumes only a simple Gaussian prior on the Gaia parallax, and thus, is completely empirical. Additionally, using the radial velocity determined from the HIRES data (−41.71 ±0.10 kms−1), and the proper motion and parallax from Gaia DR2, we derived Galactic spacevelocity components (U,V,W) = (3.99 ±0.03, −43.53 ±0.06, −32.36 ±0.08) kms−1, following the procedure detailed in Jofré et al. (2015). From the space-velocity components and the membership formulation by Reddy et al. (2006), we find that TOI-1266 has a probability of ∼96% of belonging to the thin disc population. 6https://github.com/jdeast/EXOFASTv2 A49, page 5 of 21 A&A 642, A49 (2020) 1 10 λ (μm) -14 -13 -12 -11 -10 log λFλ (erg s-1 cm-2) Fig. 3. Spectral energy distribution (SED) of TOI-1266. Red symbols represent the observed photometric measurements, where the horizontal bars represent the effective width of the passband. Blue symbols are the model fluxes from the best-fit NextGen atmosphere model (black). Table 3. Stellar characterisation. Parameter Value Source Teff/K3548 ±70 HIRES 3600 ±150 SED+Mann 3533 ±45 EXOFASTv2 3570 ±100 TRES [Fe/H] −0.24 ±0.09 HIRES −0.5 ±0.5 SED+Mann −0.03 ±0.18 TRES M∗/M0.45 ±0.03 Mann 0.447 ±0.023 EXOFASTv2 0.48 ±0.10 TRES R∗/R0.43 ±0.02 HIRES 0.420 ±0.037 SED+Mann 0.428 ±0.012 EXOFASTv2 0.43 ±0.10 TRES 0.42 ±0.02 This work log g/ dex 4.85 ±0.08 SED+Mann 4.826 ±0.024 EXOFASTv2 4.78 ±0.10 TRES 4.85 ±0.05 This work ρ∗/g cm−38.7 ±2.3 SED+Mann 8.04+0.66 −0.55 EXOFASTv2 8.99 ±1.25 This work Spectral type M3 TRES Fbol/ rm erg s−1cm−2(6.72 ±0.16) ×10−10 SED+Mann Notes. SED+Mann use the empirical relations of Mann et al. (2019). Parameters in bold are the adopted stellar values. We use the SED+Mann mass and metallicity as priors in our global analysis presented in Sect. 5.1.2, where we also derive the other parameters that appear in bold. 3.3. Activity We first searched for signs of stellar variability in the TESS PDC-SAP data from Sect. 2.1, as one might expect given the prevalence of activity in M dwarfs (e.g. Newton et al. 2016, 2018). We performed a visual inspection of the light curve and found no hints of rotational modulation nor evidence of flaring activity. We then used the tools provided by the Lightkurve Python package (Lightkurve Collaboration 2018) to compute the Lomb-Scargle periodogram (Scargle 1982) for all the photometric data and for the data of each sector, without detecting any significant peak that might indicate periodic variability. To check that the variability was not removed by the PDC pipeline, we also performed an independent data reduction from the TESS Target Pixel Files (TPFs). We used a circular aperture, tracking the star on each image, and we stitched the observations from sectors 14, 15, 21 and 22 together to produce a new lightcurve. We computed a Lomb-Scargle periodogram, and we additionally fit a Gaussian process model with a quasi-periodic kernel (see e.g. Aigrain et al. 2016). We did not find any periodic variation consistent with stellar variability, however. This could result from a rotation period much longer than the 120 day observation period, or a small spot coverage. Within the total ∼120 days of TESS observations across all four sectors, we did not observe any flares from the star, consistent with an inactive early M dwarf (Hawley et al. 2014). We note that the radius and inferred mass are compatible with an uninflated early M-dwarf, and the NUV photometry is consistent with an unspotted photosphere and negligible chromospheric activity. In summary, TOI-1266 appears to be an old, inactive, slightly metal-poor early M dwarf. 4. Target vetting tests 4.1. TESS pipeline data validation As a first step in the false-positive vetting, we closely examined the data validation (DV) report (Twicken et al. 2018;Li et al. 2019) combining all four sectors (14, 15, 21, and 22) provided by the SPOC pipeline. Both of the TOI-1266 planet candidates successfully passed all the tests, which includes a search for discrepancies between odd and even transit depths, as well as, the search for a shallow secondary eclipse that could both identify a possible eclipsing binary scenario. DV also includes a difference image centroid test to ensure that the source of the transit occurs on the target star, as well as, a “ghost diagnostic test” to discard scattered light as a source for the observed signal. These conclusive tests encouraged us to conduct further analysis to validate both planet candidates. 4.2. SAINT-EX ground-based photometry In the context of the TESS Follow-up Observing Program, we used SAINT-EX to perform observations of the transit of each planet to eliminate possible contamination from nearby stars included in the large TESS aperture. We identified five such sources located between 36 and 13200 from TOI-1266 within the Gaia DR2 catalogue (Brown et al. 2018), with measured delta magnitudes ranging between 4.6 and 7.1 in the z0band. Using AstroImageJ (Collins et al. 2017), we estimated that eclipse depths between 0.064 and 0.2 occurring on these stars would be necessary to create the 0.0015 deep transit observed on TOI-1266 for the c planet candidate. Visual inspection of the light curves A49, page 6 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 POSS I Blue: 1953 10′′ POSS II Blue: 1993 1′ Pan−STARRS i: 2012 1′ TESS Sector 14: 2019 1′ Fig. 4. Archival images of TOI-1266 to assess for current, unresolved blending. POSS I (1953), II (1993) and Pan-STARRS (2012) archival images around TOI-1266 with TESS field of view for sectors 14, 15, 21 and 22 superimposed. The red cross marks TOI-1266’s location at the 2020 epoch. did not show transit signatures with the predicted depths on the nearby stars but revealed a 0.0015 deep transit on the target star as expected. We repeated the same analysis for the deeper transit of the b planet candidate, and confirmed both the lack of deep transits on the other stars and the detection of a ∼0.003 transit on TOI-1266. This step of the analysis confirms that both transits occur on TOI-1266 without contamination from known Gaia DR2 sources. We also detected no wavelength dependence of the transit depths between the TESS, z0,r0,IC, and Vbandpasses (see Sect. 5.1.2). 4.3. Archival imagery Archival images are useful for investigating the background contamination of stars with non-negligible proper motion (PM), such as TOI-1266. None of the blue and red POSS1 images from 1953 at the current target location show the presence of a source that would be blended with the target at the epoch of the TESS observations (Fig. 4). TOI-1266’s PM, combined with the moderate spatial resolution of ground-based all-sky surveys, does not allow us to constrain background sources from more recent optical imagery such as POSS2, Pan-STARRS and SDSS. 4.4. Statistical validation using high-resolution imaging with TRICERATOPS & VESPA As an additional vetting step, we calculated the false positive probabilities of these TOIs using triceratops (Tool for Rating Interesting Candidate Exoplanets and Reliability Analysis of Transits Originating from Proximate Stars; Giacalone & Dressing 2020) and vespa (Validation of Exoplanet Signals using a Probabilistic Algorithm; Morton 2012,2015). To tighten the constraints we obtain from these calculations, we incorporated the contrast curves obtained from our high-resolution AO imaging. With triceratops, we compute false positive probabilities of 0.00310 and 0.08287 for TOI-1266.01 and TOI1266.02, respectively, and with vespa, we compute false positive probabilities of 0.00002 and 0.00993, respectively7. Because this is a multi-planet system, we were able to apply additional priors to these probabilities using the results of Lissauer et al. (2012a). This study uses data from the Kepler mission to estimate the fraction of planet candidates that are false positives when in 7Discrepancies in these false positive probabilities are due to differences in the calculations performed by the two tools. At the time of this writing, triceratops estimated probability using maximum likelihood estimation, while vespa determines probability by calculating marginal likelihoods. In addition, triceratops considers more false positive scenarios in its procedure. Regardless, both tools require a false positive probability <0.05 to validate a planet candidate. systems with multiple planet candidates. This is done by assuming that the expected number of targets with kfalse positives, E(k), is given by the Poisson distribution E(k)=λke−λ k!N≈RFP n Nk k!N,(1) where RFP is the false positive rate, nis the number of planet candidates, and Nis the number of targets from which the sample is drawn. The sample used in Lissauer et al. (2012a) contained n=230 planet candidates in candidate two-planet systems from N=160 171 total targets and assumed RFP =0.5. Thus, the expected number of candidate two-planet systems in which both planet candidates are false positives is E(k=2) ≈1and the expected number of candidate two-planet systems in which only one planet candidate is a false positive is (1−RFP)n N×E(k=1) ≈ 2. Therefore, the prior probability that a planet candidate in a candidate two-planet system is a false positive is (1 +2)/230 ≈ 0.01 (see also Guerrero et al. 2020). Multiplying the probabilities above by this prior, we find a false positive probability <0.01 for both TOIs using both triceratops and vespa. These probabilities are sufficiently low to rule out astrophysical false positives originating from both resolved nearby stars, and unresolved stars blended within the photometric aperture. In conclusion, the combination of TESS photometry, high-resolution spectroscopy, high-precision ground-based photometry, and archival imagery enable us to rule out false-positive scenarios for the observed transit signals of TOI-1266 b and c. 4.5. Possible bound stellar companions Thanks to our data, we can place some limited constraints on the presence of a binary companion. Our high-resolution imaging in Fig. 2limits any object with a ∆m>4in z-band at a distance >0.200(corresponding to a semi-major axis of 7.2 au, and orbital period of 25yr). Using the Baraffe et al. (2015) models, and assuming a system age of 5 Gyr, we can exclude any binary companion with a mass >0.10 M. Using the K-band observations, we can exclude the presence of any Hydrogen-burning object beyond 100 (>0.07 M). Inside of 7.2 au, we have to rely on spectroscopy to place any constraints. A companion could in principle be detected as an extra set of lines, which requires two conditions: it is bright enough, and its orbital velocity is distinct enough from the TOI-1266 A, the primary star. Our HIRES spectrum reached a S/N of 80. Using models from Baraffe et al. (2015) once more, stellar companions with masses >0.15 Mcould be detectable with S/N= 5. The velocity resolution of HIRES is ∼6 km s−1. All companions with orbital velocity <12 km s−1would see A49, page 7 of 21 A&A 642, A49 (2020) Fig. 5. TESS data of TOI-1266 from the four sectors in which it was observed. In all cases, the black line corresponds to the PDC-SAP fluxes obtained from SPOC pipeline, the solid-orange line corresponds to the best-detrended model, and the teal line is the final detrended lightcurve. The red triangles mark the 10.8 d period planetary candidate, while the blue triangles mark the 18.8 d candidate. their absorption lines blending with the primary’s for the majority of their orbital motion, meaning they would likely remain unnoticed. An orbital velocity difference of >12 km s−1, corresponds to orbital separations of <3.8 au (period ∼10.7yr), for a secondary mass of 0.15 M, and sin i=1. Using dynamical arguments, we can also exclude the presence of a secondary star at orbital separations <0.3 au (orbital periods around 0.2yr) or TOI-1266 c would be unstable for any orbital inclination (Holman & Wiegert 1999). Similarly, a stable and circular inner planetary system implies that an external companion has not produced Lidov-Kozai cycles that would excite the eccentricities of the planetary orbits (Lidov 1962;Kozai 1962;Wu et al. 2007). Using the Lidov-Kozai timescale derived in Kiseleva et al. (1998), we find that companions with a mass of >0.05 M, eccentricities of 0.3, and orbital periods of 1000 yr (∼82 au), would induce cycles in the eccentricity of TOI-1266 c, on timescales much shorter than the age of the system (21 Myr for the parameters we quote), but only if that companion has an orbital inclination between 40 and 140◦. In summary, we can exclude detecting the presence within the data of most stellar companions, except at orbital separations between 3.8 and 7.2 au. However, we cannot exclude any companion should that object be within the line of sight (i.e. high-angular resolution imaging does not detect it, and with a relative radial velocity near 0 km s−1). Radial velocities would not detect face-on orbits, however those are ruled out using dynamical arguments. 5. Results 5.1. Photometric analysis 5.1.1. Planet search and detection limits from the TESS photometry As mentioned previously, TOI-1266 was observed by TESS in sectors 14, 15, 21 and 22. The TESS Science Office issued two alerts for this object based on SPOC DV reports; TOI 1266.01 and TOI 1266.02 correspond to planetary candidates with periods of 10.8 and 18.8 d respectively. We performed our own search for candidates using the SHERLOCK8 (Searching for Hints of Exoplanets fRom Lightcurves Of spaCe-based seeKers) pipeline presented in Pozuelos et al. (2020). This pipeline makes use of the Lightkurve package (Lightkurve Collaboration 2018), which downloads the PDCSAP flux data from the NASA Mikulski Archive for Space Telescope (MAST), and removes outliers defined as data points >3σabove the running mean. In order to remove stellar noise and instrumental drifts, our pipeline uses WOTAN (Hippke et al. 2019) with two different detrending methods: bi-weight and Gaussian process with a Matérn 3/2-kernel. In both cases, a number of detrending approaches were applied by varying the window and kernel sizes to maximise the signal detection efficiency (SDE) of the transit search, which was performed by means of the TRANSIT LEAST SQUARES package (Hippke & Heller 2019). The TRANSIT LEAST SQUARES uses an analytical transit model based on the stellar parameters, and is optimised for the detection of shallow periodic transits. We properly recovered the two aforementioned candidates, where the best systematics model corresponded to the bi-weight method with a window-size of 0.4149 d. In addition to these two signals, we found a threshold-crossing event with a period of 12.5 d. However, after an in-depth vetting process (Heller et al. 2019), we discarded this signal, which we attributed to systematics in the data set. We show in Fig. 5the PDC-SAP flux, the best systematics model, and the final detrended lightcurve with the two planet candidates. In order to assess the detectability of other planets in the data set available from TESS, we performed an injection-recovery test, where we injected synthetic planetary signals into the PDCSAP fluxes corresponding to planets with different radii and 8SHERLOCK code is available upon request. A49, page 8 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 Fig. 6. Injection-and-recovery test performed to check the detectability of extra planets in the system. We explored a total of 1305 different scenarios. Larger recovery rates are presented in green and yellow colours, while lower recovery rates are shown in blue and darker hues. Planets smaller than 1.5 R⊕would remain undetected for almost the full set of periods explored. periods. We then detrended the lightcurve using the best method found previously that is the bi-weight approach with a window size of 0.4149 d. Before searching for planets, we masked the two known candidate planets with periods of 10.8 and 18.8 d. We explored the Rplanet–Pplanet parameter space in the ranges of 0.8–3.0 R⊕with steps of 0.05 R⊕, and 1–30 d with steps of 1 d, for a total of 1305 different scenarios. In this test, we defined an injected signal as being recovered when a detected epoch matched the injected epoch within one hour, and if a detected period matched any half-multiple of the injected period to better than 5%. It is important to note that since we injected the synthetic signals directly in the PDC-SAP lightcurve, they were not affected by the PDC-SAP systematic corrections. Therefore, the detection limits found correspond to the most optimistic scenario (see e.g. Pozuelos et al. 2020;Eisner et al. 2020). The results are shown in Fig. 6, and we reached several conclusions from this test: (1) we can rule out the presence of planets with sizes larger than ∼1.5 R⊕with orbits shorter than 10 d. However, such planets might reside in orbits with longer periods and, thus, remain undetected, as a longer period yields a lower detectability. In fact, for periods greater than 10 d, we obtained a recovery rate ranging from 30 to 70%. (2) For the full set of investigated periods, planets with sizes smaller than 1.5 R⊕would remain undetected, with recovery rates lower than 30%, and close to 0% for 1.0 R⊕. 5.1.2. Global analysis We used the full photometric dataset described in Sect. 2along with the transit signals detected in the previous step as input parameters to a global analysis of the TOI-1266 system. We also included the stellar mass and effective temperature derived in Sect. 3as Gaussian priors to convert the transit fitted parameters into physical values. We used the MCMC algorithm implementation already presented in the literature (e.g. Gillon et al. 2012,2014;Demory et al. 2012). The inputs to the MCMC are the photometric time-series obtained during the data reduction described above. For each light curve, we fit simultaneously for the instrument baseline model and a transit model of two Keplerian orbits corresponding to TOI-1266 b and c. This approach ensures that instrumental systematic noise is properly propagated to the system parameters of interest. The photometric baseline model coefficients used for detrending for each instrument are determined at each step of the MCMC procedure using a singular value decomposition method (Press et al. 1992). The resulting coefficients are then used to correct the raw photometric lightcurves. Such an approach is necessary because the input data originate from multiple instruments with different sources of systematics. We show in Table 4the baseline model used for each light curve. To derive accurate uncertainties on the system parameters, we computed two scaling factors, βwand βr, following Winn et al. (2008), to account for overor under-estimated white noise and correlated noise (Pont et al. 2006) in each dataset (these values are computed over 10 to 240-min timescales). We computed the quadratic limb-darkening (LD) coefficients u1and u2in the TESS, z0,r0,IC, and Johnson Vfilters, using the PyLDTk code (Parviainen & Aigrain 2015) and a library of PHOENIX high-resolution synthetic spectra (Husser et al. 2013). We placed Gaussian priors on each of the quadratic LD parameters, with a 5-fold inflation of the uncertainties computed from model interpolation. All LD parameters used in this analysis are shown in Table 5. For each of the two planets, we fitted for (1) the transit depth (planet-to-star area ratio) R2 P R2 ? for each instrument to assess transit depth chromaticity, (2) the transit duration T14, (3) the orbital period P, (4) the transit centre T0, and (5) the impact parameter b=acos i R?, where ais the orbital semi-major axis and ithe orbital inclination. For this MCMC fit, we assumed circular orbits, and fixed √ecos ωand √esin ωvalues to 0. We ran two chains of 100 000 steps (including 20% burn-in) each, and checked their efficient mixing and convergence by visually inspecting the autocorrelation functions of each chain, and by using the Gelman & Rubin (1992) statistical test, ensuring that the test values for all fitted parameters were <1.01. We show in Table 6the median and 1σcredible intervals of the system parameter’s posterior distribution functions. The corresponding light curves for both planets are shown in Fig. 7–10. Our analysis shows a good agreement between the combined stellar density value of ρ?=8.99 ±1.25 g cm−3derived from the photometry alone, and (1) the ρ?=8.7±2.3g cm−3derived from the SED+Mann analysis, as well as, (2) the ρ?=8.04+0.66 −0.55 g cm−3 derived from the EXOFASTv2 analysis described in Sect. 3. The transit depths measured in four different bandpasses (TESS, z0, r0, and V) for TOI-1266 b are all in agreement at the ∼2-σlevel. We find a good agreement as well for the transit depth of TOI1266 c albeit with only two bandpasses (TESS and z0). We also repeated the same MCMC analysis, this time allowing √ecos ω and √esin ωto vary, but did not find evidence for eccentric orbits for any of TOI-1266 b and c, using transit photometry alone. 5.2. Dynamical analysis 5.2.1. Mass constraints from TTVs As the system is within 6% of the second order 5:3 mean motion resonance (MMR) and within 14% of the stronger first order 2:1 MMR, we attempt to constrain planet masses from the transit timing variations (Agol et al. 2005;Holman 2005; Agol & Fabrycky 2017) measured in our combined photometric dataset. We have in total 13 transits of planet b and five A49, page 9 of 21 A&A 642, A49 (2020) 2.5 5.0 7.5 10.0 12.5 15.0 17.5 Mass [M E ] 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Radius [R E ] 100% H 2 O 32% Fe + 68% Rock 16% Fe + 34% Rock + 50% H 2 O 100% Fe VE UN TOI-1266 b TOI-1266 c Venus Earth Uranus Neptune Fig. 16. TOI-1266b and c on the mass-radius diagram based on the radii and masses derived from the transit photometry and TTV analyses. Alongside, we show mass radius curves for pure iron (brown), an Earth like composition of 32% iron, and 68% rock (red), a mixture of 50% water with an Earth like core (grey blue), and for pure water (blue). The data of the background planets were taken from the NASA Exoplanet Archive (retrieved on 2020 July 15). We show only those planets that have a relative mass and radius uncertainty of less than 20%. Taken together, our results suggest that it is likely that the outermost planet TOI-1266 c is tidally locked, but due to its small but non-zero eccentricity, which may persist through MMR excitations, it may not have the same side always facing the star. In a recent paper by Pozuelos et al. (2020), the authors adapted the general description of the flux received by a planet given by Kane & Torres (2017) for the configuration found here, for a tidally-locked planet on a non-circular orbit. It was found that in such a configuration, the orbital phase is no longer relevant, and the distribution of flux along the latitude (β) is the same for the whole orbit, which depends only on the planet’s eccentricity. In our case, assuming the nominal value of eccentricity found in our models (i.e. 0.04) we found 3805– 3242 W m−2(2.79–2.37 F⊕) at the equator, 2691–2293 W m−2 (1.97–1.68 F⊕) at β=±45◦, and ∼0 W m−2(0.0 F⊕) at the poles. These values suggest that the stellar flux received by the planet may vary by ∼15%along its orbit. Based on dynamical arguments, this variation of ∼15%may be considered as an upper limit. Indeed, we demonstrated that the nominal values of the eccentricities might be considered as upper limits, where the system tends to more stable scenarios for low eccentricities (see Fig. 13). At the 1-σuncertainty level, the minimum eccentricity of TOI-1266 c is 0.01. In such a case, the variation of the flux along the orbit is only ∼4%. Hence, a plausible set of values for the variation of the flux along the orbit ranges from 4 to 15%. 6. Discussion 6.1. TOI-1266 b and c in the mass-radius diagram In this section, we use the mass and radii constraints on TOI-1266 b and c derived from the transit photometry to investigate the location of these planets in the mass-radius diagram. We emphasise that the mass constraints will be improved once additional transit timings, or precise radial-velocity timeseries become available. However, Fig. 16 provides a first assessment on whether the derived planet properties are consistent with usual mass-radius relationships. In this figure, the composition curves were calculated with model B of Michel et al. (2020). We used for iron the equation of state of Hakim et al. (2018), for rock the model of Sotin et al. (2007) and for water the equation of state of Mazevet et al. (2019). 6.2. TOI-1266 and the radius valley With radii of about 2.4 and 1.6 R⊕, TOI-1266 b and c span the so-called radius valley (Fulton et al. 2017). Since atmospheric evaporation is a possible explanation for the origin of the valley (Owen & Wu 2013;Lopez & Fortney 2013;Jin et al. 2014), this feature is also known as the “evaporation valley”. We note, however, that the origin for the radius valley feature is currently unconstrained and that other studies advocate, for instance, for core-powered mass loss as the driving mechanism for this pattern A49, page 16 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 (Ginzburg et al. 2018;Gupta & Schlichting 2020). The recent discovery of a change in the fraction of planets above and below the valley over ∼Gyr timescales has indeed been interpreted in favour of the core-powered mass loss scenario for some solartype stars (Berger et al. 2020), although there is evidence that the formation pathway may be different in low-mass stars (Cloutier & Menou 2020). Planets in multiple systems with dissimilar radii, like TOI1266 b & c, Kepler-36 b & c (Carter et al. 2012) or TOI-402b & c (Dumusque et al. 2019), are of particular interest to understand the origin of the radius valley, as they allow to study the differential evolutionary history of the planets, and to check whether evaporation can self-consistently explain all the planets in a system (Lopez et al. 2012;Lopez & Fortney 2013;Owen & Campos Estrada 2020). To understand the implication of the hypothesis that evaporation has shaped the distinct radii of TOI-1266 b & c, we simulated their long-term thermodynamical evolution (cooling, contraction, atmospheric escape) with the Bern evolution model completo21 (Mordasini et al. 2012;Jin et al. 2014;Jin & Mordasini 2018). The model simulates the long-term evolution of the planets after the dissipation of the protoplanetary disc by solving the internal structure equations of the planets. It includes XUV-driven atmospheric photoevaporation in the radiationrecombination and energy-limited regimes (Murray-Clay et al. 2009). The planets consist of a solid core and a H/He envelope, and we assumed that their cores have an Earth-like 2:1 silicate:iron composition described by the polytropic equation of state (EOS) of Seager et al. (2007). The gaseous envelope consists of H/He, described by the EOS of Saumon et al. (1995), and the opacity corresponds to a condensate-free solar-composition gas (Freedman et al. 2014). As in Mordasini (2020), we simulated the evolution of 6000 planets on a grid of semi-major axis and mass. The initial conditions (i.e. the post-formation envelope – core mass ratio and luminosity) are the same as in the nominal case considered in Mordasini (2020), but the stellar mass is now 0.5 M. The stellar XUV luminosity as a function of time was taken from McDonald et al. (2019). Figure 17 shows the result in the plane of orbital distance versus planet radius at an age of 10 Gyr, where the evaporation valley is apparent. Planets above this threshold retain some H/He, while those below become bare rocky cores. In the top left corner, the region devoid of planets corresponds to the subNeptunian desert (Lundkvist et al. 2016;Mazeh et al. 2016; Bourrier et al. 2018). TOI-1266 b & c are also shown as black squares. Planet b is located clearly above the valley, while planet c’s median radius sits just underneath. This position corresponds to the most massive (and largest) cores at a given semi-major axis that have lost their H/He. The interesting property of TOI-1266 is that the inner planet is larger than the outer one. In the context of evaporation, this can be explained if the inner planet is also more massive than the outer one. The higher mass allows for the planet to keep its H/He envelope even at a higher XUV flux. We thus studied the masses of model planets that are compatible with TOI-1266 b & c in terms of orbital distance and radius in Fig. 17, and find that these model planets have masses of about 10 ±1.5M⊕for planet b, and 6+1 −2M⊕for planet c, which is consistent with the masses derived using our TTV measurements (Sect. 5.2.1). These specific numbers depend on model assumptions, such as the core 1 1.5 2 2.5 3 3.5 4 4.5 0 0.05 0.1 0.15 0.2 R [Re] a [AU] 0 5 10 15 20 25 Mass [Me] Fig. 17. Comparison of TOI-1266 b & c with evaporation models. The coloured dots show the position in the plane of semi-major axis versus radius of simulated planets evolving under the effect of cooling, contraction, and atmospheric escape around a 0.5 Mstar at an age of 10 Gyr. One sees the radius valley and the sub-Neptunian desert. The colours show the planet mass. Black symbols show TOI-1266 b & c. and envelope composition, the initial conditions, or the evaporation model. But the general result, namely that the inner planet should be more massive, is robust. We note that Fig. 17 has been obtained with an evaporation model that assumes a constant evaporation efficiency in the energy limited domain. It is known (Owen & Wu 2017) that this yields a steeper slope (R vs a) for the evaporation valley than the one found in models which calculate the evaporation efficiency self-consistently (Owen & Jackson 2012). This shallower slope is also in better agreement with observations (Mordasini 2020). This affects the prediction for the mass ratios of the two planets; in particular, a shallow slope means that the two masses can be more similar than estimated above. If we assume that the slope of the valley is instead the same as the slope observed for FGK stars (Van Eylen et al. 2018;Cloutier & Menou 2020), and not the one predicted in the model, then we can estimate that the inner planet must have a mass that is at least 25% higher than the mass of the outer planet (Mordasini 2020). Then, the inner planet can keep some H/He. 6.3. Potential for atmospheric characterisation The potential for atmospheric characterisation of an exoplanetary system relates directly to the size and brightness of the host: the smaller and brighter the host, the larger the signal in transmission and photon count (i.e. S/N), all other parameters being equal. TOI-1266’s relatively small size and proximity makes it a remarkably good host for the atmospheric study of super-Earth-sized and larger planets. In order to quantify and contextualise its prospect for atmospheric study, we followed the same approach as in Gillon et al. (2016), focusing here on temperate sub-Neptune-sized planets following the NASA Exoplanet Archive10. Figure 18 reports the planets’ signal in transmission, which we derive as follows: S=2Rpheff R2 ∗ ,with heff=7kT µg ,(4) 10 https://exoplanetarchive.ipac.caltech.edu A49, page 17 of 21 A&A 642, A49 (2020) 150 200 250 300 350 400 Equilibrium Temperature (K) 100 101 102 103 Expected Transmission Signal (ppm) Kepler-62 f Kepler-62 e Kepler-150 f Kepler-186 f Kepler-296 e Kepler-1652 b Kepler-438 b Kepler-436 b Kepler-1653 b Kepler-69 c Kepler-1649 b Kepler-441 b K2-72 e Kepler-61 b EPIC 212737443 c K2-3 d Kepler-446 d K2-72 c Kepler-445 d Kepler-22 b Kepler-437 b K2-118 b Kepler-296 f K2-72 d K2-239 d Kepler-296 d K2-239 c K2-264 c EPIC 211822797 b K2-152 b K2-286 b TRAPPIST-1 g TRAPPIST-1 h TRAPPIST-1 f L 98-59 d TRAPPIST-1 e Kepler-445 c K2-18 b K2-288 B b K2-3 c TRAPPIST-1 d TRAPPIST-1 c LTT 1445 A b TOI-270 d GJ 143 b LP 791-18 c LHS 1140 b 3.0 R⊕ 1.0 R⊕ TRAPPIST-1 b TOI-1266 c TOI-1266 b −1.5 −1.0 −0.5 0.0 0.5 Log10 of Relative S/N Fig. 18. Most promising sub-Neptune-sized planets for atmospheric characterisation. Point colours illustrate the S/N of a JWST/NIRSPEC observation relative to TRAPPIST-1 b. S/N below 1/100th of TRAPPIST-1 b have been removed to enhance readability of the figure. The planets for which the presence of an atmosphere could be assessed by JWST within ∼100h are encircled in red, if their atmospheric signals are above JWST’s threshold of ∼50 ppm. The size of the circle is proportional to the size of the planet. where Rpis the planetary radius, R∗is the stellar radius, heffis the effective atmospheric height, µis the atmospheric mean molecular mass, Tis the atmospheric temperature, and gis the local gravity. We assume heffto cover seven atmospheric scale heights, assuming atmospheres down to ∼0.1bar. We assume the atmospheric mean molecular mass to be 2.3 amu for planets larger than 1.6 R⊕, which we model as sub-Neptunes, and 20 amu for the smaller planets, which we model as terrestrial. We assume the atmospheric temperature to be the equilibrium temperature for a Bond albedo of 0. For the planets with missing masses, we estimated gusing the statistical model from Chen & Kipping (2017). TOI-1266 b and c’s projected signals in transmission are ∼220 and ∼100 ppm, which are significantly above JWST’s plausible noise floor level (20–50 ppm; Greene et al. 2016). In order to quantify this further, we also report in Fig. 18 a relative S/N scaling the signal amplitude with the hosts’ brightness in the J band and using TRAPPIST-1 b’s S/N as a reference. We find that TOI-1266’s planets compare favourably to TRAPPIST-1b in terms of potential for atmospheric exploration. In fact, assessing the presence of a clear hydrogen-dominated atmosphere as assumed for the purpose of this discussion would require 1 and 4 transits for planet b and c, respectively, with JWST NIRSpec Prism mode. However, this preliminary assessment could be complicated by at least two factors. A first obstacle of transmission spectroscopy is refraction, which bends starlight away from the line of the sight to the observer. In a transmission spectrum, this has a similar effect to an opaque cloud with a “cloud-top pressure” (Sidis & Sari 2010), which introduces a spectral continuum that mutes the strength of spectral features. To determine if this effect is significant, we use Eq. (14) and Table 1 of Robinson et al. (2017) to estimate the pressure corresponding to this refraction continuum. For nitrogen-dominated atmospheres, we estimate this pressure to be 1.3 and 0.8 bar for TOI-1266 b and TOI1266 c, respectively, if we assume g∼103cm s−2and T∼Teq. For carbon dioxide-dominated atmospheres, we estimate this pressure to be 0.7 and 0.4 bar for TOI-1266 b and TOI-1266 c, respectively. Since transmission spectroscopy probes pressures ∼1–10 mbar or lower, we therefore expect refraction to have a negligible effect on the shape of the transmission spectra for TOI-1266 b and TOI-1266 c. Given the low equilibrium temperatures, a second obstacle may be the presence of clouds or hazes that are likely to strongly shape the transmission spectra of these weakly-irradiated exoplanets (see e.g. Crossfield & Kreidberg 2017, and references therein). We finally note that the planets of TOI-1266 are attractive targets for transmission spectroscopy due to the small geometric size of and lack of activity of the host star. More active M dwarfs like TRAPPIST-1 pose significant challenges to transmission spectroscopy, since the stellar surface features emit different spectra from the mean photosphere, introducing a degenerate signal that one must disentangle to correctly identify the signal from the exoplanet atmosphere (Morris et al. 2018a,b,c;Ducrot et al. 2018;Wakeford et al. 2019). TOI-1266 has no photometric evidence of these confounding starspot signatures, and therefore, makes a clean case for transmission spectroscopy. 6.4. Prospects for radial velocity follow-up and additional TESS data Radial velocities (RVs) are likely to become available for this target in the future. It is critical to obtain precise masses and thus bulk densities for planets with sub-Neptune radii to identify if A49, page 18 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 they are evaporated giants or giant rocks, so that we may identify model atmospheres to apply to the transmission spectroscopy. This requirement for a reliable mass constraint will be one of the fundamental limitations in choosing sub-Neptune targets for observations with JWST (Batalha et al. 2017). For this system, we might have some certainty about whether or not these are evaporated giants, which will strengthen our interpretations of the transmission spectra. In addition, we might we be able to measure the RossiterMcLaughlin effect for this system, due to its brightness and lack of confounding stellar activity. The obliquity of the system would be a valuable addition to the recent observation by Hirano et al. (2020) that the TRAPPIST-1 planets are well-aligned with the host star’s spin. We finally note that as of writing, TOI-1266 will not be observed by TESS during Sectors 23-26 nor during Extended Mission 1. 7. Conclusions This study reports on the discovery and preliminary characterisation of the TOI-1266 system that hosts a super-Earth and a sub-Neptune around a M3 dwarf. Our analysis combines photometry obtained from spaceand ground-based facilities with a careful treatment of instrumental systematics and correlated noise for each dataset. The resulting data enable us to compute preliminary mass measurements for both planets, investigate their tidal evolution and search for additional companions in the system. Along with other recently discovered TESS exoplanets (e.g. Cloutier et al. 2020;Cloutier & Menou 2020), TOI-1266 will likely become a key system to better understand the nature of the radius valley around early to mid-M dwarfs. First, its orbital architecture, influenced by the 2:1 mean-motion resonance, will facilitate the measurement of precise planetary masses with high-precision photometry (TTV) and Doppler spectroscopy (radial velocities). Second, the host brightness is such that the system will be observable in most JWST modes, hence providing a large wavelength coverage. Third, the outer planet TOI-1266 c has an irradiation level that is similar to that of Venus and is also a favourable target for atmospheric characterisation. We also note that TOI-1266 c may be tidally-locked but with a non-circular orbit, resulting in incident stellar flux varying at a few percent level only along its orbit. Such a configuration would also lead to more homogeneous longitudinal temperature differences, in stark contrast with the bulk of small transiting exoplanets discovered so far. Acknowledgements. We warmly thank the entire technical staff of the Observatorio Astronómico Nacional at San Pedro Mártir in México for their unfailing support to SAINT-EX operations, namely: U. Ceseña, A. Córdova, B. García, C.A. Guerrero, F. Guillén, J.A. Hernández, B. Hernández, E. López, B. Martínez, G. Melgoza, F. Montalvo, S. Monrroy, J.C. Narvaez, J.M. Nuñez, J.L. Ochoa, I. Plauchú, F. Quiroz, H. Serrano, T. Verdugo. We gratefully acknowledge the support from the Embassy of México in Bern to the SAINT-EX project. We also thank the past members of the SAINT-EX team B. Courcol, E. Rose and K. Housen for their help in the course of the project. We are grateful to the anonymous referee for a thorough and helpful review of our paper. We thank Jonathan Irwin for his help on the TRES data analysis. B.-O.D. acknowledges support from the Swiss National Science Foundation (PP00P2-163967). This work has been carried out within the frame of the National Centre for Competence in Research PlanetS supported by the Swiss National Science Foundation. Y.G.M.C acknowledges support from UNAM-PAPIIT IN-107518. R.P. and E.J. acknowledge DGAPA for their postdoctoral fellowships. The research leading to these results has received funding from the ARC grant for Concerted Research Actions, financed by the Wallonia-Brussels Federation. TRAPPIST is funded by the Belgian Fund for Scientific Research (Fonds National de la Recherche Scientifique, FNRS) under the grant FRFC 2.5.594.09.F, with the participation of the Swiss National Science Fundation (SNF). TRAPPIST-North is a project funded by the University of Liege, and performed in collaboration with Cadi Ayyad University of Marrakesh. M.G. and E.J. are F.R.S.-FNRS Senior Research Associates. J.C.S. acknowledges funding support from Spanish public funds for research under projects ESP2017-87676-2-2 and RYC-2012-09913 (Ramón y Cajal programme) of the Spanish Ministry of Science and Education. This paper includes data collected by the TESS mission. We acknowledge the use of public TOI Release data from pipelines at the TESS Science Office and at the TESS Science Processing Operations Center. Funding for the TESS mission is provided by the NASA Explorer Program. Resources supporting this work were provided by the NASA High-End Computing (HEC) Program through the NASA Advanced Supercomputing (NAS) Division at Ames Research Center for the production of the SPOC data products. This work is based upon observations carried out at the Observatorio Astronómico Nacional on the Sierra de San Pedro Mártir (OAN-SPM), Baja California, México. This research has been partly funded by the Spanish State Research Agency (AEI) Projects No.ESP2017-87676-C5-1-R and No. MDM-2017-0737 Unidad de Excelencia “María de Maeztu”- Centro de Astrobiología (INTA-CSIC). This paper is based on observations collected at Centro Astronómico Hispano en Andalucía (CAHA) at Calar Alto, operated jointly by Instituto de Astrofísica de Andalucía (CSIC) and Junta de Andalucía. The research leading to these results has received funding from the European Research Council under the European Union’s Seventh Framework Programme (FP/2007-2013) ERC Grant Agreement no336480 (SPECULOOS). D.H. acknowledges support from the Alfred P. Sloan Foundation, the National Aeronautics and Space Administration (80NSSC18K1585, 80NSSC19K0379), and the National Science Foundation (AST-1717000). We finally acknowledge support from the Antoine de Saint-Exupéry Youth Foundation through their involvement in the educational programme of the SAINT-EX Observatory. References Agol, E., & Fabrycky, D. C. 2017, Handbook of Exoplanets (Cham: Springer), 7 Agol, E., Steffen, J., Sari, R., & Clarkson, W. 2005, MNRAS, 359, 567 Aigrain, S., Parviainen, H., & Pope, B. J. S. 2016, MNRAS, 459, 2408 Alam, S., Albareti, F. D., Allende Prieto, C., et al. 2015, ApJS, 219, 12 Aller, A., Lillo-Box, J., Jones, D., Miranda, L. F., & Barceló Forteza, S. 2020, A&A, 635, A128 Astropy Collaboration (Robitaille, T. P., et al.) 2013, A&A, 558, A33 Astropy Collaboration (Price-Whelan, A. M., et al.) 2018, AJ, 156, 123 Baraffe, I., Homeier, D., Allard, F., & Chabrier, G. 2015, A&A, 577, A42 Barkaoui, K., Burdanov, A., Hellier, C., et al. 2019, AJ, 157, 43 Barnes, R. 2017, Celest. Mech. Dyn. Astron., 129, 509 Batalha, N. E., Kempton, E. M.-R., & Mbarek, R. 2017, ApJ, 836, L5 Berger, T. A., Huber, D., Gaidos, E., van Saders, J. L., & Weiss, L. M. 2020, AJ, 160, 108 Blanco-Cuaresma, S., & Bolmont, E. 2017, EWASS Special Session 4 (2017): Star-planet interactions (EWASS-SS4-2017) Bolmont, E., Raymond, S. N., von Paris, P., et al. 2014, ApJ, 793, 3 Bolmont, E., Raymond, S. N., Leconte, J., Hersant, F., & Correia, A. C. M. 2015, A&A, 583, A116 Bourrier, V., Lecavelier des Etangs, A., Ehrenreich, D., et al. 2018, A&A, 620, A147 Bradley, L., Sip˝ ocz, B., Robitaille, T., et al. 2019, astropy/photutils: v0.6 Brown, A. G. A., Vallenari, A., Prusti, T., et al. 2018, A&A, 616, A1 Buchhave, L. A., Bakos, G. Á., Hartman, J. D., et al. 2010, ApJ, 720, 1118 Carter, J. A., Agol, E., Chaplin, W. J., et al. 2012, Science, 337, 556 Chambers, J. E. 1999, MNRAS, 304, 793 Chen, J., & Kipping, D. 2017, ApJ, 834, 17 Choi, J., Dotter, A., Conroy, C., et al. 2016, ApJ, 823, 102 Cincotta, P., & Simó, C. 1999, Celest. Mech. Dyn. Astron., 73, 195 Cincotta, P. M., & Simó, C. 2000, A&AS, 147, 205 Cincotta, P. M., Giordano, C. M., & Simó, C. 2003, Phys. D Nonlinear Phenom., 182, 151 Cloutier, R., & Menou, K. 2020, AJ, 159, 211 Cloutier, R., Eastman, J. D., Rodriguez, J. E., et al. 2020, AJ, 160, 3 Collins, K. A., Kielkopf, J. F., Stassun, K. G., & Hessman, F. V. 2017, AJ, 153, 77 Crossfield, I. J. M., & Kreidberg, L. 2017, AJ, 154, 261 Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, 2MASS All Sky Catalog of point sources Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2013, VizieR Online Data Catalog, II/328 Delrez, L., Gillon, M., Queloz, D., et al. 2018, SPIE Conf. Ser., 10700, 107001I Demory, B.-O., Gillon, M., Seager, S., et al. 2012, ApJ, 751, L28 Dorn, C., Venturini, J., Khan, A., et al. 2017, A&A, 597, A37 Dotter, A. 2016, ApJS, 222, 8 A49, page 19 of 21 A&A 642, A49 (2020) Ducrot, E., Sestovic, M., Morris, B. M., et al. 2018, AJ, 156, 218 Dumusque, X., Turner, O., Dorn, C., et al. 2019, A&A, 627, A43 Eastman, J. D., Rodriguez, J. E., Agol, E., et al. 2019, PASP, submitted [arXiv:1907.09480] Eggleton, P. P., Kiseleva, L. G., & Hut, P. 1998, ApJ, 499, 853 Eisner, N. L., Barragán, O., Aigrain, S., et al. 2020, MNRAS, 494, 750 Freedman, R. S., Lustig-Yaeger, J., Fortney, J. J., et al. 2014, ApJS, 214, 25 Fried, D. L. 1978, J. Opt. Soc. Am. 1917-1983, 68, 1651 Fulton, B. J., Petigura, E. A., Howard, A. W., et al. 2017, AJ, 154, 109 Furesz, G. 2008, PhD thesis, University of Szeged, Hungary Gaia Collaboration (Brown, A. G. A., et al.) 2018, A&A, 616, A1 Gavel, D., Kupke, R., Dillon, D., et al. 2014, Proc. SPIE, 9148, 914805 Gelman, & Rubin. 1992, Stat. Sci., 7, 457 Giacalone, S., & Dressing, C. D. 2020, AAS J., submitted [arXiv:2002.00691] Gillon, M., Triaud, A. H. M. J., Fortney, J. J., et al. 2012, A&A, 542, A4 Gillon, M., Anderson, D. R., Collier-Cameron, A., et al. 2013, A&A, 552, A82 Gillon, M., Demory, B.-O., Madhusudhan, N., et al. 2014, A&A, 563, A21 Gillon, M., Jehin, E., Lederer, S. M., et al. 2016, Nature, 533, 221 Ginzburg, S., Schlichting, H. E., & Sari, R. 2018, MNRAS, 476, 759 Greene, T. P., Line, M. R., Montero, C., et al. 2016, ApJ, 817, 17 Grimm, S. L., & Stadel, J. G. 2014, ApJ, 796, 23 Grimm, S. L., Demory, B.-O., Gillon, M., et al. 2018, A&A, 613, A68 Guerrero, et al. 2020, AAS J., submitted Günther, M. N., Pozuelos, F. J., Dittmann, J. A., et al. 2019, Nat. Astron., 3, 1099 Gupta, A., & Schlichting, H. E. 2020, MNRAS, 493, 792 Hakim, K., Rivoldini, A., Van Hoolst, T., et al. 2018, Icarus, 313, 61 Hauschildt, P. H., Allard, F., & Baron, E. 1999, ApJ, 512, 377 Hawley, S. L., Davenport, J. R. A., Kowalski, A. F., et al. 2014, ApJ, 797, 121 Heller, R., Hippke, M., & Rodenbeck, K. 2019, A&A, 627, A66 Hinse, T. C., Haghighipour, N., Kostov, V. B., & Gózdziewski, K. 2015, ApJ, 799, 88 Hippke, M., & Heller, R. 2019, A&A, 623, A39 Hippke, M., David, T. J., Mulders, G. D., & Heller, R. 2019, AJ, 158, 143 Hirano, T., Gaidos, E., Winn, J. N., et al. 2020, ApJ, 890, L27 Holman, M. J. 2005, Science, 307, 1288 Holman, M. J., & Wiegert, P. A. 1999, AJ, 117, 621 Hormuth, F., Brandner, W., Hippler, S., & Henning, T. 2008, J. Phys. Conf. Ser., 131, 012051 Horner, J., Wittenmyer, R. A., Wright, D. J., et al. 2019, AJ, 158, 100 Husser, T. O., Wende-von Berg, S., Dreizler, S., et al. 2013, A&A, 553, A6 Hut, P. 1981, A&A, 99, 126 Janson, M., Bonavita, M., Klahr, H., et al. 2011, ApJ, 736, 89 Jehin, E., Gillon, M., Queloz, D., et al. 2011, The Messenger, 145, 2 Jehin, E., Gillon, M., Queloz, D., et al. 2018, The Messenger, 174, 2 Jenkins, J. M. 2002, ApJ, 575, 493 Jenkins, J. S., Jones, H. R., Go´zdziewski, K., et al. 2009, MNRAS, 398, 911 Jenkins, J. M., Twicken, J. D., McCauliff, S., et al. 2016, SPIE Conf. Ser., 9913, 99133E Jenkins, J. M., Tenenbaum, P., Seader, S., et al. 2017, Kepler Data Processing Handbook: Transiting Planet Search, Kepler Science Document (California: NASA) Jenkins, J. S., Pozuelos, F. J., Tuomi, M., et al. 2019, MNRAS, 490, 5585 Jensen, E. 2013, Astrophysics Source Code Library [record ascl:1306.007] Jin, S., & Mordasini, C. 2018, ApJ, 853, 163 Jin, S., Mordasini, C., Parmentier, V., et al. 2014, ApJ, 795, 65 Jofré, E., Petrucci, R., Saffe, C., et al. 2015, A&A, 574, A50 Kane, S. R. 2015, ApJ, 814, L9 Kane, S. R. 2017, ApJ, 839, L19 Kane, S. R. 2019, AJ, 158, 72 Kane, S. R., & Torres, S. M. 2017, AJ, 154, 204 Kiseleva, L. G., Eggleton, P. P., & Mikkola, S. 1998, MNRAS, 300, 292 Kolbl, R., Marcy, G. W., Isaacson, H., & Howard, A. W. 2014, AJ, 149, 18 Kostov, V. B., Schlieder, J. E., Barclay, T., et al. 2019, AJ, 158, 32 Kozai, Y. 1962, AJ, 67, 579 Kubyshkina, D., Cubillos, P. E., Fossati, L., et al. 2019, ApJ, 879, 26 Lang, D., Hogg, D. W., Mierle, K., Blanton, M., & Roweis, S. 2010, AJ, 139, 1782 Laskar, J., & Robutel, P. 1993, Nature, 361, 608 Latham, D. W., Rowe, J. F., Quinn, S. N., et al. 2011, ApJ, 732, L24 Leconte, J., Chabrier, G., Baraffe, I., & Levrard, B. 2010, A&A, 516, A64 Li, J., Tenenbaum, P., Twicken, J. D., et al. 2019, PASP, 131, 024506 Lidov, M. L. 1962, Planet. Space Sci., 9, 719 Lightkurve Collaboration (Cardoso, J. V. d. M., et al.) 2018, Astrophysics Source Code Library [record ascl:1812.013] Lillo-Box, J., Barrado, D., & Bouy, H. 2012, A&A, 546, A10 Lillo-Box, J., Barrado, D., & Bouy, H. 2014, A&A, 566, A103 Lissauer, J. J., & Cuzzi, J. N. 1985, in Protostars and Planets II, eds. D. C. Black, & M. S. Matthews (Tucson, AZ: University of Arizona Press), 920 Lissauer, J. J., Marcy, G. W., Rowe, J. F., et al. 2012a, ApJ, 750, 112 Lissauer, J. J., Barnes, J. W., & Chambers, J. E. 2012b, Icarus, 217, 77 Lithwick, Y., Xie, J., & Wu, Y. 2012, ApJ, 761, 122 Lopez, E. D., & Fortney, J. J. 2013, ApJ, 776, 2 Lopez, E. D., Fortney, J. J., & Miller, N. 2012, ApJ, 761, 59 Lundkvist, M. S., Kjeldsen, H., Albrecht, S., et al. 2016, Nat. Commun., 7, 11 201 Maldonado, J., Affer, L., Micela, G., et al. 2015, A&A, 577, A132 Mann, A. W., Dupuy, T., Kraus, A. L., et al. 2019, ApJ, 871, 63 Marois, C., Lafreniere, D., Doyon, R., Macintosh, B., & Nadeau, D. 2006, ApJ, 641, 556 Mathis, S. 2018, Handbook of Exoplanets (Cham: Springer), 24 Mazeh, T., Holczer, T., & Faigler, S. 2016, A&A, 589, A75 Mazevet, S., Licari, A., Chabrier, G., & Potekhin, A. Y. 2019, A&A, 621, A128 McCormac, J., Pollacco, D., Skillen, I., et al. 2013, PASP, 125, 548 McDonald, G. D., Kreidberg, L., & Lopez, E. 2019, ApJ, 876, 22 McGurk, R., Rockosi, C., Gavel, D., et al. 2014, Proc. SPIE, 9148, 91483A Michel, A., Haldemann, J., Mordasini, C., & Alibert, Y. 2020, A&A, 639, A66 Mignard, F. 1979, Moon Planets, 20, 301 Modirrousta-Galian, D., Locci, D., & Micela, G. 2020, ApJ, 891, 158 Mordasini, C. 2020, A&A 638, A52 Mordasini, C., Alibert, Y., Klahr, H., & Henning, T. 2012, A&A, 547, A111 Morris, R. L., Twicken, J. D., Smith, J. C., et al. 2017, Kepler Data Processing Handbook: Photometric Analysis, Kepler Science Document (California: NASA) Morris, B. M., Agol, E., Davenport, J. R. A., & Hawley, S. L. 2018a, ApJ, 857, 39 Morris, B. M., Agol, E., Hebb, L., & Hawley, S. L. 2018b, AJ, 156, 91 Morris, B. M., Agol, E., Hebb, L., et al. 2018c, ApJ, 863, L32 Morton, T. D. 2012, ApJ, 761, 6 Morton, T. D. 2015, Astrophysics Source Code Library [record ascl:1503.011] Morton, T. D., Bryson, S. T., Coughlin, J. L., et al. 2016, ApJ, 822, 86 Morzinski, K. M., Males, J. R., Skemer, A. J., et al. 2015, ApJ, 815, 108 Murray-Clay, R. A., Chiang, E. I., & Murray, N. 2009, ApJ, 693, 23 Neron de Surgy, O., & Laskar, J. 1997, A&A, 318, 975 Newton, E. R., Irwin, J., Charbonneau, D., et al. 2016, ApJ, 821, 93 Newton, E. R., Mondrik, N., Irwin, J., Winters, J. G., & Charbonneau, D. 2018, AJ, 156, 217 Newville, M., Stensitzki, T., Allen, D. B., & Ingargiola, A. 2014, LMFIT: NonLinear Least-Square Minimization and Curve-Fitting for Python Nielsen, E. L., Close, L. M., Biller, B. A., Masciadri, E., & Lenzen, R. 2008, ApJ, 674, 466 Owen, J. E., & Campos Estrada, B. 2020, MNRAS, 491, 5287 Owen, J. E., & Jackson, A. P. 2012, MNRAS, 425, 2931 Owen, J. E., & Wu, Y. 2013, ApJ, 775, 105 Owen, J. E., & Wu, Y. 2017, ApJ, 847, 29 Parviainen, H., & Aigrain, S. 2015, MNRAS, 453, 3821 Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2011, J. Mach. Learn. Res., 12, 2825 Pont, F., Zucker, S., & Queloz, D. 2006, MNRAS, 373, 231 Pozuelos, F. J., Suárez, J. C., de Elía, G. C., et al. 2020, A&A, 641, A23 Press, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. P. 1992, Numerical recipes in FORTRAN. The art of scientific computing (Cambridge: Cambridge University Press) Reddy, B. E., Lambert, D. L., & Allende Prieto, C. 2006, MNRAS, 367, 1329 Rein, H., & Liu, S. F. 2012, A&A, 537, A128 Rein, H., & Tamayo, D. 2015, MNRAS, 452, 376 Ricker, G. R., Winn, J. N., Vanderspek, R., et al. 2015, J. Astron. Telesc. Instrum. Syst., 1, 014003 Robinson, T. D., Fortney, J. J., & Hubbard, W. B. 2017, ApJ, 850, 128 Saumon, D., Chabrier, G., & van Horn, H. M. 1995, ApJS, 99, 713 Scargle, J. D. 1982, ApJ, 263, 835 Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103 Seager, S., Kuchner, M., Hier-Majumder, C. A., & Militzer, B. 2007, ApJ, 669, 1279 Sidis, O., & Sari, R. 2010, ApJ, 720, 904 Smith, J. C., Stumpe, M. C., Van Cleve, J. E., et al. 2012, PASP, 124, 1000 Sotin, C., Grasset, O., & Mocquet, A. 2007, Icarus, 191, 337 Stassun, K. G., & Torres, G. 2016, AJ, 152, 180 Stassun, K. G., & Torres, G. 2018, ApJ, 862, 61 Stassun, K. G., Collins, K. A., & Gaudi, B. S. 2017, AJ, 153, 136 Stassun, K. G., Oelkers, R. J., Pepper, J., et al. 2018, AJ, 156, 102 Stefansson, G., Kopparapu, R., Lin, A., et al. 2020, AAS J., submitted [arXiv:2006.11180] Steffen, J. 2006, PhD dissertation, University of Washignton, Washington, USA Strehl, K. 1902, Astron. Nachr., 158, 89 Stumpe, M. C., Smith, J. C., Van Cleve, J. E., et al. 2012, PASP, 124, 985 Stumpe, M. C., Smith, J. C., Catanzarite, J. H., et al. 2014, PASP, 126, 100 A49, page 20 of 21 B.-O. Demory et al.:A super-Earth and a sub-Neptune orbiting the M3V TOI-1266 Ter Braak, C. J. F. 2006, Stat. Comput., 16, 239 Twicken, J. D., Clarke, B. D., Bryson, S. T., et al. 2010, SPIE Conf. Ser., 7740, 774023 Twicken, J. D., Catanzarite, J. H., Clarke, B. D., et al. 2018, PASP, 130, 064502 Van Eylen, V., Agentoft, C., Lundkvist, M. S., et al. 2018, MNRAS, 479, 4786 Vogt, S. S., Allen, S. L., Bigelow, B. C., et al. 1994, SPIE Conf. Ser., 2198, 362 Vrugt, J. A., Braak, C. J. F. T., Diks, C. G. H., et al. 2009, Treatment of forcing data error using MCMC sampling Wakeford, H. R., Lewis, N. K., Fowler, J., et al. 2019, AJ, 157, 11 Wang, J., Fischer, D. A., Xie, J.-W., & Ciardi, D. R. 2015, ApJ, 813, 130 Winn, J. N., Holman, M. J., Torres, G., et al. 2008, ApJ, 683, 1076 Wood, J., Horner, J., Hinse, T. C., & Marsden, S. C. 2017, AJ, 153, 245 Wu, Y., Murray, N. W., & Ramsahai, J. M. 2007, ApJ, 670, 820 Yee, S. W., Petigura, E. A., & Von Braun, K. 2017, ApJ, 836, 77 Zacharias, N., Finch, C. T., Girard, T. M., et al. 2013, AJ, 145, 44 1Center for Space and Habitability, University of Bern, Gesellschaftsstrasse 6, 3012 Bern, Switzerland e-mail: [email protected] 2Space Sciences, Technologies and Astrophysics Research (STAR) Institute, Université de Liège, Allée du 6 Août 19C, 4000 Liège, Belgium 3Astrobiology Research Unit, Université de Liège, Allée du 6 Août 19C, 4000 Liège, Belgium 4Instituto de Astronomía, Universidad Nacional Autónoma de México, Ciudad Universitaria, Ciudad de México 04510, México 5Instituto de Astronomía, Universidad Nacional Autónoma de México, Apdo. Postal 877, 22800 Ensenada, B.C., México 6Universidad Nacional de Córdoba - Observatorio Astronómico de Córdoba, Laprida 854, X5000BGR, Córdoba, Argentina 7Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Buenos Aires, Argentina 8Department of Physics, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK 9Centre for Exoplanets and Habitability, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK 10 Oukaimeden Observatory, High Energy Physics and Astrophysics Laboratory, Cadi Ayyad University, Marrakech, Morocco 11 Center for Astrophysics | Harvard & Smithsonian, 60 Garden Street, Cambridge, MA 02138, USA 12 Observatoire astronomique de l’Université de Genève, 51 chemin des Maillettes, 1290 Versoix, Switzerland 13 Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA 14 Instituto de Astrofísica de Canarias, Vía Láctea s/n, 38205 La Laguna, Tenerife, Spain 15 Department of Astronomy, 501 Campbell Hall, University of California at Berkeley, Berkeley, CA 94720, USA 16 Observatori Astronòmic Albanyà, Camí de Bassegoda s/n, Albanyà 17733, Girona, Spain 17 Department of Earth and Planetary Sciences, University of California, Riverside, CA 92521, USA 18 Centro de Astrobiología (CAB, CSIC-INTA), Dpto. de Astrofísica, ESAC campus 28692 Villanueva de la Cañada (Madrid), Spain 19 Physikalisches Institut, University of Bern, Gesellschaftsstrasse 6, 3012 Bern, Switzerland 20 Cavendish Laboratory, J.J. Thomson Avenue, Cambridge, CB3 0HE, UK 21 Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA 22 Department of Physics, University of California, Berkeley, Berkeley, CA 23 Kotizarovci Observatory, Sarsoni 90, 51216 Viskovo, Croatia 24 Department of Physics & Astronomy, Vanderbilt University, 6301 Stevenson Center Lane, Nashville, TN 37235, USA 25 School of Physics & Astronomy, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK 26 Società Astronomica Lunae, Castelnuovo Magra, Via Montefrancio 77, Italy 27 Department of Aeronautics and Astronautics, MIT, 77 Massachusetts Avenue, Cambridge, MA 02139, USA 28 Department of Astrophysical Sciences, Princeton University, NJ 08544, USA 29 NASA Ames Research Center, Moffett Field, CA, 94035, USA 30 Departamento de Física Teórica y del Cosmos, Universidad de Granada, Campus de Fuentenueva s/n, 18071 Granada, Spain 31 Instituto de Astrofísica de Andalucía (CSIC), Glorieta de la Astronomía s/n, 18008 Granada, Spain 32 Department of A&A, University of California, Santa Cruz, CA 95064, USA 33 Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, HI 96822, USA 34 Division of Geological and Planetary Sciences, California Institute of Technology, 1200 East California Blvd, Pasadena, CA 91125, USA 35 Centre for Astrophysics, University of Southern Queensland, Toowoomba, QLD, Australia 36 Department of Physics & Astronomy, University of California Irvine, Irvine, CA 92697, USA 37 SETI Institute, Mountain View, CA 94043, USA 38 Department of Astronomy, The University of Texas at Austin, Austin, TX 78712, USA A49, page 21 of 21