Abstracts of Extreme Solar Systems 4 (Reykjavik, Iceland)

American Astronomical Society

August, 2019

100 — New Discoveries100.01 — Review of TESS’s First Year Survey andFuture Plans

George Ricker11 Kavli Institute, MIT (Cambridge, Massachusetts, United States)

Successfully launched in April 2018, NASA’s Tran-siting Exoplanet Survey Satellite (TESS) is well on itsway to discovering thousands of exoplanets in orbitaround the brightest stars in the sky. During its ini-tial two-year survey mission, TESS will monitor morethan 200,000 bright stars in the solar neighborhood ata two minute cadence for drops in brightness causedby planetary transits. This first-ever spaceborne all-sky transit survey is identifying planets ranging insize from Earth-sized to gas giants, orbiting a widevariety of host stars, from cool M dwarfs to hot O/Bgiants.

TESS stars are typically 30–100 times brighter thanthose surveyed by the Kepler satellite; thus, TESSplanets are proving far easier to characterize withfollow-up observations than those from prior mis-sions. Such TESS followup observations are enablingmeasurements of the masses, sizes, densities, orbits,and atmospheres of a large cohort of small planets,including a sample of habitable zone rocky worlds.

An additional data product from the TESS missionis its full frame images (FFIs), which are collected ata cadence of 30 minutes. These FFIs provide precisephotometric information for every object within the2300 square degree instantaneous field of view of theTESS cameras. In total, nearly 100 million objectsbrighter than magnitude I = +16 will be preciselyphotometered during the two-year prime mission.

The initial TESS all-sky survey is well under-way, covering 13 observation sectors in the South-ern Ecliptic Hemisphere during Year 1, and 13 obser-vation sectors in the Northern Ecliptic Hemisphereduring Year 2. A concurrent, year-long deep surveyby TESS of regions surrounding the North and SouthEcliptic Poles will provide prime exoplanet targetsfor characterization with the James Webb Space Tele-

scope (JWST), as well as other large ground-basedand space-based telescopes coming online in the nexttwo decades.

The status of the TESS mission as it completes itsfirst year of survey operations in July 2019 will be re-viewed. The opportunities enabled by TESS’s uniquelunar-resonant orbit for an extended mission lastingmore than a decade will also be presented.

100.02 — The Gemini Planet Imager Exoplanet Sur-vey: Giant Planet and Brown Dwarf Demographicsfrom 10-100 AU

Eric Nielsen1; Robert De Rosa1; Bruce Macintosh1;Jason Wang2; Jean-Baptiste Ruffio1; Eugene Chiang3;Mark Marley4; Didier Saumon5; Dmitry Savransky6;Daniel Fabrycky7; Quinn Konopacky8; JenniferPatience9; Vanessa Bailey10

1 KIPAC, Stanford University (Stanford, California, United States)2 Jet Propulsion Laboratory, California Institute of Technology

(Pasadena, California, United States)3 Astronomy, California Institute of Technology (Pasadena, Califor-

nia, United States)4 Astronomy, U.C. Berkeley (Berkeley, California, United States)5 NASA Ames Research Center (Mountain View, California, United

States)6 Los Alamos National Laboratory (Los Alamos, New Mexico,

United States)7 Sibley School of Mechanical and Aerospace Engineering, Cornell

University (Ithaca, New York, United States)8 Astronomy & Astrophysics, University of Chicago (Chicago,

Illinois, United States)9 Center for Astrophysics and Space Science, U.C. San Diego (La

Jolla, California, United States)10 SESE, Arizona State University (Tempe, Arizona, United States)

The Gemini Planet Imager Exoplanet Survey (GPIES)has observed 521 young, nearby stars, making it oneof the largest, deepest direct imaging surveys for gi-ant planets ever conducted. With detections of sixplanets and four brown dwarfs, including the newdiscoveries of 51 Eridani b and HR 2562 B, GPIESalso has a significantly higher planet detection ratethan any published imaging survey. Our analysis

1

of the uniform sample of the first 300 stars revealsnew properties of giant planets (>2 MJup) from 3-100 AU. We find at >3 σ confidence that these plan-ets are more common around high-mass stars (> 1.5solar masses) than lower-mass stars. We also presentevidence that giant planets and brown dwarfs obeydifferent mass functions and semi-major axis distri-butions. Our direct imaging data imply that the gi-ant planet occurrence rate declines with semi-majoraxis beyond 10 AU, a trend opposite to that foundby radial velocity surveys inside of 10 AU; taken to-gether, the giant planet occurrence rate appears topeak at 3-10 AU. All of these trends point to wide-separation giant planets forming by core/pebble ac-cretion, and brown dwarfs forming by gravitationalinstability. If our power-law model that fits giantplanets around high-mass stars is also applicable tosolar-type stars, and these power-laws remain validdown to the mass of Jupiter and inward to 5 AU, thenthe occurrence rate for giant planets more massivethan Jupiter within 100 AU could be less than 40%.Looking beyond these results, we present our earlyanalysis of the full GPIES sample, whether thesetrends persist over all 521 observed stars, and impli-cations for future observations from Gemini-Northwith an upgraded GPI.

100.03 — Three Red Suns in the Sky of the NearestExoplanet Transiting an M Dwarf

David Charbonneau1; Jennifer Winters1; AmberMedina1; Jonathan Irwin1

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

The only terrestrial exoplanets whose atmosphereswill be spectroscopically accessible in the near fu-ture will be those that orbit nearby mid-to-late Mdwarfs. We present the discovery from TESS dataof LTT-1445Ab, a terrestrial planet transiting an Mdwarf only 6.9 parsecs away, making it the closestknown transiting planet with a small-star primary.Remarkably, the host stellar system is composed ofthree mid-to-late M dwarfs in a hierarchical config-uration, which are blended in a single TESS pixel.We use follow-up observations from MEarth and thecentroid offset from the TESS data to determine thatthe planet transits the primary star in the system.The planet has a radius 1.35 times that of Earth, anorbital period of 5.36 days, and an equilibrium tem-perature of 428 K, and the mass should be readilymeasurable with radial velocity observations in thecoming months. The system is particularly favorablefor ground-based observations to advance the studyof the atmospheres of terrestrial exoplanets: Such

observations are typically performed using multi-object spectrographs on large telescopes, but previ-ous studies have been limited by the need to use bluefield stars to calibrate telluric variations, which haveprovided a poor color match to the red target stars.Here, the companion stars provides an ideal telluriccalibrator, namely one of nearly equal brightness andsimilar spectral type located only 7 arcseconds fromthe target.

This work is supported by grants from the Na-tional Science Foundation and the John TempletonFoundation.

100.04 — Evidence for an additional planet in the βPictoris system.

Anne-Marie Lagrange11 Institut de Planétologie et d’Astrophysique de Grenoble (Saint

Martin d’Heres, France)

With its well resolved debris disk of dust, its evapo-rating exocomets, and an imaged giant planet orbit-ing at about 9 au, the young (∼23 Myr) β Pictoris sys-tem is a unique proxy for detailed studies of planetformation and early evolution processes as well asplanet-disk interactions. We have studied 10 yearsof ESO/HARPS high resolution spectroscopic dataon the star. After removing the δ Scuti pulsations, a∼1200 days periodic signal is observed. Within ourcurrent knowledge, we can only attribute this signalto a second massive planet orbiting at ∼2.7 au fromthe star (Lagrange et al, 2019, Nat. Astron., underminor revisions). To our knowledge, this is the firstsystem hosting a planet detected in imaging and onedetected with indirect technics. I will present the ev-idence for this additional planet, and analyse the im-pact of this result on previous results, including pre-vious analysis of GAIA astrometric data, the systemdynamical stability, the exocomets activity.

100.05 — Absence of a thick atmosphere on a ter-restrial exoplanet

Laura Kreidberg1; Daniel D. B. Koll2; Caroline Morley3;Renyu Hu9; Laura Schaefer4; Drake Deming5; KevinStevenson6; Jason Dittmann2; Andrew Vanderburg3;David Berardo2; Xueying Guo2; Keivan Stassun7; IanCrossfield2; David Charbonneau1; David Latham1;Abraham Loeb1; George Ricker8; Sara Seager2; RolandVanderspek2

1 Harvard University (Cambridge, Massachusetts, United States)2 MIT (Cambridge, Massachusetts, United States)3 UT Austin (Austin, Texas, United States)4 Stanford University (Palo Alto, California, United States)5 University of Maryland (College Park, Maryland, United States)

2

6 Space Telescope Science Institute (Baltimore, Maryland, UnitedStates)

7 Vanderbilt (Nashville, Tennessee, United States)8 Kavli Institute, MIT (Cambridge, Massachusetts, United States)9 Jet Propulsion Laboratory (Pasadena, California, United States)

I will present a thermal phase curve measurementfor a terrestrial exoplanet recently detected by theTESS mission. The planet is in a short period orbitaround a nearby M-dwarf star. The phase curve isthe first such measurement for a planet smaller than1.6 Earth radii, the size below which the existence ofan atmosphere is unknown a priori. The amplitudeof the phase variation puts strong constraints on theplanet’s atmospheric properties, which I will discuss.

100.06 — Recent Microlensing Results: IndividualSystems and Demographic Frontiers

Calen Henderson1; David Bennett4; B. Scott Gaudi2;Jennifer Yee3; Rachel Street5

1 Caltech/IPAC-NExScI (Pasadena, California, United States)2 The Ohio State University (Columbus, Ohio, United States)3 Harvard-Smithsonian CfA (Cambridge, Massachusetts, United

States)4 UMBC/NASA GSFC (Greenbelt, Maryland, United States)5 Las Cumbres Observatory (Goleta, California, United States)

Over the past several years, the field of gravita-tional microlensing has made myriad advancementswith regard to characterizing individual planetarysystems, exploring relatively unknown demographicregimes, and developing tools and resources forcommunity use. Here I will highlight a handful of re-sults that lead to precise planet masses for microlens-ing planets, including (A) measuring the microlensparallax effect (e.g., Street+ 2016); (B) using high-resolution photometry to constrain the flux of thelens (e.g., Bhattacharya+ 2018); (C) complementingmicrolensing photometry with astrometric and spec-troscopic data (Han+ 2019); and (D) deriving the Ein-stein radius through interferometry (Dong+ 2019).I will also discuss recent demographic studies, in-cluding (i) constraining the frequency of free-floatingplanets (e.g., Mróz+ 2017, 2018; Poleski+ 2014); (ii)determining the Galactic distribution of exoplanetsvia a multi-year Spitzer program (cf. Yee+ 2015); and(iii) understanding and contextualizing the planet-star mass-ratio distribution (Suzuki+ 2016, 2018; Pas-cucci+ 2018). Finally, I will conclude by describ-ing the public tools and data provided, in particu-lar by the WFIRST Microlensing Science Investiga-tion Team, to allow for the larger exoplanet commu-nity to get involved with immediate science and alsohelp prepare for the WFIRST microlensing survey.

101 — Direct Imaging101.01 — Frequency of Massive Wide-orbit Plan-ets vs. Stellar Mass: SPHERE SHINES on the ESOVLT

Michael Meyer11 Department of Astronomy, The University of Michigan (Ann

Arbor, Michigan, United States)

We describe the SpHere INfrared Exoplanet (SHINE)survey, a key part of the SPHERE GTO Program onthe ESO VLT, and present new statistical analysesof the frequency of gas giant planet occurrence asa function of host star mass. Constraining the fre-quency of the most massive planets, as well as thelowest mass brown dwarf companions, at wide or-bital separations vs. host star mass, enables us todiscern the mean outcomes of planet formation, thusdefining what is extreme in the context of planetaryarchitectures. In addition, our data provides a strongtest of predictive theories of star and planet forma-tion. This high contrast imaging survey has dis-covered and characterized dozens of very low masscompanions (1-76 MJupiter), on wide orbits (10-1000AU) around a range of host star masses (0.3-3 MSUN).Three papers in preparation (Desidera et al.; Lan-glois et al.; Vigan et al.) describe the survey sam-ple and strategy, data reduction and analysis tech-niques, and the first statistical results. Our survey,constraining the frequency of gas giants 1-10 MJupiter,as well as brown dwarf companions, from 10-100AU, suggests: 1) the frequency of gas giants aroundFGK (and other) stars peaks between 1-10 AU; 2) thegas giant planet mass function appears to be a uni-versal power-law relative to host star mass, explain-ing the trend of gas giant detection rate of with starmass; 3) the brown dwarf companion mass functionis consistent with extrapolation from a universal stel-lar companion mass ratio distribution down to theminimum mass for fragmentation; and 4) some, butnot all, relevant predictions made by D.N.C. Lin arethankfully inconsistent with these data.

101.02 — Population-Level Eccentricity Distribu-tions of Imaged Exoplanets and Brown DwarfCompanions

Brendan Bowler1; Sarah Blunt2; Eric Nielsen31 The University of Texas at Austin (Austin, Texas, United States)2 Caltech (Pasadena, California, United States)3 KIPAC, Stanford University (Stanford, California, United States)

The dominant formation channel of long-period di-rectly imaged exoplanets and brown dwarf compan-

3

ions has been challenging to unambiguously con-strain with observations because of their low occur-rence rates, limited composition measurements, anddegeneracies among theoretical predictions. Eccen-tricities offer a robust tool to assess the origin of thesepopulations because they directly trace the dynami-cal imprint after formation and any subsequent or-bital evolution.

In this talk I will discuss new results on the un-derlying eccentricity distributions of directly imagedexoplanets and brown dwarf companions. We havecarried out homogeneous orbit fits based on newhigh-contrast imaging observations together with acompilation of astrometry from the literature to as-sess the individual and population-level eccentric-ity distributions of over two dozen long-period gi-ant planets and brown dwarfs between 5-100 AU us-ing hierarchical Bayesian modeling. Each compan-ion traces out a small orbit arc which typically resultsin a broad constraint on its individual eccentricity,but together as an ensemble these systems providevaluable insight into their collective underlying or-bital patterns.

The population-level eccentricity distributions forthe subset of giant planets (2-15 Mjup) and browndwarf companions (15-75 Mjup) are significantly dif-ferent and provide compelling dynamical evidencefor distinct formation pathways. As a population,long-period planets preferentially have low eccen-tricities, suggesting formation within a disk. Thebrown dwarf subsample is dynamically hotter witha broad peak at high eccentricities, which is quali-tatively similar to binary stars. Larger samples andcontinued astrometric orbit monitoring will help es-tablish whether these eccentricity distributions cor-relate with other parameters such as stellar hostmass, multiplicity, and system age.

101.03 — Two giant Exomoons around two low-mass Brown Dwarf companions detected withSPHERE

Cecilia Lazzoni1,21 OAPD, INAF (Padova, Italy)2 Università di Padova (Padova, Italy)

It is still unclear if brown dwarfs companion detectedwith the direct imaging technique were formed asstars or planets. The analysis of their multiplicity canprovide clues on their formation mechanism. In thiscontext, we analyzed the residuals around browndwarf companions observed with SPHERE duringthe SHINE/GTO with the technique of negativefake planets to look for features around them. Wefound an extended source around one of the brown

dwarf in the sample that would suggest the pres-ence of a disk and two candidate companions, mas-sive gaseous exomoon-like objects, bound to othertwo brown dwarfs. These latter would represent thefirst triple systems ever discovered with two substel-lar companions, one in the planetary regime and thesecond just above the deuterium burning limit.

101.05 — PDS 70 b: Evidence for a circumplanetarydisc around the fIrst directly imaged protoplanet

Valentin Christiaens1; Faustine Cantalloube2; SimonCasassus3; Daniel Price1; Olivier Absil4; ChristophePinte1; Julien Girard5; Matías Montesinos6

1 Monash university (Clayton, Victoria, Australia)2 MPIA (Heidelberg, Germany)3 University of Chile (Santiago, Chile)4 University of Liege (Liege, Belgium)5 Space Telescope Institute (Baltimore, Maryland, United States)6 University of Valparaiso (Valparaiso, Chile)

The observed properties of the major moons ofJupiter — and of other gas giants — have suggestedthat they formed within a circumplanetary disc. Thisprediction has been supported by theoretical calcu-lations and numerical simulations of increasing com-plexity over the past few decades. Despite intensivesearch, circumplanetary discs had until now eludeddetection. In this talk, I will present the first obser-vational evidence for a circumplanetary disc, aroundrecently imaged protoplanet PDS 70 b. Our detectionis based on a new near-IR spectrum acquired withVLT/SINFONI. We tested several hypotheses (atmo-spheric emission alone, variable extinction, combina-tion of atmospheric and circumplanetary disc emis-sion) to explain the spectrum and show that modelsconsidering atmospheric emission alone consistentlyunderpredict the longward portion of the spectrum.Our best fit is obtained with a combined atmosphericand circumplanetary disc model, with emission fromthe circumplanetary disc accounting for the appar-ent excess IR emission.

101.06 — A Pan-STARRS and TESS Search for Dis-tant Planets

Matthew J. Holman1; Matthew J. Payne11 Center for Astrophysics | Harvard and Smithsonian (Cambridge,

Massachusetts, United States)

Several lines of evidence, both theoretical and ob-servational, indicate that additional planets in theouter solar system remain to be discovered. We re-cently developed a novel technique to search for so-lar system bodies (Holman et al. 2018). This method

4

is particularly well-suited to very slow-moving ob-jects, even those for which the motion within a daymight be to small to detect. We present the results ofour use of this method to search for distant planetsand minor planets in existing Pan-STARRS and TESSdata. Perhaps even more important than the searchitself is a detailed, quantitative analysis of the sur-vey’s detection limits and biases. This informationis essential for the rigorous interpretation of thesesurvey results. Such simulators have been devel-oped for CFEPS/OSSOS, NEOWISE, and other sur-veys, leading to detailed results on the small bodypopulations throughout the solar system. We havedeveloped a high-fidelity survey simulator for Pan-STARRS and have extended it to TESS. This simu-lator takes positions, magnitudes, and rates of mo-tion calculated from a solar system model (Grav et al2011) at the times and locations of individual expo-sures. It then inserts synthetic detections into the re-sulting exposure source catalogs, accounting for thedetails of the camera, photometric zero point, andother essential details. The source catalogs, includ-ing synthetic detections, are then run through ourfull search pipeline. This approach allows a clear,quantitative statement about the prevalence of dis-tant planets, as seen by Pan-STARRS and TESS.

101.07 — Studying the Interior Structure of an Ex-tremely Eccentric Hot Jupiter via Deep VLT Imag-ing

Sasha Hinkley1; Arthur Vigan2; Subo Dong4; Ken Rice3;Richard Nelson5; Aarynn Carter1; Julien Milli7; JulienGirard6

1 Physics, University of Exeter (Exeter, United Kingdom)2 Laboratoire de Astrophysique (LAM) (Marseille, France)3 Astronomy, Royal Observatory (Edinburgh, United Kingdom)4 Kavli Institute for Astronomy & Astrophysics (Peking, China)5 Physics, Queen Mary University of London (London, United

Kingdom)6 STScI (Baltimore, Maryland, United States)7 ESO (Santiago, Chile)

I will describe how our group at Exeter has usedthe VLT-SPHERE instrument to place constraints onthe internal structure of HD 20782b, a Hot Jupiterwith the most extreme eccentricity known to date(e∼0.96). In the dynamically-driven migration sce-nario (e.g. Kozai-Lidov cycles combined with tidaldissipation), a Jupiter mass planet is dynamically ex-cited to a high eccentricity by a third body, and itsorbit subsequently shrinks and circularizes throughtidal dissipation. Our deep observations of the HD20782 system rule out any additional (third) compan-ions with masses in the range 20-60 Jupiter masses at

orbital separations ∼10-60 AU that might be respon-sible for exciting the extreme eccentricity of the innerplanet. Our lack of detections of any additional com-panions in the system indicates that the eccentricityof the planet was gained early on and has persisteduntil the present. The apparent failure of the tidaldissipation mechanism in this system means that wecan place strong constraints on the tidal quality fac-tor “Q” of HD 20782b. Specifically, our models ofplanetary tidal evolution suggest a remarkably highvalue for the planet’s tidal Q factor of 107 – 108: twoto three orders of magnitude higher than that mea-sured for other extrasolar planets, as well as mem-bers of our own solar system. Our result suggestsa possible structural difference between HD 20782band other giant planets inside and outside our solarsystem. If time allows, I will discuss how our ap-proved 52-hour JWST Early Release Science Programwill pave the way for future observations of addi-tional systems with extremely eccentric planets start-ing in 2021. JWST will illuminate the interior struc-tures of many more eccentric Jovian mass planets go-ing forward, or possibly even image the extremelyeccentric planets themselves.

102 — Radial Velocities102.01 — HARPS and HARPS-N solar telescopes:the key to extremely precise radial-velocity mea-surements

Xavier Dumusque11 Department of Astronomy, University of Geneva (Versoix,

Geneva, Switzerland)

Detecting and measuring the masses of planets in thepresence of stellar signals is the main challenge weare facing when using the radial-velocity (RV) tech-nique. Even in the TESS era where planetary periodsare known, obtaining a precise mass, which is criticalto constrain further planetary composition and thusplanetary formation, is challenging.

Critical to a better understanding of RV varia-tions induced by stellar signals and finding correc-tion techniques is RV data with a sampling sufficientto probe timescales ranging from minutes to years.To address this challenge, our team built two so-lar telescopes that feed sunlight into HARPS-N andHARPS, which allows us to obtain Sun-as-a-star RVsat a sub-m/s precision.

In this talk, I will present the data that obtainedduring the last 4 years with HARPS-N and nearly ayear with HARPS. I will show how the two datasetsmatch at a level of 40 cm/s within a day, which

5

allows to characterize p-modes, granulation signal,and stellar activity on the rotation timescale of theSun, which we know are the main limitations to pre-cise RVs.

With these data, we start to improve our under-standing of how stellar signals affect RVs: - CCF-lineshape variability correlates with RVs with a signifi-cant time-delay that prevents using shape variationsdirectly as a proxy for stellar activity. - RV correlatesstrongly with total magnetic field strength, whichmakes sense as magnetic regions are at the origin ofmost stellar signal. - With the extreme SNR that wereach on the Sun, we see that when analyzing the RVof individual spectral lines, some are much more sen-sitive to stellar activity than others, due to differingformation height in the stellar photosphere.

All these new insights into stellar signals give usthe key to develop the techniques capable of mitigat-ing their impact down to a level that will allow thedetection and characterization of Earth-twins usingthe RV technique.

102.02 — Rocky planets from the CARMENES Sur-vey

Stefan Dreizler11 Astrophysics, University Goettingen (Goettingen, Germany)

Since the first discovery, more than 800 exoplan-ets have been detected through the radial velocitymethod, the majority orbiting solar-like stars. Al-though M-stars are the most frequent stars, very fewplanets have yet been found around M-stars of latespectral type.

CARMENES, operated since 2016, is a high-resolution visible-near-IR spectrograph dedicated tosearch for such low-mass planets around low-massstars and already doubled the number of knownplanets with host stars below 0.2 MSun. Not surpris-ingly, also this stellar parameter range has its sur-prises in terms of planetary system architectures. Wewill give an overview of exoplanet detections (pub-lished and unpublished) from the CARMENES sur-vey and then concentrate on the low-mass planets,including the very recent detection of two Earth-mass planets around Teegarden’s star highlightingthe capability of CARMENES. The planetary sys-tem is special since Teegarden’s star is only one outof three planet host stars with an effective temper-ature below 3000K. Its two planets are within theoptimistic and conservative habitable zone, respec-tively. Notably, the Earth, as well as other Solar Sys-tem planets are currently or in near future in the tran-sit visibility zone see from Teegarden’s star.

102.03 — A low-mass planet candidate orbitingProxima Centauri at a distance of 1.5 au

Mario Damasso11 INAF-Astrophysical Observatory of Torino (Pino Torinese, Italy)

By analyzing ∼17 years of radial velocity data ofProxima Cen collected with the UVES and HARPSspectrographs, we detected a signal of period P∼5years that could be explained by the presence ofa second planet, Proxima c, with minimum massm sin i∼6. Earth masses. Together with the low-mass temperate planet Proxima b, this candidateplanet would make Proxima the closest multi-planetsystem to the Sun. We will present the propertiesof the new RV signal and investigate the likelihoodthat it is related to a magnetic activity cycle of thestar. We will discuss how the existence of Proxima ccan be confirmed, and its true mass determined withhigh accuracy, by combining Gaia astrometry and ra-dial velocities. Proxima c would be a prime target forfollow-up and characterization measurements, espe-cially with next generation direct imaging instru-mentation due to the large maximum angular sep-aration of ∼1 arcsecond from the parent star. Sincethe orbit would be beyond the original location of thesnowline, Proxima c would challenge the models ofthe formation of super-Earths. Presently, this studyis under review by Science Advances.

102.04 — New insights into the keystone WASP-107system: shedding light on the formation and chem-istry of WASP-107b using Spitzer eclipse spec-troscopy and a clue to unveiling its dynamics his-tory from the detection of an outer companion withKeck/HIRES

Caroline Piaulet1,2; Björn Benneke1,2; Ryan Rubenzahl3;Andrew Howard3; Laura Kreidberg4; Michael W.Werner5; Ian Crossfield6,7; Evan Sinukoff8,9

1 Physics, University of Montreal (Montreal, Quebec, Canada)2 Institute for Research on Exoplanets (Montreal, Quebec, Canada)3 Astrophysics, California Institute of Technology (Lowville, New

York, United States)4 Harvard University (Cambridge, Massachusetts, United States)5 Jet Propulsion Laboratory, California Institute of Technology

(Pasadena, California, United States)6 Physics, Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)7 Kavli Institute for Astrophysics and Space Research (Cambridge,

Massachusetts, United States)8 California Institute of Technology (Pasadena, California, United

States)9 Institute for Astronomy, University of Hawai‘i at Manoa (Hon-

olulu, Hawaii, United States)

6

With the radius of Jupiter, the near-Neptune-massplanet WASP-107 presents a major challenge toplanet formation theories. Meanwhile, the system’sbrightness and planet’s low surface gravity makes ita keystone target for spectroscopic characterization,especially in the poorly-probed low-temperature (Teq< 800 K) regime. In this talk, we will present the mainresults of an extensive follow-up program of WASP-107b using over 2 years of Keck/HIRES radial veloci-ties as well as >60 hours of Spitzer observations. Theradial velocity data reveal an even lower planetarymass than previously thought. The inferred 1.8 Nep-tune mass indicates an extraordinarily high H/Hemass fraction of 80% accreted by a core of only 7 ±3 Earth masses. The resulting lower surface gravitymeans that all the transmission spectroscopy for thisplanet has to be reinterpreted. With Spitzer, we fur-thermore detect the thermal emission of this 720K ex-oplanet at 3.6μm, indicating substantial eccentricity(e = 0.129+0.028

−0.011) and making it the best targetfor eclipse observations with JWST in this tempera-ture regime. A puzzling brightness temperature con-trast between the 3.6 and 4.5μm bandpasses presentsdirect evidence for disequilibrium chemistry, andmakes WASP-107b a keystone target to unveil the un-derlying mechanisms of quenching and atmosphericdynamics. We show that the non-zero eccentricity ofWASP-107b could result from the presence a secondplanet in the WASP-107 system on a highly eccentric(e = 0.56+0.11

−0.14) and wide (∼2000d) orbit, whichwe also detect in the radial velocity data. Overall,the joint constraints from the secondary eclipse andRV observations shed unprecedented light on therich dynamics history of this peculiar planetary sys-tem offering an intriguing possibility for the originof close-in exo-Neptunes like WASP-107b.

102.05 — Radial Velocity Discovery of an EccentricJovian World Orbiting at 18 au

Sarah Blunt1; Michael Endl2; Lauren Weiss3; WilliamCochran2; Andrew Howard1; Phillip MacQueen2; Ben-jamin Fulton4; Gregory Henry9; Marshall C. Johnson5;Molly Kosiarek10; Kellen Lawson11; Bruce Macintosh8;Sean M. Mills1; Eric Nielsen6; Erik Petigura1; GlennSchneider12; Andrew Vanderburg7; John Wisniewski11;Robert Wittenmyer2; Erik Brugamyer2; CarolineCaldwell2; Artie Hatzes13; Lea Hirsch8; HowardIsaacson14; Paul Robertson2; Arpita Roy1; Zili Shen2

1 Caltech (Pasadena, California, United States)2 UC Santa Cruz (Santa Cruz, California, United States)3 Univ. of Oklahoma (Norman, Oklahoma, United States)4 Univ. of Arizona (Tucson, Arizona, United States)5 Thuringer Landessternwarte (Tautenburg, Germany)

6 UC Berkeley (Berkeley, California, United States)7 UT Austin (Austin, Texas, United States)8 Institute for Astronomy, University of Hawaii at Manoa (Hon-

olulu, Hawaii, United States)9 NASA Exoplanet Science Institute / Caltech-IPAC (Pasadena,

California, United States)10 Department of Astronomy, The Ohio State University (Colum-

bus, Ohio, United States)11 KIPAC, Stanford University (Stanford, California, United States)12 University of Texas at Austin (Austin, Texas, United States)13 Physics, Stanford University (Burlingame, California, United

States)14 Tennessee State Univ. (Nashville, Tennessee, United States)

We announce the discovery of the longest-periodplanet with a well-constrained orbit discovered withradial velocities (RVs). HR 5183 b, with P = 75 ± 30yr, e = 0.84 ± 0.04, and M sin i = 3.23 ± 0.14 MJ, wasdetected independently in more than two decadesof data from Keck/HIRES and McDonald/Tull. Thehighly eccentric orbit takes the planet from withinthe orbit of Jupiter to beyond the orbit of Neptuneover one period. Because of this high eccentricity, or-bital information density is strongly peaked aroundperiastron, which occurred in January 2018. By ob-serving this periastron passage event with high ca-dence, we were able to place tight constraints on theorbital parameters without witnessing an entire or-bital period.

In terms of semimajor axis and mass, HR 5183 bis most similar to a typical directly imaged planet,but its advanced age, extreme eccentricity, and solar-type primary star differentiate it from this popu-lation. This discovery probes a previously unex-plored population of exoplanets, highlighting thevalue of long-baseline RV surveys and raising in-teresting questions about the long-term evolution ofplanetary systems with massive planets.

102.06 — First Results from the SPIRou Legacy Sur-vey

Rene Doyon11 Université de Montreal (Montréal, Quebec, Canada)

SPIRou is the infrared high-resolution echelle spec-tropolarimeter currently in operation on the Canada-France-Hawaii telescope, an instrument specificallydesigned and optimized to achieve precision radialvelocity at infrared wavelengths. SPIRou features aunique polarimetric capability, high resolving power(70,000) and a very broad simultaneous wavelengthcoverage between 0.98 to 2.5 microns. SPIRou hastwo main science goals: detect small planets aroundnearby low-mass stars and explore the impact of

7

magnetic field of star/planet formation. SPIRou hasbeen allocated a 300-night Legacy Survey over a pe-riod of 4 years that was initiated in February with thefollowing main science objectives: 1) search for smallplanets around low-mass stars, 2) provide mass mea-surements for new transiting planets from TESS andother transit surveys and 3) observe a large sampleof pre-main sequence stars to detect and characterizehot Jupiters at early evolutionary stages and to inves-tigate planet formation and planet/disc interactions.Thanks to its wide wavelength range, SPIRou is also avery powerful capability for atmospheric characteri-zation of transiting exoplanets. This talk will presentan overview of the instrument and its on-sky perfor-mance along with a highlight of the first science re-sults obtained so far as part of the Legacy Survey.

103 — Transits

103.01 — Expectations vs. Reality: The ExoplanetYield in the TESS Full-Frame Images

Adina Feinstein1; Benjamin Montet1; Nicholas Earley11 Astronomy & Astrophysics, University of Chciago (Chicago,

Illinois, United States)

During its two year prime mission, the Transiting Ex-oplanet Survey Satellite (TESS) will perform a time-series photometric survey for 80% of the sky, observ-ing 26 24x96 degree sectors of the sky each for 27days. The primary objective of TESS is to find tran-siting planet candidates around 200,000 pre-selectedstars for which fixed aperture photometry is recov-ered every two minutes. However, TESS is alsorecording and delivering Full-Frame Images (FFIs)of each detector at a thirty minute cadence. Usingthe eleanor pipeline, which creates light curves for allstars in the FFIs, we have begun a uniform transitsearch for targets in multiple sectors. In this talk, Iwill highlight several of the current findings withinthe FFI data. I will discuss our reduction and vettingprocesses, specifically highlighting the most com-mon false positives found within the data, how weidentify them, and how we remove them from the fi-nal data set. As TESS has already proven a successfulmission and has been awarded an extended mission,we will continue to search the FFIs for new planetcandidates to further our understanding of the exo-planet population, especially those of longer periods,and its implications for finding new planets in thisdata set

103.02 — Newly Formed Planets within the De-bris Disk of the Nearest Pre-main-sequence StarAU Mic

Peter Plavchan11 Physics & Astronomy, George Mason University (Fairfax, Vir-

ginia, United States)

We report a two-planet system orbiting a young starwith a debris disk, one inner planet discovered us-ing data from NASA’s TESS mission and a secondplanet with multi-wavelength radial velocities. Thetwo newly identified planets in this system can beused to investigate disk-planet interactions and in-form the planet formation and migration process.

103.03 — Identifying Exoplanets with Deep Learn-ing: New Discoveries and Progress towards PlanetOccurrence Rates in Kepler, K2, and TESS

Andrew Vanderburg1; Christopher J. Shallue3; AnneDattilo1; Liang Yu2

1 University of Texas at Austin (Austin, Texas, United States)2 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)3 Google AI (Mountain View, California, United States)

Deep learning, a cutting edge machine learning tech-nique, is leading to remarkable advancements infields ranging from biomedical imaging to linguis-tics. Our team is leveraging this technology to dis-cover new exoplanets and characterize their popu-lations. We have built and tested neural networksto classify and vet transiting planet candidates fromKepler, K2, and TESS and identified new exoplanetsfrom large sets of unclassified signals. Our discov-eries include two super-Earths from the K2 mission,a new planet in a five-planet resonant chain aroundKepler 80 and an eighth planet around Kepler 90,making this the most extreme solar system knownin number of planets. I will give an overview of deeplearning, describe our newly discovered planets, anddiscuss the path forward to using these deep learn-ing classification tools to measure planet occurrencerates in Kepler, K2, and TESS.

103.04 — The HD 21749 System: A Temperate Sub-Neptune, an Earth-sized planet, and Who KnowsWhat Else

Diana Dragomir1; Chelsea X. Huang3; Stephen Kane2;Paul A. Dalba2; Maximilian N. Günther3

1 MIT/UNM (Cambridge, Massachusetts, United States)2 Department of Earth and Planetary Science, University of Califor-

nia Riverside (Riverside, California, United States)3 MIT (Cambridge, Massachusetts, United States)

8

Our understanding of multi-planet systems has beendominated by Kepler discoveries, our understand-ing of multis is nowhere near complete. With TESSwe have the opportunity to fill in missing pieces, likeprecise masses and orbital eccentricities of planets inmultis, and to search for non-transiting planets withradial velocity measurements.

I will present the recent discovery of HD 21749(jointly enabled by TESS and existing ground-basedobservations), a multi-planet system with an intrigu-ing architecture. It includes an unusually dense (7g/cm3) sub-Neptune in a mildly eccentric 35.6-dayorbit, and a 0.9 Earth radius planet with a period of7.8 days, around a K dwarf star located 16 pc awayfrom the Sun. A similar architecture (in terms of pe-riods, and non-zero eccentricity of the outer planet)has surfaced in four other systems known to hostsmall/low-mass planets. The host stars of HIP 57274,HIP 7924 and HIP 69830 are also K dwarfs, but thesesystems are not known to transit so only lower limitsare available on their masses, and no radius measure-ment. The periods of the two known low-mass plan-ets in the K2-18 system (9 and 33 days) are very simi-lar to those of HD 21749 b and c, but while K2-18b hasa radius and a mass measurement, planet c does nottransit and only has a lower mass limit. Moreover,the host star is a M dwarf, and thus planet forma-tion probably differed between those two systems.With prospects for a mass measurement of planet cin the near future, HD 21749 is poised to become thebest characterized system with this emerging archi-tecture.

We have continued monitoring this system withMagellan-PFS radial velocities, leading to improvedconstraints on the orbital eccentricity of planet b, andon the presence and properties of additional plan-ets in the system. I will also show results from adynamical analysis of the system, which provide anindependent constraint on the mass of planet c andon islands of stability where other planets could or-bit. Lastly, I will explore a few exciting follow-upavenues within reach, including prospects for atmo-spheric characterization and orbital obliquity mea-surements.

103.05 — Detecting Magnetic Fields in Exoplanetswith Spectropolarimetry in the Helium Line at 1083nm

Antonija Oklopcic1; Christopher Hirata2; Paulo MonteroCamacho2; Makana Silva2

1 Harvard University (Cambridge, Massachusetts, United States)2 Ohio State University (Columbus, Ohio, United States)

Most planets in the solar system have or previ-

ously had a global magnetic field, yet not much isknown about magnetic fields in exoplanets. Infor-mation about the presence of a magnetic field andits strength could give us valuable insights into theinterior structure and thermal evolution of an ex-oplanet. Furthermore, a global magnetic field onan exoplanet could have important consequences forthe extent, composition, and evolution of its atmo-sphere, by controlling atmospheric escape and its in-teraction with the stellar wind. In this talk, I willpresent a new method for detecting magnetic fieldsin the atmospheres of close-in exoplanets, based onspectropolarimetric transit observations at the wave-length of the helium line at 1083 nm. Strong ab-sorption signatures (transit depths on the order ofa few percent) in the 1083 nm line have recentlybeen observed for several close-in exoplanets. Mostof the work so far has been focused on measuringand interpreting the effects of extended or escap-ing planetary atmospheres on the radiation inten-sity at 1083 nm; however, a wealth of informationcan be stored in radiation polarization as well. Iwill describe how linear and circular polarizationsignals in the helium 1083 nm line arise in the pres-ence of an external magnetic field due to atomiclevel polarization induced by anisotropic stellar ra-diation, and the combined action of the Zeeman andHanle effects. This phenomenon has been well estab-lished in solar physics as a means to probe the mag-netic field properties of the solar chromosphere andcorona, and I will demonstrate how the diagnosticpower of this method can be extended to the fieldof exoplanets. Assuming exoplanetary magneticfields with strengths comparable to the magneticfields observed in the solar system planets, polariza-tion signals in the helium 1083 nm line should bedetectable with modern high-resolution spectropo-larimeters operating at these wavelengths.

103.06 — The Most Metal-Poor Planets Around theOldest Stars in Milky Way

Ji Wang11 Astronomy, The Ohio State University (Comlubus, Ohio, United

States)

The Gaia-TESS synergy opens up a new window topeer through planet formation in the galactic con-text. We can finally answer the following question:when and how did the first planet form? By se-lecting ∼27,000 halo stars, the oldest stellar popula-tion in the Milky Way, via their galactic kinematicsprovided by Gaia, we conduct the most conprehen-sive study on chemical abundance of halo stars usingSDSS/APOGEE spectra, and the search for the most

9

metal-poor planets using TESS data. The study leadsto discoveries of a few planet candidates throughTESS sector 9 and the characterization of their stel-lar enviroment. We will report the exciting discover-ies, the follow-up observations for planet confirma-tion and validation, and the implications on planetformation in extremely metal-poor enviroment in theinfant universe.

103.07 — TESS Discovery of the First Ultra HotNeptune, LTT9779b

James Jenkins11 Universidad de Chile (Santiago, Chile)

In this talk I will discuss the discovery of a su-per Neptune orbiting the bright and metal-rich star,LTT9779. The planet was first detected as a candidatein data from Sector 2 of the Transiting Exoplanet Sur-vey Satellite (TESS), and subsequent ground-basedphotometric and spectroscopic follow-up confirmedits reality and constrained its mass. With an or-bital period of only 19 hours, this world is the firstNeptune-like ultra short period planet, and with anequilibrium temperature greater than 2000 K, it canbe classed as the first Ultra Hot Neptune. I willdiscuss the detection and confirmation of this newplanet, possible origins, and highlight the unique op-portunities it presents for atmospheric characterisa-tion and further follow-up. Finally, I will briefly dis-cuss additional small planet candidates from TESSthat we are actively following up in Chile, both toconfirm them as bonafide planets and also to con-strain their radii, masses, and bulk densities.

200 — Dynamical Evolution200.01 — Low-Eccentricity Formation of Ultra-Short Period Planets in Multi-Planet Systems

Dong Lai1; Bonan Pu11 Astronomy, Cornell University (Ithaca, New York, United States)

Recent studies suggest that ultra-short period plan-ets (USPs), Earth-sized planets with subday periods,constitute a statistically distinct sub-sample of Keplerplanets: USPs have smaller radii (1-4 Earth radii) andlarger mutual inclinations with neighboring planetsthan nominal Kepler planets, and their period dis-tribution is steeper than longer-period planets. Westudy a ”low-eccentricity” migration scenario for theformation of USPs, in which a low-mass planet withinitial period of a few days maintains a small but fi-nite eccentricity due to secular forcings from exte-rior companion planets, and experiences orbital de-

cay due to tidal dissipation. USP formation in thisscenario requires that the initial multi-planet systemhave modest eccentricities (∼0.1) or angular momen-tum deficit. During the orbital decay of the inner-most planet, the system can encounter several ap-sidal and nodal precession resonances that signifi-cantly enhance eccentricity excitation and increasethe mutual inclination between the inner planets. Wedevelop an approximate method based on eccentric-ity and inclination eigenmodes to efficiently evolvea large number of multi-planet systems over Gyrtimescales in the presence of rapid (as short as 100years) secular planet-planet interactions and othershort-range forces. Through a population synthesiscalculation, we demonstrate that the ”low-e migra-tion” mechanism can naturally produce USPs fromthe large population of Kepler multis under a vari-ety of conditions, with little fine tuning of parame-ters. This mechanism favors smaller inner planetswith more massive and eccentric companion planets,and the resulting USPs have properties that are con-sistent with observations.

200.02 — Relaxation of Resonant Two-planet Sys-tems and their TTVs

Rosemary Mardling11 School of Physics and Astronomy, Monash University (Clayton,

Victoria, Australia)

Many two-planet systems reside near or inside first-order resonances, while many multi-planet systemsform resonant chains. These are normally the prod-uct of planet-disk interactions during the time of for-mation, with eccentricity damping and migration re-sulting in a relaxed system with fewer degrees offreedom than for an arbitrary two-planet system.Are most multi-planet systems in this state? Evenif they reside ‘far’ from resonance? A simple for-mulation describing two-planet systems will be pre-sented which is valid inside, across and outside res-onance. We will show that all such systems are gov-erned by a single two-parameter ordinary integro-differential equation, and that all system informa-tion (variation of eccentricities, orbital frequencies,resonance angles, apsidal orientations, transit timingvariations or TTVs) can be derived from its solution.An expression for the TTVs can be easily inverted tosolve for the planet masses (and other system param-eters) when both planets transit; if no valid inver-sion is possible (given sufficient signal to noise forthe TTVs), it is possible to infer the existence of addi-tional non-transiting planets, the signature of whichwill be imprinted on the signal.

10

200.03 — Signatures of Hidden Friends to Multi-planet systems

Smadar Naoz11 Physics and Astronomy, University of California, Los Angeles

(Los Angeles, California, United States)

Multiplanet systems seem to be abundant in ourGalaxy. These systems typically feature tightlypacked multiple super-Earths or sub-Neptunes withperiods less than a few hundred days. Moreover,these systems seemed to be dynamically calm, withnearly co-planar and circular orbital configurations.In contrast, a significant fraction of single, close in,planets found by Kepler, have larger orbital eccen-tricities. Is the single planet population a result of theinstability episode of the multiplanet population? Ifso, what triggered the instability?

A possible cause for instability is gravitational in-teractions with a distant companion. Radial veloc-ity surveys found a population of giant planets andcompanions at far distances from their host star (e.g.,Knutson et al. 2014; Bryan et al. 2016). Moreover,distant (few AUs) planetary and stellar companionswere identified to orbit some specific tightly packedmultiplanet systems (e.g., Uehara et al. 2016, Bryanet al. 2019, Mills et al. 2019). These companions co-exist with their inner multiplanet system and do nottrigger dynamical instability. Thus, we ask, what arethe allowable orbital configurations that friends forstable multiplanet systems?

I this talk I will present an analytical criterion thatspecifies the possible orbital configuration of a faraway companion to a multiplanet system. I will alsoprovide a set of predictions for the possible distantcompanion’s orbital architecture of existing systems,such as Kepler-56, Kepler-448, Kepler-88, Kepler-109,and Kepler-36. Finally, I will show that a distantcompanion can affect the planets’ obliquity with re-spect to their orbital angular momentum. In turn,this has a unique observable signature on the plan-ets’ flux incident at the top of the atmosphere as afunction of orbital phase.

200.04 — In-Situ Excitation of Warm Jupiter Eccen-tricities: Implications for Dynamical Histories &Migration

Kassandra Anderson1; Dong Lai1; Bonan Pu11 Astronomy, Cornell University (Ithaca, New York, United States)

Warm Jupiters (giant planets with orbital periods10-300 days) are a major topic in exoplanetary dy-namics, given their possible links to hot Jupiters,and unresolved puzzles regarding their dynami-cal histories and migration. Many planets show

hints of a violent past, with substantial eccentrici-ties. High-eccentricity tidal migration is a naturalmechanism for producing eccentric warm Jupiters,but struggles to reproduce other characteristics of thewarm Jupiter population. This talk discusses alter-native dynamical mechanisms for raising eccentrici-ties, starting from a low-eccentricity state consistentwith either a disk migration origin or in-situ forma-tion. First I discuss eccentricity growth due to secu-lar perturbations from an external giant planet com-panion, through an apsidal precession resonance(for low-inclination systems), or Lidov-Kozai cycles(for highly-inclined systems). Taking the sample ofwarm Jupiters with characterized giant planet com-panions, I evaluate the prospects for secular eccen-tricity excitation, and find that high mutual incli-nations (at least 40-50 degrees) are typically neededto produce observed eccentricities. The results ofthis work place constraints on possibly unseen exter-nal companions to eccentric warm Jupiters. Next Idiscuss the possibility of producing eccentric warmJupiters due to in-situ formation of three giant plan-ets, followed by planet-planet scattering. Scatteringat sub-AU distances from the host star results in acombination of planet collisions and ejections, pro-ducing comparable numbers of one-planet and two-planet systems. Two-planet systems arise exclusivelythrough planet-planet collisions, and tend to havelow eccentricities/inclinations and compact config-urations. One-planet systems arise through a combi-nation of ejections and collisions, resulting in muchhigher eccentricities. The observed eccentricity dis-tribution of solitary warm Jupiters is consistent withroughly half of systems having undergone in-situscattering, and the remaining having experienced aquiescent history.

200.05 — Signatures of a Planet – Planet Im-pacts Phase in Exoplanetary Systems Hosting Gi-ant Planets

Renata Frelikh1; Hyerin Jang1; Ruth Murray-Clay1;Cristobal Petrovich2

1 Astronomy and Astrophysics, UC Santa Cruz (Santa Cruz, Cali-fornia, United States)

2 Canadian Institute for Theoretical Astrophysics (Toronto, Ontario,Canada)

Giant planets are often found on substantially non-circular, close-in orbits. An important clue for theirdynamical histories has not yet been explained intheories for the origins of their eccentricities: mostplanets with high eccentricities (e>0.6) tend to alsobe planets of higher mass (m>1 MJ). This is surpris-ing: the orbits of the lower-mass planets in a sys-

11

tem are typically the most easily excited. Further-more, these eccentric planets are preferentially foundaround stars that are metal-rich. We propose thatthese eccentricities arise in a phase of giant impacts,during which the planets scatter each other and col-lide, with corresponding mass growth as they merge.We numerically integrate an ensemble of systemswith varying total planet mass, allowing for colli-sional growth, to show that (1) the high-eccentricitygiants observed today may have formed preferen-tially in systems of higher initial total planet mass,and (2) the upper bound on the observed giant planeteccentricity distribution is consistent with planet-planet scattering.

200.06 — AMD-stability of Planetary Systems

Jacques Laskar1; Antoine Petit11 IMCCE, Observatoire de Paris (Paris, France)

Due to the increasing large number of discoveredplanetary systems, it becomes important to set upsome framework for a rapid understanding of thedynamics of the discovered systems, without theneed of computer intensive numerical simulations.This has been the goal of our recent work on AMD-stability.

In a planetary system, the AMD (Angular Momen-tum Deficit) is the difference between the planar cir-cular angular momentum and the total angular mo-mentum. This quantity is conserved between colli-sions in the average system, and decreases duringcollisions.

This leads to the concept of AMD-stability. A plan-etary system is AMD-stable if the AMD in the sys-tem is not sufficient to allow collisions. The advan-tage of this notion is that it becomes possible to ver-ify very quickly whether a newly discovered plan-etary system is stable or potentially unstable, with-out any numerical integration of the equations ofmotion. These principles have been applied to the131 multiple planetary systems of the exoplanet.eudatabase whose orbital elements are sufficiently welldetermined (Laskar and Petit, 2017a).

AMD-stability, based on the secular evolution, ad-dresses to long time stability, in absence of mean mo-tion resonances. On the other hand, criterions forshort term stability have been established on the ba-sis of Hill radius (Marchal & Bozis 1982; Gladman1993; Pu & Wu 2015) or on the overlap of mean mo-tion resonances ( Wisdom 1980; Duncan et al. 1989;Mustill & Wyatt 2012; Deck et al. 2013). Both longand short time scales can be combined owing somemodification of the AMD-stability criterion (Petit,Laskar & Boué, 2017b). Finally, Hill stability can be

expressed in a very effective and simple way in theAMD framework ( Petit, Laskar, Boué, 2018).

Ref: Laskar, J. and Petit, A.C., 2017a, AMD-stability and the classification of planetary systems,A&A, 605, A72 Petit, A.C. Laskar, J. and Boué,G., 2017b, AMD-stability in presence of first ordermean motion resonances, A&A, 607, A35 Petit, A.C.Laskar, J. and Boué, G., 2018, Hill stability in theAMD framework, A&A, 617, A93

201 — Ultrashort Periods andPlanet-Star Interactions201.01 — Remote Sensing of Extreme Worlds:High-Resolution Spectroscopy of Exoplanet Atmo-spheres

Ray Jayawardhana1; Ernst J.W. De Mooij2; Jake Turner1;Emily Deibert3; Miranda Herman3; Andrew Ridden-Harper4; Abhinav Jindal3; Raine Karjalainen5; MarieKarjalainen5

1 Cornell University (Ithaca, New York, United States)2 School of Physical Sciences and Centre for Astrophysics and Rela-

tivity, Dublin City University (Dublin, Ireland)3 University of Toronto (Toronto, Ontario, Canada)4 Department of Astronomy, Cornell University (Ithaca, New York,

United States)5 Isaac Newton Group of Telescopes (Santa Cruz de La Palma,

Spain)

High-resolution spectroscopy, combined with theDoppler cross-correlation technique, is emergingas a powerful and robust probe of exoplanet at-mospheres. Here we will report on our wide-ranging observational program targeting Jovian-mass worlds, sub-Saturns and super-Earths usinga suite of frontline instruments in the optical aswell as the near-infrared. In particular, we willpresent new findings and on-going work related totwo of the hottest gas giants and a very hot terres-trial planet. With a dayside at ∼4000K, KELT-9b is aso-called ultra-hot Jupiter with a temperature akinto those of dwarf stars. Transmission spectra re-veal a host of metal lines in its atmosphere. Withnew Calar Alto/CARMENES observations, we notonly confirm strong Hα in its extended exosphere,but also report robust detections of the resolved CaIItriplet for the first time (paper in prep.). We havealso observed two transits and portions of six phasecurves of WASP-33b (∼3000K), with Keck/HIRESand CFHT/ESPaDOnS to obtain transmission andemission spectra at high spectral resolution, to detectmolecular signatures of TiO and water vapor (paperin prep.) Finally, we report a sensitive new search

12

for water vapor and TiO in the atmosphere of thenearby very hot super-Earth 55 Cancri e (∼2700 K)using a combination of data from Gemini/GRACES,Subaru/HDS and CFHT/ESPaDOnS (paper submit-ted). Our findings suggest that unless the signal issuppressed significantly by clouds/haze, this planetmay well be bone-dry. Moreover, we have recentlyobtained high-resolution near-infrared spectra of 55Cnc e from CARMENES at Calar Alto as well as thebrand-new SPIRou instrument on the CFHT, and ex-pect to present first results at ESS IV.

201.02 — Stellar Systems at Low Radio Frequencies:The Discovery of Radio Exoplanets

Joseph Callingham1; Harish Vedantham1; TimShimwell1; Benjamin J S Pope2,3; Megan Bedell4

1 ASTRON, Netherlands Institute for Radio Astronomy(Dwingeloo, Netherlands)

2 New York University (New York, New York, United States)3 Sagan Fellow (New York, New York, United States)4 Flatiron Institute (New York, New York, United States)

For more than thirty years, radio astronomers havesearched for auroral emission from exoplanets. WithLOFAR we have recently detected strong, highlycircularly polarised low-frequency (144 MHz) radioemission associated with a M-dwarf — the expectedsignpost of such radiation. The star itself is quies-cent, with a 130-day rotation period and low X-rayluminosity. In this talk, I will detail how the radioproperties of the detection imply that such emissionis generated by the presence of an exoplanet in ashort period orbit around the star, and our follow-up radial-velocity (RV) observations with Harps-Nto confirm the exoplanet’s presence. Our study high-lights the powerful new and developing synergy be-tween low-frequency radio astronomy and RV obser-vations, with radio emission providing a strong prioron the presence of a short-period planet. I will con-clude the talk detailing how the radio detection ofan star-exoplanet interaction provides unique infor-mation for exoplanet climate and habitability stud-ies, and the extension of our survey to other stellarsystems.

201.03 — Mass Loss from the Exoplanet WASP-12bInferred from Spitzer Phase Curves

Taylor James Bell1; Michael Zhang2; Patricio Cubillos3;Lisa Dang1; Luca Fossati3; Kamen O. Todorov4; NickB. Cowan1,5; Drake Deming6; Robert T. Zellem7;Kevin Stevenson8; Ian Crossfield9; Ian Dobbs-Dixon10;Jonathan Fortney11; Heather Knutson12; Michael Line13

1 Department of Physics, McGill University (Montreal, Quebec,Canada)

2 Department of Physics, NYU Abu Dhabi (Abu Dhabi, UnitedArab Emirates)

3 Other Worlds Laboratory, University of California, Santa Cruz(Santa Cruz, California, United States)

4 Division of Geological and Planetary Sciences, California Instituteof Technology (Pasadena, California, United States)

5 School of Earth & Space Exploration, Arizona State University(Tempe, Arizona, United States)

6 Department of Astronomy, California Institute of Technology(Pasadena, California, United States)

7 Space Research Institute, Austrian Academy of Sciences (Graz,Austria)

8 Anton Pannekoek Institute for Astronomy, University of Amster-dam (Amsterdam, Netherlands)

9 Department of Earth and Planetary Science, McGill University(Montreal, Quebec, Canada)

10 Department of Astronomy, University of Maryland (CollegePark, Maryland, United States)

11 Jet Propulsion Laboratory, California Institute of Technology(Pasadena, California, United States)

12 Space Telescope Science Institute (Baltimore, Maryland, UnitedStates)

13 Department of Physics, Massachusetts Institute of Technology(Cambridge, Massachusetts, United States)

As an exoplanet orbits its star, the thermal radiationcoming from the planet varies roughly sinusoidally,with a peak occurring when the hottest hemisphereof the planet faces the observer. A Spitzer Space Tele-scope phase curve of the ultra-hot Jupiter WASP-12bfrom 2010 showed an unexplained anomaly in thedata; unlike every other planet observed to date, theinfrared signal from WASP-12b showed two maximaper planetary orbit, rather than one (Cowan et al.2012). Stranger still, this was only seen at a wave-length of 4.5 μm, while the phase curve at 3.6 μmshowed only one maximum. At the time, the authorsdismissed this finding as being the result of detectorsystematics.

We present new work which robustly confirms thefindings of Cowan et al. (2012) through the analysisof new Spitzer phase curves taken in 2013, as well asthe reanalysis of their original data. We obtain con-sistent results at both epochs using three indepen-dent analyses. We rule out the possibility of detec-tor systematics, nor can tidal distortion of the planetor star explain these variations. We then show thatthese observations require the planet to be undergo-ing mass loss, likely in the form of a gas stream flow-ing directly from the planet to the star. While massloss has been predicted for WASP-12b, our inferredflow geometry is unanticipated. We also find strong

13

evidence for atmospheric variability, with the offsetin the phase curve maximum at 3.6 μm changing bymore than 6σ between the two sets of observations.

Our findings provide an independent confirma-tion of past claims of mass loss from near-ultraviolettransit observations. We also show that our obser-vations provide new constraints on the composition,flow geometry, and temperature of the gas strippedfrom the planet. For example, the wavelength de-pendence of the gas emission may suggest that thegas is rich in CO. Finally, many of the past findingsregarding the atmosphere of WASP-12b, one of thebest-studied exoplanets, will need to be reconsideredin light of the contamination from the escaping gas.

201.04 — Observations of Tidal Orbital Decay ofHot Jupiters

Joshua Winn11 Astrophysical Sciences, Princeton University (Princeton, New

Jersey, United States)

Soon after the discovery of 51 Peg b, Rasio et al.(1996) and Lin et al. (1996) realized that tidal in-teractions between hot Jupiters and their host starsmay lead to significant orbital evolution. In particu-lar, the orbits of almost all of the known hot Jupitersshould be shrinking due to tidal orbital decay. Thetimescale for tidal decay is unknown and dependson the mechanism by which tidal oscillations of thestar are dissipated as heat, a longstanding sourceof uncertainty in stellar astrophysics. The best op-portunity to detect orbital decay directly is throughlong-term transit timing. I will present the results ofsearch for orbital decay among the dozen most fa-vorable hot Jupiters, some of which have now beenobserved for more than a decade. WASP-12 showsa clear decrease in the transit period which seemslikely to be caused by either tidal orbital decay orapsidal precession. I will present two new seasonsof transit observations, and four new Spitzer obser-vations of eclipses, that help to distinguish betweenthese possibilities. In addition to WASP-12, two othercandidates for orbital decay have been identified, butthe evidence is not compelling and further observa-tions are needed. I will also discuss the prospects fordetecting orbital decay using data from the Transit-ing Exoplanet Survey Satellite.

201.05 — Tidally-Induced Radius Inflation of Sub-Neptunes

Sarah Millholland1; Gregory Laughlin11 Yale University (New Haven, Connecticut, United States)

Recent work suggests that many short-period super-Earth and sub-Neptune planets may have signifi-cant spin axis tilts (“obliquities”). When planetsare locked in high-obliquity states, the tidal dissipa-tion rate increases by several orders of magnitude.This intensified heat deposition within the planets’interiors should generate significant structural con-sequences, including atmospheric inflation leadingto larger transit radii. Using up-to-date radius esti-mates from Gaia Data Release 2 and the California-Kepler Survey, we show evidence for larger averageradii of planets wide of first-order mean-motion res-onances, a population of planets with theorized fre-quent occurrence of high obliquities. We investigatewhether this radius trend could be a signature ofobliquity tides. Using an adaptation of the Mod-ules for Experiments in Stellar Astrophysics (MESA)stellar evolution toolkit, we model the evolution ofthe H/He envelopes of sub-Neptune-mass planetsin response to additional internal heat from obliq-uity tides. The degree of radius inflation predictedby the models is indeed consistent with the observa-tions, suggesting that these planets have likely beeninflated. We present several case studies that areparticularly strong candidates for having undergonethis process. Broadly speaking, we find that tidal dis-sipation can affect a sub-Neptune’s radius to first or-der, yet it has not been included in previous interiorstructure models. This must be accounted for if thevalley in the super-Earth/sub-Neptune radius distri-bution is to be fully understood.

201.06 — The Life Expectancy of Hot Jupiters

Benjamin Montet11 Astronomy and Astrophysics, University of Chicago (Chicago,

Illinois, United States)

Short period giant planets are theorized to spiral intotheir host stars on relatively short timescales, withthe exact rate of inspiral dependent on the strengthof tidal forces on main sequence stars. Thus, bymeasuring the occurrence rate of hot Jupiters as afunction of time, one can effectively measure thestrength of stellar tidal dissipation. For this, onemust measure the ages of stars. I will discuss twoavenues through which we can accomplish this forplanet hosts. The first is an analysis of the NGC6791 cluster, an old (8 Gyr), metal-rich ([Fe/H] ∼+0.3) cluster observed throughout the Kepler mis-sion. While this cluster is too crowded for planetsearches through traditional pipelines, we (have) de-veloped a PSF modeling scheme to search for andfind transiting planets in and toward this cluster. I

14

will present the results from this search and the im-plications for the destruction of hot Jupiters in time.

We can also understand the ages of planethosts when they are associated with other, well-characterizable stars. Gaia is now providing us withdata on widely separated, co-moving systems acrossthe sky, including in the Kepler and K2 fields. Withtwo stars, we can use isochronal or gyrochronologi-cal information on the non-planet hosting star in thesystem to understand the system age, without wor-rying about potential tidal spin-up from the exist-ing hot Jupiter. I will present how this method ishelping us understand the ages of field stars, the rateof tidal spin-up by hot Jupiters on their host stars,and the long-term evolution of these planetary sys-tems. I will also examine how data from Kepler andK2 can be useful for understanding the prevalenceof lithium-rich red giants across the sky, potentiallyanswering a longstanding question in stellar physics.

201.07 — Key Planets for Exogeology in the 2020s:Discoveries from the Dispersed Matter PlanetProject

Carole Ann Haswell1; John Barnes1; Daniel Staab1; LucaFossati2; Guillem Anglada-Escude3; James Jenkins4

1 School of Physical Sciences, The Open University (Milton Keynes,United Kingdom)

2 Space Research Institute, Austrian Academy of Sciences (Graz,Austria)

3 School of Physics and Astronomy, Queen Mary University ofLondon (London, United Kingdom)

4 Departamento Astronomia, Universidad de Chile in Santiago(Santiago, Chile)

HST revealed a complete lack of chromosphericemission from extreme hot Jupiter (hJ) host starWASP-12. We since found ∼40% of hJ hosts haveanomalously low chromospheric emission in Ca IIH&K. We attribute this to absorption in diffuse cir-cumstellar gas, originiating from the highly irra-diated planets. Archival spectra of ∼6000 bright,nearby stars revealed 39 main sequence field starswith similar deficits in the Ca II H&K line cores.These 39 targets were not known planet hosts; wehypothesized they harboured, close-in, mass-losing,low mass, small planets. The Dispersed Mat-ter Planet Project (DMPP) makes high precision,high cadence RV measurements to find ∼Earth-massplanets in short period (<∼6 d) orbits. We havefound planets wherever we have more than 60 RVmeasurements. We will present DMPP-1, a com-pact multi-planet system containing multiple super-Earths in short period orbits; DMPP-2, a hot Saturnmass planet orbiting a pulsating star; and DMPP-3.

DMPP-3AB is a K dwarf and a star just above theminimum mass for Hydrogen burning in an e=0.59,500d orbit. DMPP-3Ab is a super-Earth planet ina 6 d orbit. DMPP-3 is the most compact knownS-type planet host. Angular momentum considera-tions suggest the mass lost from the ablating planetswill be concentrated in the orbital plane of the ab-lating planet(s), so these systems are likely to transit.Indeed DMPP was partly motivated by the search forbright, nearby analogues of Kepler 1520b; the tran-siting dust in Kepler 1520b must co-exist with metal-rich circumstellar vapour which would absorb in theresonance lines of abundant elements, producing ab-sorption exactly like the Ca II H&K line core deficitswe use to select DMPP targets. The dispersed gasin the DMPP systems has a large scale-height com-pared to a terrestrial planet atmosphere, and is henceamenable to transmission spectroscopy techniquesexploiting azimuthal column density variations toreveal the composition of the ablating planetary sur-face. Thus the DMPP systems offer unprecedentedopportunities to directly measure the mass-radius-composition relationship(s) for rocky planets outsideour Solar System.

202 — Stellar Spins and Obliquities202.01 — New developments on the obliqueness ofexoplanet systems

Simon Albrecht2; Rebekah Ilene Dawson1; JoshuaWinn3; Maria Hjorth2; Emil Knudstrup2; AndersJustesen2

1 Pennsylvania State University (University Park, Pennsylvania,United States)

2 Stellar Astrophysics Centre, Department of Physics and Astron-omy, Aarhus University (Aarhus C, Denmark)

3 Princeton University (Princeton, New Jersey, United States)

The angle between the rotation axis of a star and itsorbital angular momentum – its obliquity – conveysinformation about the formation and evolution of thestar and its planetary system. Here I report on newtrends and results we have recently obtained andtheir possible interpretations. Our results give newindications about which mechanisms are responsi-ble for generating large obliquities in some systems,and whether tidal alignment is an important factorin shaping obliquity distributions.

(1) We have found that high eccentricities and highobliquities are linked, suggesting that close-in giantplanets can be separated into dynamically cool anddynamically hot populations, in line with other re-cent results.

15

(2) Observations of protopanetary disks as wellas some models suggest that protoplanetary disksneed not be well-alighed with the stellar equator.However, out of the ten obliquity measurements inmulti transiting systems (tracing the disk plane) onlyone has coplanar planets on an oblique orbit. Andeven in this exceptional system, Kepler-56, the largetilt may be caused by a fourth body, not primordialmisalignment. We have also tentatively identified amulti-transiting system in which the planets appearto travel on retrograde orbits. Additional transit ob-servations are scheduled for June and should clarifythe situation.

(3) Tidal alignment has been invoked to explain theobserved dependence of the obliquity distributionon the host star’s effective temperature, the planet-star mass ratio, and the orbital separation. However,theoretical and observational counterarguments ex-ist. We report on two new trends that suggest tidesare indeed important: (i) Stars with retrograde plan-ets have a lower projected stellar rotation speed thanprograde stars. (ii) Effective temperature is a bet-ter predictor of high obliquity than stellar mass. Weshow that the current sample of 140 systems with re-liable obliquity measurements is generally consistentwith a picture of tidal alignment.

202.02 — The Spin-Orbit Misalignment Distribu-tions of Hot Jupiters

Marshall C. Johnson1; Aaron Rizzuto2; Daniel J.Stevens3

1 The Ohio State University (Columbus, Ohio, United States)2 University of Texas at Austin (Austin, Texas, United States)3 Pennsylvania State University (University Park, Pennsylvania,

United States)

Many hot Jupiters have orbits that are highly mis-aligned with respect to the stellar rotation; most mis-aligned planets orbit stars above the Kraft break,which is generally though to be the result of less ef-ficient tidal damping in hotter stars. Many mech-anisms have been proposed to generate misalignedorbits, but previous observational constraints havebeen unable to definitively distinguish among thesemechanisms. We present initial results from an ex-tensive program to address this problem using statis-tical analyses of the spin-orbit misalignments of hotJupiters around A and early F stars, and correlationswith other parameters of the systems.

We demonstrate that there is not a sharp breakin the spin-orbit misalignment distribution at theKraft break as typically assumed, but rather a moregradual transition; a significant population of well-aligned planets exists up to at least Teff∼6600 K,

and early F stars’ planets require a two-populationmodel. Less massive and longer-period planets tendto have more misaligned orbits, consistent with ex-pectations from tidal damping, suggesting that thetidal damping of obliquities around these stars maybe stronger than previously assumed. There is nocorrelation between metallicity and misalignment,contrary to expectations from planet-planet scatter-ing.

We present results from new Keck NIRC2 non-redundant aperture masking interferometry obser-vations, coupled with archival high-resolution imag-ing and Gaia DR2 astrometry, to perform a compre-hensive survey for stellar companions to these starsfrom a few to tens of thousands of AU. Such com-panions could drive migration and misalignmentsvia the Kozai-Lidov mechanism. We tentatively findthat hot Jupiters with stellar companions have moremisaligned orbits than those that do not, contrary toprevious results.

We are also leveraging Gaia DR2 parallaxes andTESS light curves to measure these stars’ densitiesand radii and thus infer their ages and the orbitaleccentricities. We demonstrate that we are typicallyable to measure their ages to a precision of 500 Myr,much better than is typically possible for field stars.We present initial results from this work.

202.03 — Tilting short-period planetary systems inphoto-evaporating disks.

Cristobal Petrovich1; Wei Zhu11 Canadian Institute for Theoretical Astrophysics (Toronto, Ontario,

Canada)

The current sample of nearly a dozen Neptune-massplanets with stellar obliquity measurements revealsa surprising population that move on nearly polarorbits (4 planets with obliquities of ∼90 degrees,e.g., GJ 436b, HAT-P-11b). The absence of stel-lar companions and the likely presence of nearbyplanetary companions in these systems makes thecurrent proposals to drive large obliquities in hotJupiter systems (e.g., primordial disk misalignmentsor high-eccentricity migration) inapplicable to thesesub-Jovian planets. In this talk, I will show thatlarge stellar obliquities in short-period planetary sys-tems are naturally excited by a distant (>1 AU) Jo-vian companion embedded in a photo-evaporatingdisk. This excitation is the result of secular reso-nances and often leads to polar planetary systems,even if the primordial tilts are small (∼5 degrees), re-vealing that: (i) the disk dispersal phase plays a ma-jor role at shaping the architecture of planetary or-bits; (ii) large obliquities might be intimately linked

16

to the observed misalignments between inner andouter disks in many transitional disks. Beyond thesmall sample of sub-Jovian planets with obliquitymeasurements, we show that our proposed mecha-nism fits well within the observed trends of obliqui-ties with stellar metallicities and orbital separationsfound in the Kepler sample. Finally, we provide withpredictions from this mechanism that can be imme-diately tested by ongoing follow-up campaigns withTESS, as well as future Gaia’s astrometric measure-ments.

202.04 — Nearly Polar orbit of the sub-NeptuneHD3167 c : Constraints on a multi-planet systemdynamical history

SHWETA Dalal1; Guillaume Hebrard1,2; Alain Lecave-lier des Etangs1; Antoine Petit3; Vincent Bourrier4;Jacques Laskar3; Pierre-Cecil Konig1; Alexandre C.M.Correia5,3

1 Institut d’astrophysique de Paris (Paris, France)2 Observatoire de Haute-Provence (Saint-Michel-l’Observatoire,

France)3 Observatoire de Paris (Paris, France)4 Observatoire de Genève (Geneva, Switzerland)5 University of Coimbra (Coimbra, Portugal)

We present the obliquity measurement, that is, theangle between the normal angle of the orbital planeand the stellar spin axis, of the sub-Neptune planetHD3167 c, which transits a bright nearby K0 star.We study the orbital architecture of this multi-planetsystem to understand its dynamical history. Wealso place constraints on the obliquity of planetd based on the geometry of the planetary systemand the dynamical study of the system. New ob-servations obtained with HARPS-N at the Telesco-pio Nazionale Galileo (TNG) were employed forour analysis. The sky-projected obliquity was mea-sured using three different methods: the Rossiter-McLaughlin anomaly, Doppler tomography, andreloaded Rossiter-McLaughlin techniques. We per-formed the stability analysis of the system and inves-tigated the dynamical interactions between the plan-ets and the star. HD3167 c is found to be nearly polarwith sky-projected obliquity, λ = -97 ± 23 degrees.This misalignment of the orbit of planet c with thespin axis of the host star is detected with 97% con-fidence. The analysis of the dynamics of this sys-tem yields coplanar orbits of planets c and d. It alsoshows that it is unlikely that the currently observedsystem can generate this high obliquity for planets cand d by itself. However, the polar orbits of plan-ets c and d could be explained by the presence of

an outer companion in the system. Follow-up obser-vations of the system are required to confirm such along-period companion.

202.05 — Precision Rossiter-McLaughin Observa-tions of M-Dwarf Planets in the Near-Infrared withthe Habitable-zone Planet Finder

Gudmundur Kari Stefansson11 The Pennsylvania State University (University Park, Pennsylva-

nia, United States)

Significant progress has been made in recent years inmeasuring the sky-projected obliquity distributionof FGK planet hosting systems via precise Rossiter-McLaughlin (RM) effect observations. However, cur-rently only one M-dwarf system, GJ 436 has a pub-lished obliquity constraint via the RM effect—whichinterestingly is observed to be misaligned. With onlyone published measurement, key questions remainabout the dynamical histories of M-dwarf planets.The advent of stabilized extremely precise RV spec-trographs in the near-infrared (NIR) is opening thedoors to answering these questions, capitalizing onthe large RM-effect amplitudes—large compared totheir Doppler RV amplitude—produced by transit-ing exoplanets orbiting around rapidly-rotating M-dwarfs. In this talk, we will discuss recent precisionRM effect observations of fully-convective M-dwarfswith the Habitable-zone Planet Finder (HPF), a sta-bilized NIR spectrograph recently commissioned onthe 10m Hobby-Eberly Telescope at McDonald Ob-servatory. We will discuss recent RM effect observa-tions of the fully-convective M-dwarfs K2-25b, andTRAPPIST-1b, early results of which favor alignedand misaligned orbits, respectively, yielding impor-tant insights into the formation history and evolutionof these systems.

202.06 — Constraints on the 3D Angular Momen-tum Architecture of a Planetary System

Marta Levesque Bryan1; Brendan Bowler2; SarahBlunt3; Henry Ngo4; Caroline Morley5; Dimitri Mawet6

1 Astronomy, UC Berkeley (Berkeley, California, United States)2 Astronomy, The University of Texas at Austin (Austin, Texas,

United States)3 Caltech (Pasadena, California, United States)4 NRC Herzberg (Victoria, British Columbia, Canada)5 Astronomy, University of Texas at Austin (Austin, Texas, United

States)6 Astronomy, Caltech (Pasadena, California, United States)

Studying 3D architectures of planetary systemspresents a unique window into their formation his-

17

tories. A full description of a system’s geometric ori-entation requires measuring the stellar spin, plane-tary spin, and orbital angular momentum vectors. Inthe past, studies have focused on just one or two ofthese vectors, namely the orbital plane and the stel-lar spin axis. For instance, measurements of pro-jected spin-orbit alignment have found a numberof hot Jupiter systems that are misaligned, raisingthe possibility that the misalignments resulted fromhigh-eccentricity migration and dynamical interac-tions between planets. In addition, while discov-eries of multi-planet compact coplanar systems in-dicate smooth disk migration, a handful of thesesystems exhibit spin-orbit misalignment, suggestingthe disk itself was tilted relative to the stellar spinaxis. While these partial views of 3D architecturesprovide an important perspective on planet forma-tion, we can learn even more about system forma-tion histories by characterizing all three angular mo-mentum vectors. Here we present our study char-acterizing the three angular momentum vectors ofthe directly imaged single-planet system 2M0122-2439. In this study we measure for the first time pro-jected spins for both star and planet from our NIRhigh-resolution Keck/NIRSPEC spectra. We com-bine these with our new stellar photometric rota-tion period measurement from TESS data and witha previously published planetary rotation period toobtain spin axis inclinations for both objects. Wefit multiple astrometry epochs, including two un-published epochs, to constrain the companion’s or-bital inclination. With these pieces needed to mea-sure the three angular momentum vectors, we pro-vide first measurements of the planetary obliquity,stellar obliquity, and relative inclinations betweenstar/planet spin axes, allowing us to place new con-straints on this system’s formation history.

203 — Transiting Multi-planet Sys-tems203.01 — A Family of Newborn Planets Transitinga Young Solar Analog at 20-30 Myr

Trevor David1,2; Ann Marie Cody3; Christina Hedges4;Eric Mamajek1; Lynne Hillenbrand5; David Ciardi6;Charles Beichman6,1; Erik Petigura5; B.J. Fulton6;Howard Isaacson7; Andrew Howard5; Jonathan Gagné8;Nicholas Saunders3; Luisa Rebull9; John Stauffer10;Gautam Vasisht1; Sasha Hinkley11

1 Jet Propulsion Laboratory, California Institute of Technology(Pasadena, California, United States)

2 Caltech/IPAC-SSC (Pasadena, California, United States)

3 University of Exeter (Exeter, United Kingdom)4 Flatiron Institute (New York, New York, United States)5 NASA Ames Research Center (Moffett Field, California, United

States)6 Bay Area Environmental Research Institute (Moffett Field, Cali-

fornia, United States)7 California Institute of Technology (Pasadena, California, United

States)8 Caltech/IPAC-NASA Exoplanet Science Institute (Pasadena,

California, United States)9 University of California, Berkeley (Berkeley, California, United

States)10 Université de Montréal (Montréal, Quebec, Canada)11 Caltech/IPAC-IRSA (Pasadena, California, United States)

Compact, multi-planet systems are one of the defin-ing discoveries of the Kepler mission. These plane-tary systems are ubiquitous in the galaxy yet muchabout their nature remains a mystery, includingwhether they formed in situ and what their architec-tures were when the protoplanetary disk dispersed.Theoretical models suggest that close-in Kepler plan-ets had radii that were roughly 2 to 10 times largerat the time of disk dispersal. With the recent dis-coveries of exoplanets transiting young stars (<100Myr), it is now possible to put these models to thetest and study close-in planets at a stage when con-traction, cooling, and initial atmospheric loss are stillunderway. To date, only a few exoplanets have beendiscovered transiting pre-main sequence stars, all ofwhich are currently single-planet systems. I will dis-cuss the recent detection of four transiting planetslarger than 5 Earth radii orbiting within ∼0.5 AU ofa young solar analog aged between 20-30 Myr. Theinner planets are larger than Neptune and expectedto be actively losing envelope mass through photo-evaporation. The outer planets are both Jupiter-sizedand, with separations >0.15 AU, they are expectedto be largely shielded from the effects of photo-evaporation. Consequently, the outer planets maybe particularly valuable benchmarks, with propertiesthat may more closely reflect the initial conditionsof Kepler planets. In a single system, we thus havethe opportunity to study proto-Kepler planets acrossa range of insolation fluxes shortly after the accretionof envelopes and at a time when stellar X-ray emis-sion is near its peak.

203.02 — Higher Compact Multiple OccurrenceAround Metal-Poor M-Dwarfs

Sophie Anderson1; Jason Dittmann1; Sarah Ballard1;Megan Bedell2

1 Massachusetts Institute of Technology (Cambridge, Mas-sachusetts, United States)

18

2 Flatiron Institute (New York, New York, United States)

The planet-metallicity correlation provides a poten-tial link between exoplanet systems as we observethem today and the effects of bulk composition onthe planet formation process. Many observers havenoted a tendency for Jovian planets to form aroundstars with higher metallicities. However, there is noconsensus on a trend for smaller planets. In this talk Iwill discuss my recent work investigating the planet-metallicity correlation for rocky planets in single andmulti-planet systems around Kepler M-dwarf stars.M-dwarfs are too dim to easily make direct elementalabundance measurements using spectroscopy, so weinstead used a combination of parallaxes and pho-tometry to find relative metallicities. We used color-magnitude diagrams to show that compact multi-ple systems prefer metal-poor M-dwarfs, with 73%of compact multiple systems, compared to 55% ofsingle-planet systems, in orbit around a star moremetal poor than the average star in the MK vs. GBP–GRP plane. This trend is broadly consistent withother choices of color and magnitude. Our conclu-sion is that metallicity plays a role in the architectureof rocky planet systems. Compact multiples eitherform more readily, or are more likely to survive onGyr timescales, around metal-poor stars.

203.03 — Unlocking the Interpretation of Transit-ing Multiplanet Systems using High Impact Pa-rameters

Daniel Clark Fabrycky1; Gregory Gilbert1; AaronHamann1; Benjamin Montet1; Eric Agol2; Ethan Kruse3

1 Astronomy and Astrophysics, University of Chicago (Chicago,Illinois, United States)

2 Astronomy, University of Washington (Seattle, Washington,United States)

3 Sciences and Exploration Directorate, NASA Goddard SpaceFlight Center (Greenbelt, Maryland, United States)

Multiplanet systems in which all the planets transitoffer clean individual solutions and precise parame-ter constraints. In this talk, we discuss new ideas andresults regarding planetary systems in which someof the members transit with high impact parameter:they enable even better solutions. Moreover, com-puting multiplanet statistics with impact parametersresults in a clear global view of the population andcan infer the existence or absence of non-transitingplanets. We discuss one individual system in de-tail: K2-146. In campaign 5 of K2, one planet wasknown to transit with large transit timing variations(TTVs). In campaigns 16 and 18 we found its per-turber to transit also. The two planets form a 3:2 res-

onance, and the fast precession of the nodes and pe-riastra caused the outer planet’s impact parameter todecrease from 0.99 to 0.89 in the ∼3 years spanned byK2 observations, tripling its observed transit depth.Interpreting the transit shape variations along withTTVs allows us to measure the most precise masses(Mp/MEarth = 5.77±0.18 and 7.50±0.23) for any super-Earth or sub-Neptune (Rp/REarth = 2.04±0.06 and2.19±0.07, respectively). We further infer that thesevalues support photoevaporation models. Zoomingout to the population of transiting planetary systemsof high multiplicity, we define statistics for mass par-titioning (based on radii), spacing complexity (basedon periods), and system flatness (based on impactparameters through durations and stellar densities).For the first time, we find the three to be correlated:if a multiplanet system has similar-sized planets thatare laid out regularly in period, then it tends to be ex-tremely flat with mutual inclinations <1 deg. We canturn this argument around and infer more rigorouslythan before that some systems have non-transitingplanets — we identify eight systems of 3 planets withconspicuous gaps, where a 4th planet would com-plete an evenly-spaced set. Far from a blind appli-cation of Titius-Bode’s rule, our inference of addi-tional planets derives from system layouts actuallyobserved in the population.

203.04 — The Intrinsic Distribution of PlanetarySystems: Modeling the Impact of Clustering onPlanetary Architectures

Matthias Yang He1; Eric Ford1; Darin Ragozzine21 Department of Astronomy & Astrophysics, The Pennsylvania

State University (State College, Pennsylvania, United States)2 Physics and Astronomy, Brigham Young University (Provo, Utah,

United States)

The Kepler Mission discovered thousands of exo-planet candidates and hundreds of multi-transitingsystems, enabling detailed population studies. Wepresent a forward model for generating populationsof exoplanetary systems, modeling the Kepler de-tection pipeline, comparing simulated planetary sys-tems to those observed by Kepler, and performingstatistical inference on planetary system architectures,and not just the distribution of individual planets.We show that models assuming independent planetsizes and orbital periods do not adequately repro-duce the observed population and can lead to in-accurate inferences about the intrinsic distributionsof multiplicities, period ratios, and mutual inclina-tions. In contrast, our model allowing for the clus-tering of planet sizes and periods within each systemprovides a significantly improved description of the

19

Kepler multi-planet systems, especially in modelingthe observed multiplicity and period ratio distribu-tions. Our results are consistent with previous stud-ies finding that the observed multiplicity distribu-tion implies two populations of planetary systems, amajority coming from a low mutual inclination (∼1°)population with low eccentricities (∼0.01) and a sec-ond population of systems with high mutual inclina-tions (or isolated planets) making up ∼40% of all sys-tems. However, we find that a large fraction of starsdo not harbor any planets (with Rp > 0.5 REarth and10 d < P < 300d ). Those that do tend to have many.Our model makes predictions for the distribution ofadditional non-transiting planets that can inform ob-serving strategies for the follow-up of TESS discover-ies and can be tested by the findings of upcoming ra-dial velocity campaigns. We provide large simulatedcatalogs drawn from our models for testing whetherapparent trends in the Kepler catalog could be dueto observational biases or are evidence of patterns inthe true distribution of planetary systems’ physicalproperties. We provide a public code for simulatingboth intrinsic and Kepler-observed catalogs of plan-etary systems enabling comparisons between planetformation models and observations.

203.05 — Linguistic Modeling of Kepler’s Exoplan-ets

Emily Sandford1; David Kipping1; Michael Collins21 Department of Astronomy, Columbia University (Staten Island,

New York, United States)2 Department of Computer Science, Columbia University (New

York, New York, United States)

Planets belonging to the same system are related:formed from the same protoplanetary disk aroundthe same star, and in general, size-ordered, withcorrelated radii and dynamically organized spacing.As a consequence, the individual planets in a sys-tem inform upon each other, and upon their yet-unobserved siblings.

Here, we apply the techniques of natural languageprocessing to the problems of (i) classifying planetsinto maximally informative categories, analogous toparts of speech; and (ii) predicting the next unob-served planet orbiting a star. In the case of text in-terpretation, we expect a word’s context to informupon the word itself; for example, the incompletesentence ”The plane landed in ____ yesterday” con-tains enough information to limit the set of sensiblemissing words to a relative handful.

We treat planetary systems as ”sentences” madeup of planet ”words,” ordered from the star outward.By our analogy, we expect the properties of a planet’s

host star and neighboring planets to inform us of thelikely properties of the planet itself, and we apply thetechnique of maximizing mutual information (MMI)to model the relationships between these properties.

203.06 — Compact Multi-Planet Systems WithPlanets That are “Way Out of Line”

Joseph E. Rodriguez1; Juliette Becker2; AndrewVanderburg3; Samuel Quinn1; Jason Eastman1; SamHadden1

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 University of Michigan (Ann Arbor, Michigan, United States)3 University of Texas (Austin, Texas, United States)

The Kepler Mission led to the discovery of hun-dreds of multi-planet systems. Remarkably, thesesystems tend to have relatively flat architectures (lessthan a few degrees), much smaller than the mis-alignments seen in our own Solar system. Recently,however, using observations from the K2 mission,we have discovered a compact six-planet system ofsub-Neptunes orbiting a nearby (78 pc) K-star, K2-266. Interestingly, this system contains an ultra-short-period (USP) super Earth that is significantlymisaligned (>12 degrees) to the other five planets.More recently, we discovered a system of up to fiveplanets around an early K-star TOI 125 using datafrom the TESS mission. . Similar to K2-266, TOI-125has a USP planet candidate, that if confirmed, wouldbe misaligned to the rest of the system by ∼17 de-grees. To date, these systems are the most exoplanetdiscovered in one system by K2 and TESS, respec-tively. As a result of their close proximity to theirhost stars, the USPs in each system actually transitalong our line-of-sight. One explanation for the mis-alignments is that an additional unseen companionresides in each system. In this scenario, the strengthof its dynamical coupling could vary between each ofthe inner planets, potentially resulting in a differencein the evolution of each planet’s inclination, caus-ing the observed misalignment. We will present thediscovery and characterization of each systems, andprovide a range of parameters for the unseen com-panion that could explain the misalignment seen inthe K2-266 system.

J.E.R. is supported by the Harvard Future FacultyLeaders Postdoctoral fellowship.

20

300 — Planet Formation

300.01 — Results from the New Horizons encounterwith 2014 MU69 and what they tell us about plan-etary formation

Catherine Olkin1; S. Alan Stern1; Harold A. Weaver2;John Spencer1; William B. McKinnon3

1 Space Sciences, Southwest Research Institute (Boulder, Colorado,United States)

2 JHU/APL (Laurel, Maryland, United States)3 Washington University in St. Louis (St. Louis, Missouri, United

States)

On January 1, 2019, NASA’s New Horizons mis-sion flew past the cold classical Kuiper Belt Object,2014 MU69 [1]. This was the furthest encounter ofa solar system object and our first close-up look ata cold classical Kuiper Belt Object (CCKBO). As aCCKBO with a circular, low-inclination orbit, 2014MU69 likely formed at the same heliocentric distancefrom the protoplanetary disk about 4.5 billion yearsago.

From the encounter, we learned that this object isa bilobed object composed of two ellipsoidal com-ponents. The two lobes are distinct and show nosign of significant disruption at the margin betweenthem indicating that they formed from a low-velocitymerger of the two components. The larger compo-nent is lenticular in shape with dimension of ∼22× 20 × 7 km while the smaller component is lessflattened (∼14 × 10 × 10 km). The two lobes havetheir long axis aligned which would be consistentwith the two ellipsoidal parent bodies being tidallylocked before their merger. From the color and com-positional data collected on the spacecraft, we findno significant differences in average color and noobvious compositional differences between the twolobes. The low-velocity merger and homogeneity ofthe lobes are consistent with objects forming in peb-ble cloud gravitational collapse models.

In order for the two objects to merge, the systemwould have needed to lose angular momentum. Pre-ferred mechanisms for this are either (1) interactingwith other small bodies in the vicinity and/or (2) gasdrag from the protoplanetary disk.

Acknowledgements: We thank the New Horizonsmission for supporting this work.

References [1] Stern, S.A., et al. Initial results fromthe exploration of 2014 MU69, a small Kuiper BeltObject, Science, V364, 2019.

300.02 — Why super earths are of a similar size —observational signatures of a limiting pebble accre-tion mass scale

Mickey Morley Rosenthal1; Ruth Murray-Clay11 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)

We propose a novel mass scale at which growth bypebble accretion ceases, which has far reaching ram-ifications for both modeling and observation of plan-etary systems. By noting that, at scales of order thenascent planet’s atmosphere, the flow of nebular gasis altered by the presence of the planet, we demon-strate that particles that require impact parametersfor accretion smaller than this atmospheric scale areinhibited from accreting. Not only does this processdetermine the smallest particle sizes that be capturedvia pebble accretion, but at terrestrial and super-Earth masses this “smallest” particle size is oftenlarger than the maximal sizes present in the disk, nat-urally shutting off pebble accretion. This mass scale,which we term the “flow isolation mass,” is particu-larly salient due to the rapid growth timescales viapebble accretion for planets of terrestrial and super-Earth masses. Taken at face value, these rapid rateswould predict that few planets end their growth inthis mass regime; instead planets either “stall” beforethis scale, or continue their growth and become gasgiants. Flow isolation resolves this problem, allow-ing close-in planets to naturally finish their growth atsuper-Earth mass scales. We demonstrate that flowisolation can explain several observed features of thetransiting planet population. In particular, recentwork indicating that super-Earths in the same sys-tem are correlated in size, and that the super-Earthpopulation can be produced by a single characteris-tic core mass, both point to the existence of a limitingcore mass scale. For reasonable fiducial disk param-eters, the magnitude of the flow isolation mass andits dependence on stellar mass and semi-major axisagree well with these observations.

300.03 — The Boundary between Gas-rich and Gas-poor Planets

Eve Lee1,21 Department of Physics, McGill University (Montreal, Quebec,

Canada)2 California Institute of Technology (Pasadena, California, United

States)

Sub-Saturns straddle the boundary between gas-richJupiters and gas-poor super-Earths/sub-Neptunes.

21

Their large radii (4–8Rearth) suggest that their gas-to-core mass ratios range ∼0.1–1.0. With their en-velopes as massive as their cores, sub-Saturns arejust on the verge of runaway gas accretion; they areexpected to be significantly less populous than gasgiants. Yet, the observed occurrence rates of sub-Saturns and Jupiters are comparable within ∼100days. We show that in these inner regions of plan-etary systems, the growth of sub-Saturns/Jupiters isultimately limited by local and global hydrodynamicflows — runaway accretion terminates and the for-mation of gas giants is suppressed. Within a finitedisk lifetime ∼10 Myrs, massive cores (>10Mearth)can become either gas-poor or gas-rich dependingon when they assemble but smaller cores (<10Mearth)can only become gas-poor. This wider range of pos-sible outcomes afforded by more massive cores mayexplain why metal-rich stars harbor a more diverseset of planets. We also speculate on the origin of thefast rise in the occurrence rate of gas-rich planets to-wards longer orbital periods.

300.04 — Forming rocky super-Earths with realisticcollisions

Jennifer Scora1; Diana Valencia1; AlessandroMorbidelli3; Seth Jacobson2

1 University of Toronto (Toronto, Ontario, Canada)2 Northwestern University (Evanston, Illinois, United States)3 Université Côte d’Azur (Nice, France)

Recent data on rocky super-Earths shows that theyhave a wider distribution of Fe/Mg ratios, or core tomantle ratios, than the planets in our Solar System.We show that this range is too large to be explainedby the Fe/Mg ratios of the stars that host them. In-stead, we demonstrate that realistic collisions, whichalter the composition of the colliding bodies by pref-erentially stripping debris from the mantle, can ex-plain much of this spread. Planet formation simu-lations have only recently begun to treat collisionsmore realistically in an attempt to replicate the plan-ets in our Solar System. We investigate planet for-mation more generally by simulating the formationof rocky super-Earths with varying initial conditionsusing a gravitational N-body code that incorporatesrealistic collisions. We track the maximum plausi-ble change in composition after each impact. Thefinal planets span a wider range of Fe/Mg ratiosthan the Solar System planets, but do not completelymatch the distribution in super-Earth data. The mostiron-rich planets are similar in Fe/Mg ratio to Mer-cury, and the most iron-depleted planets are foundat lower masses. This indicates that further work onour understanding of planet formation is required to

explain planets at the extremes of this Fe/Mg distri-bution.

300.05 — Atmospheric mass loss due to giant im-pacts: the importance of the thermal component forhydrogen-helium envelopes

John Biersteker1; Hilke E. Schlichting2,11 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)2 Earth, Planetary, and Space Sciences, University of California,

Los Angeles (Los Angeles, California, United States)

Systems of super-Earths and mini-Neptunes displaystriking variety in planetary bulk density and com-position. Giant impacts are expected to play a rolein the formation of many of these worlds. Previ-ous works, focused on the mechanical shock causedby a giant impact, showed that these impacts caneject large fractions of the planetary envelope, offer-ing a partial explanation for the observed compo-sitional diversity. I will describe the thermal con-sequences of giant impacts, and show that the at-mospheric loss caused by these effects can signifi-cantly exceed that caused by mechanical shocks forhydrogen-helium (H/He) envelopes. During a gi-ant impact, part of the impact energy is convertedinto thermal energy, heating the rocky core and en-velope. We find that the ensuing thermal expansionof the envelope can lead to a period of sustained,rapid mass loss through a Parker wind, partly orcompletely eroding the H/He envelope. The degreeof atmospheric loss depends on the planet’s orbitaldistance from its host star and its initial thermal state,and hence age. Close-in planets and younger plan-ets are more susceptible to impact-triggered atmo-spheric loss. For planets where the heat capacity ofthe core is much greater than the envelope’s heat ca-pacity (envelope mass fractions ≤4 percent), the im-pactor mass required for significant atmospheric re-moval is M𝑖𝑚𝑝/ Mp ∼ μ/μc ∼ 0.1, approximatelythe ratio of the heat capacities of the envelope andcore. Conversely, when the envelope dominates theplanet’s heat capacity, complete loss occurs when theimpactor mass is comparable to the envelope mass.Because of their stochastic nature, giant impacts mayprovide a natural explanation for the observed rangeof super-Earth and mini-Neptune densities.

300.06 — The diversity of Super-Earths and a newSuper-Earth-class formed from high-temperaturecondensates

Caroline Dorn1; John H. D. Harrison2; Amy Bonsor2;Thomas Hands1

22

1 University of Zurich (Zürich, Switzerland)2 University of Cambridge (Cambridge, United Kingdom)

Super-Earths do not follow a simple mass–radiustrend, but rather reveal a diversity of mass–radius re-lationships that are usually associated with compo-sitional and structural differences. We hypothesisethat differences in the temperatures at which rockymaterial condensed out of the nebula gas can lead todifferences in the composition of key rocky species(e.g., Fe, Mg, Si, Ca, Al, Na) and thus planet bulkdensity. Such differences in the observed bulk den-sity of planets may occur as a function of radial lo-cation and time of planet formation. In this talk weshow that the predicted differences are on the cuspof being detectable with current instrumentation. Infact, for HD 219134, the 10 % lower bulk density ofplanet b compared to planet c could be explained byenhancements in Ca, Al rich minerals. However, wealso show that the 11 % uncertainties on the individ-ual bulk densities are not sufficiently accurate to ex-clude the absence of a density difference as well asdifferences in volatile layers. Besides HD 219134 b,we demonstrate that 55 Cnc e and WASP-47 e are sim-ilar candidates of a new Super-Earth class that haveno core and are rich in Ca and Al minerals whichare among the first solids that condense from a cool-ing proto-planetary disc. Planets of this class havedensities 10-20% lower than Earth-like compositionsand may have very different interior dynamics, out-gassing histories and magnetic fields compared tothe majority of Super-Earths.

302 — Planet Detection — TransitsPoster Session302.01 — One Hit Wonders: recovering the longestperiod TESS planets

Carl Ziegler1; Suresh Sivanandam1; Emily Deibert11 Dunlap Institute, University of Toronto (Toronto, Ontario,

Canada)

TESS is searching for transiting planets aroundbright, nearby stars across nearly the entire sky. Witha 27-day observing cadence, TESS will only be sen-sitive to close-in planets with periods of less than14 days over the majority of the sky. We estimateTESS will observe a single-transit for approximatelya thousand planets during its initial two-year mis-sion but, with largely unconstrained periods, theseplanets will be difficult to follow-up. The One HitWonders surveys will use an autonomous half-metertelescope to observe subsequent transits of the TESS

single-transit planets in the North. Each night, up toa dozen targets with expected transits will be moni-tored at long-cadence; real-time data reduction willdetect potential transits and trigger short-cadence,continuous observations. Simulations suggest ap-proximately twenty planets may be recovered perobserving year. Once confirmed, these planets willpopulate a previously sparse parameter space: long-period planets transiting bright host stars. This willmake these planets excellent candidates for atmo-spheric characterization of cool planets, with refrac-tory elements condensed out of the upper atmo-sphere, as well as serve as a bridge between hotJupiters and the solar-system gas giants,

302.02 — TESS Science Processing Operations Cen-ter Pipeline and Data Products

Jon Michael Jenkins11 NASA Ames Research Center (San Jose, California, United States)

TESS launched 18 April 2018 to conduct a two-year,near all-sky survey for at least 50 small, nearby ex-oplanets for which masses can be ascertained andwhose atmospheres can be characterized by ground-and space-based follow-on observations. TESS justcompleted its survey of the southern hemisphere,identifying >600 candidate exoplanets and unveilinga plethora of exciting non-exoplanet astrophysics re-sults, such as asteroseismology, asteroids, and su-pernova. The TESS Science Processing OperationsCenter (SPOC) processes the data downlinked ev-ery two weeks to generate a range of data prod-ucts hosted at the Mikulski Archive for Space Tele-scopes (MAST). For each sector (∼1 month) of obser-vations, the SPOC calibrates the image data for both30-min Full Frame Images (FFIs) and up to 20,000pre-selected 2-min target star postage stamps. Dataproducts for the 2-min targets include simple aper-ture photometry and systematic error-corrected fluxtime series. The SPOC also conducts searches fortransiting exoplanets in the 2-min data for each sec-tor and generates Data Validation time series and as-sociated reports for each transit-like feature identi-fied in the search. Multi-sector searches for exoplan-ets are conducted periodically to discover longer pe-riod planets, including those in the James Webb Con-tinuous Viewing Zone (CVZ), which are observedfor up to one year. Data products also include co-trending basis vectors (CBVs) and calibration files,such as the Pixel Response Functions across the fieldof view of each of TESS’s four cameras. To maxi-mize the usability, the TESS science data productsare modeled after those for Kepler, including TargetPixel Files and Light Curve files.

23

In this talk, I describe the SPOC pipeline andthe chief differences between the TESS and the Ke-pler pipelines, and the major updates to the SPOCpipeline (4.0) available now to the community atMAST. I also discuss the documentation available tothe community to help them in properly interpretingand analyzing the TESS data products.

The TESS Mission is funded by NASA’s ScienceMission Directorate as an Astrophysics ExplorerMission.

302.03 — Confirming transiting exoplanets withTraCS

Christian Obermeier11 University Observatory Munich (USM) (München, Germany)

The number of exoplanet candidate detections bythe transit method is increasing every year. Follow-ing up each target traditionally by spectroscopy isboth expensive and, in some cases, infeasible dueto the faint host star or shallow transit depth. Oneway to alternatively confirm the planetary nature ofthose targets is multiband photometry, where theplanet’s transit color signature allows to disentanglestellar limb darkening from other parameters. Wehave built on this and expanded it with the three-channel multiband camera 3KK, mounted on the2.1m Wendelstein telescope located in the BavarianAlps. Thanks to its near-infrared capabilities, it en-ables us to observe the secondary eclipse as well toput constraints (or measure) the planet candidate’seffective temperature. I will then showcase this tech-nique by two recently confirmed exoplanets that or-bit faint (V<15mag) stars.

302.04 — Observing transits of Hot Jupiters fromBaker Observatory

Michael Reed11 Physics, Astronomy, & Materials Science, Missouri State Univer-

sity (Springfield, Missouri, United States)

Our project involves observing transits from our lo-cal observatory. Baker Observatory has three mod-est telescopes with virtually unlimited access for ourgroup with approximately 200 clear nights per year.In this poster, we will show some of our preliminarytransit observations.

We are observing Hot Jupiters and seeking to mea-sure transit timing variations (TTVs) and transit du-ration variations (TDVs). From the measurements,our goals are to improve orbital periods, find non-transiting objects, determine eccentricities and incli-nations.

While Hot Jupiters represent a small portion of allexoplanets, how they arrived at their current orbitscan be used to determine if they likely migrated orwere scattered by other objects. Additionally, thereis the issue of other, likely smaller, planets in longerorbits, which have yet to be discovered. These issuesaddress solar system formation scenarios and so areimportant for the field.

Our observatory contains a PlaneWave CDK20, aMeade LX200GPS, and a Celectron C14 telescopes,all equipped with Apogee Alta CCDs. We couldpotentially observe transits for nine different star-planet combinations during any one night.

We are members of ExoFOP and ExoFOP-TESSand are actively seeking collaborations for focusedobservations of targets of interest.

302.05 — Exoplanet Characterization using PhaseVariations Observed by TESS

Tara Fetherolf1; Stephen Kane3; Chelsea X. Huang2; AviShporer2; Ian Wong4

1 Physics and Astronomy Department, University of CaliforniaRiverside (Riverside, California, United States)

2 Kavli Institute for Astrophysics and Space Research, Mas-sachusetts Institute of Technology (Cambridge, Massachusetts, UnitedStates)

3 Earth and Planetary Sciences Department, University of Califor-nia Riverside (Riverside, California, United States)

4 Earth, Atmospheric, and Planetary Sciences Department, Mas-sachusetts Institute of Technology (Cambridge, Massachusetts, UnitedStates)

The field of exoplanets has progressed beyond planetdetection to characterization of planetary systems.This includes planetary structure and atmospheres,studies of exoplanet host star properties, and mea-surement of planet masses. While the precision pho-tometry from the Transiting Exoplanet Survey Satel-lite (TESS) is primarily directed towards the detec-tion of planetary transits, it can also be utilized fora variety of additional science applications. In par-ticular, a phase modulation present in the photom-etry can be used to study the planetary mass andalbedo, potentially providing insights into the atmo-spheric properties of the planet. We have developedan analysis program designed to detect phase varia-tions caused by the beaming, ellipsoidal, and reflec-tion (BEER) effects. In particular, we focus on thephase variations of TESS Objects of Interest (TOIs)that were observed during Cycle 1, which currentlyincludes ∼100 known planetary systems and ∼500new planet candidates. Our analysis provides esti-mates for the companion mass and albedo, and po-tentially identifies false positives in the list of TESS

24

planetary candidates. In addition to the massesbeing useful for characterizing known and candi-date planets, we also use the companion mass mea-surements to search for contaminating stellar bina-ries in the planet candidate catalog based on theircharacteristic light curve shapes and evidence for asecondary eclipse. Therefore, our phase variationanalyses complement the planet candidate catalogby identifying high-priority TOIs for spectroscopicplanet confirmation. From the albedo measure-ments, we infer atmospheric characteristics and de-termine planetary habitability more accurately thanusing the distance from the host star alone. Fi-nally, we emphasize that BEER modulations encap-sulate a specific type of phase variation. Our analy-sis additionally identifies phase variations that can-not be naturally explained by the BEER effects andare not necessarily well-understood. By identifyingseveral systems with similar unusual phase variationshapes, we begin to explore possible astrophysicalmechanisms behind these characteristic signals.

302.06 — Unbiased Determination of PlanetaryMasses and Orbital Precession from Kepler Tran-sit Variations

Oded Aharonson1,2; Aviv Ofir1; Gideon Yoffe1; YairJudkovsky1

1 Weizmann Institute of Science (Rehovot, Israel)2 Planetary Science Institute (Tucson, Arizona, United States)

The exquisite accuracy of exoplanetary transits mea-sured by Kepler allows detection of deviations fromperiodicity in the light-curves. Here we develop andapply a global fitting approach to the light-curvedata that allows constraining a smaller number ofrelevant degrees of freedoms describing the orbitalvariations. The approach enables uniform and pre-cise determinations of the masses of the perturbingplanets. This avoids the necessity of having individ-ual transits precisely determined and thus increasesthe sensitivity to lower TTV amplitude, shorter or-bital period, and shallower transit depth, removingpast biases in these variables. We fit for the eccen-tricity difference between interacting planets, reduc-ing the parameter space by two dimensions, andalleviating the intrinsic mass-eccentricity degener-acy. Taken together, these allow us to successfullydetermine exoplanetary masses for more- and eversmaller- planets, and in particular for planets in the1 Earth radius regime. Additionally, we developed atechnique for detecting secular variations in the orbitby measuring and fitting transit variations beyondthose seen in the mid-transit time. Detecting such

dynamical scenarios provides information regard-ing the possible existence of non-transiting planetarycompanions, or the non-spherical mass distributionof the host star. The variations may imply forces outof the orbital plane, and thus probe mutual incli-nations among components of the system. Becausesecular precession occurs about the system’s invari-able plane, we translate these orbital elements to sky-plane variables, and derive an analytic descriptionof the transit parameters. By applying a non-linearsearch for the best-fit parameters, we constrain thedynamical scenarios that may be responsible for thelight-curve variations, and show results for syntheticand Kepler data.

302.07 — Detecting astrophysical events in TESSdata using deep learning

Emil Knudstrup1; Simon Albrecht11 Stellar Astrophysics Centre, Aarhus University (Aarhus C, Den-

mark)

One of TESS’ primary objectives is to discover andcharacterize 50 Earth-like planets and it is the hopethat these planets will be found among the 200,000preselected targets observed with a cadence of 2 min-utes. The standard routines for automatic planet de-tection (e.g., the box least squares (BLS; Kovács etal. (2002)) method) are excellent at picking out plan-ets (both Earth-sized and bigger) transiting multipletimes and orbiting stars with modest variability, theyare not as well suited for detecting the oddballs likeevaporating planets, single transiting planets (mono-transits), etc. An oddly shaped transit or any otheratypical astrophysical phenomenon is traditionally(and often easily) identified by a trained human eye.These events are naturally very interesting, however,eyeballing all stars is a daunting task, especially asthe TESS data is not only comprised of the 200,000stars observed in 2 minute cadence, but also a set offull-frame images (FFIs) covering the entire field ofview (FOV) for a given sector, yielding light curvesfor millions of stars observed with a cadence of 30minutes. This is why an automatic way of charac-terizing these systems is desirable. We are there-fore training a convolutional neural network (CNN). Itidentifies the oddballs and will support traditionaldetection algorithms (such as BLS). It will label alight curve as one of the following: planetary can-didate, eclipsing binary, variable star, oddball (i.e.,a weirdly shaped or monotransit), or seemingly fea-tureless. Here we present our results for the availableTESS data and highlight some particular systems.

25

302.08 — New Discoveries and high precision pho-tometric follow-up from NGTS

Oliver Turner11 Geneva Observatory (Versoix, Switzerland)

The Next-Generation Transit Survey (NGTS) is awide-field photometric survey designed to discovertransiting exoplanets of Neptune-size and smalleraround bright stars (magnitude V<13) from theground. These objects will be intensely interestingfor detailed follow-up as they will allow us to in-crease our understanding of the mass-radius relationfor smaller planets and the brightness of their hoststars will make them accessible to present day andupcoming instruments for atmospheric characterisa-tion. Thanks to it’s versatility and high photometricprecision NGTS now boasts several follow-up pro-grams too which synergise with TESS discoveries.

We present a number of new discoveries all ofwhich occupy the current fringes of parameter space.NGTS-4, a sub-Neptune planet orbiting a K-starwith the smallest transit signal discovered from theground, It orbits in the so-called Neptune desert,where it is thought strong stellar irradiation wouldcause significant mass loss. NGTS-6 is an ultra shortperiod (∼21 hours) hot-Jupiter which, models sug-gest, should have lost around 5% of its gaseous atmo-sphere since its formation. By the time of the confer-ence we expect will be a detailed analysis of at least3 additional systems.

In addition, we will present the new follow-up ac-tivities of NGTS, in particular the synergies availableto us via the (soon) completed TESS southern hemi-sphere survey. NGTS is ideally situated to confirm-ing a number of the current mono-transit, and there-fore long period, TESS candidates. We have also re-cently demonstrated the ability to reach exceptionalphotometric precision (of the order 150 ppm/30 min)which will allow us to follow-up even shallow TESSdetections from the ground.

302.09 — Exoplanets and the KESPRINT collabora-tion

Carl Malcolm Fridlund1,2; Carina Persson21 Leiden Observatory, University of Leiden (Leiden, Netherlands)2 Department of Space, Earth and Environment, Chalmers Univer-

sity of Technology (Onsala, Sweden)

KESPRINT is a collaboration of around 50 exoplan-etary researchers, located in some 20 institutes cov-ering a large part of the world. This group concen-trates on carrying out follow-up observations of datafrom space missions dedicated to the observation of

exoplanetary transits. The importance of a groundbased follow-up program (FuP) when interpretingsuch results can not be overstated. The verificationof the existence of the exoplanet through detectionof radial velocity variations or through validation byexcluding all other sources (typically background bi-naries) through e.g. Adaptive Optics imaging, needto be carried out. It is also required to know the fun-damental parameters of the host star with very highprecision in order to discern the planetary data. Re-cently, asteroseismology have also started to play akey role. Here we briefly describe the work of thiscollaboration and highlight an important result —The characterization of one of the oldest planets inour Galaxy.

302.10 — Searching for transiting cold Jupitersaround bright stars with ASTEP South at Dome C,Antarctica

Nicolas Crouzet1; Djamel Mékarnia4; Tristan Guillot4;Daniel Bayliss2,3; Hans Deeg5,6; Enric Palle5,6; LyuAbe4; Abdelkrim Agabi4; Jean-Pierre Rivet4; FelipeMurgas5,6; Michaël Gillon7; Laetitia Delrez8; EmmanuëlJehin7; Néstor Espinoza9

1 European Space Research and Technology Centre, European SpaceAgency (Noordwijk, Netherlands)

2 Department of Physics, University of Warwick (Coventry, UnitedKingdom)

3 Centre for Exoplanets and Habitability, University of Warwick(Coventry, United Kingdom)

4 Laboratoire Lagrange, CNRS (Nice, France)5 Instituto de Astrofísica de Canarias (La Laguna, Tenerife, Spain)6 Departamento de Astrofísica, Universidad de La Laguna (La La-

guna, Tenerife, Spain)7 Space sciences, Technologies and Astrophysics Research (STAR)

Institute, Université de Liège (Liège, Belgium)8 Cavendish Laboratory (Cambridge, United Kingdom)9 Max-Planck-Institut für Astronomie (Heidelberg, Germany)

Much of our understanding of gas giant exoplanetscome from those transiting in front of bright stars atsmall orbital separations (P ∼ 3 days, a ∼ 0.05 au).These hot Jupiters are coupled to their host star: stel-lar irradiation impacts the chemistry and tempera-ture structure of their atmospheres and tidal interac-tions affects the orbital dynamics and may even im-pact the star itself. In contrast, gas giant exoplanetswith long orbital periods and large separations (P >30 days, a > 0.2 au) are much less coupled to theirhost star and provide ideal benchmarks to study gasgiant planets in general. However, only a few transit-ing ”cold Jupiters” orbiting bright stars are known todate. In the past years, we conducted the ASTEP ex-periment (Antarctica Search for Transiting ExoPlan-

26

ets) to search and characterize transiting exoplanetsfrom Dome C, Antarctica and to qualify this site forphotometry in the visible. One instrument, ASTEPSouth, is a 10 cm diameter lens equipped with a CCDcamera in a thermalised box pointing continuouslytowards the celestial South pole. We analysed fourwinters of data collected with this instrument andidentified about 30 transit candidates around rela-tively bright stars (9 < V < 13) with orbital periodsup to 80 days. We performed photometric follow-up with the Las Cumbres Observatory (LCO) 0.4mtelescopes to investigate these signals. Most of thesestars are also observed by TESS and their lightcurvescan be extracted from the full frame images. In thisposter, we present our set of candidates, the first re-sults of the photometric follow-up, and discuss theuse of TESS data to investigate these objects.

302.11 — MASCARA and bRing, finding brighttransiting planets and synergies with TESS

Patrick Dorval1,2; Geert Jan Talens5; Gilles Otten6;Sam Mellon7; Remko Stuik1,2; John I. Bailey8; Si-mon Albrecht3; Don Pollacco9; Enric Palle4; JamesMcCormac9; Rafael Brahm10,11; Andrés Jordán11;Steven Crawford12,13; Michael Ireland14; BlaineB.D. Lomberg13; Rudi Kuhn13; Ignas Snellen1; MattKenworthy1; Eric Mamajek7,15

1 Leiden Observatory, Leiden University (Leiden, Zuid Holland,Netherlands)

2 Center of Astro-Engineering UC, Pontificia Universidad Católicade Chile (Santiago, Chile)

3 Instituto de Astrofísica, Pontificia Universidad Católica de Chile(Santiago, Chile)

4 Space Telescope Science Institute (Baltimore, Maryland, UnitedStates)

5 South African Astronomical Observatory (Cape Town, SouthAfrica)

6 Research School of Astronomy and Astrophysics (Canberra, Aus-tralian Capital Territory, Australia)

7 Jet Propulsion Laboratory, California Institute of Technology(Pasadena, California, United States)

8 NOVA Optical/IR Instrumentation Group, ASTRON(Dwingeloo, Netherlands)

9 Department of Physics and Astronomy, Aarhus University(Aarhus, Denmark)

10 Investigacion, Instituto de Astrofisica de Canarias (La Laguna,Spain)

11 Département de Physique, Institut de Recherche sur les Exo-planètes, Université de Montréal (Montréal, Quebec, Canada)

12 Aix Marseille University (Marseille, France)13 Department of Physics and Astronomy, University of Rochester

(Rochester, New York, United States)14 Department of Physics, University of California at Santa Barbara

(Santa Barbara, California, United States)15 Department of Physics, University of Warwick (Coventry,

United Kingdom)

In this talk, I will discuss the current state of theMASCARA and bRing networks, and go into detailon how our observations can be combined with TESSto search for long period transiting exoplanets. Thesenetworks have discovered four exciting transiting ex-oplanets which are ideal for follow up atmosphericcharacterization surveys. The near continuous ob-servations of these networks provides a unique syn-ergy with TESS, where our long-term monitoringof all stars brighter than 8.4 magnitudes in the skywill provide information needed to find long periodtransiting systems when combined with the shorterterm TESS data. The MASCARA and bRing net-works are composed of a set of observatories in thenorthern and southern hemispheres with the aim offinding Hot Jupiters around bright stars using pho-tometry. They are composed of four stations, twomain MASCARA stations located in La Palma, theCanary Islands and La Silla, Chile, and two smallerbRing stations in Sutherland, South Africa and Sid-ing Springs, Australia. Each MASCARA station isequipped with five interline CCD cameras, and ob-serves the local sky down to an airmass of around 2.The smaller bRing stations are equipped with onlytwo interline CCD cameras, and observe the decli-nation range of -30 to -90 degrees down to an air-mass of around 10. The bRing stations are combinedwith the southern MASCARA station, and allows forat least one station observing at any given time formuch of the year in the bRing declination range, pro-vided weather permits. With five minute exposures,we obtain photometric observations of tens of thou-sands of bright stars over years with a window func-tion which is unprecedented for ground-based ob-servatories. All MASCARA data will soon be freelyavailable through the ESO database.

302.12 — The case for transiting warm giant ex-oplanets: from TESS discoveries to atmosphericcharacterization with JWST

Néstor Espinoza1; Rafael Brahm2,6; Andrés Jordán2,6;Thomas Henning1; Jonathan Fortney3; DanielThorngren4; Benjamin Rackham5; Diana Kossakowski1;Paula Sarkis1; Felipe Rojas2,6; Trifon Trifonov1; MatiasJones7

1 Max-Planck-Institut für Astronomie (Heidelberg, Germany)2 Instituto de Astrofísica, Facultad de Física, Pontificia Universidad

Católica de Chile (Santiago, Chile)3 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)

27

4 Physics, University of California, Santa Cruz (Santa Cruz, Cali-fornia, United States)

5 Department of Astronomy, University of Arizona (Tucson, Ari-zona, United States)

6 Millenium Institute of Astronomy (Santiago, Chile)7 European Southern Observatory (Santiago, Chile)

Transiting warm giant exoplanets (planets with equi-librium temperatures below 1000 K, or periodslonger than about 10 days) are fundamental objectsto study, as they are ideal laboratories for tests onplanet formation and evolution, which might notonly provide distinct signatures to their much bettercharacterized hotter counterparts, but also constrainour understanding of giant (exo)planet interiors andtheir accretion history through, e.g., mass-metallicityrelations. Their detection and characterization, how-ever, is very challenging due to their longer peri-ods (which makes them hard to detect from ground-based transit surveys) and cooler nature (which im-plies smaller scale-heights and, thus, smaller signalsin transmission). In fact, only about 40 of these sys-tems have been discovered to date, with only a hand-ful of them being optimal for atmospheric character-ization. In this talk, I will present the state-of-the-arton our understanding of warm giant exoplanets bothfrom a population and from an interior modellingperspective, emphasizing how the study of these ex-oplanets can help us understand previously uncon-strained properties and predictions of gas giant ex-oplanets. With this motivation at hand, I will thenpresent the ongoing work of the Chile-MPIA collab-oration, a multi-institutional effort focused on thesystematic search of these long-period systems us-ing data from the Transiting Exoplanet Survey Satel-lite (TESS) and several ground-based facilities. Em-phasis will be given both in the methodology of thesearch and on the most exciting and recent resultsof the collaboration, including a handful of transit-ing systems with the longest periods discovered bythe mission yet. Finally, I will focus on the uniquecapabilities the upcoming James Webb Space Tele-scope (JWST) has for characterizing the atmospheresof these exciting systems, and how together with cur-rent ground-based facilities, this promising observa-tory will allow us to get a panchromatic view of thesedistant worlds, which will in turn enable the firsttests of key predictions from structure modelling andplanet formation.

302.13 — A Spitzer search for transiting exoplanetsaround ultra-cool dwarf stars viewed equator-on

Stanimir Metchev1,2; Paulo Miles-Páez1,3; Enric Palle4;Maria Rosa Zapatero Osorio5; Megan Tannock1; Dániel

Apai3; Étienne Artigau6; Adam Burgasser7; GregoryMace8; Amaury Triaud9

1 Physics & Astronomy, University of Western Ontario (London,Ontario, Canada)

2 American Museum of Natural History (New York, New York,United States)

3 Astronomy, University of Arizona (Tucson, Arizona, UnitedStates)

4 Investigacion, Instituto de Astrofisica de Canarias (La Laguna,Spain)

5 Centro de Astrobiología (Madrid, Spain)6 Physics, Université de Montréal (Montreal, Quebec, Canada)7 Physics, University of California San Diego (San Diego, Califor-

nia, United States)8 Astronomy, University of Texas (Austin, Texas, United States)9 School of Physics & Astronomy, University of Birmingham

(Birmingham, United Kingdom)

Exoplanet population studies indicate that smallrocky planets may be common around very low-mass stars and brown dwarfs: ”ultra-cool” dwarfs.Temperate rocky planets around ultra-cool dwarfstars could be the best targets for detecting the atmo-spheric signatures of extrasolar life with the JamesWebb Space Telescope: because of a favourable star-to-planet contrast ratio and very short orbital peri-ods in the habitable zone that allow frequent observ-ing opportunities. The seven-rocky-planet planetsystem around the ultra-cool dwarf TRAPPIST-1 isthe best — and so far only — such known exam-ple. With James Webb launching in two years andwith a nominal mission lifetime of only five years,we urgently need to discover more temperate plan-ets around nearby ultra-cool dwarfs.

We are conducting a large Spitzer Space Telescopeprogram to search for exoplanets around 15 ultra-cool dwarfs. Our sample is optimized for transitplanet discoveries by including only stars inferredto rotate nearly equator-on. The observed near-ubiquitous spin-orbit alignment between the hoststars and the planets in multi-planet systems dic-tate that any planetary systems around our selectedultra-cool dwarfs should also be oriented nearlyedge-on. We will describe our survey and willpresent preliminary findings.

302.14 — The Occurrence Rate of Planets aroundK2’s Ultracool Dwarfs

Marko Sestovic1,2; Brice-Olivier Demory11 Centre for Space and Habitability, University of Bern (Bolligen,

Switzerland)2 Saint-Ex Group, University of Bern (Bern, Switzerland)

Over its 19 campaigns, the K2 mission has observed

28

more than 450 spectroscopically confirmed ultracooldwarfs. Using this data, we determine the occur-rence rates of planets around such stars. To that end,we build an automatic detrending and transit-searchpipeline, and we determine its completeness with in-jection recovery modelling.

Our results place upper bounds on the occur-rence rates of planets larger than ∼2 earth radii (sub-Neptunes and giant planets), suggesting they are rel-atively rare (less than ∼0.1 per star). For super-earthsand planets such as those in the TRAPPIST-1 system,our constraints are limited by the photometric preci-sion of K2. However, we do not exclude the possibil-ity that rocky planets are common.

Additionally, our Gaussian process-based de-trending pipeline automatically constrains periodic-ity in the lightcurves. We use this to study the stel-lar variability and rotation patterns of the sample ofultracool dwarfs. Of particular interest is the possi-ble correlation between flare timing and variabilityphase, found in previous works on TRAPPIST-1. Thequestion still remains whether this is a feature sharedby other ultracool stars.

302.15 — Classifying Exoplanet Candidates withConvolutional Neural Networks: Application tothe Next Generation Transit Survey

Alexander Chaushev1; Liam Raynard21 Technical University of Berlin (Berlin, Germany)2 University of Leicester (Leicester, United Kingdom)

A key bottleneck in the discovery of transiting ex-oplanets is the large number of false positives pro-duced by existing detection algorithms. Currentlythe solution to this problem is to vet the candidatesby hand, however this is time consuming and canbe inconsistent. Recently convolutional neural net-works (CNNs), a type of ‘Deep Learning’ algorithm,have been shown to be effective at this task [Shal-lue+19]. Not only can CNNs reduce the man hoursrequired to vet candidates, but they have helped toidentify planets which otherwise had been missed[Shallue+19, Datillo+19]. As the number of plane-tary candidates pile up from missions like TESS, re-ducing the false positive rate (FPR) is crucial for max-imising the yield of small planets and making thebest use of follow-up time. Additionally, loweringthe FPR, is also important for improving measure-ments of occurrence rates to better constrain the un-derlying distribution of exoplanets.

Here I will present results from the on-going ef-fort to automate the Next Generation Transit Sur-vey (NGTS) candidate vetting process using a CNN.

Currently we are able to exclude ∼50% of false pos-itives, while recovering ∼90% of our manually iden-tified candidates and all currently known planets inthe NGTS dataset [Chaushev+19, in prep]. On-goingwork is focused on reducing the number of false pos-itives further, by searching for novel CNN architec-tures and by adding additional information to thenetwork. A key goal of the project is to understandand improve the network performance in the low-est signal to noise regimes. In this regard, NGTSprovides a unique dataset as it has been continuallypushing to find planets on the edge of detectabil-ity, leading to the discovery of NGTS-4b, the shal-lowest transit discovered from the ground to date[West+19]. This makes NGTS an ideal testing groundfor CNNs and improvements made in the techniqueshere can readily be applied to space based data fromK2 and TESS currently, and PLATO in the future.

302.16 — KELT is up to the TESS: The OngoingSearch for Transiting Planets in Longer Periods andaround Hotter Stars

Michael B. Lund11 Caltech/IPAC-NExScI (Pasadena, California, United States)

The Kilodegree Extremely Little Telescope (KELT)transit survey was designed to find planets transit-ing stars from roughly 7th to 10th magnitude, fillingin the gap between brighter RV surveys and previ-ous transit surveys targeting fainter stars. To date,KELT has discovered 25 planets, many of which arebright enough to for detailed follow-up and charac-terization. In particular, KELT has found many ofthe transiting planets now known to be orbiting hot-ter (B/A) main-sequence stars, including KELT-9b,the hottest known exoplanet and KELT-5b, the firstplanet transiting a hot, pulsating star in a progradeorbit. We have also found the brightest host star witha massive transiting Jupiter, KELT-24b.

KELT overlaps with a large fraction of the TESS skyfootprint and magnitude rangewith time baselines ofup to 15 years. While the TESS cadence and sensi-tivity means that TESS is very complete in detectingtransits as they occur, the 27-day duration of TESSobservations in most of the sky will not provide theephemerides for planets with periods greater thanabout 13.5 days. The combination of short durationand high precision TESS observations and the KELTbaseline means that existing KELT planet candidatescan be found in the TESS data, and that TESS planetcandidates can be identified in KELT data enablingprecovery of single transit TOIs.

29

302.17 — Studying the Diversity of Exoplanets withLUVOIR

Eric D. Lopez11 NASA/GSFC (Washington, District of Columbia, United States)

LUVOIR is powerful and flexible observatory de-signed to revolutionize our view of the universe.In addition to searching for signs of life on habit-able worlds, LUVOIR will be capable of detectingand characterizing hundreds of non-habitable exo-planets orbiting nearby stars dramatically advancingthe field of “comparative exoplanetology”. Operat-ing at L2, with a large aperture of 8-15 m and a so-phisticated instrument suite, LUVOIR will allow forfantastic characterization of planets across parame-ter space both with direct imaging and transmissionspectroscopy. At FUV wavelengths the LUVOIR Ul-travoilet Multi Object Spectrograph (LUMOS) willobtain high signal to noisetransmission spectra athigh spectral resolution, allowing us to detect transit-ing planetary exospheres and constrain the physicsof atmospheric escape. Meanwhile, High DefinitionImager (HDI) instrument we can obtain high signalto noise, medium resolution spectra from the NUVto the NIR, allowing us to constrain the properties ofclouds, map absorption from alkali metals, and mea-sure abundances for a wide range of molecules in-cluding H2O, CO2, and O2. Finally, direct imagingand spectroscopy with the ECLIPS coronograph willenable a systematic investigation of system architec-tures and the diversity of exoplanet atmospheres atwide orbits.

302.18 — Exoplanet Science Using University ofWyoming Observatories

Hannah Jang-Condell1; Cristilyn Gardner1; DavidKasper1; Henry Kobulnicky1; Michael Pierce1; Cather-ine Pilachowski2

1 Physics & Astronomy, University of Wyoming (Laramie,Wyoming, United States)

2 Indiana University (Bloomington, Indiana, United States)

The University of Wyoming is home to the 2.3-mWyoming Infrared Observatory (WIRO) and the 0.6-m Red Buttes Observatory (RBO). These facilities en-able research in exoplanet detection and characteri-zation. Transit observations at RBO have led to thediscoveries of KELT-9b and KELT-21b. Multi-bandphotometry of HD 189733b at WIRO has helpedto characterize its atmosphere. New instrumen-tation under construction at WIRO include instal-lation of diffusers for better PSF control and anechelle spectrograph (FHiRE: Fiber-fed High Resolu-tion Echelle). FHiRE is poised to become a precision

radial velocity measurement instrument for long-term RV monitoring of candidate exoplanet hoststars. As TESS identifies new planet candidates, ourfacilities will make significant contributions towardexoplanet discovery and characterization.

302.19 — Searching for Long-Period Planets inTESS

Steven Villanueva1; Diana Dragomir2; Chelsea X.Huang1; Paul A. Dalba3; Dax Feliz4; Scott Gaudi5;Arvind Gupta9; Stephen Kane3; Belinda Nicholson8;Josh Pepper6; Joseph E. Rodriguez7; Daniel JosephStevens9; Xinyu Yao6

1 MIT (Malden, Massachusetts, United States)2 MIT/UNM (Cambridge, Massachusetts, United States)3 Department of Earth and Planetary Science, University of Califor-

nia Riverside (Riverside, California, United States)4 Vanderbilt University (Nashville, Tennessee, United States)5 The Ohio State University (Columbus, Ohio, United States)6 Lehigh University (Bethlehem, Pennsylvania, United States)7 Center for Astrophysics | Harvard & Smithsonian (Cambridge,

Massachusetts, United States)8 University of Southern Queensland (Darling Heights, Queens-

land, Australia)9 Astronomy & Astrophysics, The Pennsylvania State University

(University Park, Pennsylvania, United States)

I will discuss the progress in confirming and char-acterizing TESS long-period (P > 20 days) planets,specifically those identified as single-transit events(STE). I will discuss the efforts to identify and vetSTEs, including an overview of our identificationpipeline. STEs require more careful planning andlengthier follow-up campaigns for confirmation thanmultiply transiting or short-period planets, but canyield some of the longest-period planets that TESScan find. I will discuss some of the strategies andchallenges specific to confirming STEs, and I willgive an overview of the known STEs and give an up-date on their status.

302.20 — NGTS-6b: An Ultra Short Period Hot-Jupiter Orbiting an Old K Dwarf

Jose Ignacio Vines11 Departamento de Astronomia, Universidad de Chile (Santiago,

Región Metropolitana, Chile)

We report the discovery of a new ultra-short periodhot Jupiter from the Next Generation Transit Sur-vey. NGTS-6b orbits its star with a period of 21.17h, and has a mass and radius of 1.330+0.024

−0.028MJ, and 1.271+0.197

−0.188 RJ respectively, returninga planetary bulk density of 0.805+0.498

−0.283 g cm−3.

30

Conforming to the currently known small popula-tion of ultra-short period hot Jupiters, the planet ap-pears to orbit a metal-rich star ([Fe/H] =+0.11 ± 0.09dex). Photoevaporation models suggest the planetshould have lost 5% of its gaseous atmosphere overthe course of the 9.6 Gyrs of evolution of the system.NGTS-6b adds to the small, but growing list of ultra-short period gas giant planets, and will help us tounderstand the dominant formation and evolution-ary mechanisms that govern this population.

302.21 — A transiting circumbinary planet fromTESS

Veselin Kostov11 NASA/SETI (Greenbelt, Maryland, United States)

We report the TESS detection of a transiting cir-cumbinary planet. At the time of writing, the targetwas observed in 11 sectors of long-cadence data andin 8 sectors of short-cadence data. The host eclipsingbinary exhibits prominent primary and secondaryeclipses, the planet produced two transits of differentdurations, and is expected to produce another transitin Sector 13. We combined the precision photometryfrom TESS with ground-based observations in a nu-merical photometric-dynamical model to reproducethe observed planet transits and stellar eclipses. Thesystem demonstrates the discovery potential of TESSfor circumbinary planets, and provides further un-derstanding of the formation and evolution of plan-ets orbiting binary stars.

303 — Planet Detection — RadialVelocities, Poster Session303.01 — Search for Earth analogues in the habit-able zone around solar type stars: radial velocity orastrometry?

Nadege Meunier11 Univ. Grenoble Alpes (Grenoble — cedex, France)

Stellar activity is currently a major limitation to thedetection of very low mass planets around solar typestars using radial velocity techniques. Various tech-niques have been implemented to mitigate this prob-lem, without allowing to reach one Mearth planetsfor stars similar to the Sun yet. It is therefore cru-cial to estimate precisely the effect of activity on ex-oplanet detectability using realistic time series forvarious types of stars to overcome this problem. Iwill describe the basic processes at work and how

we extended a realistic solar model to build repre-sentative time series of radial velocity, photometry,astrometry and chromospheric emission variability.We built coherent sets of stellar parameters coveringa large range in effective temperature (K4-F6) and av-erage activity levels. Such simulations are extremelyuseful to better understand the relationship betweenRV, astrometry and activity indicators and the lim-itations of correction techniques. I will present theimpact of activity on the detectability of Earth massplanet in the habitable zones of those stars using ra-dial velocity and high precision astrometry and dis-cuss their respective performance.

303.02 — Stellar activity and wavelength-dependent radial velocity measurements

Maksym Lisogorskyi1; Hugh R.A. Jones1; Fabo Feng2; R.Paul Butler2

1 Centre for Astrophysics Research, University of Hertfordshire(Hatfield, Hertfordshire, United Kingdom)

2 Department of Terrestrial Magnetism, Carnegie Institution ofWashington (Washington, Washington, United States)

The Alpha Centauri system is the primary target forplanet search as it is the closest star system. Here welook at contaminating signals from telluric lines andactivity sensitive lines across the spectrum in HARPSobservations of Alpha Centauri B. We compile andquantify the behaviour of 345 spectral lines with awide range of line shapes and sensitivity to activityand investigate its effect on radial velocity using ob-servations from UVES. Removing parts of the spec-trum that contain activity-sensitive lines removes aradial velocity trend reaching 8 m/s. The differen-tial velocity can be used as an indicator for contam-inating signals and can benefit greatly from carefulselection of ”active” and ”inactive” echelle orders.

303.03 — An Expanded Catalog of Long-Period Ex-oplanets, Discovered with HIRES, Lick-Hamilton,and APF

Lee Jesse Rosenthal1; Benjamin J. Fulton2; Lea A.Hirsch3; Andrew Howard1

1 Astronomy, California Institute of Technology (Pasadena, Califor-nia, United States)

2 NASA Exoplanet Science Institute (Pasadena, California, UnitedStates)

3 Kavli Institute for Particle Astrophysics and Cosmology, StanfordUniversity (Stanford, California, United States)

The California Planet Search team has been conduct-ing a radial velocity survey of almost 800 nearbyF/G/K stars for the past three decades, using the

31

HIRES instrument on Keck I, and the Hamilton spec-trograph and Automated Planet Finder at the Lickobservatory. We describe a systematic search ofthis dataset for previously undetected periodic sig-nals and long-term trends, and present a catalog ofdozens of newly-detected planet candidates, rangingfrom warm sub-Neptunes to cold gas giants.

303.04 — The Beginning of the Strategic Large Ex-ploration for Exoplanets Orbiting Nearby Late-MDwarfs with the InfraRed Doppler (IRD) Spectro-graph on the Subaru Telescope

Masayuki Kuzuhara9,1; Bun’ei Sato2; MotohideTamura6,9; Takayuki Kotani9; Nagayoshi Ohashi1,3;Masashi Omiya9,1; Teruyuki Hirano2; HirokiHarakawa3,9; Wako Aoki1,5; Norio Narita9,1; Ya-sunori Hori9,1; Akitoshi Ueda1,5; Akihiko Fukui4; Hi-royuki Tako Ishikawa5,1; MASATO ISHIZUKA6;Takashi Kurokawa9,7; Nobuhiko Kusakabe9,1; TomoyukiKudo3,9; Eiichiro Kokubo1,5; Mihoko Konishi8; TadashiNakajima9,1; Jun Nishikawa1,5; Masahiro Ogihara1;Takuma Serizawa7

1 National Astronomical Observatory of Japan (Mitaka, Japan)2 Tokyo Institute of Technology (Meguro, Japan)3 Subaru Telescope (Hilo, Japan)4 University of Tokyo (Hongo, Japan)5 SOKENDAI (Mitaka, Japan)6 Department of Astronomy, University of Tokyo (Taito-ku, Japan)7 Tokyo University of Agriculture and Technology (Koganei, Japan)8 Oita University (Oita, Japan)9 Astrobiology Center, NINS (Mitaka, Japan)

The observations of the Kepler space telescope sug-gest that small planets are abundant around coolmain-sequence stars, among which late-M dwarfs(LMDs) represent the coolest objects. LMDs are thegreat targets for the exoplanet search with the radialvelocity (RV) technique due to their relatively lowmasses and inner habitable zones. However, LMDsare so faint especially at optical wavelengths that theRV technique for LMDs needs the infrared spectro-graph available on a large-aperture telescope witha stable calibration system of RV measurement. Wehave developed and operated the InfraRed Doppler(IRD) spectrograph that can be utilized with theadaptive optics of the Subaru Telescope. IRD ob-serves a laser frequency comb simultaneously withan object spectrum, enabling the stable RV calibra-tion comparable to 2 m s−1 at ∼1.0–1.7 μm. SinceFebruary of 2019, we have started a strategic cam-paign to explore planets around LMDs using IRD,which is planned to go on until 2024 with the to-tal allocation of 175 nights. This is the first large-scale survey dedicated to LMDs that is achieved by

the precision RV measurements in the infrared. TheIRD survey is expected to discover habitable planetsthat can be characterized in detail with next gener-ation telescopes. Also, the monitor of LMDs overa few years can reveal the population of rocky toice-giant planets inside and outside of snow lines.We have listed 150 targets based on the literatureand our pre-selection spectroscopic observations tofilter out active LMDs unsuitable for precision RVmeasurements. In addition, the rapid rotators andclose-separation multiple stars are removed throughthe first-year IRD observations, selecting the best 60LMDs for extensive RV measurement. In parallelwith the science observations, we are testing the pre-cision and stability of our RV measurements by ob-serving RV-stable stars and planet-host stars such asGJ 699 and GJ 436. We here present the strategy ofthe IRD planet survey and its latest progress, as wellas the results of the performance verification.

303.05 — The Magellan TESS Survey (MTS): Prob-ing the Formation and Evolution of Small Planetswith a Statistically Robust Survey

Xuesong Wang1; Johanna Teske1; Angie Wolfgang21 Carnegie Observatories (Pasadena, California, United States)2 Penn State (State College, Pennsylvania, United States)

We present the design, execution, and latest resultsof the Magellan TESS Survey (MTS), a systematicradial velocity (RV) follow-up program using thePlanet Finder Spectrograph (PFS) on the 6.5m Mag-ellan II telescope in Chile. We will characterize a sta-tistically robust sample of ∼30 super-Earths and sub-Neptunes discovered by TESS.

There are several features that make our surveyunique: (1) We designed our survey to be most ef-fective in addressing three specific science questions:How do planetary bulk densities depend on stel-lar insolation? How do planetary bulk densities de-pend on host star composition? How do planetarybulk densities depend on system architecture? (2)We have a clearly defined target selection functionand observation cadence design formula, enablingmore powerful statistical/population studies usingour sample. (3) We will publish all RV results regu-larly (∼ once per year) and at the end of our study,including non-detections and upper limits. We havealready been making our observing schedules publicon Exo-FOP.

On behalf of the MTS team, I will present: (1) Thedesign and execution of our program, including a de-scription of our target selection and cadence cover-age schemes, as well as the observation queue man-agement, which is uniquely challenging for first-year

32

follow-up observations of TESS targets. These wouldbe of general interest to groups that are running orwill run similar programs. (2) The latest highlightsand lessons learned from the first 10 months of MTS,including our battles with stellar jitter and severaluniquely interesting planetary systems (from recentpublications and papers in preparation).

303.06 — Radial Velocity Follow-up Program withFIES (Nordic Optical Telescope)

Andreea Gornea1; Lars Buchhave11 DTU Space, Technical University of Denmark (Copenhagen,

Denmark)

The high resolution Fiber-fed Echelle Spectrograph(FIES) at the Nordic Optical Telescope (NOT) is goingunder instrumental developments in 2019 to prepareit for a large radial velocity (RV) follow-up programfor TESS. The wavelength calibration for FIES willbe improved with the installation of a Fabry-Pérotcalibration source. The grating of the spectrographwill be enclosed with a pressure chamber which willdecrease the variations caused by the atmosphericchanges. Increasing the precision and stability of theinstrument will make it suitable for the RV follow-upprogram that will yield mass measurements for theexoplanet candidates that the TESS mission discov-ers. With support from NASA and MIT the programwill consists of up-to 200 nights.

303.07 — Results of the SOPHIE search for Nep-tunes and Super-Earths around bright solar-typestars

Nathan Hara1; François Bouchy1; Isabelle Boisse2; LucArnold3; Alexandre Santerne2

1 Université de Genève (Versoix, Switzerland)2 Laboratoire d’astrophysique de Marseile (Marseille, France)3 Observatoire de Haute Provence (Saint-Michel l’Observatoire,

France)

The SOPHIE spectrograph search for Neptunes andSuper-Earths around bright solar-type stars began in2011. We present the analysis of the 124 systems ob-served, with more than 7000 data points in total. Thefirst part of the presentation will be dedicated to newsignal processing techniques, regarding the correc-tion of the drift of the instrument, the selection of de-tection thresholds and the analysis of the time series,in particular the l1-periodogram. In the second part,we will present the ten new planets that have beendiscovered, most of which are detectable by TESS.

303.08 — Analysis of Exoplanetary Systems asWFIRST Targets

Zhexing Li1; Stephen Kane1; Margaret Turnbull21 Earth and Planetary Sciences, University of California, Riverside

(Riverside, California, United States)2 SETI Institute (Mountain View, California, United States)

As part of the WFIRST Coronagraph Science Inves-tigation Team (WFIRST-C SIT) to study exoplanetsaround nearby stars, we aim to characterize nearbyexoplanetary systems and provide a list of stars thatwould be suitable targets for WFIRST to carry outexoplanet direct imaging mission. To achieve that,we will be addressing two primary issues: charac-terization of stellar and orbital properties of nearbyexoplanets. Having a better understanding of hoststar characteristics and the Keplerian orbit propertiesof the known nearby exoplanets are crucial in deter-mining exoplanet targets for WFIRST direct imaging.These two aspects give us important insights suchas the presence of stellar and substellar companionin the system, planet-star separation, reflected lightfrom planets, background star fields etc. We use Ex-oCat as well as other online sources such as Sim-bad, Vizier, and Gaia DR2 to provide the best pos-sible stellar parameters for nearby exoplanet hoststars. We provide a strategy to conduct precursor ra-dial velocity observations to refine orbital ephemerisof nearby potential WFIRST exoplanet targets bythe use of major telescopes such as the AutomatedPlanet Finder. The combined effort will allow us toprogress towards the completion of target selectionfor WFIRST exoplanet observing program.

303.09 — Exoplanets orbiting giants stars: 10 yearsobservations of the EXPRESS program

Matias Jones11 European Southern Observatory (Santiago, Chile)

Evolved stars (subgiants and giants) are suited tar-gets for precision radial velocity studies by two mainreasons: 1) they are cooler and rotate slower thantheir former main-sequence progenitor, which allowus to achieve a radial velocity precision at the m/slevel for intermediate-mass stars , and 2) we can usethem to study the dynamical evolution of planetaryorbits due to the interaction with the expanding stel-lar envelope. Since 2009, we have been conducting aradial velocity survey called EXPRESS (EXoPlanetsaRound Evolved StarS) aimed at studying the pop-ulation of planets orbiting giant stars. We have ob-tained multi-epoch spectroscopic data for a sampleof 166 bright giant stars, resulting in the detection of

33

∼30 planetary systems (some of these in common withthe Pan-Pacific Planet Search), 2 brown dwarf candi-dates in the desert and 24 spectroscopic binaries, two ofthem with an astrometric orbit resolved using Hip-parcos data. Additionally, we have found that theplanet-metallicity correlation is valid for giant stars andwe have also confirmed previous results showingthat the giant planets formation efficiency increases withthe stellar mass (up to ∼ 2.0 Msun). In this talk I willdescribe our project and present the main results af-ter 10 years of observations. Finally, I will discussour findings in the context of planetary formation.

303.10 — A New Tool for Simulating Spectra Spec-tra & Validating Machine Learning Models for Ex-oplanet Discovery

Eric B. Ford1,2; Michael Palumbo1; Johannes Löhner-Böttcher3; Suvrath Mahadevan1; Jason Wright1

1 Astronomy & Astrophysics, Pennsylvania State University (Uni-versity Park, Pennsylvania, United States)

2 Institute for CyberScience, Pennsylvania State University (Uni-versity Park, Pennsylvania, United States)

3 High Altitude Observatory, NCAR (Boulder, Colorado, UnitedStates)

Currently, the planet detection sensitivity of state-of-the-art Doppler RV spectrographs is limited by in-trinsic stellar variability for most target stars. Mul-tiple groups are developing improved spectroscopicindicators (e.g., Jones et al. 2017, Zechmesiter et al.2017, Dumusque 2018, Wise et al. 2018) and pow-erful, but complex statistical models (e.g., Rajpaul etal. 2015; Jones et al. 2017) to mitigate stellar variabil-ity. These methods need to be validated and com-pared, in order to make robust and credible detec-tions of rocky planets in or near the habitable zoneof Sun-like stars. Recent results appear promising formitigating stellar variability originating from activeregions rotating across the disk, as long as observ-ing campaigns obtain dense time sampling and highspectral resolution/signal-to-noise. However, it isunclear if these same methods will be effective forstars dominated by granulation, which has a muchshorter timescale than rotationally-linked variability.

We present a new tool for generating syn-thetic high-resolution stellar spectroscopic time-series. While previous tools (e.g., SOAP, StarSim)have focused on modeling stellar activity, our toolfocuses on exploring the effects of convection andgranulation on measured radial velocities. Gener-ating synthetic datasets that include these effects iscritical for evaluating the performance of strategiesfor mitigating the stellar variability problem for Sun-like stars. Thus, our tool will play an important role

in validating machine learning algorithms and com-paring the efficacy of various approaches for mit-igating the effects of intrinsic stellar variability onDoppler planet surveys. Additionally the results arelikely to have important implications for target selec-tion in upcoming Doppler surveys. We will describeour model and show a movie of the line-profile vari-ations in our simulated data. We will evaluate theapparent radial velocities perturbations predicted byour new model and discuss the implications for up-coming Doppler planet surveys and TESS follow-upcampaigns.

303.11 — Distinguishing planets from stellar vari-ability with machine learning

Christian Gilbertson1; Eric B. Ford1; David Jones3;David Stenning4; Tom Loredo2

1 Astronomy & Astrophysics, Pennsylvania State University (Uni-versity Park, Pennsylvania, United States)

2 Astronomy, Cornell University (Ithaca, New York, United States)3 Statistics, Texas A&M University (College Station, Texas, United

States)4 Mathematics, imperial College London (London, United Kingdom)

The radial velocity method is one of the most suc-cessful techniques for the discovery and characteri-zation of exoplanets. Current RV surveys are sensi-tive to planetary signals of 1 m/s or less, but the vari-ability of stellar spectra (caused by star spots, pul-sations, convective motions, granulation, etc.) canmimic and obscure true planet signals at the samelevel. A data-driven approach for detecting plane-tary RV signals amidst stellar activity has recentlybeen proposed by Rajpaul et al. (2015) and refinedby Jones et al. (2017). This approach uses a physi-cally motivated multivariate Gaussian process (GP)to jointly model the apparent RV and multiple indi-cators of stellar activity, allowing the planetary RVcomponent to be separated from the total RV sig-nal. Our statistical framework combines spectro-scopic and temporal information to reconstruct theapparent RV perturbation and improve sensitivity tolow-mass planets. In this work, we build on previ-ous studies by simulating higher-fidelity active solarspectra time series (e.g., distribution of active regionsizes, rise and decay timescales, latitudes, and differ-ential rotation) and compare the performance of sev-eral different GP kernel functions. Our early resultssuggest specific alternative kernel functions that arelikely to improve the model and make it more sensi-tive to detecting low-mass planets. We will describesimulated datasets (which may be valuable for re-search groups testing their own approaches to mit-igating stellar variability), demonstrate the features

34

of our statistical model, and share our most recentresults on the impact of the choice covariance ker-nel on sensitivity to low-mass planets and the im-plications for planning RV surveys and follow-upcampaigns. The success of current and upcomingplanet-hunting instruments hinges on the commu-nity’s ability to overcome the stellar variability chal-lenge. This research represents important steps onthe path toward developing, validating and applyingpowerful machine-learning tools and a robust statis-tical framework for discovering and characterizinglow-mass exoplanets in the presence of stellar vari-ability.

303.12 — New Astrophysical Insights into RadialVelocity Jitter

Jacob Luhn1; Fabienne Bastien1; Jason Wright1; AndrewHoward2; Howard Isaacson3

1 Penn State University (State College, Pennsylvania, UnitedStates)

2 Caltech (Pasadena, California, United States)3 UC Berkeley (Berkeley, California, United States)

For nearly 20 years, the California Planet Search(CPS) has simultaneously monitored precise radialvelocities and chromospheric activity levels of starsfrom Keck observatory to search for exoplanets. Thissample provides a useful set of stars to better deter-mine the dependence of RV jitter on magnetic ac-tivity and stellar convection. For ∼650 stars cover-ing a wide range of stellar parameters (effective tem-perature, surface gravity, and activity, among oth-ers), there are enough RV measurements to distin-guish this astrophysical jitter from accelerations dueto orbital companions. To properly isolate RV jitterfrom these effects, we first remove the RV signal dueto these companions. We present some new resultsfrom our analysis of the CPS data, highlighting em-pirical evidence of two regimes of RV jitter – activity-dominated and convection-dominated – and the re-sulting “jitter minimum”. A more thorough under-standing of the various sources of RV jitter and theunderlying stellar phenomena that drive these in-trinsic RV variations will enable more precise jitterestimates for RV follow-up targets such as those fromthe K2 or TESS missions.

303.13 — Exoplanet Imitators: A test of stellar ac-tivity behavior in radial velocity signals

Chantanelle Nava1; Mercedes Lopez-Morales1; RaphaëlleHaywood1; Helen Giles2

1 Astronomy, Center for Astrophysics | Harvard & Smithsonian(Cambridge, Massachusetts, United States)

2 Astronomy, Observatoire de Genève (Geneva, Switzerland)

Effects from stellar activity are the largest barrierto detecting radial velocity signals of low-mass exo-planets. Radial velocity (RV) signals due to stellar ac-tivity are primarily caused by stellar magnetic activeregions on the rotating stellar surface. Current meth-ods to identify activity signals assume that the maxi-mum peak in a RV periodogram will occur at the stel-lar rotation period or a related harmonic. In this talk,we present results of simulations to test the effect ofnon-perfectly periodic activity signals in typical RVobservations. We simulate RVs with quasi-periodicsignals that account for the evolution and migrationof magnetic active regions. As test cases, we applyour analysis to two known exoplanet hosts, Kepler-20 and K2-131. Our simulations show the maximumpeak in the RV periodogram occurring at a periodunrelated to the stellar rotation period in 85% and72% of iterations, respectively for K2-131 and Kepler-20. We also show that in datasets with observationalsampling typical of current RV surveys, signals fromstellar activity can imitate those of exoplanet can-didates with orbital periods many days away fromthe stellar rotation period and any of its harmonics.These new results not only apply to small exoplanetdetection, but also have broad implications for thegeneral interpretation of periodic signals in stellarRVs.

303.14 — Two massive planets orbiting HD 25723and 17 Sco and two planet candidates around 3 Cncand 44 UMa

Marcelo Tala Pinto1; Sabine Reffert1; AndreasQuirrenbach2

1 Landessternwarte Königstuhl (Heidelberg, Germany)2 Landessternwarte, U Heidelberg (Heidelberg, Germany)

More than hundred exoplanets have been discoveredaround K and G giant stars, and their properties dif-fer considerably from those of the planets found or-biting late-type main-sequence stars. This allows usto study the properties of planetary systems afterthe host star has evolved off the main-sequence, andhelps us to constrain planetary formation and evolu-tion models. Our aim is to confirm the long-periodradial velocity variations observed in four giant starsof the Lick survey as caused by orbiting planets, andto study the properties of the planet population asa function of stellar evolutionary stage. We analyzetwelve years of precision radial velocity data for fourstars of the Lick sample. In addition, we compare theplanet occurrence rates as a function of evolutionarystage for two surveys, Lick and EXPRESS, based on

35

the evolutionary stages derived by the Bayesian In-ference method. We report the detection of two newexoplanets and two candidates orbiting giant stars.The best Keplerian fits to the data predict minimummasses of 2.3 mJ and 4.3 mJ for the planets orbitingHD25723 and 17 Sco, respectively. The minimummasses of the planet candidates around 3 Cnc and44UMa would be 20.7 mJ and 12.1 mJ, respectively.In addition, we computed planet occurrence rates forthe Lick and EXPRESS samples as a function of evo-lutionary stage. For the Lick sample the planet occur-rence rates are 5.3% and 4.3% for horizontal branchand red giant branch stars, respectively. For the EX-PRESS sample the horizontal branch and red giantbranch planet occurrence rates are 11.1% and 9.9%,respectively.

304 — Planet Detection — Mi-crolensing, Poster Session304.01 — Precise Mass Measurements of Cold Plan-ets Discovered by Microlensing: Cold planet MassFunction and Spatial Distribution in Our Galaxy

Jean-Philippe Beaulieu1,21 Institut d’Astrophysique de Paris (Paris, France)2 School of Natural Sciences, University of Tasmania (Hobart, Tas-

mania, Australia)

Microlensing is probing the unique population ofcold planets down to Earth mass orbiting aroundany kind of star, at any distance towards the galac-tic center. Relative physical parameters are knownto good precision from the modelling of the lightcurves, but it is necessary to combine the result oflight curve modeling with lens mass-distance rela-tions from additional observations and/or performa Bayesian analysis with a galactic model. Often,physical parameters are determined to 30-50 %. Re-cently, two kinds of constraints on masses have beenextensively used, coming from ground-space paral-lax Spitzer observations and high angular resolutionobservations with adaptive optics or HST. Our teamhas shown that we can derive physical parameters onknown systems to 10 % or better with mass-distancerelations obtained from high angular resolution ob-servations, either by detecting the lens flux and/orresolving source and lens and measuring the ampli-tude and direction of their relative proper motion.This work is also a pathfinder of the mass measure-ment method to be applied to WFIRST and Euclidmicrolensing programs.

We will report the results from our large observ-ing program with KECK and HST over 40+ plane-

tary systems. We revised the stellar and planetarymasses and distances for these systems, and oftenfound significant differences, even despite the initiallarge error bars. We also show some tensions withthe constraint from ground-space parallax Spitzerdata, where Spitzer lightcurves photometry seems tobe plagued by under estimated systematics for thefaint targets in very crowded field. We will discussthe impact of our analysis on the cold planet massfunction. With our revised distances, we found thatthe systems we re-visited so far tend to be located inthe Sagittarius or Scuttum-Crux arms, or at the tip ofthe bar.

304.02 — A blind search for free-floating planets inK2 Campaign 9

Iain McDonald1; Eamonn Kerins11 University of Manchester (Manchester, United Kingdom)

Free-floating planets may be as frequent in the galaxyas stars, and act as tracers of the dynamism of plan-etary systems. However, their occurrence rate, par-ticularly for low-mass planets, is very poorly known.We have conducted a blind search for microlensingsignatures caused by free-floating planets over thecontiguous K2 Campaign 9 field, towards the Galac-tic Bulge. This has revealed a number of candidatemicrolensing planets. I will present the latest analy-sis of the Campaign, and reflect on its shortfalls andthe lessons to be learnt for future surveys.

304.03 — Determining the NIR Microlensing EventRate at |b| < 2 with the United Kingdom InfraredTelescope

Savannah Renee Jacklin1; Yossi Shvartzvald2; GeoffBryden4; Sebastiano Calchi Novati2; Keivan Stassun1,6;B. Scott Gaudi3; Kiri Wagstaff4; Selina Chu4; CalenHenderson5; Matthew Penny3; Chas Beichman2

1 Physics & Astronomy, Vanderbilt University (Nashville, Ten-nessee, United States)

2 IPAC/Caltech (Pasadena, California, United States)3 The Ohio State University (Columbus, Ohio, United States)4 NASA Jet Propulsion Laboratory (Pasadena, California, United

States)5 Caltech/IPAC-NExScI (Pasadena, California, United States)6 Fisk University (Nashville, Tennessee, United States)

With the mid-2020s launch of the Wide Field InfraredSurvey Telescope (WFIRST) fast approaching, it isbecoming increasingly imperative to understand theoptimal spatial region for microlensing event detec-tion. The Galactic center (i.e. where |b| < 2) whichhas the highest density of potential source stars in

36

the Milky Way, has been historically understudieddue to the obscuring properties of its high volume ofgas and dust. The United Kingdom Infrared SurveyTelescope (UKIRT) microlensing project has succeedin mitigating some of the reddening effect of Galacticdust by observing in the near-infrared over a base-line from 2015-2018. Observations in the K and HNIR bands in unique fields have yielded hundredsof microlensing events detected via our UKIRT datareduction pipeline. We combine our microlensingdetections with image-level mock event injections inorder to determine our survey’s detection efficiency,and subsequently aim to derive the NIR microlens-ing event rate per observed square degree. Here wediscuss the methodology of our pipeline as well aspreliminary results for the NIR microlensing detec-tion efficiency and event rate. Understanding the in-trinsic NIR microlensing event rate at low Galacticlatitude is crucial for informing mission design andfield specifications for WFIRST.

304.05 — UKIRT Microlensing Survey as aPathfinder for WFIRST

Geoffrey Bryden11 Jet Propulsion Laboratory (Pasadena, California, United States)

Exoplanet microlensing surveys generally neglectthe very center of the Galaxy due to the very high op-tical extinction. The future NASA flagship mission,WFIRST, however, will operate at near-IR wave-lengths, such that its optimal target fields may be lo-cated in more central regions of higher stellar den-sity. To test this, we are using UKIRT’s wide-fieldnear-IR camera to survey the galactic bulge all theway to the center. We will present our 2017-2018 sur-vey results, both the raw number of event detectionsand the corresponding event rate maps, after correct-ing for the detection efficiency.

304.06 — Mass Measurements of Wide Orbit Exo-planets

David Bennett1,21 Code 667, NASA Goddard & U of Maryland (Greenbelt, Mary-

land, United States)2 Astronomy, University of Maryland (College Park, Maryland,

United States)

The original prediction of the core accretion theorywas that planet formation was dominated by wideorbit planets that grew beyond the snow line. Thediscovery of a large number of planets in short pe-riod orbits by the transit and radial velocity meth-ods may have changed these expectations somewhat,

but an understanding of the wide orbit planet pop-ulation is needed to advance our understanding ofthe panet formation process. The gravitational mi-crolensing method has unique sensitivity to wide or-bit planets down to an Earth mass, beyond the snowline, so it is our most promising method to study thedemographics of wide orbit planets over a large massrange. However, microlensing light curves usuallyreveal only the planet-star mass ratio and not theplanet or host star mass. (Masses are rarely mea-sured for wide orbit radial velocity planets, but thehost star mass is almost always knwon.) I presentnew results on the measurement of microlens planetand host star masses using the microlensing paral-lax method and high angular resolution follow-upobservations, and show how this difficulty will beresolved for microlens planetary systems discoveredby ground-based surveys and WFIRST.

305 — Planet Detection — Imaging,Poster Session305.01 — Kernel-Phase Interferometry for Super-Resolution Detection of Faint Companions

Samuel Factor1; Adam Kraus11 Astronomy, The University of Texas at Austin (Austin, Texas,

United States)

Filling out the dearth of detections between directlyimaged and radial velocity planets will test theoriesof planet formation across the full range of semi-major axes, connecting formation of close to wideseparation gas giants, and also substellar compan-ions. Direct detection of close-in companions is no-toriously difficult: coronagraphs and point spreadfunction (PSF) subtraction techniques are signifi-cantly limited in separation and contrast. Non-redundant aperture masking interferometry (NRMor AMI) can be used to detect companions well in-side the PSF of a diffraction limited image, thoughthe technique is severely flux-limited since the maskdiscards ∼95% of the gathered light. Kernel-phaseanalysis applies similar interferometric techniquesto an unobscured diffraction limited image. Kernel-phases are constructed by simulating a redundant in-terferometer as a grid of subapertures superimposedon the full telescope aperture and calculating phase-like observables (similar to closure-phases used withNRM). I have developed a new faint companion de-tection pipeline which analyzes kernel-phases utiliz-ing Bayesian model comparison. I break open theblack box of interferometry by demonstrating the useof this pipeline on archival HST/NICMOS images of

37

nearby brown dwarfs. I refine astrometry of previ-ously known companions and search for new com-panions, in order to constrain formation models atau scales. I also present contrast curves to demon-strate the strength of this technique at separations in-accessible to classical imaging techniques. Using thismethod, it is possible to detect companions downto flux ratios of ∼102—reaching the planetary-massregime for young targets—at half the classical λ/Ddiffraction limit while using a fraction of the tele-scope time as NRM. I am now preparing to use thistechnique to search for planetary mass companionsusing HST/ACS imaging of the young star-formingregions of Taurus and Upper Scorpius. Since JWSTwill be able to perform NRM and unobscured imag-ing, further development and characterization ofkernel-phase analysis will allow efficient use of com-petitive JWST time.

305.02 — Direct imaging of exoplanets in the midinfrared with VISIR

Dominique Petit Dit de La Roche1; Mario E. van denAncker1; Markus Kissler-Patig2; Valentin D. Ivanov1;Davide Fedele4; Sascha Patrick Quanz3

1 ESO (Garching, Germany)2 ESA (Madrid, Spain)3 ETH Zurich (Zurich, Switzerland)4 INAF (Firenze, Italy)

Direct imaging is a tried and tested method of detect-ing exoplanets in the near infrared, but has so far notbeen extended to longer wavelengths. Large ground-based telescopes are capable of routinely producingdiffraction limited images at mid-IR (8-20 micron)wavelengths. We have used the VISIR instrument onthe VLT to image the close vicinity of the young starsHD 100546 (10 Myr) and HR 8799 (60Myr) systemsin the mid infrared at 8.7micron. We use two differ-ent methods to reduce the data, angular differentialimaging and the subtraction of a circularised psf, andpresent the best mid-IR images to date of the diskaround HD 100546. We derive the most stringent up-per limits to date for the 8.7 micron flux of planets inboth systems.

305.03 — Initial results from BEAST: The B-star Ex-oplanet Abundance Study

Ruben Asensio Torres1; Markus Janson11 Stockholm University (Stockholm, Sweden)

In the last decade, a large number of direct imag-ing surveys have targeted hundreds of young and

nearby stars in the near-infrared, looking for self-luminous giant planets at separations >10AU. Theseobservations have proven that substellar compan-ions on wide orbits are rare, but seem to be morecommon with increasing host stellar mass. How-ever, the more massive B-type stars (>3MSun) havenot been studied to the same level of scrutiny asAFGKM types, and it is not clear whether this trendholds for the most massive stars or there is an over-turn, as suggested by the indirect methods at shortseparations. To address this issue, the B-star Ex-oplanet Abundance Study (BEAST) survey has re-cently been started with the goal of detecting giantplanets, brown dwarfs and disks around 83 B-typestars in Scorpius Centaurus with SPHERE. Here, wedescribe the layout of the survey and the current sta-tus of the initial exploratory observations. We alsopresent the first result yielded by BEAST, the discov-ery of a ∼20 Mjup circumbinary brown dwarf in Up-per Scorpius with a mass ratio <1%, i.e., consistentwith being formed through a planet-like mechanism.We will discuss the spectral properties of this objectand the importance of common proper motion whenclaiming physical association.

305.04 — Pushing the Limits of Exoplanet Discov-ery via Direct Imaging with Deep Learning

Kai Hou Yip1; Nikolaos Nikolaou1; Piero Coronica3;Angelos Tsiaras1; Billy Edwards2; Quentin Changeat2;Mario Morvan1; Beth Biller4; Sasha Hinkley6; JeffreySalmond3; Matthew Archer3; Paul Sumption3; ElodieChoquet5; Remi Soummer7; Laurent Pueyo7; IngoWaldmann1

1 Department of Physics and Astronomy, University College Lon-don (London, United Kingdom)

2 Physics and Astronomy, University College London (London,United Kingdom)

3 Research Software Engineering„ University of Cambridge (Cam-bridge, United Kingdom)

4 Centre for Exoplanet Science„ University of Edinburgh (Edin-burgh, United Kingdom)

5 Aix Marseille Univ (Marseille, France)6 Department of Physics and Astronomy, University of Exeter

(Exeter, United Kingdom)7 STScI (Baltimore, Maryland, United States)

Further advances in exoplanet detection and charac-terisation require sampling a diverse population ofextrasolar planets. One technique to detect these dis-tant worlds is through the direct detection of theirthermal emission. The so-called direct imaging tech-nique, is suitable for observing young planets farfrom their star. These are very low signal-to-noise-ratio (SNR) measurements and limited ground truth

38

hinders the use of supervised learning approaches.In this paper, we combine deep generative and dis-criminative models to bypass the issues arising whendirectly training on real data. We use a GenerativeAdversarial Network to obtain a suitable dataset fortraining Convolutional Neural Network classifiersto detect and locate planets across a wide range ofSNRs. Tested on artificial data, our detectors exhibitgood predictive performance and robustness acrossSNRs. To demonstrate the limits of the detectors, weprovide maps of the precision and recall of the modelper pixel of the input image. On real data, the mod-els can re-confirm bright source detections.

305.05 — New Spatially Resolved Observations ofthe SR 21 Transition Disk

Stephanie Sallum1; Andy Skemer1; Josh Eisner2; Nienkevan der Marel3; Patrick Sheehan4; Laird Close2; MichaelIreland5; Jared Males2; Katie Morzinski2; VanessaBailey6; Runa Briguglio7; Alfio Puglisi7

1 UC Santa Cruz (Santa Cruz, California, United States)2 University of Arizona (Tucson, Arizona, United States)3 NRC Herzberg (Victoria, British Columbia, Canada)4 NRAO (Charlottesville, Virginia, United States)5 Australian National University (Canberra, Australian Capital

Territory, Australia)6 Jet Propulsion Laboratory (Pasadena, California, United States)7 INAF - Observatorio Astrofisico di Arcetri (Firenze, Italy)

We present new 0.6 - 4 micron imaging of the SR 21transition disk from Magellan / MagAO and Keck /NIRC2. The protoplanetary disk around SR 21 has a30 - 40 AU dust clearing first inferred from its spec-tral energy distribution and later confirmed in sub-millimeter imaging. The gas and small, micron-sizeddust grains are known to have a different morphol-ogy, with a truncation in CO at ∼ 7 AU and H bandscattered light detected within the millimeter clear-ing. The observations presented here probe tighterangular separations than previous studies, placingnew constraints on the geometry of the small-graindisk. Reproducing the imaging data requires amisaligned inner disk or azimuthal asymmetries inthe dust distribution. Furthermore, reconciling theimaging with the spectral energy distribution mayrequire grain growth to ∼2-5 microns. These fea-tures can be connected to dynamical shaping by anunseen, giant-planet mass companion, a situationsupported by previous observational and theoreticalstudies.

305.06 — Giant planet formation in the near-infrared: single large telescopes are not enough

Jens Kammerer1,2; Alexander Wallace1; Michael Ireland11 Australian National University (Canberra, Australian Capital

Territory, Australia)2 European Southern Observatory (Garching, Germany)

Giant planets are thought to form at large orbital sep-arations (>3 au), which is why direct imaging is cru-cial to detect young gas giants and study their forma-tion process. PDS 70 b remains the only fascinatingcase of such an object for which accretion is clearlyevident.

Using Keck/NIRC2, we search for companionsaround 33 members of the ∼1 Myr old Taurusstar-forming region. Fourier plane imaging (kernelphase) and PSF subtraction allow us to probe a largerange of orbital separations down to Solar Systemscales (∼6 au = 0.5 λ/D), which has not been pos-sible before in the near-infrared. Together with oursimulation of giant planet formation via core accre-tion and the latest planet distribution from radial ve-locity surveys, our observations put some constraintson the peak luminosity and the timescale of the run-away accretion.

Considering future observations, we show thateven with optimistic assumptions, the number of“normal” core accretion giant planets detectable by8-10 m telescopes is of order 1. This is due to evi-dence for a turnover in planet frequency at ∼3 au,the small number of ∼solar-mass young stars har-bouring gas disks within 200 pc, and the insufficientbrightness of truly “Jovian” planets. We make thecase that in order to probe the formation of planetslike Jupiter, a high-contrast interferometric instru-ment (such as Hi-5/VIKiNG) is required.

305.07 — PDS70, witnessing a young solar systemanalog in formation

Dino Mesa11 INAF - OAPD (Padova, Italy)

PDS70 is a young (∼5 Myr) star hosting a knowntransition disk with a large gap resolved with ob-servations at NIR and (sub-)millimiter wavelengthswith SMA and ALMA. Recent observations withSPHERE allowed us to detect a planetary companion(5-9 MJup) at a separation of ∼22 au (PDS70b), wellinside the disk gap. This companion was then con-firmed by observation with other instruments (e.g.MagAO, SINFONI, MUSE) which furthermore re-vealed ongoing accretion on PDS 70 b. In addi-tion, recent Hα observations with MUSE also sug-

39

gested the presence of a second planetary compan-ion at larger separation from the star. Re-analysisof archive SPHERE data and new data taken withthis aim allowed us to confirm and to characterizethis object (PDS70c). Furthermore, we reporton thetentative dtection of a third source (PDS70d) locatedat shorter separation than PDS70b. Its spectrum iscompatible with the presence of dust and thereforethe existing data do not allow to confirm its plane-tary nature. We will present here some hypothesison its nature. Finally, we will describe the possiblestructure of the PDS70 planetary system as emerg-ing from our current knowledge. PDS70 is a uniquecase where the properties of a planetary system canbe characterized during its formation stage.

305.08 — Thermal-Infrared spectroscopic studies ofplanets and protoplanets with ALES

Jordan Stone1; Phil Hinz2; Andy Skemer2; CharlesWoodward3; Travis Barman4; Mike Skrutskie5

1 Astronomy, University of Arizona (Tucson, Arizona, UnitedStates)

2 Astronomy, UC Santa Cruz (Santa Cruz, California, UnitedStates)

3 University of Minnesota (Minneapolis, Minnesota, United States)4 Lunar and Planetary Lab, University of Arizona (Tucson, Ari-

zona, United States)5 Astronomy, Univeristy of Virginia (Charlottesville, Virginia,

United States)

Understanding the physical state of gas-giant exo-planet atmospheres is challenging due to persistentdegeneracies between effective temperature, cloudi-ness, and dis-equilibrium chemistry. Broad wave-length spectroscopic studies are essential to breakthese degeneracies, yet, until recently, all extremeadaptive optics systems focused on 1-2.5 micronsensitivity. To break model degeneracies and im-prove our understanding of gas-giant atmospheres,we built an adaptive optics-fed integral field spectro-graph with sensitivity out to 5 microns — the Ari-zona Lenslets for Exoplanet Spectroscopy (ALES). Iwill present initial results including 2.8-4.2 micronspectra of gas-giant exoplanets. I will also provide anupdate on the status of recent instrument upgradesincluding increased spectral resolution and the ad-dition of high-performance coronagraphic technolo-gies.

305.09 — How to Detect Long-Period Neptunesthrough Direct Imaging

Daniel Tamayo1; Loic Nassif-Lachapelle2

1 Astrophysical Sciences, Princeton University (Princeton, NewJersey, United States)

2 University of Toronto (Toronto, Ontario, Canada)

Modeling of the gap structures in protoplanetarydisks revealed by the Atacama Large Millimeter Ar-ray suggests that sub-Jupiter-mass planets may becommon at large orbital separations. If true, this im-poses fundamental constraints on theories of planetformation to both form giant planet cores quicklyat large orbital separations, and to sufficiently sup-press runaway gas accretion to yield the inferredplanet mass function. Unfortunately, while futurehigh-contrast instruments will push toward detect-ing ever-smaller planets closer to their host star, di-rectly imaging this putative population of distantsub-Jupiters at typical contrast ratios of 10−10 - 10−12

will remain out of reach for the foreseeable future.However, building on models of circumplanetary

debris disks (CPDDs) by Kennedy & Wyatt (2011),we show that debris from collisions between irregu-lar satellites can increase young planets’ contrast ra-tios by several orders of magnitude. Dozens of irreg-ular satellites are found around each of our own gi-ant planets, and they indeed represent the most col-lisional evolved population in the solar system, be-traying bright CPDDs in the past. We find that exo-planetary CPDDs would have been below the detec-tion threshold of the GPIES direct imaging survey.However, depending on irregular satellites’ capturemechanism and efficiency, we show that by opti-mizing the target selection for finding such struc-tures, the CPDDs of this population of long-period,Neptune-mass planets may be detectable with cur-rent instrumentation, and demonstrate that it shouldbe an important science case for the next generationof direct imaging with ELTs and WFIRST coronog-raphy. We argue that such structures can be unam-biguously identified through their strong polariza-tion signatures and discuss their implications for ourunderstanding of not only planet formation but alsoour own Solar System’s history.

306 — Planet Detection — Other,Poster Session306.01 — How to detect forming planets?

Judit Szulagyi11 Institute for Computational Science, University of Zurich

(Zurich, Switzerland)

Giant- and immediate mass planets are surroundedby their circumplanetary disk during the last stage

40

of their formation. In order to detect nascent plan-ets, therefore, we need to understand the character-istics of this disk, since this is what we are going toobserve. The planet is embedded within the circum-planetary disk, therefore we would not be able tosee that directly, as simulations show. I create mockobservations on various wavelengths by combining3D radiative hydrodynamic simulations and Monte-Carlo radiative transfer to create synthetic images,and spectral energy distributions (SEDs). In my talkI will show how these images look like at sub-mmand radio wavelengths, at near/mid-infrared andat polarized scattered light for the various currentand near-future instrumentation, such as ALMA,SPHERE/GPI, ERIS/NaCo etc. The spectral energydistribution of the circumplanetary disk will be alsodiscussed and compared with the circumstellar diskSED, in order to identify the wavelength range wherethe best contrast can be achieved to detect the circum-planetary disk and the forming planet. Finally, I willtalk about what line fluxes we can expect regardingdetecting H-α from these sources. I will try to givean answer why previous attempts of detection of cir-cumplanetary disks often failed, what are the diffi-culties to face with, and what systems we could de-tect with current/near future instrumentation basedon my simulations. To understand how formingplanets should look like on observations, also help usdistinguishing forming planets from other circum-stellar disk features. I will highlight, that in the for-mation phase, unfortunately we cannot estimate theplanet mass based on the observed brightness, be-cause the fluxes are always contaminated by the cir-cumplanetary disk contribution, and therefore thebrightness will depend mainly on the disk proper-ties (temperature, dust-to-gas ratio, density, viscos-ity, etc.), less about the planet luminosity.

306.02 — Astrometric orbits of tight substellar bi-naries

Johannes Sahlmann11 Space Telescope Science Institute (Baltimore, Maryland, United

States)

We present new results from the high-precisionastrometric monitoring of nearby very-low-massstars brown dwarfs with Gemini/GMOS andVLT/FORS2. The goals of these projects are thecharacterisation of known spectral binaries andthe discovery of companions down to sub-Jupitermass, respectively. We will give an overview ofthe program, report on the orbit determination ofspectral binaries, and present an update on ourplanet-search results. We will put these results

into the context of efforts to determine the tightbinary fraction of ultracool dwarfs and to explorethe occurrence of planets around these objects.

306.03 — What we’ll see when we’ve seen what wesee that we can see

Zephyr Penoyre1; Emily Sandford21 University of Cambridge, Institute of Astronomy (Cambridge,

United Kingdom)2 Department of Astronomy, Columbia University (New York, New

York, United States)

Improving instrumental precision is like a recedingtide, revealing the geological shapes underneath thewater’s surface. New signals rise out of the noise,take shape and gain familiarity, until it is hard toimagine ever not having known them.

Out-of-transit effects — tides, beaming and reflec-tions in particular — are one such family of signalswhich will transition from near-invisible to common-place. As photometric precision drops from 100sof parts per million to 10s and below, these signalswill go from being occasional to ubiquitous in lightcurves, especially for massive, close-in or eccentricplanets.

We can leverage these signals as tools for con-straining planetary properties, confirming candi-dates and detecting new planets — but doing so re-quires a detailed, intuitive and accurate theoreticalunderstanding of the physics at play and the obser-vational signatures.

Here we present analytic models of these effects, ofsufficient simplicity to allow easy intuition and cal-culation, whilst encoding a full physical picture ableto capture the behaviour of the broad exoplanet zoo.

306.04 — The potential of direct detection of exo-planets by optical interferometry

Sylvestre Lacour11 LESIA, Observatoire de Paris (Meudon, France)

With over 4000 exoplanets discovered, the focus ofexoplanet research progressively shifts from censusto characterization. Direct imaging targets a differ-ent planet population than transit spectroscopy: it ispossible to obtain spectra of young, far-out exoplan-ets. And in that field, optical interferometry is on theverge of playing a major role: with baselines of hun-dred meters, its spectral and differential imaging ca-pacities surpass by order of magnitudes those of sin-gle dish telescopes. During this talk, I will presentthe interferometric technique which enables direct

41

observations of exoplanets. I will present the detec-tion of HR8799e by the GRAVITY instrument, anddiscuss the capability of the technique for future de-tections.

306.05 — Alkaline Signatures of an Active Exo-moon

Apurva V. Oza1; Robert E. Johnson2; EmmanuelLellouch3; Carl Schmidt4; Nick Schneider5; ChenliangHuang6; Diana Gamborino1; Andrea Gebek1,7; Aure-lien Wyttenbach8; Brice-Olivier Demory9; ChristophMordasini1; Prabal Saxena11; David Dubois10; ArielleMoullet10; Nicolas Thomas1

1 Physikalisches Institut, Universität Bern (Bern, Switzerland)2 AMES Research Center, NASA (Moffett Field, California, United

States)3 Goddard Space Flight Center, NASA (Greenbelt, Maryland,

United States)4 Engineering Physics, University of Virginia (Charlottesville,

Virginia, United States)5 LESIA, Observatoire de Paris (Meudon, France)6 Center for Space Physics, Boston University (Boston, Mas-

sachusetts, United States)7 LASP, University of Colorado Boulder (Boulder, Colorado, United

States)8 Physics and Astronomy, University of Las Vegas (Las Vegas,

Nevada, United States)9 Physik, Eidgenossische Technische Hochschule Zurich (Zurich,

Switzerland)10 Leiden Observatory, Leiden University (Leiden, Netherlands)11 Center for Space and Habitability, Universität Bern (Bern,

Switzerland)

Exomoons are generally too small to be detectedby nominal searches. By analogy to the most ac-tive body in the Solar System, Io, we describe howsodium (Na I) and potassium (K I) gas could be asignature of the geological activity venting from anotherwise hidden exo-Io. Analyzing a dozen close-in gas giants hosting robust alkaline detections, weshow that an Io-sized exomoon can be stable againstorbital decay below a planetary tidal Qp < 1011.This tidal energy is focused into the satellite driv-ing ∼105 times more mass loss than Io’s supply toJupiter’s Na exosphere, based on a simple atmo-spheric loss model. The remarkable consequenceis that several exo-Io column densities are on av-erage more than sufficient to provide the 1010±1Nacm−2required by the equivalent width of exoplanettransmission spectra. Furthermore, the benchmarkobservations of both Jupiter’s extended ( ∼1000 RJ)Na exosphere and Jupiter’s atmosphere in transmis-sion spectroscopy yield similar Na columns that arepurely exogenic in nature. As a proof of concept,

we fit the “high-altitude” Na at WASP 49-b with anionization-limited cloud identical to the precise Naprofile about Io. Moving forward, we strongly en-courage time-dependent ingress and egress monitor-ing along with spectroscopic searches for other vol-canic volatiles.

306.06 — Detection of a New Planet in a ResonantOrbit Using Transit Timing Variations

Chris Fox11 Physics & Astronomy, University of Western Ontario (London,

Ontario, Canada)

The vast amount of data from the Kepler Space Tele-scope has provided more than just transiting plan-ets. By analyzing the timing of transit, the exis-tence of more planets, as well as the properties of al-ready known planets, can be determined. Here welook at the case of Kepler-159, a system of two tran-siting planets, one of which shows significant tran-sit timing variations. Using orbital modeling andBayesian Inference, we determine the existence of athird planet in a resonant orbit, as well as mass esti-mates and orbital parameters for the two interactingplanets.

307 — Transit Timing, Poster Ses-sion307.01 — Unravelling the Hidden Features of Time-Varying Signals in a Photocentric Model of Transit-ting Exoplanets with Moon

Pongpichit Chuanraksasat1; Supachai Awiphan11 National Astronomical Research Institute of Thailand (Chiang

Mai, Chiang Mai, Thailand)

There have been several attempts to use the TransitTiming Variations (TTV) and Transit Duration Vari-ations (TDV) techniques to infer the presence of ex-omoons from existing observational data, but nonehas successfully been able to accomplish this. Partof the reasons is the unexplained deviations of sig-nals in the theoretical sinusoidal shape, which re-sult from light curves fitted by a photocentric model.In this work, the deviation of time-varying signalsis characterised by considering the fitting of syn-thetic star-planet-moon light curves using a photo-centric model. The magnitude of signal deviationdepends on the position and length of moon transitcomponent at different moon phases. This results inunique features in the phase evolution of TTV, TDVand effective transit depth signals in and around

42

the primary and secondary transits of the moon in-side the planet. Taking into account the aforemen-tioned effect and neglecting limb darkening, the de-pendency of features’ magnitude on masses, radiiand orbital separations is described for the cases ofmoons at large planet-moon orbital separations. Forlarge moons with non-negligible transit components,this demonstrates that the contribution of moon tran-sit component to the resulting TTV, TDV and effec-tive transit depth needs to be accounted for in orderto confirm exomoon detections.

307.02 — Transit Timing Variation Refinement ofthe Long-period Exoplanet Kepler-167e

Paul A. Dalba1; Patrick Tamburo21 Department of Earth and Planetary Science, University of Califor-

nia Riverside (Riverside, California, United States)2 Department of Astronomy, Boston University (Boston, Mas-

sachusetts, United States)

Kepler-167e is a long-period (P=1,071 days), Jupiter-size exoplanet that was discovered in transit obser-vations by the Kepler spacecraft during its primarymission. Many properties of Kepler-167e includ-ing its eccentricity, stellar insolation, and equilib-rium temperature are strikingly similar to those ofJupiter, making this exoplanet an excellent candidatefor comparative planetology. Kepler observed onlytwo transits of Kepler-167e, which left the existenceof transit timing variations (TTVs) unknown. Manylong-period exoplanets and candidates have shownTTVs of up to 40 hours in duration. Until the exis-tence of such TTVs are constrained, follow-up transitobservations (e.g., for atmospheric characterization)are extremely risky. We present new Spitzer obser-vations of Kepler-167 that recover a partial transit ofKepler-167e. These observations constrain the extentof TTVs in the system, and enable accurate and pre-cise predictions of future transits through the antici-pated era of JWST.

307.03 — Maintenance of Transit Timing Errors us-ing Real and Simulated Telescope Networks

Hamish Elliot Caines11 University College London (London, United Kingdom)

The Atmospheric Remote-sensing Infrared Exo-planet Large-survey (ARIEL) will observe 1000 ex-oplanets spectroscopically during transit to charac-terise their atmospheres. Optimal use of the obser-vation time will require accurate transit times for allof the selected targets. The current uncertainty inthe transit time will propagate significantly between

now and the mission launch in 2028. The calculateduncertainty at launch in many cases is too large foroptimal scheduling of ARIEL observation time, andin some it is larger than the duration of the tran-sit itself, so these transit timings are effectively lost.Therefore, ground-based follow-up to obtain moretransit times for each target is required, as this re-duces the amount of time the uncertainties are prop-agated over. In this work we determine a set of cri-teria that provides optimal prioritisation of transitevents that maintains timing accuracy for all ARIELtargets to within seconds from today until launch.We present a telescope network simulator that willbe used to evaluate potential criteria sets. The size ofthe network in the simulation can be adjusted, so theminimum number of telescopes required to best ex-ecute the follow-up can be also determined. In par-allel, high priority targets will be observed in tran-sit using a real network of telescopes, yielding newtiming data. We present sets of real light curves andephemerides obtained using this telescope network,and demonstrate the effect of adding observations toa data set on the timing uncertainty of a given target.In addition, we provide an estimate of the size andcost of a network able to provide the amount of tele-scope time needed to constrain the transit timing forall targets until and beyond launch.

308 — Stellar Spins and Obliqui-ties, Poster Session308.01 — Finding Waldo: The Rossiter-McLaughlineffect of π Men c hidden within stellar oscillations

Vedad Kunovac-Hodzic1; Amaury Triaud11 School of Physics and Astronomy, University of Birmingham

(Birmingham, United Kingdom)

Context. The formation and dynamical evolution ofsuper-Earths is subject to an intense debate. About50% (70%) of FGK (M) stars host super-Earths, mak-ing them one of the most frequent types of plan-ets. Yet, we do not understand whether the close-in super-Earths we can detect formed close to thestar, or further out followed by an inwards migrationprocess, such as dynamical scattering with an outercompanion. In the latter case, the scattering pro-cess would leave an observable trace in the form ofhigh obliquity, which can be measured through theRossiter-McLaughlin (R-M) effect and can thus dis-tinguish between different formation mechanisms.The first TESS planet, π Men c, offers a great test casegiven that the G dwarf also hosts an outer massive

43

planetary companion on an eccentric orbit, reminis-cent of past dynamical interactions within the sys-tem.

Methods. R-M observations for π Men c are chal-lenging since its signal is expected to be dwarfed bystellar pulsations. However, the bright host star al-lows for short cadence observations that can resolvethe oscillation frequency, yet retain the R-M signal.We have obtained spectroscopic transits of π Men con ESPRESSO. The data show high frequency peak-to-peak variation of 4-5 m/s, masking the expected0.7 m/s R-M amplitude. In a novel framework, weemploy Gaussian processes to perform asteroseis-mology in the time-domain on a main-sequence starto successfully extract the dominant frequency, νmax,and uncover the projected obliquity, λ. Moreover,we use TESS data to constrain the stellar rotation andextract the 3D obliquity, ψ. We verify our analysis byattempting to extract the planet-occulted light of thestar for an independent retrieval of ψ.

In this talk. I will present our efforts towardsthe first detection of the R-M effect from a super-Earth, showing some of the first science results fromESPRESSO on the obliquity of pi Men c. In the eraof extreme precision RVs, our analysis demonstratesthe need for incorporating asteroseismology in themodelling of stellar variability to push R-M observa-tions towards small planets in order to study theirdynamical histories.

308.02 — Doppler Tomographic Analysis for Plan-etary Orbital Precession of WASP-33b

Noriharu Watanabe1; Norio Narita2; Marshall C.Johnson3

1 Astronomical Science, SOKENDAI (Graduate University forAdvanced Studies) (Mitaka, Tokyo, Japan)

2 Astrobiology Center (Mitaka, Tokyo, Japan)3 Department of Astronomy, The Ohio State University (Columbus,

Ohio, United States)

Apparent orbital obliquity λ is one of the importantparameters to understand orbital evolutions. If aplanet has followed an orbital evolution like thosein our solar planets, its orbit will be aligned withthe stellar spin. It is called a prograde orbit (|λ| <90 deg) and many exoplanets orbit in this direction.However, there are few exoplanets with retrogradeorbits (|λ| > 90 deg). Doppler tomography (DT) isone of the methods to measure λ. When a planet cov-ers part of the stellar surface during transit, a plan-etary shadow appears in the broadened line profile.Then, λ can be derived by the track of the shadow.Moreover, impact parameter b can be also measuredby DT.

Johnson et al. (2015), henceforth as J+15, foundthat the transit chord of WASP-33b, which has a 1.2-day period retrograde circular orbit around a rapidlyrotating and pulsating A-type star, changed slightlyfrom 2008 and 2014. They detected its orbital preces-sion due to its slightly flattened central star. How-ever, only two observational epochs, from 2008 and2014, were used in J+15. We aim to confirm and moreprecisely measure the precession using not only thedataset of 2008 and 2014 but also a previously un-published dataset from 2011.

In our research, we used observational data ofWASP-33 which was obtained using the High Dis-persion Spectrograph (HDS) on the 8.2m Subarutelescope on 19th October 2011 (UT), as well as datasets of J+15 which has already been analyzed upto their line profiles. We got the planetary shadowshowing a retrograde orbit and a component fromstellar pulsations. In order to make the measure-ment of the planetary parameters more easily, we ex-tracted only the planetary shadow by Fourier filter-ing used in J+15.

In order to make the measurement of λ and b, weadopted an MCMC analysis for the datasets with ourplanetary shadow model with Fourier filtering. Thenwe found that our measured values did not followthe equations of a long-term orbital precession fromIorio (2016). This may imply that WASP-33b’s pre-cession has a short unclear variation or our measurederrors are underestimated.

308.03 — Reloaded RM with ESPRESSO: NewPlanet Architectures and Stellar Activity Character-isation

Heather Cegla11 University of Geneva (Versoix, Geneva, Switzerland)

Stellar surface phenomena (spots, faculae, granula-tion etc.) alter the observed stellar spectra and can in-ject spurious signals into a variety of planet detectionand characterisation methods. As such, our knowl-edge of other solar systems depends strongly on ourunderstanding of their host stars. For these reasons,we ‘Reloaded’ the Rossiter-McLaughlin (RM) effectusing transiting planets to directly probe stellar sur-faces and 3D planetary dynamical histories. Withthe ‘Reloaded RM’, we can isolate the starlight be-hind the planet without making any assumptionson the local absorption line profiles, or underly-ing stellar radial velocities. We have successfullyapplied this new technique to planets orbiting GKdwarfs, pinning down 3D geometries and demon-strating that classical RM analysis may bias our plan-etary interpretations. Moreover, with the Reloaded

44

RM we have also unveiled the first planet architec-ture around a cool M dwarf. Here, I will presentthe first Reloaded RM results obtained from theESPRESSO spectrograph, including both a planet or-biting a magnetically active K dwarf and anothermisaligned planet orbiting a hot F dwarf. Initial anal-ysis reveals striking divergences from the literature,owing to the increased instrumental precession ofESPRESSO and advances from this new technique;in particular, we find evidence for significantly dif-ferent star-planet obliquities and potentially contam-ination from starspot penumbral flows.

309 — Population Statistics andMass-Radius Relations, Poster Ses-sion

309.01 — Threshold Radii of Volatile-rich Planets

Michael Lozovsky1; Ravit Helled1; Caroline Dorn1; JuliaVenturini1

1 University of Zurich (Zurich, Switzerland)

We use a statistical analysis to determine the charac-teristic maximum radii (“threshold radii”) for vari-ous compositions for exoplanets with masses up to25 Earth masses. We constructed a series of plane-tary models in order to characterize exoplanets bytheir compositions. Our models correspond to ho-mogeneous Earth-like composition planets, planetsof pure water or rock, along with planets with com-plex structure of a rocky core surrounded by a lightatmosphere containing water. We confirm that mostplanets with radii larger than 1.6 Earth radii (Re) arenot rocky, and must consist of lighter elements, asfound by previous studies (Rogers, 2014). We findthat planets with radii above 2.6 Re cannot be pure-water worlds, and must contain significant amountsof light gases, such as hydrogen and helium (H–He).We find that planets with radii larger than about 3 Re,3.6 Re, and 4.3 Re are expected to consist of at least2%, 5%, and 10% of H–He, respectively. We showthat the atmospheric composition, the mass fractionof H–He in the planet, and the distribution of the el-ements play a significant role in the determination ofthe threshold radius. We conclude that, although theexact planetary composition cannot be inferred frommass and radius alone, it is possible to put limits onthe range of possible compositions for planets withwell-measured mass and radius.

309.02 — The Typical Earth-mass Planet Discov-ered by Transit Surveys and Its Implications forPlanet Formation and Evolution

Kevin C. Schlaufman11 Physics and Astronomy, Johns Hopkins University (Baltimore,

Maryland, United States)

All mass-radius relations for low-mass planets pub-lished to date have been affected by observational bi-ases. Since planet occurrence and primordial atmo-spheric retention probability increase with period,the ”typical” planet discovered by transit surveysmay bear little resemblance to the short-period plan-ets sculpted by atmospheric escape ordinarily usedto calibrate mass-radius relations. An occurrence-weighted mass-radius relation for the typical low-mass planets in the Galaxy observed so far by tran-sit surveys requires the typical Earth-mass planet tohave about 1% of its mass in a H/He atmosphereto explain its observed radius. Unlike the terrestrialplanets in our own solar system that finished form-ing long after the protosolar nebula was dissipated,these Earth-mass planets discovered in transit sur-veys must have formed early in their systems’ histo-ries. The existence of significant H/He atmospheresaround Earth-mass planets confirms an importantprediction of the core-accretion model of planet for-mation. It also implies that such planets can retaintheir primordial atmospheres and requires an order-of-magnitude reduction in the fraction of incidentXUV flux converted into work usually assumed inphoto-evaporation models. Because the short-periodplanets likely to be discovered by NASA’s TransitingExoplanet Survey Satellite (TESS) are not representa-tive of the Galaxy’s planet population, it will be im-portant to use occurrence-weighting when consider-ing the implications for models of planet formationof the masses and radii of TESS discoveries.

309.03 — Getting Better at Measuring the GalacticDistribution of Planets with Spitzer

Lisa Dang1; Sebastiano Calchi Novati2; Sean Carey21 Physics, McGill University (Montréal, Quebec, Canada)2 IPAC, California Institute of Technology (Pasadena, California,

United States)

Gravitational microlensing is a powerful tool thatallows us to discover planets through the gravita-tional effect they have on light from more distantsources. Unlike most other planet detection meth-ods, gravitational lensing does not rely on the detec-tion of photons from the planet or its host star. There-fore, this method allows us to probe planets well out-

45

side of the Solar neighborhood. In addition, grav-itational microlensing is most sensitive to detectingplanets at and beyond the snowline and Solar Systemanalogs, a category of planets that is extremely chal-lenging to detect with other methods. Since 2015, theSpitzer team is leading a microlensing observationalcampaign towards the Galactic Bulge following upmicrolensing events alerted by ground-based sur-veys. Near-simultaneous observations of microlens-ing event from two distant observatories allow forthe measurement of microlens parallax which al-lows us to obtain robust measurements of both thelens’ mass and distances. The main scientific driveof this campaign is to build the galactic distribu-tion on planets towards the bulge of the Milky Way.As microlensing event are mostly unpredictable, sur-veys toward the bulge are favorable, however, pho-tometry extraction in crowded fields is challenging.In this talk, I will present results from this SpitzerMicrolensing campaign and our efforts in obtainingexquisite level of photometric precision.

309.04 — Transiting Planets Around Red GiantStars

Samuel Grunblatt1; Daniel Huber1; Eric Gaidos21 Institute for Astronomy, University of Hawaii (Honolulu, Hawaii,

United States)2 Department of Geology and Geophysics, University of Hawaii

(Honolulu, Hawaii, United States)

Every Sun-like star will eventually evolve into a redgiant, a transition which can profoundly affect theevolution of a surrounding planetary system. Thetimescale of dynamical planet evolution and orbitaldecay has important implications for planetary hab-itability, as well as post-main sequence star andplanet interaction, evolution and internal structure.We demonstrate how photometric surveys such asKepler, K2, and TESS are vastly improving our un-derstanding of evolved planetary systems. We de-scribe results from the first estimate of planet occur-rence around evolved stars using light curves, andcompare the observed planet populations of evolvedand main sequence stars. We then discuss the popu-lation of red giant planetary system candidates dis-covered by transit surveys to date, and use this pop-ulation to derive constraints on the timescales ofplanet inflation, inspiral and engulfment. Finally,we illustrate the potential of full frame images fromthe full 2-year primary TESS mission to increase theknown transiting planet population of red giants bymore than an order of magnitude, and advocate forcontinued all-sky coverage at a 30 minute cadence or

higher to maximize the potential to understand late-stage planetary evolution.

309.05 — “Dynamically Hot” Stars Prefer HigherPlanet Fraction and Smaller Planet?

Huigen Liu11 School of Astronomy and Space Science, Nanjing University

(Nanjing, China)

The correlations between stellar physical propertiesand planetary system have been explored based onKepler planets, e.g. stellar mass, metallicity. How-ever, the stellar velocities may reveal the dynamicalhistory and influence the formation and evolution ofplanetary systems. Here we utilize the data of GaiaDR2 to calculate the motion of Kepler stars includ-ing planet hosts, to check if there is any correlationsbetween stellar motion and the planetary systems.The KIC stars are divided into two samples accord-ing to the deviation of the stellar velocities, i.e. stan-dard stars and “dynamically hot” stars. We find the“dynamically hot” stars have a higher planet frac-tion. Based on the 355 CKS multi-planet systems,the radius of the outer most planet around “dynam-ically hot” stars are smaller than the standard stars.To explain the interesting correlations, we comparethe distribution of stellar mass and metallicity forthe two star samples. The “dynamically hot” sam-ple prefers stars with smaller masses and metallici-ties. Based on the MC simulations, planet fractionof Dynamically hot star have a larger planet fractionin 1-σ confidential level for Kepler samples. The con-clusions induced that dynamical history of star couldinfluence planetary formation and evolution via ob-servations.

309.06 — The TESS Follow-up Observing Programand the Characterization of Small Planets

Samuel N. Quinn1; David Latham1; Karen Collins1;David Ciardi3; Diana Dragomir2

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 Hubble Fellow, MIT Kavli Institute (Cambridge, Massachusetts,United States)

3 Caltech/IPAC-NASA Exoplanet Science Institute (Pasadena,California, United States)

Over the course of its two-year primary mission,TESS will survey most of the sky in search of smallplanets transiting the nearest stars, the brightnessof which enables studies of planetary compositionsand atmospheric properties. The efficient deploy-ment of ground- and space-based observing facil-ities in pursuit of such characterization, however,

46

requires a community effort to vet planet candi-dates and coordinate resources. The TESS Follow-upObserving Program (TFOP) is a mission-organized,community-driven working group, the primary goalof which is to deliver such coordination. We describethe organization of TFOP resources, summarize thescale of the community effort, and highlight oppor-tunities for involvement in TFOP.

309.07 — A Planet Hunters study of the long-periodexoplanet population around Kepler M-dwarfs

Emily Safron1; Tabetha Boyajian11 Physics & Astronomy, Louisiana State University (North

Ridgeville, Ohio, United States)

M-dwarfs are the most common stars in our galaxy,and efforts to characterize the population of exo-planets orbiting them are numerous and ongoing.Among some of the most difficult of these exoplanetsto study are those with long periods (P > ∼600 days).Surely many single and double transit events causedby such exoplanets reside in the four years of Ke-pler lightcurves, undiscovered by traditional searchpipelines, as several groups have recently come tofind using deep learning techniques. In this work, weutilize a different, yet unexplored resource to combthe Kepler lightcurves for these signals — the brainsof hundreds of thousands of volunteer citizen sci-entists, through the interface of the Planet Hunterswebsite. By crowd sourcing the identification ofpromising transit-like features and designing carefulvetting techniques, both automated and subjective,we aim to broaden the population of known long-period M-dwarf exoplanets and further constrain thestatistical properties of this under-represented pop-ulation.

309.08 — The Sub-Saturn Mass-Radius Relation-ship from K2 and a NASA-Keck Key Project

Ryan Rubenzahl1; Andrew Howard1; Evan Sinukoff21 Cahill Center for Astronomy & Astrophysics, California Institute

of Technology (Pasadena, California, United States)2 Institute for Astronomy, University of Hawaii (Honolulu, Hawaii,

United States)

We now know of thousands of exoplanets with sizesbetween that of the Earth and Saturn, thanks to theKepler mission, which mapped out a distributionof planet radii and orbital periods. However, mostof these planets do not have mass measurements,which are fundamental to understanding their com-positions and formation histories. An accurate pre-dictor of planet mass as a function of radius and

host star properties will be valuable for estimatingpopulation-wide distributions from transit surveys.We have constructed a catalog of 139 exoplanets withprecisely measured masses and radii between 1 and8 REarth. This sample represents the largest and mostprecise catalog of small planet masses and radii, aswell as orbital and host star properties. We explorea number of empirically and physically motivatedmodels to quantify the small-planet mass-radius re-lationship. We use a hierarchical Bayesian model-ing approach to characterize the intrinsic model scat-ter and parameter uncertainties. The hierarchicalmodel naturally incorporates population-level infer-ences, allowing distinct sub-populations such as thesuper-Earths and sub-Neptunes to be distinctly char-acterized. We compare various models and explorethe multi-dimensional dependence of planet mass onplanet radius, orbital period, and various host-starproperties.

309.09 — Exoplanet Population Synthesis in the Eraof Large Exoplanets Surveys

Gijs Dirk Mulders1; Christoph Mordasini3; IlariaPascucci2; Fred Ciesla1; Alexandre emsenhuber2,3;Dániel Apai2

1 University of Chicago (Chicago, Illinois, United States)2 University of Arizona (Tucson, Arizona, United States)3 University of Bern (Bern, Switzerland)

The Bern planet population synthesis models (e.g.Mordasini 2018) represent a decade long effort toinvestigate the integrated effects of the processes atwork during planet formation and make predictionsfor exoplanet populations, planetary system archi-tectures, and planet compositions. Over the last fewyears new physical mechanisms have been incorpo-rated and adjusted to reflect the lessons learned fromKepler, in particular on atmospheric loss shaping theplanet radius distribution and N-body interactionssetting the architectures of planetary systems. Bycomparing the synthetic planet populations to ob-served exoplanet systems we can constrain planetformation mechanisms to inform predictions of plan-etary compositions.

This poster shows simulated planet populationsfrom the latest version of the Bern planet populationsynthesis model. I will make detailed, quantitativecomparisons between what the synthetic populationwould look like compared to exoplanet survey datausing the Exoplanet Population Observation Simu-lator (EPOS), which takes into account the uniqueobservation biases in both transit and radial veloc-ity surveys. While the synthetic populations repro-duce many key features seen in the known popula-

47

tions of radial velocity giant planets and systems ofclose-in super-earths observed with Kepler, we alsosee key differences in other diagnostics. These dif-ferences inform the setup for the Next Generation ofPlanet Population Synthesis (NGPPS) models.

309.10 — Are Planets with Close Siblings System-atically Less Dense Than Those Without?

Angie Wolfgang1; Eric Ford1; Daniel Jontof-Hutter21 Astronomy & Astrophysics, Pennsylvania State University (Uni-

versity Park, Pennsylvania, United States)2 University of the Pacific (Stockton, California, United States)

Compositions of individual exoplanets inform thephysical nature and possible habitability of planetsorbiting other stars; population trends with compo-sition offer a valuable window into exoplanet for-mation and evolution. Mass and radius measure-ments provide the constraints on which these trendsare based, and the relationship between them de-scribe the empirical, model-independent distribu-tion of exoplanet compositions as a function of sizeor mass. In the era of TESS and Gaia, individualplanetary masses and radii can be measured withunprecedented precision, yet selection effects anddetection biases continue to confuse efforts to ex-tract insight on a population-wide level. With up-dated mass constraints from transit timing variationsobserved by Kepler, I will present new results onthe completeness-corrected mass-radius relationshipof planets in tightly spaced multi-planet systems.In comparing these results to the typical densitiesof single-transiting planets, I will discuss which re-gions of parameter space would particularly bene-fit from both additional observations and a differentapproach to conducting radial velocity follow-up oftransiting planets. Finally, I will discuss the implica-tions of these results for the formation and evolutionof low-mass planets.

309.11 — Refining the occurrence rate of inner com-panions to hot Jupiters using TESS full-frame im-age data

Lizhou Sha1; Chelsea X. Huang1; Andrew Vanderburg21 Kavli Institute of Astrophysics and Space Research, Massachusetts

Institute of Technology (Cambridge, Massachusetts, United States)2 University of Texas at Austin (Austin, Texas, United States)

To date, WASP-47 e and Kepler 730 c remain the onlyknown inner companions to hot Jupiters (Becker etal. 2015, Zhu et al. 2018). This apparent scarcityis broadly in line with the hypothesis that most

hot Jupiters form beyond the ice line and move in-wards via high-eccentricity migration (HEM). How-ever, statistical evidence based on the dearth ofsuper-eccentric hot Jupiters in Kepler data suggeststhat HEM does not explain the formation of all hotJupiters, leaving the possibility for some hot Jupitersto have close-in planets (Dawson et al. 2015). TESSbrings a golden opportunity to study the occur-rence rate of hot Jupiter companions in terms ofboth quality and quantity: not only does TESS haveenough photometric precision to detect super-earthsaround bright stars, but its full-frame images willalso provide more than double the number of high-precision light curves for hot Jupiters compared toKepler and K2 combined. Using the first eight sec-tors of TESS full-frame images, we generate lightcurves for ∼80 confirmed and candidate hot Jupitersbrighter than the 11th TESS magnitude with the MITQuick Look Pipeline. Having carefully removed theknown planet signal, we perform a uniform BLSsearch for companions in order to derive a constrainton the occurrence rate of such planets. Combiningthis constraint with the ones derived from Kepler andK2 hot Jupiters, we arrive at a refined upper limiton the occurrence rate of inner companions to hotJupiters. We then discuss the implications of thisnewly calculated occurrence rate and how it informscurrent discussions on the formation theories of hotJupiters.

309.12 — Phase Curve Analysis of a Brown Dwarfand its Stellar Host: Increasing the Aridity of theBrown-Dwarf Desert with TESS

Tiffany Jansen1; David Kipping21 Astronomy, Columbia University (New York, New York, United

States)2 Columbia University (New York, New York, United States)

Observations have shown there is a paucity ofbrown-dwarfs found orbiting within 5 AU of solar-type stars, otherwise known as the “brown-dwarfdesert”. In this study we revisit the discovery of abrown-dwarf desert member to examine the likeli-hood that the radial velocity signals which lead tothis discovery were indeed due to the existence of ashort period brown-dwarf, or if they were instead theresult of stellar variability. To do this, we use time se-ries photometry of the host star to measure the am-plitude of coherent photometric variations observedby the Transiting Exoplanet Survey Satellite (TESS)and compare these measurements to the range oftheoretical ellipsoidal variation and Doppler beam-ing amplitudes induced by the mass of the specula-tive brown-dwarf companion. Our preliminary re-

48

sults show that the observed photometric variabilityamplitudes are much too large to be explained by thepresence of a brown-dwarf, thus strengthening thevalidity of the brown-dwarf desert.

309.13 — M Dwarf Planet Occurrence Rates De-pend on Metallicity at all Planet Radii

Cicero Xinyu Lu1; Sihao Cheng1; Kevin C. Schlaufman11 Physics and Astronomy, Johns Hopkins University (Baltimore,

Maryland, United States)

There must be a threshold for the amount of solidmaterial in a protoplanetary disk below which evensmall planets cannot form. Since the amount ofplanet-making material in a protoplanetray disk isproportional to host star metallicity and mass, thebest way to search for this effect is to calculate theplanet occurrence for metal-poor, low-mass stars.For late K and early M dwarfs, we have calculatedthe effect of stellar metallicity on planet occurrence inthe DR25 Kepler KOI list. For planets in the range 2.5REarth < Rp < 5 REarth, we find that a 0.5 dex decreasein [M/H] decreases planet occurence by an order ofmagnitude. For planets in the range 0.5 REarth < Rp< 2.5 REarth, we find that a similar decrease in metal-licity decreases planet occurrence by a factor of two.This result demonstrates the importance of metallic-ity in the calculation of small planet occurrence ratesand therefore for TESS yield calculations. We predictthat for early M dwarfs at [M/H] = -1, even super-Earth planets should be rare.

310 — Multiple-Planet Systems,Poster Session310.01 — TROY: surveying a new type of extremeplanetary systems

Jorge Lillo-Box1,21 Chile, European Southern Observatory (ESO) (Madrid, Madrid,

Spain)2 Astrophysics, Center for Astrobiology (CAB) (Madrid, Madrid,

Spain)

Theoretical works of planetary system formation andtheir early evolution predict the existence of co-orbital planets (two planets sharing the same orbitalperiod) with occurrence rates up to 30% in multi-planetary systems. Trapped either in the Lagrangianpoints of more massive planets or in other type of1:1 mean motion resonances, these bodies keep thedynamical and chemical properties of the formationof the planetary system and are thus fossils of these

processes. Looking for these exotic configurationsrepresents a new viewpoint to study planet forma-tion and migration mechanisms. The TROY project(Lillo-Box et al., 2018a,b) is a multi-technique effortin the hunt for these celestial fossils. In this talkI will show the latest results of the project regard-ing the valuable data provided by TESS and by ourown new, dedicated, re-analysis of the Kepler data.In particular, I will discuss the results presented inLeleu et al. (2019) on the analysis of similar-periodplanet candidates detected by both TESS and Kepler,with a special focus on the co-orbital candidate TOI-178. Three planets were detected in this system, withthe external two components practically sharing thesame orbital period. I will also show the latest resultson other TESS candidates and the Bayesian analy-sis of the full set of Kepler light curves in a dedi-cated search for these bodies accounting for differentco-orbital configurations. The results of this analy-sis provides, for the first time, statistically significantand observational measurements of the occurrencerate of a new type of extreme planetary systems: twoplanets sharing the exact same orbit.

310.02 — Modeling light curves of the multi-transiting system Kepler-20 using Blender

Holger Matthias Müller1; Panagiotis Ioannidis1; JürgenH. M. M. Schmitt1

1 Hamburg Observatory (Hamburg, Germany)

Transiting multi-planet systems can hold additionalinformation about their orbital configurations. Thesesystems can show multi-transits where at least twoplanets are eclipsing the star at the same time. If theorbital alignments are favorable, these systems alsoprovide planet-planet occultations (PPOs). The pres-ence or absence of these events gives constraints onthe alignment of the orbits in question. We presenta comprehensive study of the multi-transiting plan-etary system Kepler-20. The solar-like host star is or-bited by six planets, while five of them perform tran-sits. Their small sizes range from roughly 1 to 3 Earthradii, clearly detected by Kepler. In our approach wesynthesize a grid of multi-transit light curves usingthe orbital parameters of planets b and c, varyingthe angle α between their orbits, while keeping theirtransit impact parameters constant. For that purposewe are the first to utilize the publically available 3Danimation software Blender. This allows us to usearbitrary surface brightness distributions of the starlike model limb darkening or spots. The resultinglight curves show PPOs depending on the angle α,which are then compared to the Kepler data. In this

49

way we are able to statistically exclude orbital ge-ometries, and we can identify which are the mostfavorable. Besides the Rossiter-McLaughlin effectwhere spectral data is needed, this method is able toacquire orbital alignment information from the opti-cal light curve alone.

310.03 — Towards Photodynamical Modeling of AllKepler Multi-Transiting Systems

Darin Ragozzine1; Sean M. Mills3; Vatsala Sharma1;Daniel Clark Fabrycky2; Rochelle J. Steele1

1 Physics and Astronomy, Brigham Young University (Provo, Utah,United States)

2 Astronomy & Astrophysics, The University of Chicago (Chicago,Illinois, United States)

3 Caltech (Pasadena, California, United States)

Kepler’s 700 systems with multiple transiting plan-ets are incredibly valuable. These systems have pro-vided large numbers of precise exoplanetary densi-ties by characterizing planet-planet dynamical inter-actions (Transit Timing Variations) caused by transit-ing planets. We are pursuing the most detailed anal-ysis of these systems to date, improving on previousstudies in several ways. We will fit all 700 multi-transiting systems with our new PhotoDynamicalMulti-planet Model (PhoDyMM) which directly fitsn-body integrations to the light curve (skipping thestep of measuring individual transit times) in orderto infer the most precise orbital and physical prop-erties for all known planets. Our study will provideBayesian posterior distributions for these propertiesin two sets: a homogeneous analysis of all systems toprovide support for later meta-analyses and a best-case analysis of all systems that includes improvedstellar parameters, and hitherto underutilized shortcadence data. We will present a detailed explanationof our methods from detrending to DEMCMC andpreliminary analyses demonstrating our technique.

310.04 — Chaos in Three-Planet Systems

Jeremy Rath1; Yoram Lithwick2; Sam Hadden31 Physics and Astronomy, Northwestern University (Evanston,

Illinois, United States)2 Northwestern University (Evanston, Illinois, United States)3 Harvard-Smithsonian CfA (Cambridge, Massachusetts, United

States)

We describe a simple analytic theory for chaos andits onset in eccentric, three-planet systems. Previouswork on this topic has relied almost entirely on nu-merical simulations. Here we show that by properly

accounting for the overlapping of mean motion res-onances and resonant combinations of these meanmotion resonances, one can predict to surprisinglygood accuracy the boundary between chaos and sta-bility for a three-planet system.

311 — Earths and Super-Earths,Poster Session311.01 — A water budget dichotomy of rocky pro-toplanets from 26Al-heating

Tim Lichtenberg1,4; Gregor J. Golabek3; Remo Burn5;Michael Meyer2; Yann Alibert5,6; Taras Gerya4;Christoph Mordasini5,6

1 Atmospheric, Oceanic and Planetary Physics, University of Ox-ford (Oxford, Oxfordshire, United Kingdom)

2 Department of Astronomy, The University of Michigan (AnnArbor, Michigan, United States)

3 Bayerisches Geoinstitut, University of Bayreuth (Bayreuth, Ger-many)

4 Institute of Geophysics, ETH Zurich (Zurich, Switzerland)5 Physikalisches Institut, University of Bern (Bern, Switzerland)6 Center for Space and Habitability, University of Bern (Bern,

Switzerland)

In contrast to the water-poor planets of the innersolar system, stochasticity during planetary forma-tion and order of magnitude deviations in exoplanetvolatile contents suggest that rocky worlds engulfedin thick volatile ice layers are the dominant familyof terrestrial analogues among the extrasolar planetpopulation. However, the distribution of compo-sitionally Earth-like planets remains insufficientlyconstrained, and it is not clear whether the solar sys-tem is a statistical outlier or can be explained by moregeneral planetary formation processes. Here we em-ploy numerical models of planet formation, evolu-tion, and interior structure, to show that a planet’sbulk water fraction and radius are anti-correlatedwith initial 26Al levels in the planetesimal-basedaccretion framework. The heat generated by thisshort-lived radionuclide rapidly dehydrates plan-etesimals prior to accretion onto larger protoplanetsand yields a system-wide correlation of planet bulkabundances, which, for instance, can explain the lackof a clear orbital trend in the water budgets of theTRAPPIST-1 planets. Qualitatively, our models sug-gest two main scenarios of planetary systems’ forma-tion: high-26Al systems, like our solar system, formsmall, water-depleted planets, whereas those devoidof 26Al predominantly form ocean worlds, where themean planet radii between both scenarios deviate byup to about 10%.

50

311.02 — The Snowball Bifurcation on TidallyLocked Planets

Dorian Abbot11 The University of Chicago (Chicago, Illinois, United States)

The ice-albedo feedback on rapidly-rotating ter-restrial planets in the habitable zone can lead toabrupt transitions (bifurcations) between a warmand a snowball (ice-covered) state, bistability be-tween these states, and hysteresis in planetary cli-mate. This is important for planetary habitability be-cause snowball events may trigger rises in the com-plexity of life, but could also endanger complex lifethat already exists. This raises the question of howthe Snowball Bifurcation might work on tidally influ-enced planets in the habitable zone orbiting M andK dwarf stars. We investigate this question using an-alytical theory, an ocean-atmosphere global climatemodel, and an intermediate complexity global cli-mate model coupled to an active carbon cycle. Wefind that planets locked in a 1:1 synchronous rota-tion state are likely to experience a smooth transitionto global glaciation rather than a bifurcation. This isimportant because it means that tidally locked plan-ets with an active silicate-weathering feedback loopshould not tend to stay in the snowball state (theywould just pop out of it if they ever entered it be-cause weathering would go to near zero while CO2outgassing would continue).

311.03 — Terrestrial Planets from CARMENES: Ex-tremely Close, Extremely Interesting

Andreas Quirrenbach11 Landessternwarte, U Heidelberg (Heidelberg, Germany)

The CARMENES consortium is conducting a sur-vey of more than 300 nearby M dwarfs (average dis-tance only 13pc), with the goal of finding terrestrialplanets in their habitable zones. To make this sur-vey possible, we have built a pair of spectrographsoptimized for measuring precise radial velocities ofcool stars; together they cover 520 to 1710 nm withresolution > 80,000. The instrument has been op-erational since January 2016 at the 3.5m telescopeon Calar Alto, Spain. So far more than 13,000 spec-tra have been taken, covering all spectral subtypesfrom M0V to M7V. 22 new planets have been dis-covered by CARMENES, almost all of them withmasses in the Super-Earth and Earth-like range. Inaddition, a number of previously known planetscould be confirmed. Among the CARMENES dis-coveries are a cold Super-Earth orbiting Barnard’sstar, and two planets with 1.05 and 1.1 Earth masses

in the habitable zone of their host star. At theother end of the mass range, Jovian planets chal-lenge planet formation models based on the peb-ble accretion paradigm. Because of their proximity,the CARMENES planets offer unique follow-up op-portunities with space missions and extremely largeground-based telescopes. Some CARMENES plan-ets are easily detectable astrometrically by Gaia, andGaia should also reveal the outer planets in those sys-tems where CARMENES detects the inner ones. If“scaled-down” twins of the Solar System (with an Mdwarf host, one or two habitable-zone Earths, andicy or gaseous planets at a few AU) are common,they should be found in the combined CARMENESand Gaia data. Among the CARMENES planetsare also some of the best targets for follow-up spec-troscopy, which can characterize their atmospheres.Since the stars in our survey are typically only afew pc away, their habitable zones can be resolvedwith adaptive optics on 30m class telescopes, mak-ing them accessible to high-resolution spectrographscoupled to coronographs. In addition, a subprojectof the CARMENES survey that is performing follow-up observations of TESS planets has already discov-ered the transiting Earth-like planet best suited forcharacterization with JWST.

311.04 — A geophysical model for 55 Cancri e

Laura Schaefer11 Geological Sciences, Stanford University (Stanford, California,

United States)

55 Cancri e is one of the closest known super-Earths,orbiting a bright host star that makes it a tantaliz-ing prospect for observations. Measurements of theplanet’s phase curve and secondary eclipse suggesta possible lava world with >1000 K temperature con-trast from day to night. Observations have failed todetect water vapor or signs of escaping hydrogen,consistent with a planet lacking most of the cosmi-cally abundant volatiles. However, the mass and ra-dius measurements likely require a relatively signif-icant atmosphere. Observations have found hints ofNa and Ca+ in an extended exosphere, which maybe consistent with the lava world interpretation, butother measurements suggest the presence of HCN,which would likely require a much more volatile-rich atmosphere. In this presentation, we will re-view the models that have been proposed for 55 Cnce and which observations those models are consis-tent with. We will propose two preliminary mod-els, depending on either the presence or absence ofHCN, that are consistent with both mass-radius con-

51

straints, as well as atmospheric composition and heatredistribution properties.

311.05 — Multi-season optical modulation phasedwith the orbit of the super-Earth 55 Cnce

Sophia Sulis11 Space Research Institute, Austrian Academy of Sciences (Graz,

Austria)

The 55 Cancri system contains five known planets,with only one transiting the star: 55 Cnc e. Witha bright host star and an orbital period of only ∼17hours, this Super-Earth is an ideal target for charac-terization. The planet’s nature still remains largelyunknown, but plausible scenarios include strong vol-canism and/or a thick atmosphere. In 2011, Winnet al. (2011) detected a quasi-sinusoidal variation atthe orbital period of the planet in optical-light withthe MOST satellite. The amplitude of this modula-tion was too large to be explained by light reflected oremitted by the planet. From 2011 to 2015, we contin-ued to monitor this target for several weeks per years,totalizing around 140 transits events. Through inde-pendent analyses, we confirm the quasi-sinusoidalvariation observed at the orbital period of the planetand we detect this modulation throughout the sub-sequent years. Intriguingly, the amplitude and thephase of the maximum light are seen to vary fromyear to year. While we can only speculate about theexact nature of this optical modulation, we arguethat additional observations with TESS and CHEOPSwill be extremely valuable for our understanding ofthis mysterious planet.

311.06 — Extreme Cassini states of planets with aliquid core

Gwenaël Boué11 IMCCE (Paris, France)

Most terrestrial exoplanets detected so far have beenfound in compact multiplanetary systems where in-terplanetary interactions are strong. The perturbedtrajectories have eccentricity, mutual inclination andare subject to precession motions. Although eccen-tricities and inclinations in these systems are lim-ited for stability reason, they are sufficiently largeto affect the long term evolution of the spin axes(e.g., Saillenfest, Laskar & Boué 2019). This can havestrong implications on the habitability of these plan-ets. For instance, high-obliquity planets undergo se-vere seasonal variations (Spiegel et al. 2009) but theirhability is also enhanced towards the outer limit ofthe Habitable zone (Colose et al. 2019).

In this paper I revisit the equilibrium configura-tions — known as Cassini states — of the spin-axisof planets subject to orbital precession. It is nowwell known that a planet can have up to three stablefixed orientation and an unstable one. This has beenshown for axisymmetric bodies by Colombo (1966)and generalised for triaxial planets by Peale (1969).But in both cases, the planet is assumed to rotate as arigid body. Here I consider the effect of a liquid coreon the rotation axis of a planet mantle. In this situa-tion the number of fixed points is much larger thanin the rigid case. I will show that extreme equilib-rium obliquities do exist even in systems where theclassical Cassini states are marginally inclined.

311.07 — An asynchronous rotation scenario for 55Cancri e

Alexis Brandeker11 Astronomy, Stockholm University (Stockholm, Sweden)

The surface temperature distribution of the transit-ing hot super-Earth planet 55 Cancri e has recentlybeen measured by Spitzer. The hottest point onthe surface is surprisingly found to be offset fromthe substellar point, necessitating efficient horizon-tal heat transport. Suggested mechanisms includelava streams or a relatively massive atmosphere withstrong lateral winds. Here we propose an alterna-tive scenario where the planet is rotating at an asyn-chronous rate. In particular, we study the case wherethe planet is in a 3:2 spin-orbit resonance, giving asynodic rotation period of 35.4 h. From a modelof the planetary surface heat distribution we findthat the thermal inertia resulting from rock evapo-ration with subsequent condensation could be suffi-cient to explain the observed shift of the hot spot.The3:2 asynchronous rotation also serves to naturallyexplain strong short-term variability, as the visibleplanetary surface will rotate by 180 degrees betweensubsequent eclipses.

311.08 — Two Terrestrial Planet Families With Dif-ferent Origins

Mark Swain1; Raissas Estrela1; Christophe Sotin1; GaelRoudier1; Robert T. Zellem1

1 JPL (Pasadena, California, United States)

We propose an oral presentation to present worksubmitted for publication that highlights small plan-ets spanning the planet occurrence rate deficit re-ported by Fulton et al. 2017. We analyze a sample ofsmall planets (R < 3.5 Rearth) in a three-dimensionalspace, incorporating radius, density, and insolation,

52

and identify physically motivated trends. The po-tentially important role of stellar irradiation in enve-lope removal for planets with diameters of < 2 Rearthhas been inferred both through theoretical work andthe observed bimodal distribution of small planet oc-currence as a function of radius. We find, terres-trial planets divide into two distinct families basedon insolation. The lower insolation family mergeswith terrestrial planets and small bodies in the so-lar system and is thus Earth-like. The higher inso-lation terrestrial planet family forms a bulk-densitycontinuum with the sub-Neptunes, and is thus likelyto be composed of remnant cores produced by pho-toevaporation. However, these potential remnantcores also have evidence for collisional processing,inferred from the positive density-mass relation forhigher insolation terrestrial planets. The implicationis that the average mass for the high insolation fam-ily of terrestrial planets, 4.8±1.8 Mearth, may repre-sent an upper bound on the typical mass needed toproduce the onset of the rapid gas accretion phaseassociated with envelope assembly during the planetformation process.

311.09 — Achieving Sub-Millimagnitude Precisionfrom the Ground: the Capabilities of ARCTIC andthe LHS 1140 System

Jessica Elizabeth Roberts1; Carlos E. Cruz-Arce1; Za-chory Berta- Berta-Thompson1

1 Astrophysics and Planetary Sciences, University of ColoradoBoulder (Lafayette, Colorado, United States)

As TESS observes most stars for only 28 days, manyTESS planetary candidates will require future obser-vations by other facilities in order to be properly vet-ted. Ground-based observations of these candidatescan reject false positives, update mid-transit times,refine planetary parameters, and provide long-termmonitoring of interesting systems. Ground-basedtelescopes achieve these science goals in part due totheir larger size compared to TESS’s 0.1m diameterlens. However, most observations from the groundstruggle to achieve precisions better than 1 millimag-nitude. The new CCD imager ARCTIC, installed onthe 3.5m Apache Point Observatory Telescope, at-tains extreme precision by combining its large col-lecting area (1000× larger than TESS) with a diffuserthat spreads the stellar PSF into a stable top-hat. Wetest the performance of this instrument by observingmultiple transits of LHS 1140b and LHS 1140c. LHS1140 is a nearby M-dwarf orbited by two rocky, nearEarth-sized planets, including one in the habitablezone. This system therefore presents a unique op-portunity to study two rocky planets in very differ-

ent temperature regimes around the same star. Ourobservations double the number of published LHS1140b and 1140c transits, and we use these to up-date the ephemeris and better constrain the plane-tary parameters. We find ARCTIC achieves a RMSof 150ppm on LHS 1140 for data binned to 20 minutetimescales. Based on our success with the LHS 1140system, we predict that ARCTIC will prove a use-ful instrument for future TESS follow-up on bothsmaller and fainter planet candidates as TESS movesinto the northern hemisphere this year.

311.10 — A Planet of Ice and Fire — Barnard Star b: Life beyond the Snowline?

Edward Guinan1; Scott G. Engle1; Ignasi Ribas21 Astrophysics & Planetary Science, Villanova University (Vil-

lanova, Pennsylvania, United States)2 Institut d’Estudis Espacials de Catalunya (IEEC) (Barcelona,

Cataylania, Spain)

Barnard’s Star (M3.5V; 5.98 ly) was recently discov-ered to host a super-Earth with a minimum massof 3.23ME, Porb = 233-d, and a = 0.40 AU (Ribas etal. 2018). Barnard b is at nearly the same distanceas Mercury (0.39 AU) from the Sun. However, theM-star is faint (L/Lsun = 0.0033) so that the planetreceives only 2% of the solar radiation received byEarth. This corresponds to the equivalent amount ofradiation received at roughly ∼7 AU from the Sun.Barnard b is cold (Teq ∼105 ±5 K; -168°C) and orbitswell beyond the M-star’s liquid water habitable zone(HZ), near the snowline.

Barnard b is a probable terrestrial (rocky) planet,and if water is present, it would be ice-covered andinhospitable to most life. However, as pointed outby Guinan et al. 2018, all hope for life on Barnardb may not be lost. Ehrenreich et al. (2006) studiedthe possibility that geothermal energy from a coldsuper-Earth planet (similar to Barnard b) could besufficient to permit liquid water under its icy sur-face at least while young. Although the cold icymoons Europa and Enceladus have subsurface wa-ter, they are heated primarily from tidal energy. Acloser analog for Barnard b may be be here on Earth.The sub-glacial Antarctic lakes (e.g. Lake Vostok)contain large inventories of liquid water (and maybelife?) and are heated by geothermal energy from theEarth’s interior.

We discuss possibilities of liquid water (and poten-tial life) on Barnard b and on other cold rocky plan-ets.The amount of geothermal energy depends onseveral factors (most of which are not well known).These include mass, radius, age and composition &internal structure of the planet. Also very important

53

are the concentrations of radiogenic elements suchas 238-U, 235-U, 232-Th & 40-K and the primordialheat from the planet’s formation. Because of its oldage (∼9 Gyr) and depletion of some radiogenic ele-ments, the best chance for liquid water (and nichesfor life) on Barnard b would be in subsurface lakes.We report on the initial results and its applications toother cold super-Earth planets.

This research is supported by grants from NASAfor HST, Chandra and XMM-Newton.

312 — Neptunes and Mini-Neptunes, Poster Session312.01 — The Orbital Damping Effect on Neptu-nian Trojans

Yuehua Ma11 Purple Mountain Observatory,Chinese Academy of Sciences

(Nanjing, China)

Using test particle simulations, the orbital elementdistributions of Neptune Trojans affected by theplanetary migration and the orbital damping ofUranus and Neptune was investigated. We examinethe stability of primordial Neptune Trojans, objectsthat were initially Trojans with Neptune prior to mi-gration and Trans-Neptunian objects captured intoresonance with Neptune and becoming NeptuneTrojans during planet migration. We find that mostprimordial Neptune Trojans were unstable and lost ifeccentricity and inclination damping took place dur-ing planetary migration. With damping, secular res-onances with Neptune can increase a low eccentricityand inclination population of Trans-Neptunian ob-jects increasing the probability that they are capturedinto 1:1 resonance with Neptune, becoming high in-clination Neptune Trojans. These suggest that theresonant trapping scenario is a promising and moreeffective mechanism to explain the origin of NeptuneTrojans if Uranus and Neptune had orbital dampingduring planetary migration.

312.02 — The origin of the obliquity of Uranus dueto the giant impact

Kenji Kurosaki1; Shu-ichiro Inutsuka11 Physics, Nagoya University (Nagoya, Aichi, Japan)

Our solar system has two ice giants, Uranus andNeptune. Those planets have similar mass and ra-dius but different obliquity and the intrinsic lumi-nosity. Differences between Uranus and Neptune

suggest origins of those planets. Since the obliq-uity difference is caused by impact events, we in-vestigate the giant impact on the ice giant to repro-duce the obliquity of the planet. In this study, weuse the Godunov-type Smoothed Particle Hydrody-namic simulation to calculate the giant impact sim-ulation on the rotting planet and calculate the obliq-uity of the ice giant after the impact directly. We findthat the obliquity of the present Uranus is able repro-duced by an Earth mass impactor. Thus, the impactevent disappears the rotation of the pre-impact tar-get and give other direction of planetary spin in toreproduce the large obliquity. Moreover, we formu-late the simple approximation for the obliquity vari-ation due to the impact and calculate the probabilitydistribution of the planetary obliquity after the im-pact. We also find that the slow rotating planet havean advantage to reproduce the large obliquity. If weconsider the two impacts, the planetary obliquity be-come larger than single impact. We conclude thatlarger than Earth mass impactor collided on slow-rotating proto-Uranus. Our study will be useful toconsider the probabillity of the large obliquity of ex-oplanets.

313 — Hot Jupiters and Ultra-ShortPeriods, Poster Session313.01 — Transmission Spectroscopy of WASP-79bfrom 0.6 to 4.5 μm with Predictions for JWST Ob-servations

Kristin Showalter Sotzen1; Kevin Stevenson2; HannahWakeford2; Joseph Filippazzo2; Jonathan Fraine3; NikoleLewis4; Sarah Hörst1,2; David Sing1; Mercedes Lopez-Morales5; Brian Kilpatrick2; Rahul Jayaraman6

1 Earth and Planetary Sciences, Johns Hopkins University (Hamp-stead, Maryland, United States)

2 Space Telescope Science Institute (Baltimore, Maryland, UnitedStates)

3 Center for Extra-solar Planetary Systems, Space Science Institute(Boulder, Colorado, United States)

4 Department of Astronomy, Cornell University (Ithaca, New York,United States)

5 Harvard-Smithsonian Center for Astrophysics (Cambridge, Mas-sachusetts, United States)

6 Department of Physics, Brown University (Providence, RhodeIsland, United States)

As part of the PanCET program, we have conducteda spectroscopic study of WASP-79b, a Jupiter-size ex-oplanet orbiting an F-type star in Eridanus with a pe-riod of 3.66 days. Building on the original WASP andTRAPPIST photometry of Smalley et al (2012), we

54

have performed reduction and light curve fitting onHST/WFC3 (1.125 – 1.650 μm) and Magellan/LDSS-3C (0.6 – 1 μm) observations for WASP-79b, and wehave conducted photometric extraction on Spitzerobservations (3.6 and 4.5 μm). Additionally, we havevalidated our light curve against the transit depthsestimated from the Sector 4 and Sector 5 TESS obser-vations of this exoplanet. We will present our lightcurve results, which show indications of a water fea-ture at 1.4 μm. Finally, we will discuss the resultsof an atmospheric retrieval analysis and simulatedJWST data based on best-fit retrieval models for thesedata. The suggested water feature makes WASP-79ba target of interest for the approved JWST Director’sDiscretionary Early Release Science (DD ERS) pro-gram, with ERS observations planned to be the firstto execute in Cycle 1. Transiting exoplanets were re-cently approved for 78.1 hours of data collection, andwith the delay in the JWST launch, WASP-79b is nowa target for the Panchromatic Transmission program.This program will observe WASP-79b for 42 hoursin 4 different instrument modes over 0.8-5.0 μm, atwhich time WASP-79b will be the best-characterizedexoplanet to date.

313.02 — Full-orbit phase curvers of known transit-ing systems with TESS

Ian Wong1; Avi Shporer2; Björn Benneke31 EAPS, MIT (Cambridge, Massachusetts, United States)2 Kavli Institute for Astrophysics and Space Research, MIT (Cam-

bridge, Massachusetts, United States)3 iREx, Université de Montréal (Montréal, Quebec, Canada)

The TESS Mission promises to be a watershed mo-ment for exoplanet science. In addition to thehuge yield of new planet candidates, TESS will pro-vide many interesting avenues of study for knownplanets. Following the legacy of similar work car-ried out with Kepler data, we seek to take advan-tage of the continuous, long-baseline photometryprovided by the TESS mission to study the full-orbit light curves of transiting systems. From thesedatasets, we can place constraints on such physicalquantities as: (1) the planet’s dayside temperatureand Bond albedo, through measurement of the sec-ondary eclipse depth in the TESS band; (2) the ef-ficiency of day-night recirculation of incident stel-lar irradiation, via the amplitude and phase shiftof the planet’s atmospheric brightness modulationacross the orbit; and (3) the response of the hoststar to the gravitational interaction with the orbit-ing planet, which are expressed in the phase curvesignal via Doppler boosting of the star’s light and

tidal distortion of the stellar surface. As an exem-plary demonstration case, we recently published ouranalysis of the phase curve of WASP-18b from TESSSectors 2 and 3 (Shporer, Wong, et al. 2019), whichreveals a strong secondary eclipse signal and highsignal-to-noise phase curve modulations attributableto atmospheric brightness variation, ellipsoidal dis-tortion, and Doppler boosting. Since then, we haveexpanded our efforts into a unified and systematiclight curve analysis of all known transiting systemscontained in TESS fields with predicted detectablesecondary eclipses and/or phase curve signals. Iwill present the latest results from this work and dis-cuss them in the context of emergent trends in exo-planet atmospheric dynamics and comparisons withthe predictions of theoretical phase curve modeling.

313.03 — A model for Hot Jupiter with bottom ther-mal perturbation

Yuchen Lian11 Atmospheric and oceanic sciences, School of physics, Peking Uni-

versity (Beijing, Beijing, China)

Hot Jupiters are one of the few types of plan-ets that can currently be observationally charac-terized. Under strong external radiation forcing,Hot Jupiters show a variety of atmosphere circu-lation pattern. Due to spectrum exhibits informa-tion and our knowledge in atmosphere dynamic ofthe Hot Jupiter, some researchers create the atmo-sphere models, for understanding the observationsand the physics of these planets generally, but almostall models ignore — or treat in a very simplified fash-ion — the interaction of the interior convection zonewith the deep, thick stratified atmosphere. Here,we take the small-scale thermal perturbation into ac-count, and import a spatially and temporally noise,which is horizonally isotropic, as forcing parameterinto 3D primitive equation model for influencing thetemperature at convective-radiation boundary layer.In some cases, the random thermal perturbation givelimited impact on pattern — a significant tempera-ture increase near the hot spot, while the others doesnot have much influence on the top layer pattern,which shows hot spot shift with super-rotation onequator at top layers, when star radiation becomesstronger.We draw a conclusion that the flow does notvary in time until the bottom perturbation is overthe bottom heating rate(0.001K/sec). We assume themodel relax the radiation forcing by Newtonian cool-ing scheme to equivalent temperature in various ra-diative time scale on different levels.Perhaps we pre-dict more time variability if atmospheres are strongly

55

enough forced by convection, and we hope get newinformation from spectrum to prove the result.

313.04 — High resolution transmission spectrum ofultra-hot Jupiters

Fei Yan11 Institute for Astrophysics, University of Göttingen (Göttingen,

Germany)

Ultra-hot Jupiter is a new class of exoplanets emerg-ing in the recent years. Their extremely hot temper-atures cause thermal dissociation of molecules andeven ionisation of atoms. We have detected an ex-tended hot hydrogen atmosphere around KELT-9b— the hottest exoplanet discovered so far. The detec-tion was achieved by measuring the atomic hydrogenabsorption during transit with the Balmer Hα lineusing the CARMENES spectrograph. The obtainedHα transmission spectrum has a strong extra absorp-tion of 1.15% at the line centre. The observation im-plies that the effective radius at the line centre is ∼1.64 times the size of the planetary radius, indicat-ing that the planet has a largely extended hydrogenenvelope close to the size of the Roche lobe and isprobably undergoing dramatic atmosphere escape.

313.05 — Insights into Terrestrial Planet Com-positions and Geophysics from Observations ofMagma Worlds

Andrew Ridden-Harper11; Ignas Snellen1; ChristophKeller1; Paul Mollière1; Ernst J.W. De Mooij2; RayJayawardhana3; Remco de Kok4; H. Jens Hoeijmakers5,6;Matteo Brogi7; Carl Malcolm Fridlund8; BertVermeersen9; Wim Westrenen10

1 Leiden Observatory, Leiden University (Leiden, Netherlands)2 Vrije Universiteit Amsterdam (Amsterdam, Netherlands)3 Cornell University (Ithaca, New York, United States)4 School of Physical Sciences and Centre for Astrophysics and Rela-

tivity, Dublin City University (Dublin, Ireland)5 Cornell University (Ithaca, New York, United States)6 Universiteit Utrecht (Utrecht, Netherlands)7 Observatoire astronomique, l’Université de Genève (Geneva,

Switzerland)8 Center for Space and Habitability, University of Bern (Bern,

Switzerland)9 University of Warwick (Warwick, United Kingdom)10 Leiden Observatory, University of Leiden (Leiden, Netherlands)11 TU Delft (Delft, Netherlands)

There exists a remarkable population of short pe-riod transiting rocky exoplanets with temperatures>2,000 K, and masses ranging from about 8 Earthmasses, such as the hot super-Earth 55 Cancri e, to

that of Mercury or smaller, such as K2-22b. Theseplanets are thought to have mineral atmospheres thatare produced by the vaporisation of their magmasurfaces, or large exospheres that are produced bysputtering of their atmospheres or exposed sur-faces by intense stellar winds. Additionally, thesmaller, low surface gravity hot rocky exoplanetshave been found to be actively disintegrating andforming ‘comet-like’ dust tails.

Since their atmospheres and released gas and dustcan be observationally constrained, these planetspresent the tantalising prospect of directly probingthe composition of rocky planets. Sodium and cal-cium are promising species to detect given their lowsublimation temperatures, large absorption cross-sections, likely presence in terrestrial planet compo-sitions, and presence in Mercury’s exosphere.

This poster presents the insights we gained fromusing high-resolution transmission spectroscopy tosearch for Na and Ca around 55 Cnc e and K2-22 busing several ground based telescopes. For 55 Cnee, we detected a tantilizing ∼5 σ signal of Ca+ onone night of observation, but a similar signal has notbeen detected since (despite our unprecedented lim-its). This may be related to variability of the star-planet system and the planet’s magnetic field.

For K2-22 b, we did not detect absorption by Naor Ca+, but found lower-limits that are smaller thanthe expected magnitude of the signal based on theplanet’s estimated mass-loss rate and assuming a ter-restrial composition. We attribute this non-detectionto the probed gases being accelerated by the stellarwind and radiation pressure to high velocities, re-sulting in very broad Doppler shifted absorption sig-nals that are hard to detect.

The implications of these results on probingrocky exoplanet compositions, constraining plane-tary magnetic fields, and understanding the environ-ment around short-period rocky exoplanets are alsooutlined.

313.06 — MuSCAT2 validation of a USP giant-planet-sized object around an M-dwarf

Norio Narita1,21 Astrobiology Center (Tokyo, Japan)2 JST (Tokyo, Japan)

MuSCAT2 is a 4-color simultaneous camera on theTelescopio Carlos Sanchez 1.52m in the Teide obser-vatory in Tenerife, Canaries, Spain. On behalf of theMuSCAT2 team, I present a latest result from theMuSCAT2 Consortium: a multi-color validation ofa TESS planet candidate TOI 263.01.

56

TOI 263.01 was first released as a planet candi-date in the TESS alert for the sector 3 in December2018. According to the TESS alert, TOI 263 is a mid-Mdwarf and TOI 263.01 has a period of about 13 hoursand a radius of 5.22 REarth, suggesting it is a possibleultra-short-period giant planet around an M dwarf.Such a planet is extremely rare.

We immediately targeted this planet candidatewith MuSCAT2. At that time MuSCAT2 could ob-serve only in 3 bands (r, i, zs) simultaneously asone CCD camera was removed for upgrading into adeep-depletion CCD camera. Although the host starwas almost setting at the time of the TESS alert, weaimed to validate this planet candidate by taking ad-vantage of the multi-color capability of MuSCAT2.We observed 3 full transits of TOI 263.01 with MuS-CAT2 on December 18 and 19, 2018 and January 2,2019.

The multi-color transit light curves of TOI 263.01confirmed achromaticity of transit depths. We fur-ther utilized a code PyTransit v2 (Parviainen 2015)which models a transit with a possible light contam-ination from unresolved sources. Consequently, theMuSCAT2 multi-color photometry excluded suchcontamination, meaning transits of TOI 263.01 arenot caused by unresolved eclipsing binaries. ThusTOI 263.01 is validated as an object orbiting aroundTOI 263.

We have constrained the radius of TOI 263.01 as0.78 ± 0.15 RJup. The MCMC posterior distributionsuggests the object is smaller than 1.5 RJup with over99.99% confidence level. Thus TOI 263.01 is an ultra-short-period giant-planet-sized object around an M-dwarf.

At this point, we cannot say TOI 263.01 as a planetbecause the mass of TOI 263.01 is still unknown. Thefaintness of the host star (V=18.97, J=14.08) make itdifficult to measure the mass of this interesting ob-ject. However, the 4-VLT mode of ESPRESSO willenable a measurement of the mass of TOI 263.01 inthe future.

313.07 — On the radiative effects on the thermalbulge of a hot Jupiter

Pin-Gao Gu1; Da-Kai Peng2,3; Chien-Chang Yen5,41 Institute of Astronomy and Astrophysics, Academia Sinica

(Taipei, Taiwan)2 Institute of Astronomy and Astrophysics, Academia Sinica

(Taipei, Taiwan)3 National Taiwan University (Taipei, Taiwan)4 Institute of Astronomy and Astrophysics, Academia Sinica

(Taipei, Taiwan)5 FuJen Catholic University (New Taipei City, Taiwan)

We investigate the influence of radiative effects onthe thermal bulge of an asynchronous hot Jupiterdriven by the semidiurnal component of stellar ir-radiation. The background states of the planet areassumed to be in thermal equilibrium maintained bya uniform internal heating per unit mass. On top ofan optically thick interior, we adopt the two-streamapproximation to model the atmosphere in the radia-tive equilibrium between the incoming optical stellarirradiation and outgoing thermal infrared emission.We then apply a semidiurnal thermal forcing to theinterior-atmosphere system to solve for the resultingthermal bulge. We find that while the radiative cool-ing damps the thermal bulge, self-absorption of ther-mal emissions can significantly enhance it depend-ing on the optical and near-infrared opacities in theatmosphere.

313.08 — A comprehensive survey of exoplanet at-mospheres with Spitzer/IRAC

Claire Baxter1; Jean-Michel Desert2; Kamen O.Todorov3; Jacob Bean4; Michael Line5; VivienParmentier6; Adam Burrows7; Heather Knutson8; DrakeDeming9; Jonathan Fortney10; Adam Showman11

1 University of Amsterdam (Amsterdam, Netherlands)2 UC Santa Cruz (Santa Cruz, California, United States)3 University of Arizona (Tucson, Arizona, United States)4 Anton Pannekoek Institute for Astronomy (API), University of

Amsterdam (UvA) (Amsterdam, Netherlands, Netherlands)5 Anton Pannekoek Institute for Astronomy, University of Amster-

dam (Amsterdam, Netherlands)6 University of Chicago (Chicago, Illinois, United States)7 Arizona State University (Tempe, Arizona, United States)8 University of Oxford (Oxford, United Kingdom)9 Princeton University (Princeton, New Jersey, United States)10 California Institute of Technology (Pasadena, California, United

States)11 University of Maryland - College Park (College Park, Maryland,

United States)

Studying exoplanets affords us the opportunity tounderstand the sheer diversity of planets in differ-ent physical regimes. We use the two IRAC band-passes from the warm Spitzer mission and the sta-tistical power of a transit survey to test the theoret-ical predictions of exoplanet atmosphere propertiesacross a broad regime of parameter space. Our sur-vey ranges from the coolest gas giant planets to theultra-hot Jupiters. It spans a range of mass, radii,equilibrium temperature and differing atmosphericcompositions. In particular, we probe the carbonmonoxide, methane and water content of these at-mospheres from emission and transmission spectra.

57

We classify the sample into groups and use the prop-erties of individual planets within these groups tounderstand their collective diversity. We see severaltrends emerging in our data that hint towards dif-ferent atmospheric properties and behaviors in theseatmospheres. We discuss the possibilities for suchtrends. Ultimately, our work places new constraintson the diverse families of exoplanet atmospheres, aswell as on possible formation and evolution scenar-ios.

313.09 — A Comprehensive Spitzer Study of GJ436b

Ryan Challener1; Joseph Harrington11 University of Central Florida (Orlando, Florida, United States)

GJ 436b is one the most observable transitingNeptune-sized planets, with hundreds of hours ofSpitzer observations, including 16 transits and 24eclispes over 6 photometric channels, some of whichhave not been published. We jointly fit all theseobservations, using advances in correlated-noise re-moval techniques to achieve the best, but realistic,signal-to-noise ratios. We then determine updatedorbital parameters, atmospheric composition, andthermal structure, and discuss these results in thecontext of past work. Spitzer is operated by the JetPropulsion Laboratory, California Institute of Tech-nology, under a contract with NASA. This workwas supported by NASA Astrophysics Data AnalysisProgram grant NNX13AF38G.

313.10 — Hot Jupiters are Destroyed While TheirHost Stars are on the Main Sequence

Jacob Howard Hamer1; Kevin C. Schlaufman11 Physics and Astronomy, Johns Hopkins University (Baltimore,

Maryland, United States)

On their extreme orbits, hot Jupiters represent theperfect testing ground for our understanding of tidaldissipation. While cooler giant planets are often ob-served with non-zero eccentricities, the small, circu-lar orbits of hot Jupiters suggest that tidal dissipa-tion plays a significant role in their formation. Onceon these circular orbits, tidal dissipation should al-low the rapidly orbiting planet to transfer angularmomentum to the slowly spinning host star, whichmay result in the inspiral of the planet. However,we do not yet know if hot Jupiters survive the mainsequence of their host stars, as the efficiency of thisprocess is uncertain by orders of magnitude, and be-cause tidal decay has never been unambiguously ob-served. If tidal decay causes hot Jupiters to be de-

stroyed while their host stars are on the main se-quence, then hot Jupiter hosts should be relativelyyoung compared to a sample of similar field starsnot hosting hot Jupiters. We use data from Gaia DR2to show that hot Jupiter hosts have a smaller Galac-tic velocity dispersion than similar stars without hotJupiters. As Galactic velocity dispersion is correlatedwith the age of a population, this implies that hotJupiter hosts are a relatively younger population dueto the inspiral of their planets. This observation re-quires that the tidal quality factor, Q′

∗ , be in therange Q′

∗ < 106.2.

313.11 — The Current State of Spitzer SecondaryEclipse Analyses: HD 209458 b

Kathleen McIntyre1; Joseph Harrington1; RyanChallener1; Matthew Reinhard1; M.R. Green1; Zaccha-eus Scheffer1; Cody Jordan1; Parker Jochum1; CatherineMillwater1

1 Physics, University of Central Florida (Jensen Beach, Florida,United States)

The Spitzer Space telescope has been the workhorsefor exoplanet secondary eclipse observations formore than a decade. Despite this, we are still uncov-ering and understanding new methods for moderat-ing systematics in our analyses. Here we present thetest case of HD 209458b, one of the most observed,published, and highest signal-to-noise exoplanetsdiscovered to date. We compare the effect of differ-ent methods, for example, BiLinearly-InterpolatedSubpixel Sensitivity (BLISS) and Pixel-Level Decor-relation (PLD), on the resulting light curves acrossall of Spitzer’s IRAC channels. One such systematicthat can mimic the features of a secondary eclipseor small primary transit is vibrations of the space-craft that manifest as changes in the point spreadfunction. Previously utilized metrics of interpretingnoise pixel do not account for the systematic’s pres-ence when there is no accompanying increased con-tribution to the noise. This omission with will im-pact published results. Spitzer is operated by the JetPropulsionLaboratory, California Institute of Tech-nology, under a contract with NASA. This work wassupported by NASA Planetary Atmospheres grantNNX12AI69G and NASA Astrophysics Data Analy-sis Program grant NNX13AF38G.

58

314 — Planets around Compact Ob-jects, Poster Session314.01 — The tidal interaction between a whitedwarf and a planet

Dimitri Veras11 University of Warwick (Coventry, United Kingdom)

I present a collection of 3 new papers which all modeltidal interactions between a white dwarf and differ-ent types of planets (gas giants, ice giants, terres-trial planets, magnetic planetary cores). Before 2019,there was almost a complete dearth of papers whosesole focus was to model star-planet tidal interactionswith a white dwarf. The models in these 3 papers aretimely given the increasing evidence of major planetsorbiting white dwarfs.

315 — Planets in Star Clusters,Poster Sessopm315.01 — Destruction of circumstellar disks by sur-rounding stars during star clusters formation

Taku Hasegawa11 Astronomy, The University of Tokyo (Tokyo, Japan)

Over 3500 planets have been discovered since firstdetection of the exoplanet. Although most stars areconsidered to be formed in clustering environments,only dozen of planets have been found in star clus-ters. In clustering environments such as star clus-ters, circumstellar disks can be broken by their sur-rounding stars due to external far-ultraviolet (FUV;6.0-13.6 eV) radiation from their surrounding O, B-stars (photoevaporation) and dynamical disk trunca-tion by stellar encounters. Although recent improve-ment of the resolution of observations enables us toinvestigate the efficiency of the dissipation of circum-stellar disks by those two effects in detail, largenessof contribution of those effects to disk dissipation isstill debatable. In addition, initial conditions of starclusters are one of the still discussed issues. Recenttheories and observations showed that initial statesof star clusters are dynamically cool and clumpy, butnot spherical. In this study, we performed N-bodysimulations of star clusters with clustering initialconditions, constructed from the results of hydrody-namic simulations of the morecular clouds with tur-bulence, and investigated the destruction of circum-stellar disks due to photoevaporation and dynamicaltruncations by stellar encounters in clustering envi-ronments. We regarded the clumps with the number

of stars > 300 and the half-mass density > 100 MSunpc−3 as clusters and investigated the influence of thetwo disk dissipation effects. We also investigated thedependency of the disk truncations effect due to stel-lar encounters on stellar density of star clusters. Wefound that photoevaporation is dominant for diskdissipation, especially in the central region, in moststar clusters. Only in the case that clusters experienceextremely dense phase (stellar density > 107 pc−3)in the early age, dynamical truncations contribute todisk dissipation, which affect strongly at the outsideof clusters.

315.02 — The fate of planetesimal discs in youngopen clusters: implications for 1I/’Oumuamua, theKuiper belt, the Oort cloud and more

Thomas Hands1; Walter Dehnen2,3; Amery Gration2;Joachim Stadel1; Ben Moore1

1 University of Zurich (Zürich, Switzerland)2 University of Leicester (Leicester, United Kingdom)3 Universitäts-Sternwarte der Ludwig-Maximilians-Universität

(Munich, Germany)

We perform N-body simulations of the early phasesof open cluster evolution including a large popula-tion of planetesimals, initially arranged in Kuiper-belt like discs around each star. Using a new, 4th-order and time-reversible N-body code on GraphicsProcessing Units (GPUs), we evolve the whole sys-tem under the stellar gravity, i.e. treating planetesi-mals as test particles, and consider two types of ini-tial cluster models, similar to IC348 and the Hyades,respectively. In both cases, planetesimals can be dy-namically excited, transferred between stars or liber-ated to become free-floating (such as A/2017 U1 or’Oumuamua) during the early cluster evolution. Wefind that planetesimals captured from another starare not necessarily dynamically distinct from thosenative to a star. After an encounter both native andcaptured planetesimals can exhibit aligned perias-trons, qualitatively similar to that seen in the Solarsystem and commonly thought to be the signature ofPlanet 9. We discuss the implications of our resultsfor both our Solar system and exoplanetary systems.

315.03 — Observational consequences of close en-counters in stellar clusters

Melvyn Davies1; Daohai Li1; Alexander Mustill11 Lund Observatory, Lund University (Lund, Sweden)

The birth environments of stars are hazardous to anyplanets they may possess. Close encounters in stel-lar clusters can perturb planetary systems, destabil-

59

ising them, causing orbits to cross, planets to scat-ter, and be ejected, leaving those behind on morebound, and eccentric, orbits. Hot Jupiters may beproduced as massive planets from further out areplaced on extremely eccentric orbits leading to tidalcircularisation very close to their host star. In ex-tremis, planets may collide with their host star, po-tentially leading to an observable enhancement inmetallicity. The study of close encounters involv-ing planetary systems enjoys a rich literature. We,and others, have previously considered the effectsof encounters between planetary systems and pass-ing stars. In this talk, we go beyond earlier work,presenting new research considering encounters be-tween two stars where both of the stars host plan-etary systems. Such encounters will be relativelycommon when one reflects how observations revealthat a large fraction of stars possess planetary sys-tems. Here, we present the results of extensive N-body simulations where we study fly-by encountersinvolving two planetary systems. We model both theinitial fly-by stage but also consider the post-fly-byevolution of both planetary systems. We consider abroad range of planetary systems and host stars. En-counters are found to be damaging to a large fractionof planetary systems. Many stable planetary sys-tems are destabilised: planets are then either ejected,placed on eccentric orbits, or collide with each other.Planets are also found to transfer from one planetarysystem to the other. These interlopers often desta-bilise the planetary system. We find that one half ofthe surviving captured planets are on retrograde or-bits. Thus, we show that encounters between twoplanetary systems within stellar clusters representan intriguing channel to place planets on wide retro-grade orbits that may be detectable through astrom-etry or direct imaging.

315.04 — Survival rates of planets in open clusters:The Pleiades, Hyades, and Praesepe clusters

Yasunori Hori2,1; Michiko S. Fujii31 Exoplanet Detection Project, Astrobiology Center (Mitaka, Tokyo,

Japan)2 Subaru telescope, National Astronomical Observatory of Japan

(Mitaka, Tokyo, Japan)3 Departmet of Astronomy, The University of Tokyo (Bunkyo-ku,

Tokyo, Japan)

In clustered environments, stellar encounters can lib-erate planets from their host stars via close encoun-ters. The detection probability of planets suggeststhat the planet population in open clusters resem-bles that in the field. Only a few dozen planet-hosting stars, however, have been discovered in open

clusters. We explore the survival rates of planetsagainst stellar encounters in open clusters similar tothe Pleiades, Hyades, and Praesepe and embeddedclusters. We performed a series of N-body simula-tions of star clusters, modeling the three open clus-ters and embedded clusters. We find that less than1.5 % of close-in planets within 1 AU and at most7 % of planets with 1–10 AU are ejected by stellarencounters in clustered environments after the dy-namical evolution of star clusters. We expect no sig-nificant difference between the frequency of short-period planets in open clusters and that in the field.Besides, our simulations imply that most of planets(within 10au) around FGKM-type stars are likely tosurvive against stellar encounters in open clusters.If a planet population from 0.01–100 AU in an opencluster initially follows the observed planet distribu-tion in the field, the production rate of free-floatingplanet per star is 0.0096–0.18, where we have as-sumed that all the stars initially have one giant planetwith a mass of 1–13 MJup in a circular orbit. Thesevalues are compatible with the observed fraction offree-floating planets, ∼0.25 per main-sequence star.

315.05 — TESS Planet Candidates in Open Clusters

Luke G. Bouma1; Joel Hartman1; Waqas Bhatti1; JoshuaWinn1; Gaspar Bakos1

1 Department of Astrophysical Sciences, Princeton University(Princeton, New Jersey, United States)

The TESS full-frame images of star clusters are a goldmine for stellar astrophysics and exoplanet science.Because of their known ages, exoplanets in clusterscan be used to resolve questions regarding the for-mation, dynamical evolution, and long-term fates ofexoplanets. We have made over 100,000 light curvesfor candidate member stars of more than 100 openclusters and moving groups, based on TESS obser-vations of the galactic plane.

I will present the most interesting planet candi-dates, eclipsing binaries, and rotational variablesfound in these light curves. These candidates havesurvived numerous vetting procedures and will bemade available to the community for follow-up ob-servations. Ultimately, we hope this work will probethe exoplanet size and separation distributions forplanets younger than 1 Gyr, and will shed light onthe origins of close-in giant planets.

I will also describe the image subtraction and lightcurve processing techniques we have used to obtainprecise photometry in crowded fields. The result-ing light curves have many other uses — notably gy-rochronology and studies of eccentricity damping in

60

eclipsing binaries — and will be accessible for gen-eral use through MAST.

315.06 — Radial velocity confirmation of K2-100b:a young transiting hot Neptune with a likely evap-orating atmosphere

Oscar Barragán11 University of Oxford (Oxford, United Kingdom)

The tens of young transiting exoplanets discoveredto date by K2, and those yet to be found by TESS, arevaluable tests of planet formation and early evolutionscenarios. However, their scientific impact so far hasbeen limited because none of them have mass mea-surements. We present an exhaustive analysis of RVmeasurements for the star K2-100, a member of thePraesepe cluster for which K2 photometry revealed aclose-in planet with a period of 1.67 days. We apply amulti-dimensional Gaussian-Processes approach toHARPS-N radial velocity and activity indicators inorder to measure the Doppler signal of the planetin this active star. We detect a Doppler signal witha semi-amplitude of 10.6 ± 3.2 m/s, consistent withthe transit ephemeris, and corresponding to a massof 22.8 ± 7.0 Earth masses for the planet. This is thefirst mass measurement of a planet orbiting a star ina young active star. The radius of K2-100b, at 3.9± 0.1 Earth radii, implies a significant volatile enve-lope. However, the planet receives is ∼1900 timesmore heavily irradiated than the Earth, and photo-evaporation is expected to play a significant role inits evolution. As the first young transiting planetwith both radius and mass measurements, K2-100bprovides valuable insights into the physical mecha-nisms that shape early planet evolution. This resultalso demonstrates the feasibility of RV follow-up fortransiting planets around active stars, given enoughmeasurements, and bodes well for the young planetsyet to be discovered by TESS.

315.07 — Understanding the early evolution ofplanetary systems with the Next Generation Tran-sit Survey

Edward Gillen11 University of Cambridge (Cambridge, United Kingdom)

We present the first results from our systematic sur-vey of nearby young open clusters with the NextGeneration Transit Survey (NGTS), a wide-field pho-tometric facility based at ESO’s Paranal Observatory.Our aim is to understand the early evolution of plan-etary systems. By characterising young transiting

planets, we can quantify timescales for migration, or-bital circularisation, evolutionary cooling and atmo-spheric loss. I will present the NGTS young clustersprogram, describe how we address the evolving ac-tivity signals of young stars using dedicated Gaus-sian process regression models, and highlight newunpublished results from a comprehensive study ofBlanco 1 (∼120 Myr) and the Orion star forming re-gion (1-10 Myr). Finally, I will conclude with an out-look towards future prospects for the survey and ourunderstanding of young planetary systems.

316 — Planets in and around BinaryStars, Poster Session316.01 — Orbital evolution of a circumbinaryplanet in a protoplanetary disk

Akihiro Yamanaka1; Takanori Sasaki11 Astronomy, Kyoto University (Kyoto, Japan)

Sub-Jupiter classed circumbinary planets (CBPs)around close-in stellar binaries discovered by Keplerhave orbits just beyond the dynamically unstable re-gion, which is determined by the eccentricity andmass ratio of the host binary stars. These planets areassumed to have formed beyond the snow line andmigrated to the current orbits rather than formingin situ. Reproducing orbits of observed CBPs havenot succeeded and the origin of such orbits is yet un-clear. In order to reproduce a stable orbit just overthe instability boundary, we proposed a new sce-nario in which a planet formed beyond the snow lineand migrated to the inner edge of the circumbinarydisk, which was within the unstable area, and thenmoved to the current orbit through outward trans-portation. We carried out N-body simulations with adissipating circumbinary protoplanetary disk for bi-nary systems with different eccentricities and massratios. We find CBPs can maintain a stable orbit nearthe instability boundary if they enter the unstablearea while enough amount of gaseous disk remains.CBPs are more likely to achieve a stable orbit justbeyond the unstable region in binary systems withsmaller eccentricities and mass ratios. These depen-dencies are consistent with the data from observedbinary systems hosting circumbinary planets. Wefind CBPs’ orbits just over the instability boundariesare explained by our orbital evolution scenario.

316.02 — Habitability of S-type tidally locked plan-ets: effects of a binary companion star

Ayaka Okuya1; Yuka Fujii2; Shigeru Ida2

61

1 Earth and Planetary Sciences, Tokyo Institute of Technology(Tokyo, Meguro-ku, Japan)

2 Earth-Life Science Institute (Tokyo, Japan)

Planets in the Habitable Zones around M-type starsare important targets for characterization in futureobservations. Due to their proximity to the host stars,they are likely to be tidally locked in synchronousspin-orbit rotations and to have a hot dayside anda cold nightside (Kasting et al. 1993). On the coldnightside, water vapor transferred from the day-side can be frozen in (“cold trap”) or the major at-mospheric constituent could also condense (“atmo-spheric collapse”) if the atmosphere is so thin thatthe heat re-distribution is not efficient. This is one ofthe serious problems for the habitability of a planetaround a single M-type star (e.g., Joshi et al. 1997;Leconte et al. 2013). Motivated by the abundance ofbinary star systems (Raghavan et al. 2010), we inves-tigate the effects of irradiation from a G-type com-panion star on the climate of a tidally locked planetaround an M-type star. We calculate the surface tem-perature distribution through simulations of the 2Denergy balance model (e.g., North 1975). While themass difference between G-type stars and M-typestars is not so large, the luminosity of G-type starsare a few orders of magnitude brighter. This en-ables a G-type star to warm the cold nightside ofthe planet around the M-type star without destabi-lizing the planetary orbit. On top of it, we find thatthe irradiation from the G-type star is more effec-tive at warming up the nightside of the planet thanthe dayside. This contributes to the prevention ofthe irreversible trapping of water and atmosphere onthe cold nightside, broadening the parameter spacewhere tidally locked planets can maintain surfaceliquid water. Tidally locked ocean planets with ≤∼0.3 bar atmospheres or land planets with ≤ ∼3 baratmospheres can realize temperate climate with sur-face liquid water when they are also irradiated bya companion star with a separation of 1 - 4 au. Wealso demonstrate that the total irradiance is not a suf-ficient measure for the planetary climate in binarysystems, as planets with given properties can be inthe Earth-like temperate climate regime or in a com-pletely frozen state under the same total irradiation.

316.03 — Understanding the Multiplicity of TESSExoplanet Host Candidates

Catherine Clark1,2; Gerard van Belle2; Elliott Horch3;Kaspar von Braun2

1 Northern Arizona University (Flagstaff, Arizona, United States)2 Lowell Observatory (Flagstaff, Arizona, United States)

3 Southern Connecticut State University (New Haven, Connecticut,United States)

While at first glance multi-star systems seem quiteextreme, they are in fact the most common typeof star system in our galaxy, throughout the stellarmass distribution. In particular, 40 to 50% of exo-planet host stars reside within multiple star systems.Given the degree to which initially undetected mul-tiplicity has skewed Kepler results, high-resolutionimaging of our nearby low-mass neighbors is neces-sary for both accurate characterization of transitingexoplanets, as well as a better understanding of stel-lar astrophysics. To address this frequent gap in ourknowledge of exoplanet hosts, we will utilize speckleinterferometry to directly image TESS exoplanet hostcandidates to complete our knowledge of individ-ual star multiplicity. Our investigation will expandupon the speckle observations taken as a part of thePOKEMON speckle survey of nearby M-dwarfs tobetter constrain the multiplicity of low-mass TESSexoplanet host candidates, and to constrain M-dwarfmultiplicity by subtype across the entire M-dwarf se-quence.

316.04 — Forming Polar Planets Around Binaries

Stephen Lubow1; Rebecca Martin21 Space Telescope Science Intsitute (Baltimore, Maryland, United

States)2 Physics and Astronomy, University of Nevada Las Vegas (Las

Vegas, Nevada, United States)

One of the key discoveries of the Kepler mission wasthe detection of transiting circumbinary planets. Bythe nature of the detection technique, these plan-ets are preferentially found on nearly coplanar orbitswith the binary. We (Martin & Lubow 2017) haveshown that a mildly misaligned circumbinary pro-toplanetary disk can naturally evolve to a highly in-clined polar orbit that is perpendicular to the binaryorbital plane. Planets formed in such a disk wouldorbit around the semimajor axis of the binary, in-stead of in the binary orbital plane. Such a disk wasrecently discovered in HD98800 by Kennedy et al.(2019) using ALMA. We have recently extended ourwork to explore how the disk angular momentumand radius can affect this process (Martin & LubowarXiv:1904.11631). We find that disk breaking and in-clined nonperperdicular orientations can occur. Thelatter can provide a constraint on the disk mass. Wewill also comment on the following questions: Is theexistence of such planets compatible with their lackof detection to date by Kepler? What types of bi-nary systems are most likely to harbor polar planets?

62

How does the dynamics of polar disks and their ac-cretion onto the stars differ from coplanar circumbi-nary disks? What would we learn about the forma-tion of polar planets by studying the properties oftheir orbits?

316.05 — Instabilities in Multi-Planet Circumbi-nary Systems

Adam P. Sutherland1; Kaitlin M. Kratter11 Astronomy, University of Arizona (Tucson, Arizona, United

States)

The majority of the discovered transiting circumbi-nary planets are located very near the innermoststable orbits permitted, raising questions about theorigins of planets in such perturbed environments.Most favored formation scenarios invoke formationat larger distances and subsequent migration to theircurrent locations. Disk-driven planet migration inmulti-planet systems is likely to trap planets in meanmotion resonances and drive planets inward into re-gions of larger dynamical perturbations from the bi-nary. We demonstrate how planet-planet resonancescan interact with the binary through secular forcingand mean-motion resonances, driving chaos in thesystem. We show how this chaos will shape the ar-chitecture of circumbinary systems, with specific ap-plications to Kepler 47 and the Pluto-Charon system,limiting maximum possible stable eccentricities andindicating what resonances are likely to exist. We arealso able to constrain the minimum migration ratesof resonant circumbinary planets. Furthermore, wedemonstrate how circumbinary planets can be af-fected by tidal evolution of the binary, limiting theefficiency of tidal evolution, constraining the age ofcircumbinary systems, and possibly explaining thelack of circumbinary planets around short period bi-naries.

316.06 — On the multiplicity of TESS planet hosts:Early results from a ground-based AO follow-upcampaign

Elisabeth Matthews1; Ian Crossfield1; David Ciardi2;Steve Howell3; Charles Beichman4; Erica Gonzalez5;Rachel Matson3; Joshua Schlieder6

1 MIT (Somerville, Massachusetts, United States)2 IPAC (Pasadena, California, United States)3 NASA AMES (Mountain View, California, United States)4 JPL (Pasadena, California, United States)5 UC Santa Cruz (Santa Cruz, California, United States)6 NASA GSFC (Greenbelt, California, United States)

We are using 8m class telescopes to study the multi-plicity of TESS planet hosts. TESS has been finding

candidate planets for nearly a year, and we have beentaking snapshot images of the most promising tar-gets to identify nearby stars at separations as smallas a few 10s of milliarcseconds, using both AO andspeckle imaging. In the short term, this work is vi-tal to check for the presence of blended stars whichcan dilute the light curve, which can bias the mea-sured radius of the observed planet, or even be re-sponsible for false positives (Ciardi et al. 2015). Mea-suring the radius correctly is crucial for accuratelydetermining the density of small planets, and there-fore fully understanding the mass-radius relation-ship. Although GAIA provides excellent data on vi-sual companions that are either bright or separatedby several arcseconds, it is not sensitive to the faint,close visual companions that we are able to detect.When visual companions are detected, we are work-ing to determine whether these objects are bound,and thereby understand how binarity affects planetformation. TESS typically detects planets roughlytwice as close as those identified by Kepler, and so weare able to probe for binary companions at smallerangular separations and determine the most com-pact binary systems where planets can be found. Inthis talk I will present some of our early results fromthe survey, where we have already identified ∼15 vi-sual companions to ∼40 TESS candidate hosts, andit appears likely that many of these are bound com-panions. This work will facilitate a broader study ofthe multiplicity of TESS planet hosts, for which weplan to combine GAIA and our high resolution data.

316.07 — The Perilous Lives of Planets in BinaryStar Systems

Adam Kraus11 UT-Austin (Austin, Texas, United States)

The majority of solar-type stars form with binarycompanions, and they should profoundly sculpt theformation and evolution of planetary systems. How-ever, most searches for extrasolar planets have con-centrated exclusively on single stars, avoiding closebinary systems where the companion might compli-cate the observations and analysis. I will discuss sta-tistically robust samples that outline the influence ofstellar multiplicity at different stages of planet for-mation and evolution, enabled by a high-resolutionimaging technique (nonredundant aperture-maskinterferometry) that opens the discovery space for bi-nary companions on solar-system scales (5-50 AU)even around relatively distant stars (100-500 pc).From the combined census of stellar multiplicity andprotoplanetary disk occurrence in star-forming re-gions, binary companions have a ruinous effect upon

63

protoplanetary disks; at ages when disks are stillubiquitous among single stars and wide binaries, thecorresponding disk fraction among close binaries issuppressed by a factor of 3-4. The correspondingcensus of stellar multiplicity and transiting planetarysystems at old ages, as assessed in the Kepler sam-ple, shows a similar suppression rate among closebinaries. However, some planetary systems do sur-vive even in very dynamically perilous configura-tions, and I will outline the factors that potentiallycontribute to planetary survival and destruction. Fi-nally, I will discuss how these results change the in-terpretation of planet formation and disk dissipationaround single stars, given that almost all of them ap-pear to host disks for at least 2-3 Myr, and what thismight mean for the timescales and mass budgets forassembling planetary systems.

316.08 — Understanding the Impacts of StellarCompanions on Planet Formation and Evolution:A Survey of Stellar and Planetary Companionswithin 25 pc

Lea Hirsch1; David Ciardi2; Andrew Howard31 Physics, Stanford University (Burlingame, California, United

States)2 IPAC, Caltech (Pasadena, California, United States)3 Astronomy, Caltech (Pasadena, California, United States)

Nearly half of all solar-type stars have at least onestellar companion, and planets around G- and K-type stars appear to be quite common. Yet theimpact of stellar multiplicity on planets is not yetwell understood. Stellar companions may truncatedisks and destabilize planetary orbits, and may con-tribute to planet migration scenarios. To test thesetheories, we have conducted a uniform survey forstellar and planetary companions around sun-likestars within 25 pc of the Sun. This survey madeuse of high-resolution imaging observations fromShaneAO at Lick Observatory and speckle interfer-ometry from DSSI at WIYN, as well as precise radialvelocities from HIRES at Keck and the AutomatedPlanet Finder at Lick. From these observations, weconfirm the stellar multiplicity statistics of previoussurveys and uncover several new high-contrast bi-nary systems. We also perform a thorough investi-gation of the nearby giant planet sample, includingthe detection of several new planets. By dividing thesample based on the presence or absence of stellarcompanions, the planet occurrence rates can be di-rectly compared between single and binary systems.We conclude that binary companions with separa-tions of hundreds of AU do not seem to strongly im-pact the planet formation process. Follow-up work

with larger sample sizes and detailed attention to ob-servational sensitivity to planets in close binary sys-tems is needed to address the hypothesis that bina-ries with separations 10 - 100 AU more strongly im-pact planets. Since these “close” binaries compriseover a quarter of sun-like stellar systems, this follow-up work is essential to reach a complete understand-ing of planet occurrence rates and distributions.

316.09 — Stability Boundary and Stopping Orbitsof Circumbinary Planets

Nader Haghighipour1; Frederic Masset21 Institute for Astronomy, University of Hawaii (Honolulu, Hawaii,

United States)2 Institute for Physical Sciences, Universidad Nacional Autonoma

de Mexico (Cuernavaca, Morelos, Mexico)

Not only did the detection of circumbinary plan-ets (CBPs) confirm theoretical predictions of theirexistence, it also revealed new findings, some ofwhich challenging our understanding of major phys-ical processes such as planet migration. One of thesefindings is the apparent clustering of CBPs at or closeto the boundary of stability. It has been stated thatthis clustering is not a detection bias and that the sta-bility boundary is a preferred stopping location formigrating CBPs. Given that the physical processesgoverning the stopping of a migrating planet andthat of the appearance of the boundary of stabilityare unrelated, the above statement, if correct, impliesthat CBP migration may involve new and unknownphysics. It is, therefore, imperative to understand thenature of this clustering and determine its implica-tions. Motivated by the latter, we launched a com-prehensive project on studying the stopping of mi-grating CBPs. A migrating planet stops where nega-tive Lindblad torques are balanced by positive coro-tation torques, a process that depends on the phys-ical properties of the disk such as its mass and vis-cosity. We, therefore, considered a parameter spaceconsisting of plausible ranges for the mass, viscosityand scale-height of the disk. The boundary of or-bital stability is a celestial mechanics entity that isthe result of the gravitational interactions betweenthe binary and planet. We, therefore, considered bi-naries with different semimajor axes, eccentricitiesand mass-ratios, and varied the mass of the planet,as well. We carried out close to 1000 simulationsof planet migration and identified the CBP’s finalorbits. Results confirmed, unequivocally, that cir-cumbinary planets stop at a variety of different or-bits, and there is no logical connection between theirstopping location and the boundary of orbital stabil-ity. This is a significant finding that fully agrees with

64

theoretical predictions, clarifies the nature of the ap-parent clustering, and prevents it from being mistak-enly labeled as a physical characteristic of CBPs. Werespectfully request time for an oral presentation topresent and discuss results of our study.

316.10 — Planetary–Stellar Orbit Alignment in Bi-nary Systems

Trent Dupuy1; Adam Kraus2; Kaitlin M. Kratter4;Aaron Rizzuto2; Andrew Mann5; Michael Ireland6;Daniel Huber3

1 Gemini Observatory, Northern Operations (Hilo, Hawaii, UnitedStates)

2 UT-Austin (Austin, Texas, United States)3 IfA/UH (Honolulu, Hawaii, United States)4 Steward/Arizona (Tucson, Arizona, United States)5 UNC-Chapel Hill (Chapel Hill, North Carolina, United States)6 ANU (Canberra, New South Wales, Australia)

Most planetary systems only offer the possibility tomeasure either the initial conditions of planet forma-tion (e.g., protoplanetary disks) or the final outcome(e.g., demographics of mature field samples). Planet-hosting binaries offer the rare opportunity to studyboth concurrently. We will present results from ourKeck adaptive optics program to monitor the stel-lar orbits of KOIs that have binary companions atsolar-system scales of 20–200 AU. The astrometric or-bital arcs that we measure enable a fundamental test:whether or not the stellar orbits are seen edge-on andthus co-aligned with the transiting planets in the sys-tem. This orbit–orbit alignment test allows us to crit-ically examine the possible formation pathways forthese systems. We find that stellar and planetaryorbits tend to be aligned, which raises the questionof whether this may be one of the keys to success-ful planet formation and subsequent survival in theotherwise inhospitable environment of such binarysystems. Projection effects and the underlying eccen-tricity distribution are limiting factors in our currentanalysis, and I will discuss how future observationswill actually turn this challenge into an opportunityto jointly constrain eccentricities and semimajor axes,as well as alignment, in the sample of Kepler planet-hosting binaries.

317 — Planet Formation Theory,Poster Session317.01 — Establishing the Diversity of Super-EarthSystems with a Continuum of Formation Condi-tions

Mariah MacDonald1; Sarah Morrison1; RebekahDawson-Rigas1

1 Astronomy & Astrophysics, Pennsylvania State University (StateCollege, Pennsylvania, United States)

Multi-planet systems observed by Kepler that con-tain super-Earths exhibit a diversity of orbital andcompositional properties. Here we investigate whatplanetary system outcomes arise from a range of pro-toplanetary disk solid surface densities and dissipa-tive conditions shortly before disk dispersal, throughsimulating the giant impact phase of planet forma-tion and subsequent dynamical evolution. We alsocompare the orbit distributions of these outcomes tothe multi- transiting systems observed by the Keplermission. For the same degree of dissipation froma gaseous disk and with no orbital migration, wefind that larger solid surface densities lead to moretightly packed, flatter systems than smaller solid sur-face densities. We find that the spread in mass-radius relation observed in the Kepler populationcan also be explained with a wide range of solid sur-face densities, where small solid surface densitieslead to rocky, dense planets and large solid surfacedensities lead to larger, gaseous planets. The distri-butions of the period ratios, spacings in mutual Hillradii, and transit duration ratios of adjacent planets— as well as the distribution of planet multiplicity— arising from these solid surface densities in con-junction with moderate gas damping (correspondingto a protoplanetary disk depleted by a factor of 100in mass before disk dispersal) agree with the distri-butions of observed systems. These disk conditionscan also produce super Earth systems with resonantchains, successive pairs near and in mean motion res-onances.

317.02 — Suppression of pebble accretion byplanet-induced gas flow: The implication for theformation of super-Earths

Ayumu Kuwahara1; Hiroyuki Kurokawa21 Earth and Planetary Sciences, Tokyo Institute of Technology

(Tokyo, Japan)2 Earth-Life Science Institute, Tokyo Institute of Technology (Tokyo,

Japan)

The ubiquity of short-period super-Earths remains a

65

mystery on the planet formation, because these plan-ets are expected to become gas giants via runawaygas accretion within the lifetime of protoplanetarydisk. Super-Earths cores should be formed in thelate-stage of the disk evolution to avoid the runawaygas accretion. Previous studies have found the three-dimensional (3D) structure of the gas flow aroundembedded planets (e.g., Ormel et al. 2015). Disk gasenters at high latitude of the Bondi/Hill sphere of theplanet and exits through the midplane region. Thisoutward gas flow was considered to suppress the ac-cretion of ∼mm—cm-sized particles, called pebbles,and delay the core growth (Kurokawa & Tanigawa2018; Kuwahara et al. 2019).

We calculated the trajectories of pebbles accretingonto a planet in the 3D gas flow field obtained fromnon-isothermal hydrodynamical simulations.

The efficiency of pebble accretion is lower in theplanet-induced flow than in the unperturbed Kep-lerian shear flow. In the midplane, pebbles comingfrom a window between the horseshoe and shear re-gions can accrete onto the core if the pebble size issufficiently large. Otherwise the outflow prohibitspebble from accreting. Because the horseshoe flowextends well above the midplane, pebbles comingfrom high altitudes are also influenced. We analyti-cally derived that the pebble accretion is suppressedwhen m ≥ √(St), where m is the dimensionless plane-tarty mass expressed by the ratio of the Bondi radiusto the disk scale height and St is the Stokes numberof pebbles (the stopping time times the Keplerian fre-quency). This means that, for a given St, a growingproto-core starts to suppress the accretion of thosepebbles as the core reaches the mass m = √(St).

We suggest a scenario for the formation of super-Earths as follows. 1. Proto-cores form in the proto-planetary disk under the influence of the flow field.Due to the planet-induced gas flow field, the growthof proto-cores may halt when m ∼ √(St). 2. When thegrowth of the proto-cores halts, they begin to migrateinward. 3. Super-Earths are formed by giant impactduring disk dispersal.

317.03 — Collisional growth of organic-mantledgrains and formation of rocky planetesimals

Kazuaki A. Homma1; Satoshi Okuzumi1; TaishiNakamoto1; Yuta Ueda1,2

1 Earth and Planetary Sciences, Tokyo Institute of Technology (Me-guro, Japan)

2 Earth and Planetary Science, The University of Tokyo (Hongo,Tokyo, Japan)

It is believed that the collisional growth of silicatedust grains is restricted by their poor stickiness. In

this study, we explore the possibility that the sticki-ness of silicate grains in protoplanetary disks is en-hanced by organic mantles. Silicate grains coated byorganics can be commonly found in interplanetarydust particles, and previous laboratory experiments(Kudo et al. 2002) showed that such organic-coatedparticles are sticky in warm environments. To studyin more detail how the stickiness of organic-mantledgrains depends on temperature and mantle thick-ness, we construct a simple grain adhesion modelthat gives the binding energy of core-mantle grainsin contact. Our model shows that the stickiness oforganic-mantled grains increases with temperature.This occurs because as the temperature increases, theelasticity of organic mantles decreases and the con-tact area increases. We find that aggregates made oforganic-coated grains are able to break through thefragmentation barrier in the inner part of protoplan-etary disks where temperature is above > 200 K. Wealso simulate the growth and radial drift of organic-mantled grains in a disk, finding that they indeedgrow into planetesimal-sized objects in a warm in-ner region of the disk. We will discuss a new scenariofor terrestrial planet formation based on our results.Reference: Homma, A. K. et al. 2019, ApJ, 877, 128(DOI: 10.3847/1538-4357/ab1de0)

317.04 — Pebble-driven planet formation aroundstars of different masses

Liu Beibei1; Anders Johansen1; Michiel Lambrechts11 Lund University (Lund, Sweden)

Observational breakthrough has been achieved incharacterizing the properties of protoplanetary disksand extrasolar planets in the last decade. We thusgain a better understanding on both the birth condi-tions and the end products of planets. Meanwhile,more advanced theoretical and numerical modelsare required to establish the bridge between thesetwo based on evolving planet formation theories. Wedevelop the pebble-driven core accretion model tostudy the formation and evolution of planets aroundstars in the range of 0.08 MSun and 1 MSun. By MonteCarlo sampling of their initial conditions, the growthand migration of a large number of individual pro-toplanetary embryos are simulated in a populationsynthesis mannar. Two hypothesis are proposed forthe birth locations of embryos, at the water ice lineor log-uniformly distributed over distance in proto-planetary disks. Two types of disks with differentturbulent viscous parameters αt of 10−3 and 10−4

are also investigated. The forming planet popula-tion is statistically compared with the observed ex-oplanets in terms of mass, semimajor axis, metallic-

66

ity and water content. We find that massive planetsare likely to form when the characteristic disk sizesare larger, the disk accretion rates are higher, thedisks are more metal rich and/or their stellar hostsare more massive. Our model shows that 1) the char-acteristic planet mass is set by the pebble isolationmass. It increases linearly with the stellar mass, cor-responding to one Earth mass around a Trappist-1star and 20 Earth mass around a solar-mass star. 2)The low-mass planets up to 20 ME can form aroundstars with a wide range of metallicities, while mas-sive gas giant planets are preferred to grow aroundmetal rich stars. 3) The super-Earth planets mainlycomposed of silicates with relatively low water frac-tions can form from the seeds at the water ice line inless turbulent disks. Altogether, the model succeedsin quantitatively reproducing several important ob-served properties and correlations among exoplan-ets.

317.06 — Unified model of formation and atmo-spheric evolution of super-Earths and Neptune-mass planets

Masahiro Ogihara1; Yasunori Hori21 National Astronomical Observatory of Japan (Tokyo, Japan)2 Astrobiology center (Mitaka, Tokyo, Japan)

According to the theoretical study of atmosphericaccretion, super-Earths and Neptune-mass planets(SENs) accumulate massive H/He atmospheres ina runaway fashion within the lifetime of protoplan-etary disks. In contrast, several observational evi-dences suggest that most SENs avoided accretion ofmassive atmospheres. Many ideas have been pro-posed to solve this discrepancy. As a possible solu-tion, we focus on the heating of atmospheres by peb-ble accretion, which would suppress the accretion ofatmospheres. In addition, the accreted atmospherecan be lost by collisional erosion during late-stage gi-ant impacts and by long-term photoevaporation dueto stellar XUV. In this study, we perform unified nu-merical calculations for N-body simulations of theformation of SENs formation and their atmosphericevolution. In other words, using our N-body simula-tions, the planetary growth and atmospheric evolu-tion can be consistently calculated. In this presenta-tion, we will reveal whether the amount of accretedatmospheres can be limited by the above mecha-nisms (i.e., heating by pebble accretion, atmosphericescape). We also discuss a few other observed prop-erties (e.g., orbital distribution, sub-Neptune desert)of SENs that can be reproduced by the results of oursimulations. We can theoretically predict observable

properties (e.g., orbital property, mass, and atmo-sphere) and their mutual correlation, which help un-derstand the results of ongoing and future observa-tional projects (e.g., TESS, CHEOPS).

317.07 — Formation of compact system of super-Earth via dynamical instabilities and giant impacts

Sanson Poon1; Richard Nelson11 Astronomy Unit, Queen Mary University of London (London,

United Kingdom)

NASA’s Kepler mission discovered ∼700 planets thatreside in multisystems containing 3 or more tran-siting planets, many of which are super-Earths andmini-Neptunes in compact configurations whose ori-gins are not yet understood. Using N-body simula-tions, we examine the final stage assembly of mul-tiplanet systems through the collisional accretion ofprotoplanets. Our initial conditions are constructedusing a subset of the Kepler 5-planet systems as tem-plates, and apply to the epoch after gas disc disper-sal. Two different prescriptions for the outcomes ofplanetary collisions are adopted. The simulationsaddress a number of questions: do the results de-pend on the accretion prescription?; do the resultingsystems resemble the Kepler systems and do they re-produce the observed distribution of planetary mul-tiplicities when synthetically observed?; do colli-sions lead to significant modification of protoplanetcompositions, or to stripping of gaseous envelopes?;do the eccentricity distributions agree with those in-ferred for the Kepler planets? We find the accretionprescription is unimportant in determining the out-comes. On average, the final planetary systems looksimilar to the Kepler templates we adopted, but thesimulations do not reproduce the observed distribu-tions of planetary multiplicities or eccentricities, be-cause gravitational scattering does not dynamicallyexcite the systems sufficiently. In addition, we findthat approximately 1% of our final systems containa co-orbital planet pair in horseshoe or tadpole or-bits. Post-processing the collision outcomes suggeststhey would not lead to significant changes of the wa-ter fractions of initially ice-rich protoplanets, but sig-nificant stripping of low mass gaseous atmospheresappears likely. Hence, it may be difficult to recon-cile the observation that many of the low mass Ke-pler planets appear to have H/He envelopes with aformation scenario that involves giant impacts afterdispersal of the gas disc.

67

317.08 — Evolution and growth of dust grainsin protoplanetary disks with magnetically drivendisk wind

Tetsuo Taki1,2; Koh Kuwabara2; Hiroshi Kobayashi3;Takeru K. Suzuki2

1 National Astronomical Observatory of Japan (Mitaka, Tokyo,Japan)

2 University of Tokyo (Tokyo, Japan)3 Nagoya University (Nagoya, Japan)

Magnetically driven disk winds (DWs) are one ofthe promising mechanism of dispersal processes ofprotoplanetary disks (Suzuki et al. 2010, Bai 2013).When the DWs play a key role, the gaseous compo-nent of protoplanetary disks evolves in a differentmanner from that of the classical viscous evolution.As a result, the subsequent planet formation is alsoaffected by the DWs. In this work, we investigate theeffects of the DWs on the radial drift of solid parti-cles with the size of 0.1μm - 1km. We propose thatthe DWs is a possible solution to the ”radial drift bar-rier” of collisionally growing dust grains, which is asevere obstacle to the planet formation (e.g., Naka-gawa et al.1986). In order to study the evolution ofdust grains in the disks, we calculate the advectionand the collisional growth of dust particles in evolv-ing protoplanetary disks under the 1+1 D (time +radial distance) approximation. We solve a coagu-lation equation of solid particles under a single-sizeapproximation (Sato et al. 2016) for various condi-tions of turbulent viscosity, the mass loss by the DW,and the magnetic braking by the DW. We found thatrapid grain growth occurs in the inner region of theprotoplanetary disks. The DWs disperse the pro-toplanetary disk from inside to outside. On suchthe process, the region where a pressure gradient issmaller than the typical value of the protoplanetarydisks appears. At the same time, the Stokes numberof dust grains tend to be larger than the case withoutsuch region. The growth timescale of dust grains be-comes shorter than the radial drift timescale of themin such the flat and gas dispersed region. In addition,when the disk gas is mainly lost by the DWs ratherthan by the accretion, an outwardly moving pressurebump is formed. The pressure bump can halt thedust radial migration (e.g., Taki et al. 2016). We con-firmed that the dust grains trapped in the pressurebumps in our simulations. It is a potential advantagefor the planetesimal formation.

317.09 — Inertial concentration of dust particles inaccretion disks

Pascale Garaud1; Sara Nasab1

1 Applied Mathematics, UC Santa Cruz (Santa Cruz, California,United States)

Turbulence in protostellar disks can cause denseconcentrations of dust particles to form through aprocess called inertial concentration. Using DirectNumerical Simulations in the two-fluid formalism(where the particles are treated as a continuum cou-pled with the gas through a linear drag term), wedemonstrate the existence of a scaling law relatingthe maximum particle concentration observed at anygiven time to the particle Stokes number, the particlediffusion coefficient, and the rms velocity of the tur-bulent fluid. This law can be explained using simpledimensional arguments. We apply our findings todusty disks, to predict what the largest possible dustconcentrations may be at any given point in the disk.

317.10 — Fragmentation favours protoplanetarydiscs around high mass stars

James Cadman11 School of Physics and Astronomy, University of Edinburgh (Edin-

burgh, United Kingdom)

Recent observations suggest that, when we observewide-orbit gas giants, they are primarily foundaround high mass stars. A possible explanation tothis may be provided by planet formation throughfragmentaion. Fragmentation most likely occurs inthe outer regions of protoplanetary discs where theircooling times are smallest, whilst also likely onlyforming massive gas giant planets and brown dwarfsrather than terrestrial planets, thus preferentiallyforming wide-orbit gas giants. The work done inthis project aims to show that the conditions neces-sary for a disc to be unstable against fragmentationare more readily satisfied around higher mass stars,therefore potentially providing explaination of thisobserved planet population.

317.11 — Pebble-driven planet formation forTRAPPIST-1 and other compact systems

Djoeke Schoonenberg1; Beibei Liu2; Chris W. Ormel1;Caroline Dorn3

1 Anton Pannekoek Institute, University of Amsterdam (Amster-dam, Netherlands)

2 Department of Astronomy and Theoretical Physics, Lund Obser-vatory (Lund, Sweden)

3 University of Zurich (Zürich, Switzerland)

Two years ago, a spectacular planetary system wasdiscovered around the M-dwarf star TRAPPIST-1.Seven Earth-sized planets are circling this star with

68

very short periods: their orbits would all fit wellwithin Mercury’s orbit in the Solar System. Threeout of the seven planets are located in the habit-able zone; the temperate conditions may support thepresence of liquid water. Thanks to transit-timingvariations, the planets’ masses and therefore compo-sitions have been constrained. Planet internal mod-elling suggests that the TRAPPIST-1 planets havemoderate water fractions of a few to tens of masspercent, which is much more than that of the Earth.These values suggest that the TRAPPIST-1 planetsformed outside the snowline and migrated inwardwhile they were still growing. We have connecteda model of the evolution of dust and pebbles andplanetesimal formation with a model of the growthfrom planetesimals to planets. This strategy enablesus to self-consistently model the assembly of theTRAPPIST-1 planets, keeping track of their compo-sition. The model may also be applied to other com-pact planetary systems. One of our key predictionsis that the planet water fraction shows a V-shapedtrend with planet order. Future (more precise) ob-servational measurements of planet water fractionsin compact systems could therefore be used to con-strain our planet formation model.

317.12 — Photodissociation-Driven Mass Loss fromYoung and Highly-Irradiated Exoplanets

Alex Howe1; Fred Adams2,3; Michael Meyer41 Goddard Space Flight Center (Ann Arbor, Michigan, United

States)2 Department of Physics, University of Michigan (Ann Arbor,

Michigan, United States)3 Department of Astronomy, University of Michigan (Ann Arbor,

Michigan, United States)4 Department of Astronomy, The University of Michigan (Ann

Arbor, Michigan, United States)

The most widely-studied mechanism of mass lossfrom irradiated exoplanets is photoevaporation viaXUV ionization. However, lower-energy FUV disso-ciation of hydrogen molecules can also theoreticallydrive atmospheric evaporation on low-mass planetsbecause the dissociation energy of hydrogen is an or-der of magnitude greater than the escape energy perproton from the gravity well of an Earth-sized planet.This implies that a significant fraction of a star’sblackbody flux can contribute to photoevaporation,potentially to a greater degree than ionizing radia-tion. For temperate planets such as the early Earth,impact erosion is expected to dominate over photoe-vaporation in most formation models, but for highlyirradiated planets such as those near the “evapora-tion valley” observed in Kepler planets, or for peb-

ble accretion formation models, they could plausiblybe sculpted primarily by photodissociation. I presentresults of a survey of various mass loss processes andtheir relative contributions to mass loss from an earlyEarth-like planet. In particular, we find that pho-todissociation could strip an atmosphere up to 0.5%of the mass of the planet even at Solar levels of irradi-ation. I then apply this prescription for mass loss tomodels of highly irradiated super-Earths and mini-Neptunes and discuss the implications of these re-sults for rocky planet formation and the interpreta-tion of the evaporation valley.

317.13 — Planetesimal formation at the inner edgeof the dead zone: Implication for the diversity inplanetary systems

Takahiro Ueda1; Satoshi Okuzumi21 Division of Science, National Astronomical Observatory of Japan

(Tokyo, Japan)2 Department of Earth and Planetary Sciences, Tokyo Institute of

Technology (Tokyo, Japan)

We perform simulations of the dust and gas diskevolution to investigate the planetesimal formationat the dead-zone inner edge. We show that the to-tal mass of planetesimals is sensitive to the turbu-lence strength in the dead zone because of the com-bined effect of turbulence-induced particle fragmen-tation and turbulent diffusion. For a typical criticalfragmentation velocity of silicate dust particles of 1m s−1, the stress to pressure ratio in the dead zoneneeds to be lower than 3×10−4 for dust trapping tooperate. if the stress to pressure ratio in the deadzone is around 10−3, planetesimals with a total massof ∼ 2 Earth masses is formed at the dead-zone innerboundary, which is preferable to form the solar sys-tem terrestrial planets. If the stress to pressure ratiois lower, the total planetesimal mass is larger, whichis preferable to form more massive planets such assuper-Earths. The strong dependence of the totalplanetesimal mass on the turbulent strength mightbe linked with the diversity in the mass of planetarysystems

317.14 — Elemental Abundances of Planetary At-mospheres and Planet Formation

Niloofar Khorshid2,1; Michiel Min1; Jean-Michel Desert21 SRON (Utrecht, Netherlands)2 Anton Pannekoek Institute for Astronomy (API), University of

Amsterdam (UvA) (Amsterdam, Netherlands, Netherlands)

Knowing elemental abundances of a planet is keyto understanding its formation. This is because the

69

composition of a planet atmosphere is heavily influ-enced by the planet formation process, and subse-quent evolution. Close-in giant planets are currentlythe best candidates to study the question of their for-mation from their atmospheric composition. In thiswork we develop a framework that uses models forprotoplanetary disks and for planet formation to pa-rameterize the inputs in simple manners and forma giant planet with its atmosphere. Ultimately wewill produce model spectra of our modeled planetand compare these to observations to put limits onthe inputs for the planet formation model previouslyconstructed.

317.15 — Planetesimal formation for planetarypopulation synthesis

Oliver Voelkel11 Max Planck Institute for Astronomy (Heidelberg, Germany)

Planetesimal formation is a key process in the for-mation of planets and up to this day there is no di-rect observational data to constrain the existing the-oretical models. Since the number of detected exo-planets has increased drastically in the last decadesit has become possible to use the properties of thispopulation of exoplanets to form constrains for ourmodels in a statistical sense. The framework of pop-ulation synthesis is therefore a unique and power-ful tool to test models for planetesimal formation andbridge the gap of its non observability by connectingthe observable initial dust density to the final observ-able population of planets. Latest results on the for-mation of planetesimals indicate a much steeper col-umn density profile of planetesimals in protoplane-tary disks than the previously assumed one, whichoriginates from the widely used minimum mass so-lar nebula model. First studies using the Bern modelof planet population synthesis indicate that the ac-cretion of large planetesimals (100km) alone is a veryefficient growth mechanism that can account for thelarge diversity in the population of exoplanets.

317.16 — The first self-consistent modeling of colli-sional water transport during late-stage planet for-mation

Christoph Burger1,2; Christoph M. Schäfer2; AkosBazso1; Thomas I. Maindl1

1 Institute of Astrophysics, University of Vienna (Vienna, Austria)2 Institut für Astronomie und Astrophysik, University of Tübingen

(Tübingen, Germany)

The final phase of terrestrial planet formation, whereplanetary embryos and remaining planetesimals ac-

crete into planets, is a dynamically stochastic pro-cess, marked by giant collisions among protoplan-ets and radial mixing of material over wide dis-tances, where the origin and abundance of wateris strongly tied to collisional transfer and loss pro-cesses, and of central importance for their habitabil-ity. Several approaches for treating collisions be-yond (over-)simplified perfect inelastic merging havebeen developed in recent years, but none has beendesigned nor applied for modeling the collisionalevolution and delivery of water to growing terres-trial planets, even though it is particularly suscep-tible to collisional transfer and erosion. To eventu-ally close this gap we have successfully developeda new hybrid framework to directly combine long-term N-body integrations with Smooth Particle Hy-drodynamics (SPH) simulations of individual col-lision events, which allows us to self-consistentlymodel collisional water transport on a system-widescale for the first time. This includes full treatmentof frequent hit-and-run events, where not only waterlosses but also transfer between the colliding bod-ies can be crucial. In a solar-system-like setting asthe first application, our simulations include vary-ing gas giant architectures, and a bi-modal initial dis-tribution of embryos and smaller bodies, where wefind that with a realistic collision treatment final wa-ter contents are reduced by a factor of 2 or more.Our results show that water delivery is dominatedby very few decisive accretionary and frequentlyalso hit-and-run encounters, with embryo-sized oreven larger impactors, and only rarely smaller bod-ies. Even though our model includes their collisionalevolution, they seem to play only a minor (direct)role in water transport to potentially habitable plan-ets. Beyond this first application and exciting results,we believe that this methodology offers a solid basisfor including further decisive physical processes, to-wards a deeper understanding of water transport toterrestrial planets throughout the galaxy.

317.17 — Pebble Accretion in Massive Disks

Tyler Takaro11 Astronomy & Astrophysics, University of California Santa Cruz

(Santa Cruz, California, United States)

A better understanding of protoplanetary disks iscrucial to help astronomers recreate the diversity ofexoplanets that we see in our galaxy. An increasingnumber of studies indicate that these disks may bemore massive than was previously thought, suggest-ing more material available for planetary formation.We apply one such observationally motivated model

70

of protoplanetary disks to our state-of-the-art turbu-lent pebble accretion model, in order to probe plan-etary formation in this new parameter regime. Test-ing a wide range of turbulence strengths in the outerdisk, we are able to rapidly grow injected protoplan-etary cores, exploring the conditions under whichthese protoplanets reach the masses needed for run-away gas accretion. This model predicts cores which,upon reaching a minimum mass, grow extremelyefficiently to their flow isolation masses, a naturalmass at which pebble accretion halts. Depending onwhen and where in the disk these small protoplanetsare injected, some grow to terrestrial planet masses,while others are able to reach masses required to run-away gas accretion.

317.18 — Overcoming the Meter-Size Barrier inLarge Protoplanetary Disks

Elizabeth Yunerman1; Diana Powell1; Ruth Murray-Clay1

1 University of California, Santa Cruz (Santa Cruz, California,United States)

The meter-size barrier is a persistent problem in cur-rent planet formation models, where particles on theorder of a meter in size fail to grow because they ei-ther fragment or drift into the star. Dust grain evo-lution at this size can be summarized by three char-acteristic timescales: growth, drift, and fragmenta-tion. Accurate models of these timescales that can re-solve the meter-size barrier are important to improv-ing our understanding of how planets form. Recentobservational and theoretical studies of protoplane-tary disks indicate that they are more massive thanpreviously assumed. Using our analytic model, wefind that with a more massive disk, the dust grainsin the outer disk are initially dominated by the drifttimescale, and as they move inward become dom-inated by the growth timescale. These grains areable to drift into the inner disk before collisionallyfragmenting, and then grow unimpeded. We adaptthe two-population dust evolution numerical modelfrom Birnstiel et al. (2012;15) to include necessarydrag regimes and verify that with larger protoplan-etary disks, particles can survive the meter-size driftand fragmentation barriers, and continue to grow.

317.19 — Delivery of ammonia ice to Ceres by peb-ble accretion

Yuto Nara1; Satoshi Okuzumi1; Hiroyuki Kurokawa21 Tokyo Institute of Technology (Tokyo, Japan)2 Earth-Life Science Institute, Tokyo Institute of Technology (Tokyo,

Japan)

Ceres is the largest asteroid in the solar system, com-prising almost one-third of the total mass of the aster-oid belt. The spectrometer onboard the Dawn space-craft revealed the presence of ammoniated phyl-losilicates across the surface of Ceres (De Sanctis etal. 2015), suggesting that Ceres contained ammoniawhen it differentiated. However, ammonia alone isunstable on the surface of present-day Ceres, wherethe maximum temperature (≈ 240 K) is well abovethe sublimation temperature of ammonia ice. Thiscould imply that Ceres was born in the cold outerpart of the solar nebula and subsequently migratedto the current orbit. Another possibility is that Ceresformed in situ but the solar nebula was cold enoughto preserve ammonia ice even at the current orbit ofCeres. In this study, we examine the latter scenarioby quantifying how much ammonia ice could havebeen delivered to Ceres in the solar nebula. We usea standard viscous accretion disk model to infer howthe temperature of the solar nebula decreased withtime. We also simulate the coagulation, radial in-ward drift, and sublimation of ammonia-bearing icyparticles in the background gas disk to compute theradial mass flux of the particles at 3 au as a func-tion of time. The mass flux is then converted intothe accretion rate of ammonia-bearing ice by Ceresand smaller asteroids using the state-of-the-art an-alytic formula for pebble accretion (Visser & Ormel2016). We find that the thickness of the ammonia-bearing ice layer forming on Ceres depends signifi-cantly on the initial mass Mdisk and dimensionlessviscosity parameter α of the solar nebula. A layerthick enough to globally cover the surface of Ceresforms when Mdisk < 10−2Msun and α > 10−3. Ourresults thus provide unique constraints on the fun-damental parameters of the protoplanetary disk thatformed the solar system. We also find that the layerthickness is highly sensitive to the initial asteroidmass, possibly explaining why ammoniated phyl-losilicates are observed in the largest asteroid Ceres.

317.20 — How scales of streaming and Kelvin-Helmholtz instabilities regulate particle over-densities in protoplanetary disks

Konstantin Gerbig1,2; Ruth Murray-Clay1; HubertKlahr2

1 UC Santa Cruz (Santa Cruz, California, United States)2 PSF, Max-Planck-Institut for Astronomy (Heidelberg, Germany)

The formation of planetesimals in protoplanetarydisks is an exciting and still unsolved problem inplanet formation theory. A promising scenario toovercome dust growth barriers is the spontaneousformation of planetesimals via gravitational collapse

71

of locally over-dense clumps. As these regions areregulated by numerous fluid-dynamical instabilitiesthat result from the frictional coupling of dust andgas, the interplay of the scales of these mechanismsis of great interest. We employ the Pencil Code to nu-merically investigate the scales of Kelvin-Helmholtzinstability (KHI) in the particle layer and streaminginstability (SI), which was recently identified as theepicyclic resonant drag instability. While the KHIis known for setting the vertical scale of the settleddust layer and thus preventing planetesimal forma-tion in a thin, gravitationally unstable disk, the SIwas previously shown to lead to significant particleover-densities, sufficient for local gravitational insta-bility. In addition to the global pressure gradient,which drives both KHI and SI by providing a rela-tive dust-to-gas velocity, our study focuses on the ef-fect of dust-to-gas ratio, which in combination withthe particle Stokes number, quantifies frictional cou-pling essential to both instabilities. Provided SI is ac-tive, i.e.for large enough particle Stokes numbers, wefind that the vertical extent of the dust layer is pre-dominantly set by SI-induced zonal flows in lieu ofthe KHI scale. Moreover, for both very high and verylow dust-to-gas ratios KHI-induced particle stirringis relatively weak, therefore allowing particles to set-tle very thin, and thus potentially leading to a grav-itationally unstable disk mid-plane despite KHI tur-bulence. Further, we find that KHI and SI are sur-prisingly similar in nature. In particular, both thedust extent set by KHI and the fastest growing SImode scale linearly in gas pressure gradient. Finally,we discuss implications for initial planetesimal sizesand how and under which circumstances these cancorrelate with the scales of KHI and SI.

317.21 — Warm-start planets from core accretion,and H α from accreting planets: Thermal and ra-diative properties of the accretion shock

Gabriel-Dominique Marleau1,2; Yuhiko Aoyama3; RolfKuiper1; Masahiro Ikoma3; Christoph Mordasini2

1 Universität Tübingen (Tübingen, Germany)2 Universität Bern (Bern, Switzerland)3 Department of Earth and Planetary Science, University of Tokyo

(Tokyo, Japan)

In the core-accretion formation scenario of gas gi-ant planets, most of the gas accreting onto a planetis likely processed through an accretion shock. Thisshock is key in setting the forming planet’s structureand thus its observable post-formation luminosity,and the radiative feedback can change the thermaland chemical structure of the circumplanetary and

local circumstellar disc. Also, direct evidence for on-going accretion has been provided very recently forPDS 70 b and c, and more forming planets are ex-pected in the near future thanks to ongoing and newsearches with e.g. SPHERE or MUSE.

We present the first dedicated radiation-hydrodynamical simulations of the planetaryaccretion shock, using non-equilibrium radiationtransport with up-to-date opacities (Marleau etal. 2017, 2019). We derive shock properties fora large grid of parameters. We find that usually,the temperature of the shock is given by the ”free-streaming” limit. At very high accretion rates, themassive Rosseland opacity of the gas raises theshock temperature dramatically, an effect which hasnot been discussed explicitly before. We comparethese results to original semi-analytical derivations.Additionally, we compute the fraction of the totalaccretion energy that is brought into the planetand find it is significant compared to the internalluminosity, supporting the hot-start scenario andsuggesting that young planets are luminous.

Finally, using the non-LTE radiation-hydrodynamics code of Aoyama et al. (2018),we present the first predictions of hydrogen-lineemission (H α, Pa beta, Br gamma, etc.) from the ac-cretion shock on the surface of the planet (Aoyama,Marleau et al., in prep.). We compare with PDS 70b and c and derive joint constraints on each planet’smass and accretion rate.

317.22 — An interpretation of exoplanet masses andorbital radii with a theoretical model of gas giantformation

Takayuki Tanigawa1; Kiyoka Murase2,3; HidekazuTanaka3

1 National Institute of Technology (Ichinoseki, Japan)2 National Institute of Polar Research (Tokyo, Japan)3 Tohoku University (Sendai, Japan)

About 4000 exoplanets have been detected and theirstatistical properties such as mass and orbital distri-butions become clear. However, the origin of thesedistributions is still uncertain. In this study, apply-ing a recent core accretion model to data of exoplan-ets, we clarify whether it is possible to explain theobserved distributions of masses and orbital radii,and in what protoplanetary disk they can be explain.We adopt the following theoretical model of gas gi-ant formation in this study. We use the model ofTanigawa & Tanaka (2016) for the mass growth rate(i.e., the gas accretion rate onto a planet) and the lat-est model of Kanagawa et al.(2018) for the planetary

72

migration speed. Since the growth rate and the mi-gration speed of a planet are proportional to the disksurface density nearby, a model for global evolutionof protoplanetary disk is also necessary. We also takeaccount of disk dissipation due to the photoevapora-tion. From this model, we obtained evolution curveson the plane of the planet mass and orbital radius.The evolution curves are also expressed in an ana-lytical form. Since both the growth rate and migra-tion speed of a planet are proportional to the disksurface density nearby, the curves do not depend onthe detail of the disk model. Such exoplanets do notmigrate much during their formation. Even for thelargest exoplanet (of which mass is 2% of the starmass), its orbital radius is reduced moderately (to1/10). Many gas giants are located at around 2AUfrom their stars. This can be explained as solid coresare easily formed near snow lines of disks at ∼3AU.The termination of growth of each gas giant planet ismainly determined by the initial disk mass and thedisk dissipation rate due to photoevaporation. Thusfor a given disk dissipation rate, we can obtain initialdisk masses corresponding to each planetary masses.We find that the peak of our initial disk mass dis-tribution agrees with the most common disk massin observations (∼0.02 solar masses, Andrews et al.2010) if the disk dissipation rate is set to be ∼3x10−9

solar mass per year.

317.23 — Planetesimal Population Synthesis: arePlanetesimals formed in Pressure Bumps?

Christian Lenz11 Max Planck Institute for Astronomy (Heidelberg, Germany)

Planetesimals, the smallest building blocks of plan-ets that are gravitationally bound, are believed to beneeded in order to form planets. The spacial distri-bution of these objects that are typically 100 km indiameter is important for the outcome of planet for-mation. There exist an entire zoo of different disk in-stabilities that can cause vortices or zonal flow withinwhich pebbles can be trapped and form planetesi-mals via gravitational instability. We parameterizedthese traps and implemented a pebble flux-regulatedplanetesimal formation rate into a dust and gas evo-lution code. We found that the radial planetesi-mal distribution spans from the regions of terrestrialplanet formation up to the Kuiper belt and is steeperthan the initial profile of dust and gas. The latterfinding indicates that the feeding zone, i.e. the re-gion from which the material forming planetesimalsoriginates, can be large. The final spacial planetesi-mal distribution strongly depends on the turbulence

strength from which we can conclude that the SolarNebula was not very turbulent (α<10−2).

317.24 — Planet Population Synthesis: the Cradleof the TRAPPIST-1 Multiplanet System

Martin Schlecker11 Max Planck Institute for Astronomy (Heidelberg, Germany)

Planet Population Synthesis is a statistical approachthat serves as a bridge between theoretical planetformation and the observed population of exoplan-ets. It has led to testable predictions, such as thenow-confirmed minimum in the planetary mass dis-tribution between a few Earth masses and 40 Earthmasses. In order to apply this technique to low-mass host stars, we have extended the Bern modelof planet formation (Mordasini et al. 2009) to thedifferent conditions in their protoplanetary disks.Changes to the original setup include a smaller in-ner disk radius and a down-scaled disk mass distri-bution.

We present a population of systems with a hoststar mass of 0.1 Solar masses which we compare toobservables of the TRAPPIST-1 multi-planet system(Gillon et al. 2017). We find that most of its fea-tures can be robustly reproduced. Using the meanplanetary mass as a metric, we find a domain in ini-tial disk solid mass and disk extent favorable forthe formation of similar systems. The fact that awell-established formation model can produce sim-ilar systems with little additional assumptions sug-gests that TRAPPIST-1 is not an exotic outlier but arather typical outcome for very-low-mass systems.This raises important implications for exoplanet de-mographics at the limit of detectability. ”

317.25 — The Formation of Jupiter’s Diluted Coreby a Giant Impact

Shangfei Liu6; Yasunori Hori1; Simon Müller5; Xi-aochen Zheng2; Ravit Helled5; Doug Lin4; AndreaIsella3

1 Astrobiology center (Mitaka, Tokyo, Japan)2 Department of Astronomy, Tsinghua Univeristy (Beijing, China)3 Department of Physics and Astronomy, Rice University (Houston,

Texas, United States)4 Department of Astronomy and Astrophysics, University of Cali-

fornia Santa Cruz (Santa Cruz, California, United States)5 Center for Theoretical Astrophysics and Cosmology, University of

Zurich (Zurich, Switzerland)6 School of Physics and Astronomy, Sun Yat-sen University

(Zhuhai, China)

The Juno Mission is designed to measure Jupiter’sgravitational field with an extraordinary precision.

73

Structure models of Jupiter that fit Juno gravity datasuggest that Jupiter could have a diluted core anda total heavy-element mass ranging from ten to twodozens of Earth masses. In that case the heavy el-ements are distributed within an extended regionwith a size of nearly half of Jupiter’s radius. Planetformation models indicate that most of the heavy el-ements are accreted onto a compact core, and that al-most no solids are accreted during runaway gas ac-cretion (mainly hydrogen and helium, hereafter H-He), regardless to whether the accreted solids areplanetesimals or pebbles. Therefore, the inferredheavy-element mass in the planet cannot signifi-cantly exceeds the core’s mass. The fact that Jupiter’score could be diluted, and yet, the estimated to-tal heavy-element mass in the planet is relativelylarge challenges planet formation theory. In thiswork, we show that sufficiently energetic head-oncollisions between additional planetary embryos andthe newly emerged Jupiter can shatter its primordialcompact core and mix the heavy elements with theouter envelope. This leads to an internal structureconsistent with the diluted core scenario which isalso found to persist over billions of years. A similarevent may have also occurred for Saturn. We suggestthat different mass, speed and impact angle of theintruding embryo may have contributed to the struc-tural dichotomy between Jupiter and Saturn.

318 — Orbital Dynamics andPlanet-Planet Interactions, PosterSession318.01 — The dynamics of the planetary system ofKepler-90

Silvia Giuliatti-Winter1; Daniel Gaslac2; Othon Winter31 Mathematics, UNESP (Guaratingueta, São Paulo, Brazil)2 Physics, UNESP (Guaratinguetá, São Paulo, Brazil)3 Mathematics, UNESP (Guaratinguetá, São Paulo, Brazil)

The planetary system of Kepler-90 presents somesimilarities to our solar system. This system hasseven planets, b, c, d, e, f, g and h, in increasing dis-tance from the star. The outer planet has an orbitaldistance equals to 1 AU, it is a compact system. Whileplanets g and h are gas giants, the planets d, e and fare super-Earths and planets b and c have sizes belowto 2 Earth radii. Small planets are closer to the starand the larger ones are distant from the star. Numer-ical simulations performed by Cabrera et al (2014)have shown that some of these planets are in meanmotion resonances (MMR) between them. Planets b

and c are in 4:5 MMR, and planets d, e, and f are closeto 2:3:4 MMR.

Through frequency analysis and long term evo-lution of the planets we will analyse their stabilityand the region surrounding them for a sample ofparameters of the planets, such as their mass, semi-major axis and eccentricity. Preliminary results haveshown that the system is stable for a period of 105

years when the eccentricities are assumed zero andthe masses of the planets are derived from the paperby Granado Contreras & Boley (2018), while with theparameters derived from the work by Cabrera et al(2014) one of the planets is ejected from the system.

318.02 — Dynamical evolution of extrasolar plane-tary systems HD 141399, HD 160691 and HD 39194

Alexander Perminov1; Eduard Kuznetsov11 The chair of astronomy, geodesy, ecology and environmental mon-

itoring, Ural Federal University (Ekaterinburg, Sverdlovsk oblast,Russian Federation)

The orbital evolution of four-planetary systems HD141399, HD 160691 and three-planetary system HD39194 is considered in this work. The motion equa-tions of planetary problem are constructed analyti-cally up to the second degree of planetary masses.The Hamiltonian of the planetary problem is writ-ten in Jacobi coordinates, and it is expanded into thePoisson series in the second system of Poincare ele-ments. The Hamiltonian expansion is constructed upto the fourth degree of eccentric and oblique Poincareelements. The averaging process of the Hamiltonianis performed by Hori-Deprit method and the motionequations are constructed in averaged elements.

The numerical integration of motion equations isperformed on time interval 1 Myr for the set of ini-tial conditions in which unknown and known withuncertainties orbital elements vary within allowablelimits. All planets in these systems are discovered bydetecting of Doppler shift of their radial velocities.So, only lower limits of values of planetary massesknown for studied extrasolar planetary systems. Thevalues of orbital eccentricities and pericenters knownfrom observations with uncertainties. Orbital in-clinations and ascending nodes are not known andvary. Dynamical features of chosen extrasolar plan-etary systems are studied. Stability and resonantproperties of these systems are defined by analysisof the integration results.

The limits of change of the orbital elements are de-termined depending on the initial conditions of themodeling process. The assumption about the stabil-ity of observed planetary systems allows us to elim-inate the initial conditions leading to the extreme

74

growth of orbital eccentricities and inclinations. It isshown a way to identify initial conditions in whichthe orbital elements remain small in the whole inter-val of the modeling. It becomes possible to limit therange of possible values of unknown orbital elementsand determine their most probable values in terms ofstability.

The work is founded by Russian Foundation forBasic Research, grant No. 18-32-00283 (A. Perminov)and Act 211 of Government of Russian Federation,contract No. 02.A03.21.0006 (E. Kuznetsov).

318.03 — About Properties of Retrograde Co-Orbital Motion

Vladislav Sidorenko11 Keldysh Institute of Applied Mathematics RAS (Moscow, Russian

Federation)

Most of objects in the Solar system move around theSun in the anticlockwise manner when seen fromabove the north ecliptic pole. And only a small num-ber of celestial bodies move in opposite direction(retrograde motion). The first retrograde asteroidDioretsa was discovered only in 1999. From year toyear the number of detected retrograde objects in-creased rapidly, but even now MPC data base con-tains only about a hundred of such bodies. Some ofthem can be classified as Centaurs, the other are ei-ther Main Belt asteroids (MBA) or Trans-Neptunianobjects (TNO).

Similar to prograde motion, the retrograde motionof a celestial body can also be in resonance with oneof the major planets. For example, the asteroid 2015BZ509 is in retrograde 1:1 mean motion resonance(MMR) with Jupiter [1]. Theoretical studies demon-strated that such a resonance can prevent collisionwith the planet and ensure a long stay of the aster-oid in this mode of motion.

Three dynamical processes can be distinguished atMMR: ”fast” process corresponds to planet and as-teroid motions in orbit, ”semi-fast” process is vari-ation of the resonance argument (which describesthe relative position of the planet and the asteroidin their orbital motions), and, finally, ”slow” processis the secular evolution of the orbit shape (character-ized by the eccentricity) and orientation (it dependson the ascending node longitude, inclination and ar-gument of pericenter).

By means of numerical averaging over the ”fast”and ”semi-fast” motions. we construct the evolution-ary equations that desribe the long-term behavior ofthe asteroid’s orbital elements (the ”slow” process) inthe case of the retrograde 1:1 MMR. These equations

allow us to reveal new properties of the retrogradeco-orbital motion.

[1] Wiegert, P., Connors, M., Veillet, C.: A retro-grade co-orbital asteroid of Jupiter. Nature, Vol. 543,pp. 687-689, 2017.

318.04 — Tidal Formation of Binary Planets DuringPlanet-Planet Scattering

Makiko Nagasawa1; Sota Arakawa2; Shigeru Ida21 Kurume University (Kurume-city, Fukuoka, Japan)2 Tokyo Institute of Technology (Meguro-ku, Tokyo, Japan)

When more than three Jovian planets are formed ina relatively close distance, their orbits become unsta-ble and planet-planet scatterings are caused. Duringthe unstable stage, two planets can approach withinseveral times of their physical radius. If enough tidaldissipation occurs between two planets, the planetsbecome a binary and they start to orbit around thecentral star rotating around each other. It is knownfrom numerical simulations assuming the dynamictide of about 100 systems by Ochiai et al. (2015) thatthe binary planets are formed with a probability ofabout 10% regardless of the semimajor axes. How-ever, it is still unknown why it does not depend onthe semimajor axis, why it is about 10%, and how itchanges when planetary size or the tidal models arechanged. We found that these questions can be ex-plained from the nature of the planet-planet scatter-ings. Unlike the HJ formation, once the planets arescattered strongly, it becomes almost impossible tocapture the planet by tidal force. The ability of tidalcapture of a planet is expressed as a function of thedistance between the planets. The distance betweenthe planets in the early stage of scattering is deter-mined from gravitational orbital evolutions of point-mass planets, without depending on the tidal modeland the planetary size. Therefore, if the relative ve-locity and mutual distance are recorded, probabilityof binary formation can be relatively well estimatedeven after the gravitational simulations. We per-formed many numerical simulations of early stageof planet-planet scatterings, and normalized the re-sults using physical radius and hill radius. Our re-sults make it possible to determine the probabilityof planet-planet collision, formation of binary plan-ets, and scatterings following HJ formation and col-lision to the star as a function of tidal strength. Wewill report how the probability of formation of bi-nary planet changes with respect to the initial condi-tions of the system such as the tides, initial eccentric-ity, and planetary size. We will also see that the rateof binary planets relative to the numbers of HJs.

75

318.05 — Stability of closely packed multiple exo-planetary systems

Su Wang1; Doug Lin2; Jianghui Ji11 Purple Mountain Observatory (Nanjing, China)2 University of California Santa Cruz (Santa Cruz, California,

United States)

There are more then 4000 exoplanets have been ob-served up to now and hundreds of multiple plan-etary systems among them. Many planet pairsare in the configuration of mean motion resonances(MMRs) or near MMRs. We investigate the stabil-ity of multiple planetary systems to find out if theMMRs configuration is more stable for planets to sur-vive. Through numerical simulation on equal-massmultiple planetary systems, we find out that (1) Ifall planets undergo inward type I migration, planetpairs tend to be more stable with the increase ofthe relative separation between them; (2) If there areboth inward and outward migration exist in the sys-tem, the separation between planet pairs will shrinkto smaller Hill Radius and planet pairs can find theirstable configuration if they are captured into MMRs.But the stable region is very narrow; (3) The densityof the gas disk is related to the final separation be-tween planet pairs and when they are captured intoMMRs. If planet pairs migrate to small separationwith larger gas density, the system are easy to becrossing. The resonances are deeper in the systemwith larger depletion timescale and the system aremore stable (4) When the libration timescale of planetpairs larger than the migration timescale which arecaused by the mass loss of planets in the system, thestable configuration can be destroyed.

318.06 — Can a flyby induce misalignment in aplanet-hosting disc?

Rebecca Nealon11 Physics and Astronomy, University of Leicester (Leicester, United

Kingdom)

We now have several observational examples ofstrongly misaligned broken discs, where the orien-tation of the disc can change rapidly as a function ofdistance from the central star. Current models sug-gest that such discs are generated through the influ-ence of a stellar or planetary companion that is or-biting in a plane misaligned to the mid-plane of thedisc. Such a companion causes the disc to separateinto an inner and outer disc, both of which responddifferentially to the presence of the misaligned com-panion leading to a relative misalignment betweenthe discs. Here we explore whether a strong mis-alignment between the inner and outer disc could

be formed without a misaligned inner companion.Instead, we invoke an aligned planetary companionin the disc that carves a gap (essentially separatingthe disc into two regions) and use a flyby encounterto disturb the discs. We find strong misalignmentscan be generated and examine these in the context ofexisting observations of strongly misaligned brokendiscs.

318.07 — Are Closely-spaced RV Planets the Prod-uct of Smooth Migration?

Sam Hadden1; Matthew J. Payne11 Harvard-Smithsonian CfA (Cambridge, Massachusetts, United

States)

A number of RV-discovered planetary systemsaround A stars host closely-spaced (period ratios <2),Jovian-mass planet pairs. These planets are not tra-ditionally expected to form in such close proxim-ity and therefore are thought to have experiencedpost-formation migration. Capture into mean mo-tion resonances (MMRs) is a natural dynamical out-come of convergent planetary migration and radialvelocity modeling suggests that many of these plan-ets are indeed in resonance. In the simplest smoothmigration scenarios, dissipative forces should driveplanets into a very specific resonant configuration,sometimes referred to as an ‘apsidal corotation reso-nance’ or ACR, that represents a fixed point of theresonant dynamics. Determining whether the dy-namical states of RV planets are consistent with theoutcomes of simple smooth migration is a naturalstarting point for better understanding the role ofmigration in planet formation. We assess whetherclosely-spaced Jovian planet pairs can be explainedas the product of smooth migration using a Bayesianmodel comparison framework to compare a tradi-tional two-planet model of each pairs’ RV observa-tions (that does not assume resonance) to a modelthat assumes the planet pair to be in an ACR, thusrestricting the number of free parameters. We findthat most RVs can be satisfactorily explained by ACRconfigurations and discuss the implications of plan-ets’ inferred dynamical states for their formation his-tories.

318.08 — Tidal evolution and formation for Prox-ima b

Yao Dong11 Purple Mountain Observatory,Chinese Academy of Sciences

(Nanjing, China)

76

Recently, Anglada et al. (2016) discovered a terres-trial planet Proxima b orbiting around Proxima Cen-tauri in the habitable zone, with the semi-major axisof about 0.05 au and a possible eccentricity less than0.35. Proxima Centauri is the Sun’s closest neighborat present, therefore, Proxima b will provide goodopportunities to investigate Earth-like planets out-side our solar system. Almost overnight, many in-vestigations have been focused on this planetary sys-tem mainly including planet formation, the interiorsand the habitable zone et al. We carry out the numer-ical simulations of the dynamical evolution of Prox-ima b based on the equilibrium tidal model. Theorbital circularization during tidal decay is within20 Myr when assumed the modified tidal dissipa-tion parameter Q=6. The planetary orbit-spin rota-tion reaches synchronous in ten thousand years. Var-ious orbital parameters are considered in studyingthe spin-orbit 1:1 resonance. To explain the observ-able eccentricity of Proxima b, we explore the dy-namical evolution using a modified secular theory,assuming that there is one candidate planet aroundProxima Centauri. The candidate planet excites asteady eccentricity of Proxima b before its tidal expe-rience. Finally, the planet formation of Proxima b viatype I migration in the gaseous disk is investigated,as the result shows that Proxima b reaches the innerdisk with a period of 14 days, then suffers tidal decayto current orbit.

318.09 — Trojan Terrestrial Planets: stability andformation

Othon Winter1; Luana Mendes11 UNESP (Guaratingueta, São Paulo, Brazil)

Up to the present moment, no co-orbital planet hasbeen found. Trojan terrestrial planets sharing theirmean orbit with a giant planet in the habitable zonemight harbor conditions of habitability. Therefore,this is a relevant subject of study. In the presentwork we explore the size and shape of the stabilityregion co-orbital to a giant planet. Massless parti-cles initially displaced in the whole co-orbital regionwere integrated for 7x105 orbital periods of the giantplanet. The initial positions of those particles that re-mained in the co-orbital region along the whole sim-ulation determined the location and size of the stableregion. A wide range of giant planet’s mass was con-sidered. The highest limiting mass value that allowsstable horseshoe trajectories is found and the anal-ysis of the results is divided into two blocks: tad-pole and horseshoe stable regions. From the resultsof the tadpole regions were measured the minimumand maximum angular location, and the radial width

around the triangular equilibrium points. In the caseof the horseshoe regions are measured the minimumand maximum angular location, and the radial widtharound the equilibrium point L3. All measurementsgenerated empirical expressions as a function of thegiant planet mass. Adopting the stable regions, sim-ulations with massive particles whose collisions re-sulted in perfect merging were performed. The fi-nal outputs of such simulations generated larger co-orbital bodies that, in many cases, reached terrestrialplanet masses.

318.10 — Possible 3D configurations of RV-detectedextrasolar systems

Anne-Sophie Libert1; Mara Volpi1; Arnaud Roisin11 University of Namur (Namur, Belgium)

Due to the limitations of the radial velocity tech-nique, no indication on the spatial architecture of gi-ant planetary systems can usually be given by theobservations. We aim to constrain the 3D configu-ration of several RV-detected two-planet extrasolarsystems. Through an analytical study based on afirst-order secular Hamiltonian expansion and nu-merical explorations performed with a chaos detec-tor, we identify ranges of values for the orbital incli-nations and the mutual inclinations which ensure thelong-term stability of the system. We find that long-term regular evolutions of 3D configurations existfor all the selected systems, either at low mutual in-clinations, or at high mutual inclinations preferen-tially if the system is in the Lidov-Kozai resonance.A rapid destabilization of highly mutually inclinedorbits is commonly observed, due to the significantchaos that develops around the stability islands ofthe Lidov-Kozai resonance. Finally, for planetarysystems with close-in planets, we show how the rela-tivistic effects influence the extent of the Lidov-Kozairesonance region.

318.11 — Cometary shape and spin-axis evolutiondue to long term solar driven outgassing

Yuhui Zhao11 Purple Mountain Observatory, CAS (Nanjing, China)

In this work we investigate the role of long termsublimation effects on reshaping the cometary nu-cleii using 3D shapes coupled with realistic spin-orbit evolution. We try to classify the typical mor-phological changes that can result from solar drivenoutgassing starting from various initial conditions.Our model includes different 3D shapes account-ing for shadowing and self-heating, orbital elements,

77

orientation parameters and CO sublimation drivencometary activity. These results will provide con-straints on the limits pure sublimation activity canprovide in terms of nucleus shape and large scalemorphology changes during long term evolution of asmall icy body in the Kuiper Belt. These simulationsprovide evidence that certain shape structures (suchas observed on the 67P/CG) cannot be produceddue to sublimation alone, indicating that other shapedefining processes at work (e.g. Jutzi et al, 2017).Similarly, we investigate evolution of spin rates andaxis orientations for different nuclei shapes.

318.12 — Transit and Radial velocity Interactive Fit-ting tool for Orbital analysis and N-body simula-tions: The Exo-Striker

Trifon Trifonov11 MPIA (Heidelberg, Germany)

I present a new very powerful and fast GUI toolfor exoplanet orbital analysis and N-body simula-tions. It uses a brand new RV fitting library called”RVmod”, which can model the stellar reflex mo-tion caused by dynamically interacting planets inmulti-planetary systems. The ”Exo-Striker” tool of-fers a broad range of tools for detailed analysis oftransit and Doppler data. Some of the key fea-tures of the tool are: – Period search via powerspectrum analysis (GLS for RVs & TLS for transitdata) – Keplerian and dynamical RV modeling ofmulti-planet systems – Transit, or joint Transit andRV modeling – Gaussian Processes modeling – Pa-rameter optimization (Simplex, LM, TNC, Powelland many more) – MCMC (via emcee) and Nestedsampling (via dynesty) – Long-term stability checkof multi-planet systems – Fully interactive, high-quality, exportable plots ready for publication – Im-port/export of working sessions – Export of ready-to-use LaTeX tables with best-fit parameters, errorsand statistics – Text editor, Bash-shell (Linux only)and a fully Integrated Jupyter shell – ”RVmod” en-gine can be used as stand alone standard Python li-brary (i.e. without the Exo-Striker GUI, simply as“import RVmod as rv”) The tool is cross-platformcompatible (works on MAC OS (10.6+), Linux (Suse,Mint, Ubuntu, etc.) and Windows 10), and it com-bines Fortran efficiency and Python flexibility. The”Exo-Striker” can be found on the ”github” underhttps://github.com/3fon3fonov/trifon.

318.13 — Updates on the GPU N-body codeGENGA and TTV calculations

Simon Lukas Grimm1

1 Center for Space and Habitability, University of Bern (Bern,Switzerland)

We present updates on the N-body code GENGA.These include symplectic treatment of non-Newtonian forces for orbital dynamics, especiallyalso for small bodies. We also present modificationsof the hybrid symplectic integration scheme, whichallows the integration of large close encountergroups. This is necessary when simulations of fullyself graviting disk with 10’000 to 100’000 bodies areperformed. It is also shown how GENGA is used toperform TTV analysis calculations.

318.14 — Comparison of machine learning tech-niques for emulating collisions in planet formation

Miles Timpe1; Maria Han Veiga1; Mischa Knabenhans11 Institute for Computational Science, University of Zurich

(Zurich, Switzerland)

Collisions between planetary embryos are a funda-mental agent of planet formation. However, the sig-nificant computational cost of simulating collisions,coupled with the high-dimensionality of the param-eter space, has so far frustrated a comprehensive un-derstanding of their role in planet and satellite for-mation. Indeed, only limited regions of the parame-ter space have been explored, with most studies fo-cused on narrow problems in planetary science, suchas the origin of Earth’s moon (Benz et al. 1986) orMercury’s large core (Chau et al. 2018). Follow-ing the recent success of emulation techniques ap-plied to high-dimensional problems in cosmology(Knabenhans et al. 2019) and ongoing attempts toclassify and predict collision outcomes in planet for-mation (Cambioni et al. 2019, Valencia et al. 2019),we investigate the ability of different uncertaintyquantification and machine learning techniques toaccurately emulate pairwise collisions. In order tocarry out this study, we have simulated an unprece-dented set of 10,000 pairwise collisions. This datasetis an order-of-magnitude larger than any previousdataset used in similar studies and explores param-eters largely overlooked in previous studies, such asthe core mass, rotation rate, and orientation of thepre-impact bodies. In addition, we present prelimi-nary results of N-body simulations using our emula-tion technique to predict collision outcomes on-the-fly.

318.15 — Comparing long-stability criteria ofhorseshoe coorbital planets

Agueda Paula Granados Contreras1; Aaron C. Boley2

78

1 Institute of Astronomy and Astrophysics, Academia Sinica(Taipei, Taiwan)

2 Department of Physics and Astronomy, The University of BritishColumbia (Vancouver, British Columbia, Canada)

Coorbital planets, should they exist, have the poten-tial to provide new constraints on the early stagesof planet formation and the subsequent dynami-cal evolution of systems. A number of recent nu-merical studies have analyzed possible transit sig-natures of coorbiting planets, as well as their sta-bility. Searches have also been carried out usingarchival Kepler (Hippke & Angerhausen 2015; Jan-son et al. 2013) and radial velocity data (e.g., theTROY project, Lillo-Box et al. 2018a,b). A few can-didate systems have been identified to have plan-ets in a 1:1 MMR using follow ground-based ob-servations, but so far there are no confirmed detec-tions. Detecting coorbital planets, especially those ina horseshoe configuration, is challenging due to thesemi-major axis exchange between the two planets,which occurs on timescales longer than the orbitalperiod. A key factor in developing tools for coorbitalsearches or in confirming possible detections is thelong-term stability of any given proposed coorbital.In this regard, different studies have shown using N-body simulations that coorbital configurations withzero eccentricity and planet-to-star mass ratio, i.e.,μ=(m1+m2)/M∗, between 2×10−6 to 2×10−3 can bestable for at least 1010 orbits. However, these simu-lations have only considered a constant timestep tointegrate their realizations and might not be able toresolve the semi-major axis exchange of the coorbitalplanets. We explore the long-term stability of two-planet systems initially at conjuction, with planet-to-star mass ratios between μ=10−4 to 10−3 and initialplanetary semi-major axis ratios from 1 to 1.065 us-ing the 15th-order Integrator with Adaptive Time-Stepping (IAS15). We compare the results with thoseof Leleu et al. (2019), Cúk et al. (2012) and, Laughlin& Chambers (2002).

318.16 — Observable Predictions from Perturber-driven High-eccentricity Tidal Migration for WarmJupiters

Jonathan Jackson1; Rebekah Ilene Dawson11 Astronomy and Astrophysics, The Pennsylvania State University

(State College, Pennsylvania, United States)

The origin of hot Jupiters (Jupiter mass planets withperiods less than 10 days) is an open question inexoplanet formation and evolution and has signif-icant ramifications for exoplanet populations, sys-tem architecture, and early disk conditions. If we as-

sume hot Jupiters migrated from larger semi-majoraxes, we can treat warm Jupiters (periods between10 and 200 days) as intermediaries between newlyformed gas giants and hot Jupiters. Warm Jupitersthus provide a test bed for assessing migration theo-ries. Our new work investigates the validity of theperturber-modulated high-eccentricity tidal migra-tion scenario, in which a migrating Jupiter is peri-odically secularly perturbed by a companion planetor star to an eccentricity large enough for its or-bit to tidally shrink and circularize. We developan observationally motivated population of warmJupiters and assign a corresponding perturber capa-ble of inducing precession faster than that of gen-eral relativity, a minimum requirement for this typeof migration, to each planet. We then calculate thedistributions of observational signals (transit timingvariations, transit duration variations, radial veloc-ities, and astrometry) these perturbers would pro-duce and compare them to current and near futuredetection limits. We show that a small percentageof these perturbers are detectable within current ob-servational limits by their transit timing variationsand that many are detectable at the 1 m/s level intheir radial velocities. With a longer time baseline,most of the perturbers would be detectable at the 10m/s level. Most perturbers within 200 pc will be de-tectable by Gaia within the primary mission lifetime.If TESS RV followup and Gaia astrometry do not dis-cover a significant number of massive companionsin warm Jupiter systems, it may suggest this mech-anism is not a common means of warm Jupiter pro-duction.

318.17 — High order regularised symplectic inte-grator for collisional planetary systems

Antoine C. Petit1; Jacques Laskar21 IMCCE, Paris Observatory (Paris, France)2 IMCCE, Observatoire de Paris (Paris, France)

We present a new mixed variable symplectic (MVS)integrator for planetary systems, that fully resolveclose encounters. The method is based on a time reg-ularisation that allows keeping the stability proper-ties of the symplectic integrators, while also reduc-ing the effective step size whenever two planets en-counter. We use a high order MVS scheme such thatit is possible to integrate with large time steps faraway from close encounters. We show that this al-gorithm is able to resolve almost exact collisions (i.ewith a mutual separation of a fraction of the physicalradius) while using the same time-step as in weaklyperturbed problem such as the Solar System. Wedemonstrate the long term behaviour on systems of

79

six super-Earths experiencing strong scattering for 50kyr. We compare our algorithm to hybrid methodssuch as MERCURY and show that for an equivalentcost we obtain much better energy conservation.

318.18 — Can a hot Jupiter host exomoons?

Alessandro Alberto Trani1; Adrian Hamers2; AaronGeller3

1 University of Tokyo (Tokyo, Japan)2 Institute for Advanced Studies (Princeton, New Jersey, United

States)3 Northwestern University (Chicago, Illinois, United States)

All the giant planets in the solar system host a largenumber of natural satellites. Moons in extrasolar sys-tems are difficult to detect, but a Neptune-sized can-didate exomoon has been recently found around aJupiter-sized planet in the Kepler 1625 system. Ofthe many extrasolar Jupiters detected so far, aboutone-tenth lies on a short period orbit. Whether thehot Jupiter population can host (or may have hosted)exomoons is still unknown. We investigated if thepresence of exoomons can allow the formation of hotJupiters, and if the exomoon can survive the tidalmigration process, by means of direct N-body sim-ulations including tidal interactions. Here we con-sider the Kozai-Lidov secular interaction induced bya distant companion star as a means to trigger thehigh-eccentricity tidal migration process. Our re-sults show that it is unlikely that a hot Jupiter canhost exomoons, since the moon will either crash intothe planet or escape from it during the migrationprocess. In some extreme cases, perturbations fromthe stellar companion can trigger dynamical insta-bilities that lead to the ejection of the planet alongwith its moon. This mechanism can explain theexomoon candidate MOA-2011-BLG-262L, a candi-date exomoon of a free-floating planet, and can ex-plain the future detections of similar systems via mi-crolensing events.

319 — Star-Planet Interactions andTides, Poster Session319.01 — How Stellar Flares and Storms RegulateAtmospheric Losses from the TRAPPIST-1 Planets

Chuanfei Dong1; Meng Jin3; Manasvi Lingam4; KevinFrance2

1 Princeton University (Plainsboro, New Jersey, United States)2 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)3 SETI Institute (Mountain View, California, United States)

4 Harvard-Smithsonian Center for Astrophysics (Cambridge, Mas-sachusetts, United States)

Stellar flares have been observed to produce a burstof radiation over a wide range of wavelengths,among which X-rays and EUV constitute the majorionizing stellar radiation for planetary atmospheresat low and high altitudes, respectively. Stellar flaresare considered an impediment to habitability, es-pecially in the case of close-in exoplanets aroundM-dwarfs since these stars are highly active. Atthe same time, there has been a growing awarenessthat coronal mass ejections (CMEs) — sometimestermed as stellar storms — associated with stellarflares pose severe threats to planetary atmosphericretention. It is evident that understanding atmo-spheric escape is vital from the standpoint of hab-itability since atmospheric evolution influences theclimate and the fluxes of ionizing radiation reachingthe surface, among other factors.

Until now, there have been no systematic studiesof the impact of stellar flares and associated stormson exoplanetary atmospheric losses despite their in-dubitable occurrence and pertinence. Here, we carryout sophisticated 3D MHD simulations (that in-cludes important photochemistry) to assess how theatmospheric escape rates of the TRAPPIST-1 planetsevolve during 1) a 1033 erg flare (based on observa-tions) without a CME (where the CME may be sup-pressed or deviated from the planet) and, 2) a 1033

erg flare with a CME, where the CME is initializedand modeled according to the flare energy by usinga stellar wind model. We found that the atmosphericescape rates are enhanced by 1-3 orders of magnitudecompared to our previous study that used normalstellar wind conditions. For the outmost TRAPPIST-1h, if such flares occur at a frequency of ∼1 per day,a 1-bar atmosphere will be scavenged on the timescale of ∼ 100 million years. This time scale reducesto ∼ 1 million years for the innermost TRAPPIST-1b.This represents the first study where the roles of stel-lar flares and storms on exoplanetary atmosphericescape for the TRAPPIST-1 planets are clearly eluci-dated. The new results obtained herein would be ofconsiderable interest to a wide audience and thus de-serving an oral presentation.

319.03 — Magnetospheres of the TRAPPIST-1 plan-ets

Adam Boldog1,2; Vera Dobos1,2; László L. Kiss1,31 Konkoly Observatory, Research Centre for Astronomy and Earth

Sciences (Budapest, Hungary)2 MTA-ELTE Exoplanet Research Group (Szombathely, Hungary)

80

3 Sydney Institute for Astronomy (Sydney, New South Wales,Australia)

The number of exoplanets in the habitable zone ofM dwarfs has increased in the last few years thanksto ground-based observatories and space telescopes.M dwarfs are among the most active stars produc-ing frequent and strong flares, strong stellar windand high energy radiation. The habitability of theseworlds strongly depends on their capability of retain-ing their atmospheres. Planetary magnetospheresplay a crucial role in reducing atmospheric loss andthus providing a potentially habitable enviroment.The M dwarf star TRAPPIST-1 is of particular in-terest because it hosts three Earth-sized rocky plan-ets in its habitale zone. We calculated the magneticproperties of these planets, such as the suface dipolarfield strength and magnetic dipole moment, usinga method based on the example of the early Earth,which assumes a process involving the exsolution ofMgO as the source of the planetary dynamo. In or-der to follow the evolution of these properties, we ap-plied a thermal evolution model for the TRAPPIST-1planets. The sizes of the magnetospheres, describedby the magnetoshperic standoff distance (i.e. the dis-tance from the planet to the point where the stellarwind is balanced by the planetary magnetic field),were derived using previously modeled stellar windparameters. Additionally, we calculated the polarcap area, which indicates the fraction of a planet’ssurface where magnetic field lines are open and at-mosperic escape is possible. Based on our results,we will estimate the atmospheric mass loss, whichcan significantly limit habitability on the planets.

319.04 — It’s raining hot Jupiters: 3D MHD simu-lations of planetary atmospheric escape

Simon Daley-Yates1; Ian Stevens11 University of Birmingham (Paris, France)

We present 3D MHD simulations of the wind-windinteractions that occur between a solar-type star anda short period hot Jupiter exoplanet. A planetaryoutflow results from atmospheric escape induced bythe host stars incident radiation. Circumstellar andcircumplanetary material which accretes onto thestellar surface in a form of coronal rain, we charac-terise this interaction for a representative hot Jupiterhosting system and predict the accretion point, sizeand extent. The nature of this accretion is variable inboth location and rate, with the final accretion pointoccurring at 133 degrees west and 53 degrees east ofthe subplanetary point. The size of the accretion spotitself has been found to vary with a period of 67 ks

(approximately 1/5 of the orbital period). The resultsare highly dependent on the magnetic fields of boththe star and the planet and on the atmospheric con-ditions of the hot Jupiter. We characterise this be-haviour as Star-Planet-Wind Interaction (SPWI).

319.05 — Interactions of tidal flows and convec-tion: frequency dependence and indications ofanti-dissipation

Craig D. Duguid1; Adrian John Barker1; Chris A. Jones11 University of Leeds (Leeds, United Kingdom)

A key mechanism in the orbital and spin evolutionof close proximity bodies is the dissipation causedby the interactions of tidal flows with convection. Itis expected that the effective viscosity of this inter-action (νE) would depend on the tidal frequency (ω)but to what extent is a matter of debate, particularlyin the regime of fast tides (e.g. Zahn 1966; Goldre-ich and Nicholson 1977). It is essential to resolvethis in order to correctly predict the tidal evolutionof hot Jupiters. We have performed the most com-prehensive investigation to date of this mechanismby way of hydrodynamical simulations and exten-sions to existing theory, building upon prior work byPenev et al. 2009 and Ogilvie and Lesur 2012. Our re-sults provide a clear scaling law for the dependenceof the effective viscosity on tidal frequency and alsoconvincing evidence which suggests that this mecha-nism can operate as anti-dissipation (which could re-sult in outward migration or excitation of eccentric-ities, contrary to prior expectations). These resultscan help guide the correct implementation of tidaldissipation for planet-star interactions. We will alsodiscuss the consequences of our results for the orbitaldecay of hot Jupiters.

319.06 — Measurements of the Ultraviolet SpectralCharacteristics of Low-mass Exoplanetary Systems(Mega-MUSCLES)

David John Wilson1; Cynthia Froning1; Kevin France2;Allison Youngblood3; Girish M. Duvvuri2; AlexanderBrown2; P. Christian Schneider4; Adam Kowalski2; R.O. Parke Loyd5; Zachory Berta- Berta-Thompson2; J.Sebastian Pineda2; Jeffrey Linsky2; Sarah Rugheimer6;Elizabeth Newton7; Yamila Miguel8; Aki Roberge3; An-drea P. Buccino9; Jonathan Irwin10; Lisa Kaltenegger11;Mariela Vieytes12; Pablo Mauas9; Seth Redfield13;Suzanne Hawley15; Feng Tian14

1 McDonald Observatory, University of Texas at Austin (Austin,Texas, United States)

2 Smithsonian Astrophysical Observatory (Cambridge, Mas-sachusetts, United States)

81

3 Cornell University (Ithaca, New York, United States)4 I. Astronomia y Fis Espacio (Buenos Aires, Argentina)5 Wesleyan University (Middletown, Connecticut, United States)6 Chinese Academy of Sciences (Beijing, China)7 University of Washington (Seattle, Washington, United States)8 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)9 Goddard Space Flight Center (Greenbelt, Maryland, United

States)10 Hamburger Sternwarte (Hamburg, Germany)11 School of Earth and Space Exploration, Arizona State University

(Tempe, Arizona, United States)12 University of Oxford (Oxford, United Kingdom)13 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)14 Leiden University (Leiden, Netherlands)15 Universidad de Buenos Aires (Buenos Aires, Argentina)

M dwarf stars have emerged as ideal targets for ex-oplanet observations. Their small radii aids plane-tary discovery, their close-in habitable zones allowshort observing campaigns, and their red spectraprovide opportunities for transit spectroscopy withJWST. The potential of M dwarfs has been under-lined by the discovery of remarkable systems suchas the seven Earth-sized planets orbiting TRAPPIST-1 and the habitable-zone planet around the closeststar to the Sun.

However, to accurately assess the conditions inthese systems requires a firm understanding of howM dwarfs differ from the Sun, beyond just theirsmaller size and mass. Of particular importance arethe time-variable, high-energy ultraviolet and x-rayregions of the M dwarf spectral energy distribution(SED), which can influence the chemistry and life-time of exoplanet atmospheres, as well as their sur-face radiation environments.

The Measurements of the Ultraviolet SpectralCharacteristics of Low-mass Exoplanetary Systems(Mega-MUSCLES) Treasury project, together withthe precursor MUSCLES project, aims to producefull SEDs of a representative sample of M dwarfs,covering a wide range of stellar mass, age, and plan-etary system architecture. We have obtained x-rayand ultraviolet data for 13 stars using the Hubble,Chandra and XMM space telescopes, along withground-based data in the optical and state-of-the-artDEM modelling to fill in the unobservable extremeultraviolet regions. Our completed SEDs will beavailable as a community resource, with the aim thata close MUSCLES analogue should exist for most Mdwarfs of interest.

In this presentation I will overview the Mega-MUSCLES project, describing our choice of targets,

observation strategy and SED production methodol-ogy. I will also discuss notable targets such as theTRAPPIST-1 host star, comparing our observationswith previous data and model predictions. Finally,I will present an exciting by-product of the Mega-MUSCLES project: time-resolved ultraviolet spec-troscopy of stellar flares at multiple targets, spanninga range of stellar types, ages and flare energies.

319.07 — Eating Planets for Breakfast, Lunch, andDinner: Signatures of Planetary Engulfment at allPhases of Stellar Evolution

Alexander Patrick Stephan1; Smadar Naoz1; B. ScottGaudi2; Jesus M. Salas1

1 Physics and Astronomy, University of California, Los Angeles(Los Angeles, California, United States)

2 Astronomy, Ohio State University (Columbus, Ohio, UnitedStates)

Most, if not all, TESS target stars can be expectedto reside in binaries, as these stars are more mas-sive than the Sun. Gravitational perturbations froma companion can drive a planet closer to its host star,potentially plunging the planet all the way into thestar. While it is challenging to observe a planet dur-ing its plunge, we have predicted that, prior to itsdemise, such a planet will appear as hot as a HotJupiter (Stephan et al. 2018). This new class of ’Tem-porary Hot Jupiters’ has recently been confirmed byTESS observations (e.g., HD 202772A b). As theplanet is eventually eaten, it can impart distinct sig-natures onto the star. We follow the engulfment ofplanets by their host stars during different stellar lifephases and calculate the changes in stellar param-eters, such as stellar spin or luminosity, caused bythis process. Our predictions for the observable sig-natures of these engulfment events will enable futureand current endeavors to find post-engulfment stars,thus, advancing our understanding of planetary sys-tem architectures and dynamical evolution.

319.08 — Impact of Stellar Magnetism on Star-planet Tidal Interactions

Aurélie Astoul1; Stéphane Mathis1; Clément Baruteau2;Florian Gallet3; Antoine Strugarek1; Kyle Augustson1;Allan Sacha Brun1; Emeline Bolmont4

1 DAP, CEA/Saclay (Bures sur Yvette, France)2 IRAP (Toulouse, France)3 IPAG (Grenoble, France)4 Département d’Astronomie, Université de Genève (Genève,

France)

Over the last two decades, about 4000 exoplanetshave been discovered around low-mass stars. For the

82

shortest period exoplanets, star-planet tidal interac-tions are likely to have played a major role in the ul-timate orbital evolution and on the spin axis evolu-tion of the host stars. Although low-mass stars aremagnetically active objects, the question of how thestar’s magnetic field impacts the excitation, propaga-tion and dissipation of tidal waves remains open.

In this work, we have derived the magnetic con-tribution to the tidal force and estimated its am-plitude all along the structural and rotational evo-lutions of low-mass stars (from M to F-type). Forthis purpose, we have used detailed grids of rotat-ing stellar models computed with the stellar evolu-tion code STAREVOL. The amplitude of dynamo-generated magnetic fields is estimated via physicalscaling laws at the base and the top of the convectiveenvelope. We find that the star’s magnetic field haslittle influence on the excitation of tidal waves in nearcircular and coplanar Hot-Jupiter systems, but thatit has a major impact on the waves dissipation. Ourresults therefore indicate that a full MHD treatmentof the propagation and dissipation of tidal waves isneeded to assess the impact of star-planet tidal inter-actions for all low-mass stars along their evolution.

319.09 — Magnetic fields of hot Jupiters calculatedfrom star-planet interactions

Paul Wilson Cauley1; Evgenya L. Shkolnik4; Joe Llama2;Antonino Lanza3

1 LASP, University of Colorado at Boulder (Boulder, Colorado,United States)

2 Lowell Observatory (Flagstaff, Arizona, United States)3 Observatorio Astrofisico di Catania, Instituto Nazionale di As-

troFisica (Catania, Italy)4 School of Earth and Space Sciences, Arizona State University

(Tempe, Arizona, United States)

Planetary magnetic fields have a critical impact onatmospheric physics, damping winds on hot, short-period planets and potentially creating the neces-sary conditions for habitability on temperate terres-trial worlds by deflecting stellar wind particles. De-spite their importance, exoplanet magnetic field de-tections remain elusive. For the first time, we re-port the derivation of the magnetic fields of a sampleof hot Jupiters using flux-calibrated signals of mag-netic star-planet interactions (SPI). We find that thesurface magnetic field values for the hot Jupiters inour sample range from 20 G to 120 G, 10 - 50 timeslarger than the values predicted by pure dynamotheories for planets with rotation periods of 2 to 4days. Such large field strengths should have severeconsequences for velocity flows in the planets’ at-mospheres and exhibit peak frequencies of electron-

cyclotron emission in the range of facilities such asLOFAR.

319.10 — Exploring the Stellar CME-flare Relation:from Historic Events’ Analysis to Stellar ActivityModeling

Sofia Moschou1; Jeremy J. Drake1; Ofer Cohen2; Ce-cilia Garraffo1; Julian D. Alvarado-Gomez1; FedericoFraschetti1

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 University of Massachusetts Lowell (Lowell, Massachusetts,United States)

A crucial aspect of habitability is the space environ-ment of the planet, which can be extreme and violentfor short-orbit planets or planets orbiting active M-dwarfs. CMEs, flares and energetic particles are themost dramatic stellar activity events. How the storedmagnetic energy gets distributed between these dif-ferent energetic coronal phenomena remains a vitalfield of research. In the Sun, more energetic solar X-ray flares are associated with faster and more mas-sive CMEs (e.g. Drake et al 2013). While highly ener-getic flares are continuously observed in active stars,such as M-dwarfs, no stellar CME has been defini-tively observed yet (e.g. Crosley & Osten 2018). Inthis presentation we discuss our most recent unpub-lished results (paper accepted by ApJ) on the stellarCME-flare relation by examining the most probablehistoric CME candidates, all of which were observedon magnetically active stars. We use the CME conemodel to infer masses and kinetic energies from ob-served quantities, and convert the associated emis-sion to the GOES 1–8 Å band. When the inferredCME masses, found in the range ∼1015 to 1022 g,are presented as a function of associated flare en-ergy they lie on the solar extrapolated trend. Thekinetic energies, found in the 1031 to 1037 erg, how-ever, lie below the extrapolated relation used on so-lar events. This is an indication that in the stellarregime there is an energy partition between flare X-ray and CME kinetic energies, contrary to the solarcase where X-ray flares have 100 times less energythan their associated CMEs. A possible mechanismresponsible for constraining the CME speeds morethan their masses is the effect of strong large-scaleoverlying magnetic fields. Stellar CME with lower ki-netic energies present an optimistic scenario for theimpact of mass ejecta on close in exoplanets relativeto their stellar hosts.

83

319.11 — Precise characterisation of the 55 Cnc andHD219134 transiting exoplanetary systems

Roxanne Ligi1; Caroline Dorn2; Aurélien Crida3,4;Francesco Borsa1

1 INAF - Osservatorio Astronomico di Brera (Merate, Italy)2 University of Zurich (Zürich, Switzerland)3 Observatoire de la Côte d’Azur (Nice, France)4 Institut Universitaire de France (Paris, France)

The harvest of exoplanets detection has led to thequest of their characterisation. In particular, the de-termination of exoplanet masses, radii and densitiesis mandatory to derive their bulk composition, itselfrelated to their formation and potential habitability.However, these parameters totally rely on their hoststar parameters. Indeed, the transit method (resp.radial velocity measurements) provide the ratio ofplanetary to stellar radius (resp. mass). When bothmethods can be applied to a system, then the plan-etary density can be obtained. Unfortunately, mostof transiting exoplanets hosts are too faint for theirmass and radius to be accurately determined, lead-ing to approximate or imprecise parameters. We willpresent two systems that host transiting exoplanets,55 Cnc and HD219134. Since the stars are bright, weperformed interferometric observations to measuretheir angular diameter, leading to accurate radii R.Using the transit light curves, we also directly derivetheir density. We thus obtain an independent mea-surement of their mass M. More precisely, we com-puted the joint probability density function of theseparameters, and the correlation between R and M.This allows to derive the planetary parameters, inde-pendently from any stellar evolution model. Usingthe R-M correlation, the planetary density can be re-fined, and subsequently the internal composition ofthe transiting exoplanets. Contrary to what previousstudies claim, the transiting exoplanet 55 Cnc e mayonly have a thin atmosphere and its interior struc-ture might not be dominated by carbon. The systemof HD219134 hosts two transiting exoplanets of sim-ilar properties, but HD210134 b shows a higher den-sity while smaller radius and mass than HD219134 c.This could be explained by a molten interior possiblyinduced by tidal heating caused by a high eccentric-ity during its formation. Those systems are bench-marks to investigate both exoplanet and stellar prop-erties in detail thanks to our method that providesthe best characterisation of these systems so far. Itwill be generalised with the coming bright TESS andPLATO targets.

319.12 — Evryscope & K2 Constraints onTRAPPIST-1 Superflare Occurrence and Plan-etary Habitability

Amy Louise Glazier1; Ward Howard1; Henry Corbett1;Nicholas Law1; Jeffrey Ratzloff1; Octavi Fors4; Daniel delSer4; Anna Hughes2; Robert Quimby3

1 Physics & Astronomy, University of North Carolina, Chapel Hill(Chapel Hill, North Carolina, United States)

2 Physics & Astronomy, University of British Columbia, Vancouver(Vancouver, British Columbia, Canada)

3 Astronomy / Mount Laguna Observatory, San Diego State Uni-versity (San Diego, California, United States)

4 Institut de Ciències del Cosmos (ICCUB), Universitat deBarcelona, IEEC-UB (Barcelona, Spain)

TRAPPIST-1 is an ultracool dwarf (UCD) of spec-tral type M8V with seven terrestrial planets, threeof which have equilibrium temperatures that maysustain surface liquid water. The TRAPPIST-1 sys-tem is a compelling target for habitability stud-ies. However, like many UCDs, TRAPPIST-1 fre-quently erupts in flares. High-energy flares can de-plete or even destroy the ozone column of exoplane-tary atmospheres, allowing ultraviolet radiation andhigh-energy particles to bombard the planet’s sur-face. Thus, detailed knowledge of TRAPPIST-1’sflare rate is important for studies of the system’shabitability. The Evryscope at CTIO observes theentire southern sky at two-minute cadence, and isthus especially well suited to characterize the oc-currence rate of superflares—flares of energy ≥ 1033

erg—for ultracool dwarves such as TRAPPIST-1. Wecombine Evryscope observations with Kepler obser-vations to constrain the high-energy flare rate ofTRAPPIST-1. The Evryscope complements short-cadence K2 data by providing data on TRAPPIST-1’s long-term average activity. Although no flareswere confirmed in Evryscope data, we calculate theminimum detectable flare energy for TRAPPIST-1with Evryscope, and thereby recompute the flarefrequency distribution (FFD) of TRAPPIST-1 usingflares observed by K2 and non-detections of high-energy flares from Evryscope. We place new con-straints on the high-energy flare rate for TRAPPIST-1, finding that the annual superflare rate is expectedto be 13.4+3.3

−0.3 per year. Converting bolometricflare energy to energy emitted in the biologically rel-evant UV-B bandpass and comparing to early-EarthUV-B insolation, we find the top-of-atmosphere UV-B flux at each TRAPPIST-1 planet due to high-energyflares is insufficient to rule out life, assuming in-tact atmospheres. We further analyze the impact ofTRAPPIST-1’s flare rate on pre-biotic chemistry andozone depletion. Our preliminary results do not rule

84

out habitability for TRAPPIST-1’s planets.

319.13 — Star-planet tidal interactions in theWASP-12 system

Gracjan Maciejewski11 Nicolaus Copernicus University (Torun, Poland)

To date, the exoplanet WASP-12 b is the only hotJupiter for which the shortening of its orbital periodwas detected. A mechanism that drives this orbitaldecay remains a puzzle. Equilibrium stellar tideswere found to be too weak to explain the observedrate of inspiral and dynamical stellar tides or plan-etary obliquity tides have been considered. Usingthe new radial velocity (RV) measurements, we showthat the orbital eccentricity of WASP-12 b is non-zeroat a 5-σ level, and the longitude of pericentre of thisapparently eccentric orbit is close to -90 degrees. Thisorbital configuration is compatible with a solutioncontaining a circular orbit and an RV signal inducedby the tidal fluid flow in the star. The amplitude ofthe RV tides was found to be consistent with a valuecalculated using the equilibrium tide approximation.We expect that further precise RV measurements willallow us to refine the phase lag of the stellar tides.

319.14 — Chaotic Dynamical Tides in Eccentric GasGiants and Hot Jupiter Formation

Michelle Vick1; Dong Lai11 Astronomy, Cornell University (Ithaca, New York, United States)

High-eccentricity (high-e) migration is an importantchannel for the formation of hot Jupiters (HJs). Inthis scenario, a giant planet is excited onto a veryeccentric orbit that decays and circularizes on Gyrtimescales due to tidal effects. Previous studiesof high-e migration have used parameterized treat-ments of weak tidal friction and overlooked criticalcontributions from dynamical tides. These earlierworks have suggested that high-e migration can re-produce multiple features of the observed HJ pop-ulation (e.g. a pile-up at a few Roche-radii and thelack of planetary companions), but that it strugglesto explain the frequency of observed HJs withoutassuming that young gas giants are many times asdissipative as Jupiter. We demonstrate that the in-clusion of chaotic dynamical tides can help resolvethis issue. In particular, we focus on high-e migra-tion due to Lidov–Kozai (LK) oscillations of orbitaleccentricity/inclination induced by a distant plane-tary or stellar companion. When the planet’s orbitis in a high-eccentricity phase, the tidal force fromthe star excites oscillatory f-modes and i-modes in

the planet. For sufficiently large eccentricity andsmall pericenter distance, the modes can grow chaot-ically over multiple pericenter passages and eventu-ally dissipate non-linearly, drawing energy from theorbit and rapidly shrinking the semimajor axis. Westudy the effect of such chaotic tides on the planet’sorbital evolution. We find that this pathway pro-duces very eccentric (e > 0.9) warm Jupiters (WJs) onshort time-scales (a few to 100 Myr). These WJs ef-ficiently circularize to become HJs due to their per-sistently small pericenter distances. Chaotic tidescan also save some planets from tidal disruption bytruncating LK eccentricity oscillations, significantlyincreasing the HJ formation fraction for a range ofplanet masses and radii. Chaotic tides endow LKmigration and other flavors of high-e migration withseveral favorable features to explain observations ofHJs.

319.15 — 3D hydrodynamic simulations of theplanetary system GJ436: constraining the plane-tary wind parameters through spectroscopic obser-vations.

Carolina Villarreal D’Angelo1; Aline Vidotto11 School of Physics, Trinity College Dublin (Dublin, Ireland)

GJ436b was the first detection of atmospheric escapefrom a warm-Neptune around an M-dwarf. The ex-treme absorption observed in the Ly-α line duringtransit indicates the existence of a very long tail ofplanetary neutral material trailing the planet, not be-ing completely swept by the interaction with the stel-lar wind. Previous work based on particle simula-tions (Bourrier et al. 2016, 2015; Ehrenreich et al.2015) have shown that the Ly-α absorption profilecan be reproduce with a high planetary wind ve-locity (50-70 km/s) and a very low stellar wind ve-locity and temperature (80 km/s and 2×103 K). AsGJ436b lies in the edge of the sub-Jovian desert, thecharacterisation of the stellar and planetary windparameters will help to understand the process in-volved in the erosion of the planetary atmosphere,and will give us a hint on the possibles scenarios thatlead to the formation of such desert. In this talk wewill present the unpublished results of the first 3Dhydrodynamics simulations of GJ436 planetary sys-tem. These simulations take into account the stel-lar and planetary wind interaction and include theprocess of charge-exchange, radiation pressure andphotoionization. With the output of the simulationswe can compute synthetic Ly-α transit profiles thatwe compare with the observations. We will showthat observed absorption profiles in Ly-α can be re-produced in a hydrodynamic scenario where the in-

85

teraction of both winds forms a shock ahead of theplanet. We will also show the constrained param-eters obtained for the planetary wind and the in-fluence that charge-exchange and radiation pressurehave on the escaping planetary material. The newderived values are not in agreement with the onespresented in the work of Bourrier et al. (2016), show-ing that the observations in Ly-α alone can not beused to constrain the planetary wind parameters. Inthis sense, we also explore synthetic observations ofthe H-α line during transit.

320 — Disk-Planet Interactions andMigration Theory320.01 — The 4:7 resonant population and adilemma of the inclination distribution in theKuiper belt

Jian Li11 School of Astronomy and Space Science, Nanjing University

(Nanjing, Jiangsu, China)

The 4:7 mean motion resonance (MMR) locates at theheart of the classical Kuiper belt, it can provide un-precedented clues about the history of planet migra-tion. By making use of a semi-analytical model basedon the simplified disturbing function, we would sup-ply new insights into the dynamics of the 4:7 MMR.For this high-order resonance, there are two differentmodes: (1) The first one is the eccentricity-type reso-nance, inside which the peculiar eccentricity and in-clination distribution of the resonators could be per-fectly constrained by a theoretical limiting curve. Ac-cordingly, a survey strategy of the highly inclinedcomponents is proposed; (2) The second one is onlyassociated to the high-inclination resonators, whilethe number ratio of simulated to observed samplesraises a dilemma of the inclination distribution of theprimordial Kuiper belt objects.

320.02 — Early-phase simulations of circumplane-tary disk formation

Yuri I. Fujii1; Oliver Gressel2; Kengo Tomida3; UdoZiegler2

1 Nagoya University (Nagoya, Japan)2 Leibniz Institute for Astrophysics Potsdam (Potsdam, Germany)3 Osaka University (Osaka, Japan)

Regular moons such as Galileans are thought to formin gaseous disks around gas giant planets. Study-ing the gas flow into the vicinity of planets is im-portant not only to learn the formation of plane-

tary atmospheres but also to discuss satellite for-mation. Gas around a sufficiently massive planetis thought to form a circumplanetary disk insteadof falling directly onto the planet. Recent studies,however, have shown that the circumplanetary gasforms an expanded envelope rather than a thin, rota-tionally supported Keplerian disk in cases where thegas temperature is very high. Thus, calculating thetemperature of the accretion flow properly is impor-tant to determine the disk structure. We performedthree-dimensional radiation hydrodynamic simula-tions of the formation of circumplanetary disks withan equation of states that considers effects such as hy-drogen dissociation and helium and hydrogen ion-ization. The region within the deep potential of theplanet reached very high temperature, yet, we ob-serve a disk to form in our simulations.

320.03 — Debris Disks in Multi-Planet Systems:Are Our Inferences Compromised by Unseen Plan-ets?

Jiayin Dong1,2; Rebekah Ilene Dawson1,2; AndrewShannon1,2; Sarah Morrison1,2

1 Astronomy & Astrophysics, The Pennsylvania State University(University Park, Pennsylvania, United States)

2 Center for Exoplanets and Habitable Worlds (University Park,Pennsylvania, United States)

Resolved debris disk structures (e.g., warps, off-sets, edges and gaps, azimuthal asymmetries, thick-ened rings, scale heights) contain valuable informa-tion about the underlying planetary systems, suchas the posited planet’s mass, semi-major axis, andother orbital parameters. Most existing models as-sume a single planet is sculpting the disk feature,but recent observations of mature planetary systems(e.g., by radial velocity surveys or Kepler) have re-vealed that many planets reside in multi-planet sys-tems. Here we investigate if/how planet propertiesinferred from single-planet models are compromisedwhen multiple planets reside in the system. For eachdisk feature, we build a two-planet model that in-cludes a planet b with fixed parameters and a planetc with a full range of possible parameters. We inves-tigate these two-planet systems and summarize theconfigurations for which assuming a single planet(i.e., planet b) leads to significantly flawed inferencesof that planet’s properties. We find that althoughdisk features are usually primarily dominated by asingle planet, when using single-planet models weare at risk of misinterpreting planet properties by or-ders of magnitude in extreme cases. Specifically, weare at high risk of misinterpreting planet propertiesfor the warp feature; at moderate risk for the edge

86

and gap feature, the thickened ring feature, and thescale height feature; and at low risk for the offset fea-ture and the azimuthal asymmetry feature.

320.04 — It can be hard to become a Gas Giant: 3DRadiation Hydro Simulations of CircumplanetaryEnvelopes

Hubert Klahr1; Doug Lin3; Ian Dobbs-Dixon21 PSF, Max-Planck-Institut for Astronomy (Heidelberg, Germany)2 Department of Physics, NYU Abu Dhabi (Abu Dhabi, United

Arab Emirates)3 Department of Astronomy & Astrophysics, University of Califor-

nia Santa Cruz (Santa Cruz, California, United States)

Why do find so many 10 Mearth planets in the uni-verse close to their host star? What prevented themfrom going into runaway gas accretion? Here wepresent long-term monitoring simulations of plan-etary cores interacting with the nebula gas. Wemeasure the mass, entropy, heat and radiation fluxacross the Hill Sphere and across sub-shells of theHill Sphere to high precision. We investigate coresof 5, 10 and 30 Earth masses. We find the Hill Sphereto be highly turbulent. This seems to be a result ofthe interaction with the Keplerian Sheer into whichthe core is embedded. The dissipation of kinetic en-ergy leads to a luminosity from the Hill Sphere thatcan prevent any net accretion of gas from the solarnebula for a wide range of planet masses and loca-tions. This finding helps to understand the survivalof super-Earth planets close to their host star.

320.05 — How flat can a planetary system get? Thecase of TRAPPIST-1.

Matthew Heising1; Dimitar Sasselov1; Lars Hernquist11 Astronomy, Harvard University (Cambridge, Massachusetts,

United States)

The seven planets orbiting the M dwarf TRAPPIST-1in a compact near-resonant chain offer a unique caseto study in planet formation theory. Of particularinterest is the remarkable flatness of the system, ex-ceeding that of any other known planetary system.We use 3D hydrodynamic simulations to study in-teractions between the planets and the gaseous diskin which they formed. We demonstrate with thesesimulations that the system’s flatness is an impor-tant constraint on the formation of the system, andcan be used to place an upper bound on the massof the disk. This result favors models for the forma-tion of the TRAPPIST-1 system that do not requirevery massive disks, such as that proposed by Ormelet al. (2017), and disfavors models that require more

massive disks, such as in situ formation and certainmodels of long-range migration.

320.06 — A young planetary system modified by anear-coplanar stellar flyby

Paul Kalas1,2; Robert De Rosa31 University of California, Berkeley (Berkeley, California, United

States)2 SETI Institute (Mountain View, California, United States)3 Stanford University (Palo Alto, California, United States)

Close stellar encounters have the potential to signif-icantly alter the architecture of planetary systems.Stars passing close to our solar system have been in-voked to explain the formation of the Oort cloud,comet showers, the disruption of the Kuiper Belt,and the distant detached orbits of dwarf planets suchas 90377 Sedna, as well as the hypothetical PlanetNine. Such stellar flybys have also been invoked toexplain the orbital properties of extrasolar planets.However, direct empirical evidence for these hypo-thetical encounters is lacking.

Here we show that the 15 Myr-old planet-hostingbinary star HD 106906 underwent a close stellar en-counter that can explain the system’s current archi-tecture. Using the exquisite precision of ESA’s Gaiasatellite for measuring stellar motions we have dis-covered a pair of external stellar perturbers that ap-proached within 1 pc of the HD 106906 system in aflyby geometry that is coplanar with the observed,highly asymmetric circumbinary disk. This flyby isconsistent with the scenario that the massive planetHD 106906 b formed in a disk near the binary star,was ejected from the inner system through inter-actions with the eccentric binary, and was subse-quently stabilized onto its current wide orbit (∼740au) by the perturbations of the passing stars.

This work was supported by NSF AST-1518332,NASA NNX15AC89G, and NNX15AD95G/NEXSS.

321 — Interior Structure Modeling,Poster Session321.01 — Modeling the evolution of ice-rich planets

Dina Prialnik1; Attay Kovetz21 Department of Geosciences, Tel Aviv University (Tel Aviv, Israel)2 School of Physics, Tel Aviv University (Tel Aviv, Israel)

Ice-rich bodies in the Solar System and beyond areof great interest to the pursuit of extraterrestrial lifeand present a challenge to our understanding of theformation and evolution of planetary systems. If

87

formed at the outskirts of the planetary disk, theymust have grown slowly, by accretion of planetes-imals, without extended atmospheres. Among thehundred and more rocky exoplanets recently discov-ered in planetary systems, a fraction must be of sim-ilar ice-rich composition and structure. By means ofa numerical code, we follow continuously the long-term growth of a planet by accretion of ice-rich mate-rial, as well as its subsequent evolution up to presenttime. The assumed composition includes silicaterock and ice in a compressed or porous structure, de-pending on local pressure. We use a second-orderBirch-Murnaghan equation of state, with an adap-tive coefficient, depending on the total mass of theplanet, and include radiogenic heating, latent heatand gravitational potential energy in the energy bal-ance equation. We allow for flow of water throughthe pores.

Two cases are discussed, for two different accre-tion rates. The typical structure that develops dur-ing the long-term evolution is an ice-depleted core,a liquid-filled middle layer (where saturation is al-most unity and temperatures are in the liquid waterrange), and a top ice-rich cold layer. The outer layeris porous, with porosity increasing from close to zeroat the lower boundary, to almost 0.6 near the surface.During the accretion phase, the surface temperaturerises well above the local equilibrium temperature,allowing the planet to cool, but also allowing subli-mation of some of the accreted ice. Hence, in the finalmodel, the ice/rock ratio varies widely throughoutthe planet, and is different from the ratio assumed forthe accreted material. In conclusion, we show thatthe emerging structure of such a planet, grown froma small embryo, should be differentiated at the end ofa 1-2 Gyr accretion phase, with an ice-depleted rockycore, an intermediate region filled with liquid waterand an outer ice-rich layer.

321.02 — Hot and Steamy, Cold and Icy, or Temper-ate and Habitable: Modeling the Early Evolution ofWater World Exoplanets

Nadejda Marounina1; Leslie A. Rogers11 Department of Astronomy and Astrophysics, University of

Chicago (Chicago, Illinois, United States)

Waterworlds are water-rich (>1% water by mass) ex-oplanets that never attained masses sufficient to ac-crete or retain large amounts of H/He nebular gas.Waterworlds are especially timely given the discov-ery and characterization of the TRAPPIST-1 plane-tary system, which hosts several planets that maycontain several percents to several tens of percent ofwater by mass.

To date, studies of the interior structure of water-world exoplanets have either assumed that the plan-ets are cold and icy (with interiors structures similarto scaled-up versions of Jupiter’s moon Ganymede)or that the planets are hot and steamy (with mostof their water in extended envelopes of supercriticalsteam). Models have not yet demonstrated the evo-lution of waterworlds from an initial post-accretionhot and steamy state to habitable and temperate con-ditions (with surface or subsurface water oceans).

We have performed the most detailed calculationsto date of the post-accretional thermal evolution ofwaterworlds with pure water envelopes. We ac-count for the condensation of liquid water and high-pressure ices as the planets cool, along with the tem-perature dependence of water opacities in the in-frared and visible. In this presentation, we will de-lineate the regions of the parameter space (orbitalseparation, planet mass, radius, and water envelopemass) wherein waterworlds are likely to be hot andsteamy, cold and icy, and temperate and (potentially)habitable.

Our results have important implications for boththe habitability of waterworlds and its observablecharacteristics — i.e. the apparent transit radius,mean planet density, and atmospheric spectra.

321.03 — How magmatic degassing of C, O, and Haffects Earth’s early atmosphere

Frank Sohl1; Gianluigi Ortenzi1,2; Lena Noack2; ClaireGuimond2; Julia Schmidt2; Sara Vulpius2

1 Institute of Planetary Research, German Aerospace Center (DLR)(Berlin, Germany)

2 Department of Earth Sciences, Freie Universität (Berlin, Ger-many)

The build-up of the Earth’s early atmosphere is a keypoint to investigate the planet’s evolution and poten-tial habitability. In this research we investigate thevolcanic outgassing of C, O, and H and the relateddevelopment of the atmosphere using the equilib-rium and mass balance method for volatile specia-tion. We estimate the outgassing process soon af-ter the magma ocean crystallization, simulating bothmagma production and lithostatic pressure effect onvolatile solubility. Considering the volume of meltproduced, we calculate the composition and pres-sure variation in the accumulated outgassed atmo-sphere during the early Earth’s evolution. Our re-sults indicate that the outgassed chemical compo-sition is mainly affected by the oxidation state ofthe mantle and by the pressure dependence on thevolatile solubility. The early Earth history was char-acterized by core-mantle segregation. During this

88

process, the oxidation state of the mantle evolvedfrom an initially reduced state to the more oxidiz-ing present-day value. The different melt redoxstates would have affected the evolution and chemi-cal composition of the early Earth atmosphere. Thevolatile species vary from H2 and CO in reducingstate to H2O and CO2 in oxidizing conditions similarto present days. We find that both chemical and pres-sure variations of the atmosphere are directly linkedto the evolution of the mantle. The abundances of theoutgassed species are also affected by the differentsolubility of the volatiles. Considering the pressureincrease of magmatic reservoirs due to formation ofthe overlying crust, the outgassed volatile specieschange according to their different solubility, influ-encing the degassing process and thus atmospherecomposition. Furthermore, planetary mass and ra-dius may affect melt production and atmosphericthickness. The coupling of the volatile speciationto the melt production suggests that this techniqueis useful to describe the early Earth evolution but ithas also the potential to investigate the habitabilityof planets other than the Earth including rocky exo-planets.

321.04 — The Limits on Interior Pressures and Tem-peratures of Likely Super Earths

Wendy Panero1; Cayman Unterborn21 School of Earth Sciences, Ohio State University (Columbus, Ohio,

United States)2 Arizona State University (Tempe, Arizona, United States)

The interior composition of structure, composition,and dynamics of exoplanetary Super Earths are notobservable. Recently described observational trendssuggest that rocky exoplanets, that is, planets with-out significant volatile envelopes, are likely limitedto ≤1.5 Earth radii. This likely upper limit in theradii of purely-rocky Super-Earth exoplanets, themaximum expected core-mantle boundary pressureand adiabatic temperature is relatively moderate, 630GPa and 5000 K, while the maximum central corepressure varies between 1.5 and 2.5 TPa. We fur-ther find that for planets with radii less than 1.5Earth radii, core-mantle boundary pressure and adi-abatic temperature are mostly a function of planetradius and insensitive to planet structure. The pres-sures and temperatures of rocky exoplanet interiors,then, are less than those explored in recent shock-compression experiments, ab-initio calculations, andplanetary dynamical studies. We further show thatthe extrapolation of relevant equations of state doesnot introduce significant uncertainties in the struc-tural models of these planets. Mass-radius models

are more sensitive to bulk composition than any un-certainty in the equation of state, even when extrap-olated to TPa pressures.

321.05 — On the Concept of Multi-Stable TectonicStates: The Un/Inevitability of Plate Tectonics

Matthew Weller1; June Wicks21 Earth, Environmental and Planetary Sciences, Brown University

(Providence, Rhode Island, United States)2 Earth & Planetary Sciences, Johns Hopkins University (Balti-

more, Maryland, United States)

With the plethora of recently discovered exoplanets,it is a natural question to consider how many of thesebodies could operate within a plate tectonic regime,and host life, as we observe for the Earth. However,our understanding of Earth’s evolution is far fromcomplete. Geologic evidence suggests that plate tec-tonics may not have operated on the early Earth, withboth the timing of onset and the length of activityfar from certain. Uncertainty about the initiation ofplate tectonics, and the initial tectonic state for Earthhas been extended to extra-solar planets. It is anopen question of whether terrestrial planets largerand more massive than Earth are more or less likelyto have plate tectonics, with groups arguing that asingle plate regime should be favoured, while oth-ers argue that plate tectonics should dominate. Re-cently, tectonic bi-stability (multiple stable, energet-ically allowed solutions) has been shown to be dy-namically viable, both from analytical analysis andthrough numeric experiments in two and three di-mensions. From scaling analysis, high-temperatureplanets with a large contribution from internal heat-ing (radiogenics or tidal sources) will operate in dif-ferent velocity-stress scaling regimes compared tocooler-temperature planets that may have a largerrelative contribution from core heating. Thus, differ-ences in predictions for plate tectonics on exoplan-ets may in part result from different model assump-tions being more appropriate to different times in theevolution of a terrestrial-type planet. This indicatesthat multiple tectonic modes may operate on a singleplanetary body at different times within its temporalevolution. It can then be shown that identical planetsat similar stages of their evolution may exhibit differ-ent tectonic regimes due to random fluctuations. Wewill discuss a new framework of planetary evolutionthat is based on general physical principals, as op-posed to particular rheologies, that further incorpo-rates the potential of tectonic regime transitions andmultiple tectonic-state viability at equivalent physi-cal and chemical conditions.

89

321.06 — New Giant Planet Physics from StatisticalPopulation Models

Daniel Thorngren1,4; Peter Gao3; Jonathan Fortney21 Physics, University of California, Santa Cruz (Santa Cruz, Cali-

fornia, United States)2 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)3 Astronomy, University of California, Berkeley (Berkeley, Califor-

nia, United States)4 Physics, Université de Montréal (Montréal, Quebec, Canada)

The dramatic rise in the number of well-characterized giant exoplanets has enabled apowerful approach to understanding their physics:combining detailed physical models with statisticalpopulation models to infer their physical proper-ties. Here I will present two new results from thisapproach. The first is that hot Jupiters quickly re-inflate in response to their parent stars brighteningon the main sequence. This reinflation can onlyoccur this rapidly if the inflationary mechanismis depositing heat below the radiative-convectiveboundary (RCB) of the planet’s atmosphere. Thusdelayed cooling mechanisms are ruled out, as is ourpresent understanding of Ohmic dissipation, whichrelies heavily on the insulating effects of shallowlydeposited heat. The second key result is that inorder to explain their radii, the interior entropies ofhot Jupiters must be so high that the RCB will bepushed up to much lower pressures than previouslyappreciated. A typical hot Jupiter will have its RCBat around ten bars, rather than the hundreds of barsto 1 kbar usually considered. This has importantimplications for atmosphere modelling, including3D general circulation models, 1D models used foranalyzing spectroscopic data, and cloud models thatrely heavily on an understanding of cold traps andvertical mixing.

322 — Planets around Young Stars,Poster Session322.01 — Search for Hα from accreting protoplanetswith Subaru/SCExAO+VAMPIRES

Taichi Uyama1,2; Barnaby Norris3; Olivier Guyon4; Mo-tohide Tamura5

1 IPAC, California Institute of Technology (Pasadena, California,United States)

2 NExSCI (Pasadena, California, United States)3 the University of Sydney (Sydney, New South Wales, Australia)4 Subaru Telescope (Hilo, Hawaii, United States)5 the University of Tokyo (Tokyo, Japan)

The Visible Aperture Masking Polarimetric Inter-ferometer for Resolving Exoplanetary Signatures(VAMPIRES) has newly started operation with Sub-aru/SCExAO. This instrument enables variety ofimaging modes of polarization differential imaging(PDI), aperture masking, and spectral differentialimaging with a Hα filter (SDI). The main purposeof the instrument is high-resolution imaging of cir-cumstellar disks but the Hα SDI mode can provideanother aspect of high-contrast imaging; accretionsignatures. As reported accreting protoplanet can-didates around LkCa 15 or PDS 70 hydrogen emis-sions from protoplanets will directly benefit discus-sion of planet formation mechanisms. We introduceHα high-contrast imaging observation and reduc-tion schematics in the presentation.

322.02 — Formation and atmospheric constraints ofthe youngest hot Jupiter.

Joe Llama1; Christopher Johns-Krull2; Lisa Prato1;Larissa Nofi1,3; Daniel Jaffe4; Lauren Biddle1,5; LauraFlagg2; Gregory Mace4

1 Lowell Observatory (Flagstaff, Arizona, United States)2 Rice University (Houston, Texas, United States)3 Institute for Astronomy, University of Hawaii (Manoa, Hawaii,

United States)4 McDonald Observatory, University of Texas (Austin, Texas,

United States)5 Northern Arizona University (Flagstaff, Arizona, United States)

The last three years have ushered in the era of youngexoplanets. Despite young stars exhibiting extremelevels of variability, a handful of newly formed exo-planets have been detected through transits and ra-dial velocity. We will present the latest results fromour young planet survey. For the first time, usinghigh-resolution infrared spectra, we have a direct de-tection of CO in the atmosphere of the youngest ex-oplanet CI Tau b, confirming the planet, and pro-viding evidence for a hot start mechanism. Our dis-covery shows that hot Jupiters either form incrediblyclose to their parent star, or, that migration occurswithin the first 2 Myr.

322.03 — The nature of a low-mass companion inan edge-on protoplanetary disk system

Karl Stapelfeldt1,2; Deborah Padgett4,2; GaspardDuchene3; Quinn Konopacky6; William Fischer8; JiWang7; Dimitri Mawet2,4; Francois Menard5; VirginieFaramaz4,2

1 NASA Exoplanet Exploration Program Office, Jet PropulsionLaboratory (Pasadena, California, United States)

90

2 California Institute of Technology (Pasadena, California, UnitedStates)

3 University of California, Berkeley (Berkeley, California, UnitedStates)

4 Jet Propulsion Laboratory (Pasadena, California, United States)5 Universite Grenoble Alps (Grenoble, France)6 University of California, San Diego (San Diego, California,

United States)7 Astronomy, The Ohio State University (Comlubus, Ohio, United

States)8 Space Telescope Science Institute (Baltimore, Maryland, United

States)

We report on the properties of the low-masscompanion to the Ophiuchus young stellar objectSSTc2dJ162221.0-230402. The companion is pro-jected at 130 AU separation along the plane of theprimary star’s edge-on protoplanetary disk seen inHubble Space Telescope images. Optical and near-IRphotometry from HST and Keck show that the com-panion spectral energy distribution is very similar tothe young brown dwarf GQ Lup B but with 10× lessluminosity. Narrowband HST images show that thecompanion has strong Hα emission. The companionis also detected as an unresolved source by ALMA in1.3 mm dust continuum, consistent with only ∼0.01Mjup of surrounding material if the emission is opti-cally thin. While this set of results is consistent withan accreting planetary mass companion, recent anal-ysis of our HST grism and Keck OSIRIS K-band spec-tra indicate a mid- to late- M spectral type. The latterimplies a strongly underluminous object just belowthe hydrogen burning limit, perhaps seen indirectlyvia scattered light from its own edge-on disk. Waysto confirm this scenario and its implications for thesystem properties will be discussed.

322.05 — Spectroscopic search for circumplanetarymaterial during the Beta Pictoris b Hill Spheretransit

Ernst J.W. De Mooij1; Matt Kenworthy2; Paul A.Wilson3; Maggie Celeste1; Blaine B.D. Lomberg11,12;Lennart Van Sluijs2; Carlo F. Manara13; AlexisBrandeker14; Alan Fitzsimmons4; Neale P. Gibson5;Flavien Kiefer7,8; Sam Mellon10; Eric Mamajek6; An-drew Ridden-Harper9; Alfred Vidal-Madjar7,8; Chris A.Watson4; Konstanze Zwintz15

1 School of Physical Sciences and Centre for Astrophysics and Rela-tivity, Dublin City University (Dublin, Ireland)

2 Department of Physics & Astronomy, University of Rochester(Rochester, New York, United States)

3 South African Astronomical Observatory (Cape Town, SouthAfrica)

4 Department of Astronomy, University of Cape Town (Cape Town,South Africa)

5 European Southern Observatory (Garching bei Munchen, Ger-many)

6 Deptartment of Astronomy, Stockholm University (Stockholm,Sweden)

7 Institut für Astro- und Teilchenphysik, Universität Innsbruck(Innsbruck, Austria)

8 Leiden Observatory, Leiden University (Leiden, Netherlands)9 University of Warwick (Coventry, United Kingdom)10 Queen’s University Belfast (Belfast, United Kingdom)11 Trinity College Dublin (Dublin, Ireland)12 Jet Propulsion Laboratory (Pasadena, California, United States)13 CNRS, UMR 7095, Institut d’Astrophysique de Paris (Paris,

France)14 UPMC Univ. Paris 6, Institut d’Astrophysique de Paris (Paris,

France)15 Department of Astronomy, Cornell University (Ithaca, New

York, United States)

During 2017-2018, the Hill sphere of the directly im-aged planet Beta Pictoris b transited its stellar host.This provided a unique opportunity to probe the gasand dust in the circumplanetary environment of ayoung, recently formed gas-giant planet. We presentthe results from a high-resolution monitoring cam-paign with the UVES spectrograph (R∼100,000) atthe VLT. Over the course of 160 epochs distributedacross one year we obtained observations of boththe Hill sphere transit and an out-of-transit baseline.This cadence is more than sufficient to detect ringcrossings as well as monitor any variations in gas ab-sorption along the line of sight, in particular in the CaH & K lines, where there are also signs of the circum-stellar gas and exocomets. In addition, this uniquedataset provides strong constraints on the presenceof transiting giant planets on short (<200 day) orbitalperiods.

322.06 — Results from the Beta Pictoris b HillSphere Transit Campaign

Matthew Kenworthy1; Konstanze Zwintz4; SamMellon5; Tristan Guillot6; Paul Kalas7; Eric Mamajek8,5;Iva Laginja9; Remko Stuik1; Steven Crawford9; MichaelIreland10; Jason Wang2; Ernst J.W. De Mooij3; Anne-Marie Lagrange15; Lyu Abe6; Nick Suntzeff12; DjamelMékarnia6; Sylvestre Lacour13; Zhang Hui14; LifanWang14; Mathias Nowak15; Flavien Kiefer13; AlainLecavelier des Etangs13; Kevin Stevenson9; Alfred Vidal-Madjar13; Paul A. Wilson16; Blaine B.D. Lomberg11

1 Leiden Observatory (Leiden, Zuid-Holland, Netherlands)2 Research School of Astronomy and Astrophysics, Australian Na-

tional University (Canberra, Australian Capital Territory, Australia)3 SAAO (Cape Town, South Africa)

91

4 Texas A&M University (College Station, Texas, United States)5 LESIA, Observatory of Paris (Paris, France)6 Key Laboratory of Modern Astronomy and Astrophysics (Nan-

jing, China)7 Institut de Planétologie et d’Astrophysique de Grenoble (Grenoble,

France)8 University of Warwick (Warwick, United Kingdom)9 Astronomy, California Institute of Technology (Pasadena, Califor-

nia, United States)10 School of Physical Sciences and Centre for Astrophysics and

Relativity, Dublin City University (Dublin, Ireland)11 Institut für Astro- und Teilchenphysik, Universität Innsbruck

(Innsbruck, Austria)12 Department of Physics and Astronomy, University of Rochester

(Rochester, New York, United States)13 OCA, Université Côte d’Azur (Nice, France)14 Department of Astronomy, University of California (Berkeley,

California, United States)15 Jet Propulsion Laboratory (Pasadena, California, United States)16 STScI (Baltimore, Maryland, United States)

Beta Pictoris b is the only extrasolar gas giant planetthat has been directly imaged and has an edge-onorbit that causes the planet’s Hill sphere to tran-sit approximately every 22 years. Based on multi-epoch direct imaging observations from the GeminiPlanet Imager, the Hill sphere began transiting inApril 2017, with closest approach to within 20% ofthe 1.2au Hill sphere radius in August 2017.

We present the results from the combined spaceand ground based photometric campaigns that coverthe entirety of the Hill sphere transit during 2017 andthe first half of 2018, to search for signs of circum-planetary dust and rings. These include a dedicatedmonitoring instrument in South Africa and Aus-tralia (the bRing project), the BRITE-Constellationnanosatellites, two telescopes in Antarctica (ASTEPand AST3) and HST/COS data.

322.07 — Mass Limits and Helium Variability forYoung Transiting Planets in the Hyades with theHabitable-Zone Planet Finder

Daniel Krolikowski1; Adam Kraus1; Aaron Rizzuto1;Caroline Morley1; Andrew Vanderburg1; AndrewMann2

1 University of Texas at Austin (Austin, Texas, United States)2 University of North Carolina at Chapel Hill (Chapel Hill, North

Carolina, United States)

While the demography of old planetary systems caninform models of their formation and evolutionarypathways, the characterization of young planets andtheir host stars is needed to directly trace the evo-lution of planetary atmospheres and orbits. Re-

cently, the K2 mission has discovered many youngplanetary systems in star clusters, where it is easi-est to characterize the host star’s age, distance, andmetallicity. While transits yield radius measure-ments, spectroscopic observations of these stars areneeded to determine planetary masses and atmo-spheric compositions. The high activity level ofyoung stars, however, introduces large stellar jitterto their radial velocities, making it harder to extracta planetary signal. To mitigate this issue, it is nec-essary to observe young stars at high cadence inthe infrared, where RV noise due to stellar activityis smaller than in the optical. The Habitable-zonePlanet Finder on the HET is a new, high precisionNIR spectrograph uniquely suited to sift throughstellar noise and detect planetary signals aroundyoung stars. The HPF bandpass also includes theHe 10830 triplet, which can be used to detect evapo-rating exospheres around transiting planets withoutthe need for expensive and challenging space-basedUV observations. We will present observations ofplanet hosting Hyades members K2-25 and K2-136,with radial velocity time series and measurements ofthe variability of the He triplet. Our results demon-strate HPF’s unprecedented near-infrared RV preci-sion, and from these data we will present mass limitsfor the transiting planets in these two systems. Weshow that the helium profile is stable to roughly 2%in the young K-dwarf K2-136, meaning it is possi-ble to detect young helium exospheres if the signalis comparable to previous detections (3-8%). Lastly,we present the current status of our large program toobserve other young planetary systems, and discusslimits on non-transiting long period planets in thosesystems.

322.08 — Planetary Magnetic Response to YoungStar Stellar Wind Environment

Anthony Sciola1; Frank Toffoletto1; David Alexander11 Physics and Astronomy, Rice University (Houston, Texas, United

States)

Young stars are known to exhibit strong magnetic ac-tivity in the form of both steady and eruptive out-flow. This, combined with the majority of knownEarth-like planets orbiting their stars closer thanMercury orbits the Sun, results in an extreme stel-lar wind environment at the planet. We employ a3D coupled MHD model to simulate the planetaryresponse to this environment with an emphasis onthe exchange of plasma between the planetary mag-netosphere, ionosphere and stellar wind. We willpresent the results of two cases: the case of extreme,

92

steady stellar wind and the case of successive Sun-like Coronal Mass Ejection (CME) impacts over shorttimescales. These results demonstrate how exoplan-etary environments differ from Earth’s and provideinsight into how early stage star-planet interactionsmay impact future evolution of the planetary envi-ronment.

322.09 — Searching for Wide Companions andIdentifying Circum(sub)stellar Disks through PSF-Fitting of Spitzer/IRAC Archival Data

Raquel Martinez1; Adam Kraus11 Astronomy, The University of Texas at Austin (Austin, Texas,

United States)

The last decade has seen the discovery of a grow-ing population of planetary-mass companions (∼<20MJup; PMCs) to young stars which are often still inthe star-forming regions where they formed. Theseobjects have been found at wide separations (>100AU) from their host stars, challenging existing mod-els of both star and planet formation. Demographictrends with mass and separation should distinguishbetween these formation models.

The extensive Spitzer/IRAC data set of major star-forming regions and associations within 300 pc hasgreat potential to be mined for wide companionsto stars. I will present new results from my auto-mated pipeline to find wide companions of stars viapoint spread function (PSF) subtraction in IRAC im-ages. A Markov Chain Monte Carlo algorithm is thebackbone of this PSF subtraction routine that effi-ciently creates and subtracts χ2-minimizing instru-mental PSFs, measuring IR photometry of the sys-tems across the four IRAC channels (3.6 μm, 4.5 μm,5.8 μm, and 8 μm). I will present a re-analysis ofarchival IRAC images of 11 low-mass (44 MJup–0.88M⊙; K3.5–M7.5) stars in 3 nearby star-forming re-gions (Chameleon, Taurus, and Upper Scorpius; d ∼150 pc; τ ∼ 1–10 Myr) known to host faint compan-ions over a range of projected separations (1.7”–7.3”).I will discuss the characteristics and disk-hosting po-tential of the systems found to have low-mass com-panions with non-zero [3.6] − [8] colors, includingthe confirmation of a ρ = 4. 66” (540 AU), M = 20 MJupcompanion to [SCH06] J0359+2009, a young browndwarf in Taurus.

My survey is sensitive to companions with massesnear that of Jupiter at orbital radii of a few hundredAU, discovering wide PMCs in their birth environ-ments and revealing their circum(sub)stellar disks. Iwill present my newest results in measuring the mid-IR photometry of directly-imaged substellar com-panions and an automated companion search of all

known young stars with existing Spitzer/IRAC data,concluding with my ongoing follow-up of candidatewide PMC systems with ground-based telescopesand the outlook for future observations with space-based telescopes.

322.10 — Deep Asymmetric Eclipse of V928 Tau

Dirk Van Dam1; Matthew Kenworthy1; Trevor David2;Eric Mamajek2,4; Lynne Hillenbrand3; Ann MarieCody5; Andrew Howard3; Howard Isaacson6; DavidCiardi7; Luisa Rebull8; John Stauffer9

1 Leiden Observatory (Leiden, Netherlands)2 Jet Propulsion Laboratory (Pasadena, California, United States)3 Astronomy, California Institute of Technology (Pasadena, Califor-

nia, United States)4 Physics & Astronomy, University of Rochester (Rochester, New

York, United States)5 NASA Ames Research Center (Moffet Field, California, United

States)6 Astronomy, University of California (Berkley, California, United

States)7 Exoplanet Science Institute, Caltech/IPAC-NASA (Pasadena,

California, United States)8 Caltech/IPAC-IRSA (Pasadena, California, United States)9 Caltech/IPAC-SSC (Pasadena, California, United States)

V928 Tau is a previously known astrometric weak-lined T Tauri binary, with an orbital period inferredto be greater than 58 years (Schaefer et al. 2014). Inaddition to confirming for the first time that V928Tau A+B are physically associated on the basis ofnearly identical spectra, obtained from adaptive-optics resolved spectroscopy, we report the detectionof a single, deep, asymmetric eclipse from K2 data.We suggest this is due to a previously unknown com-panion at intermediate separation (orbital period >80 days). From modeling of the eclipse shape wefind evidence that the transiting or eclipsing com-panion is surrounded by a circumsecondary diskor a vast ring system on an eccentric orbit arounda single component of the binary. This modelingis done by new software developed specifically forthe fitting of inclined and tilted circumsecondarydisks or ring systems to light curves. Photometryfrom ground-based time-domain surveys reveal ad-ditional eclipses of the young star system, which pro-vide period constraints, which in turn provide massand eccentricity constraints of the orbit of the sec-ondary. We find several possible periods and reportthe corresponding predictions of the next transits,which will be monitored to fully determine the or-bital period of the companion. Once this has beendone an observing campaign can be organised to

93

fully characterise the physical and chemical proper-ties of the circumsecondary disk allowing us to betterunderstand gas giant formation.

323 — Protoplanetary Disks — Ob-servations, Poster Session323.01 — The Early Chemical Evolution of PlanetForming Material

Kamber Schwarz11 Lunar and Planetary Laboratory, University of Arizona (Tucson,

Arizona, United States)

There is growing observational evidence that evolu-tion in protoplanetary disks begins early. Gaps in thedust disk, an indication of grain growth, now appearto be ubiquitous. Many disks have low CO-to-dustratios, suggesting evolution of the gas phase disk aswell. Kinematic evidence even points to the pres-ence of fully formed giant planets within these gasrich disks. It is looking more and more likely thatmost planet formation has already occurred by theClass II phase of disk evolution. Within this contextI will discuss new observations of N2H+ and HCO+

in the envelopes of four Class I protostars which havepreviously been shown to have low CO abundances.Combined with a new physical-chemical model ofenvelope+disk systems, I will use these observationsto constrain the earliest chemical evolution of thevolatile gas available to forming planets.

323.02 — A young cousin of our solar system: Newresults from HD169142

Gesa Bertrang11 Max Planck Institute for Astronomy (Heidelberg, Germany)

The disk around the closest Herbig Ae star,HD169142, has been studied extensively at multiplewavelengths. At an almost face-on inclination, itreveals three major components: a yet unresolvedring/halo at only ∼0.1 au from the star, a brightring around ∼20 au, and an outer disk stretchingfrom ∼55-122 au. The gaps are emptied fromdust and gas, which is a strong indicator for thepresence of planets. In our series of observations(2012-2018) we report for the first time the presenceof brightness variations inside the major ring at 20au (Bertrang+2018, Bertrang in prep.). AssumingKeplerian velocity, we determine that the brightnessdip is consistent with a shadow cast by a Jupiter-mass planet or brown dwarf surrounded by dust ata radial distance of only 12au, an orbit comparable

to Jupiter’s. For this talk I will reveal new multi-wavelength observations together with radiativetransfer models and hydrodynamical simulationswhich support the picture that HD169142 might bea young version of a multi-planet system such asHR8799 or our own solar system.

323.03 — New constraints on the dust and gas dis-tribution in the LkCa 15 disk

Sheng Jin11 Purple Mountain Observatory (Nanjing, China)

We search a large parameter space of the LkCa 15’sdisk density profile to fit the observed radial in-tensity of 12CO obtained from ALMA. The best-fitmodel within the parameter space has a disk massof 0.1 M (using an abundance ratio of 12 CO/H2 =1.4 ×10−4 in mass), an inner cavity of 45 AU in ra-dius, an outer edge at ∼ 600 AU, and the disk surfacedensity profile follows a power-law that ∝ r−4. Wefound that for the disk density profiles that can leadto a small reduced χ2, there is a clear linear corre-lation between their disk masses and the power-lawindex γ in their disk density profiles. This means the12CO disk of LkCa 15 is optically thick and we can fitits 12CO radial intensity profile using either a lowerdisk mass with a smaller γ or a higher disk mass witha bigger γ. However, such a degeneracy in disk massand γ can be broken using the morphology of the12CO channel maps, and we find that only modelshave a disk mass of ∼ 0.1 M can reproduce the ob-served morphology of the 12CO channel maps. Thedust continuum map of the LkCa 15 disk shows thatalthough the mm-size dust disk has a similar innercavity of the gas disk, its out edge is at ∼ 200 AU,much smaller than the fitted gas disk. Such a dis-crepancy between the outer edges of the gas and dustdisks is what would be expected from dust driftingmodels.

323.04 — You are what you eat: the elemental com-position of disks and planets

Mihkel Kama11 Institute of Astronomy, University of Cambridge (Cambridge,

United Kingdom)

The carbon-to-oxygen ratio, C/O, is often touted asa tracer of the formation location of a giant planet.To determine if this hypothesis is true, and interpretor predict the elemental composition of any planetreliably, we require an understanding of the com-position of planet-forming gas and solids. In paral-lel with the high-profile efforts to measure the C/O

94

ratio and overall composition of exoplanet atmo-spheres, a matching effort in disks is thus needed.I will present a brief review of what we know of theelemental composition of disks so far, the techniqueswe are employing to learn more, and present new re-sults on the abundances and time-evolution of car-bon, oxygen, and other elements in several proto-planetary disks. To conclude, I will outline a classof planetary systems where many pieces of the com-position puzzle are beginning to come together.

324 — Protoplanetary Disks — The-ory, Poster Session324.01 — Disk Population Synthesis

Apostolos Zormpas1; Til Birnstiel1; Giovanni Rosotti21 Universitäts-Sternwarte, Ludwig-Maximilians-Universität

München (Munich, Bavaria, Germany)2 Leiden Observatory, Leiden University (Leiden, Netherlands)

Recently, a sub-arcsecond resolution survey of thedust continuum emission from nearby protoplane-tary disks, conducted with the Submillimeter Arrayshowed a strong correlation between the sizes andluminosities of the disks. Performing models of gasand dust disk evolution, we recreate this relation us-ing a large grid of models that varies the initial con-ditions. We calculate the disk continuum emissionand the effective radius for all models as a function oftime. We simulate two different cases: a smooth disksurface density profile, and one that includes pres-sure bumps. By selecting only the disks that lie onthe observed size-luminosity relation we constrainthe parameter range and search for trends betweenthe initial conditions and the survival frequency ofevery disk. By applying neural networks, we deter-mine the influence of every parameter on the finaloutcome, showing significant results for the initialdisk mass, the turbulence-parameter α, and the stel-lar mass.

324.02 — The thickness of dusty protoplanetarydisks depend on metallicity

Min-Kai Lin11 ASIAA (Taipei City, Taiwan)

The sharpness of dust rings commonly observed inprotoplanetary disks can be explained by havingsolids settle to a thin layer at the disk midplane. In-deed, dust settling is widely considered as a precur-sor to planet formation. On the other hand, the-oretical considerations suggest that protoplanetary

disks should develop hydrodynamic turbulence thatwould stir up dust particles. How can dust sedimentin spite of turbulence? I perform numerical simu-lations of dusty protoplanetary disks to settle thisconundrum. I focus on large radii in protoplane-tary disks, accessible by ALMA, where turbulence ischaracterized by large-scale vertical motions, whichtends to loft particles. I show that enhancing the totalsolid abudance above a few times the solar value canweaken turbulence and allow dust to settle. Thus,in addition to particle size and turbulence strength,dust settling in realistic protoplanetary disks is alsoaffected by the disk metallicity. Conversely, obser-vations of thin dust layers may provide an indirectconstraint on the local solid abundance. As a corol-lary, I predict dust rings should be thinner than dustgaps. This can be tested by measuring the dust layerthickness as a function of radius. I also show that theinteraction between newly born protoplanets and itssurrounding dust-rich disk differ significantly frompure gas disks. This may have implications for shap-ing the final architecture of planetary systems. Fi-nally, I present the latest 3D simulations aimed attesting whether or not the settled dust rings seen inprotoplanetary disks are compatible with sculptingby planets.

324.03 — Evolution of the Water Snow Line in Mag-netised Protoplanetary Disks

Shoji Mori1; Xuening Bai3; Satoshi Okuzumi21 Astronomy, The University of Tokyo (Bunkyo-ku, Tokyo, Japan)2 Earth and Planetary Sciences, Tokyo Institute of Technology

(Meguro-ku, Tokyo, Japan)3 Tsinghua Center for Astrophysics, Tsinghua University (Beijing,

China)

Understanding the location of the water snow lineprovides the formation region and time of rockyplanets, which probably formed inside the snowline. In this talk, we present the migration of thesnow line in magnetised protoplanetary disks, basedon nonideal magnetohydrodynamic (MHD) simu-lations (Mori et al., 2019) which produce laminardisks due to nonideal MHD effects. The simulationshowed that magnetic accretion heating is signifi-cantly less efficient than conventional viscous heat-ing. We build a new temperature model that repro-duces our simulation results well, and then revisitthe time evolution of the snow line location. The re-sult shows that for instance, the snow line passes 1AU at 0.4 Myr after the star formation, whereas thetime for the conventional turbulent disk is 3 Myr. ForEarth-like planets to form with sustaining low wa-ter content, the protoplanet should have formed in a

95

young solar nebula with gas inflow from a surround-ing envelope.

324.04 — Orbital Reshuffle of the Asteroids due tothe IN SITU Emergence of Jupiter and Clearing ofthe Main Belt During the Depletion of the SolarNebula

Xiaochen Zheng11 Department of Astronomy, Tsinghua University (Beijing, China)

The distinctive compositional differences between C-type and S-type asteroids suggest that they wereoriginated from two well-separated regions in theprimordial solar nebula, howerve, their overlappingspatial distributions implies a substantial radial mix-ing of S and C-type asteroids after their birth. Weadopt the hypothesis that S-type asteroids are nativein the main belt, whereas C-types originate from abroad, more distant (∼ 3.3−9 AU) region outside theice line. We also assume that they formed before theemergence of gas giants and consider the possibil-ity that the orbits of the C-types’ planetesimal pro-genitors were destabilized by the rapid mass increaseof Jupiter and Saturn during the advanced stages oftheir in situ formation. We perform a series of N-body simulations and confirm that it is possible fora significant fraction of C-type asteroids’ progenitorplanetesimals to be scattered from the proximity ofJupiter’s birth site into the extended main belt re-gion. We also show that this effect alone cannot leadto the substantial removal of most native planetesi-mals. But, during the subsequent depletion of thesolar nebula, Jupiter and Saturn’s secular resonancessweep through a wide region and excite the planetes-imals’ eccentricity along its path. The diminishinggas also circularizes the orbits of these planetesimalsand causes their semi major axis to decay. This sce-nario resolves the issues of mass deficit in the mainbelt. Due to the size-dependence of the drag force,the clearing process leads to a knee in the size dis-tribution among those planetesimals retained in themain belt. This effect can account for the observedsize-distribution of both C and S-type asteroids. Thisin situ dynamical shake-up scenario provides a solu-tion for both the collocation of compositionally dis-tinctive C and S-type asteroids and the mass distribu-tion in the inner solar system without invoking the“grand-tack” scenario that Jupiter has meanderedextensively through the solar nebula.

324.05 — Dust back-reaction stops the gas accretionat the snowline

Matías Gárate1; Til Birnstiel1; Joanna Drazkowska1;Sebastian Markus Stammler1

1 Ludwig-Maximilians-Universität München (München, Germany)

In protoplanetary disks, the water snowline can actas a traffic jam for drifting dust particles, leadingto high dust-to-gas ratios in the inner regions. Us-ing numerical simulations that include dust growth,evaporation and condensation of water, and the dustback-reaction onto the gas, we find that the dust con-centration around the snowline can stop the accre-tion of light gases (such as hydrogen and helium),and further enhance the accumulation of solid mate-rial in the inner disk. The back-reaction effects stoponce the reservoir of dust particles from the outerdisk is exausted (in approximately 1 Myr), and nevertake place if the disk turbulence is high, or if the pri-mordial dust content is low.

324.06 — Can a dust gas mixture be modelled as asingle fluid?

Francesco Lovascio1; Sijme-Jan Paardekooper11 Queen Mary, University of London (London, United Kingdom)

In the formation of planets, overcoming the meterbarrier requires a mechanism to speed up planetesi-mal growth and slow down migration. This is espe-cially true in the formation of super earths and othermassive planets, where a large amount of solids arepresent. For all these cases, migration timescales canbe extremely short making it exceptionally difficultto form massive planets. Vortices are a promisingcatalyst for planet formation, collecting and retain-ing dust, allowing for planetesimals to grow with-out quickly migrating towards the star. The stabil-ity of such dust-laden vortices has been shown, bysome analytical and computational work, to becomecompromised as dust collects inside the vortex; po-tentially making vortices short lived in protoplane-tary discs. Despite making up only 1-2% of proto-planetary discs by mass, dust is very important to thedynamics and evolution of the disc. For a completeunderstanding of this phenomenon, a model of thedusty gas is required. The terminal velocity approx-imation is one of these models, it is a way to sim-plify the complex and computationally costly equa-tions involved in modelling dust and gas as two flu-ids coupled by drag terms. By treating the dust asbeing at terminal velocity in the gas, only the evolu-tion of a single fluid has to be solved instead. This

96

approximation makes the equations simpler to ap-proach numerically and more tractable using ana-lytic methods, as it only requires advancing a sin-gle additional equation for the evolution of the dustto gas ratio. We have studied the behaviour of thelocally isothermal terminal velocity approximation,a special case of the terminal velocity approxima-tion where the dust evolution equations becomes amodified equation of state. Through our analyticalstudy we show that the model can trivially be ex-tended to α viscous discs, and that this model breaksdown around shocks. We also have produced an im-plementation of this model within FARGO3D usingan unconditionally stable integrator to advance thenonlinear dust cooling term. With this we set thegroundwork for our study of the dynamics of dustyvortices in protoplanetary discs.

324.07 — Multiple Gaps and Rings Produced byPlanets in Protoplanetary Disks: The Role of Ther-modynamics

Ryan Miranda1; Roman Rafikov2,11 Institute for Advanced Study (Princeton, New Jersey, United

States)2 DAMTP, University of Cambridge (Cambridge, United Kingdom)

Multiple concentric gaps and rings have been re-vealed in a number of protoplanetary disks by sub-mm observations using ALMA. An exciting possi-bility is that these features are produced by planets.In particular, it has been recently suggested that asingle planet in a disk may produce multiple gapsand rings, as a result of the complex propagationand dissipation of the multiple spiral density wavesit excites in the disk. Numerical efforts to verify thisidea and infer the properties of the putative planetshave largely utilized the so-called locally isothermalapproximation, using a prescribed disk temperatureprofile. However, in protoplanetary disks this ap-proximation does not provide an accurate descrip-tion of the density wave dynamics on scales of tensof AU where rings and gaps are observed. Instead, amore realistic treatment of the disk thermodynam-ics, including explicit cooling, is necessary. I willdiscuss how these considerations modify the detailsof planet-disk interaction, using both linear pertur-bation theory and numerical simulations. Locallyisothermal simulations tend to overestimate the con-trast of ring and gap features, as well as misrepresenttheir shapes and positions, when compared to sim-ulations which explicitly solve the energy equationand include cooling. This can be traced to the differ-ent behavior of the wave angular momentum flux,

a key quantity characterizing the strength of planet-excited waves, under different thermodynamic as-sumptions. Caution should be exercised in using lo-cally isothermal simulations to explore planet-diskinteraction, and more generally in other studies ofwave-like phenomena in astrophysical disks. I sug-gest an improved numerical setup for planet-disk in-teractions, which could be used to more accuratelydetermine the plausibility of the observed featuresbeing produced by planets.

324.08 — Constraints on the stickiness of icy aggre-gates in the protoplanetary disk around TW Hya

Takayuki Matsuura1; Satoshi Okuzumi11 Department of Earth and Planetary Science, Tokyo Institute of

Technology (Tokyo, Japan)

Understanding how dust grows in protoplanetarydisks is crucial for understanding how planet for-mation begins. Theoretically, how far dust growthcan proceed is highly uncertain because the sticki-ness of dust aggregates is largely unconstrained. Pre-vious models assumed that water ice is sticky andfacilitates dust growth in the outer part of proto-planetary disks. However, it is now under debatewhether water ice grains are really sticky at low tem-peratures (Gundlach et al. 2018; Musiolik & Wurm2019). It is also possible that some nonsticky mate-rials like CO2 ice cover the grains and prevent theircollisional growth (Musiolik et al. 2016). Elucidat-ing whether dust grains in the outer regions of disksare sticky or not is particularly important for under-standing how icy planets and small solid bodies likecomets form. In this study, we derive constraintson the stickiness of icy aggregates from observationsof the protoplanetary disk around TW Hya, whichhas a massive protoplanetary disk. Recently, high-resolution observations with the ALMA telescope re-vealed that the dusty disk has circular gaps at 25 auand 41 au (Andrews et al. 2016; Tsukagoshi et al.2016). Based on the scenario that the gaps are cre-ated by two sub-Neptune-sized planets, we simulatehow dust aggregates grow and radially drift in thegapped disk assuming that the aggregates fragmentupon collisions at velocities above a given threshold.We find that the fragmentation threshold of as low as0.5 m s−1 gives the best match to the ALMA observa-tions. Higher fragmentation thresholds lead to sig-nificant dust accumulation at the outer edges of theplanetary gaps and to dust depletion interior to the25 au gap, both inconsistent with the observationalappearance of the TW Hya disk. The derived frag-mentation threshold is considerably lower than pre-viously anticipated for aggregates made of 0.1 μm-

97

sized water ice grains (≈ 80 m s−1; Wada et al., 2013).Possible explanations for this include (1) water icegrains are indeed not as sticky as previously thought,(2) the icy grains are larger than 10 μm, and (3) thegrains are covered by nonsticky CO2 ice.

324.09 — Creating Extreme Solar Systems throughstellar flybys

Nicolás Cuello1; Daniel Mentiplay2; Fabien Louvet4;Christophe Pinte2; Francois Menard5; Daniel Price2;Giovanni Dipierro3; Rebecca Nealon3; MatíasMontesinos6; Valentin Christiaens2; Jorge Cuadra1;Guillaume Laibe7

1 Institute of Astrophysics, Pontificia Universidad Católica de Chile(Santiago, RM, Chile)

2 Monash university (Clayton, Victoria, Australia)3 Physics and Astronomy, University of Leicester (Leicester, United

Kingdom)4 Departamento de Astronomía, Universidad de Chile (Santiago,

Chile)5 IPAG, Univ. Grenoble Alpes (Grenoble, France)6 Instituto de Física y Astronomía (Valparaíso, Chile)7 ENS Lyon, Centre de Recherche d’Astrophysique de Lyon (Lyon,

France)

It is now well established that stellar flybys occurwithin stellar cradles. Therefore, the process ofplanet formation around young stellar objects doesnot occur in isolation. More specifically, this kindof encounters dramatically affect the protoplanetarydiscs where planets born. Here we show in detail thedynamical and observational signatures of flybys:warps, spirals, shadows & misalignments. These re-sults allow us to interpret several very recent (un-published) ALMA and VLT observations of discs ex-hibiting mysterious asymmetries (e.g. FU Ori, AS205, UX Tau, SR 24). We have strong reasons tobelieve that we are currently witnessing interactingdiscs. This very fact raises others fundamental ques-tions: What is the effect of the environment on planetformation? How frequent are these encounters? Didthe Solar System experience such a flyby? To con-clude, we discuss the importance of flybys in produc-ing Extreme Solar Systems.

324.10 — The interplay between MRI and GI: vig-orous dynamo and Neptune-mass fragmentation.

Hongping Deng1,21 University of Zurich (Zurich, Switzerland)2 University of Cambridge (Cambridge, United Kingdom)

The gravitational instability (GI) and magneto-rotational instability (MRI) are two major insta-bilities in driving accretion of astrophysical disks.

We performed ultrahigh resolution global three-dimensional magnetohydrodynamic (MHD) simula-tions of self-gravitating disks with outflow bound-ary condition to study the interplay between GI andMRI. For comparison, we also reproduced globalMRI turbulence using the same meshless finite mass(MFM) scheme (Deng et al. 2019) similar to previ-ous finite volume method simulations. In magne-tized self-gravitating disks, the GI spirals becomefuzzy when MHD turbulence fully develops, and su-perthermal fields puff up the GI disk driving strongoutflow. On the other hand, the vertical circulationbesides the spirals drags the mean toroidal fieldsto form poloidal fields which regenerate amplifiedtoroidal fields due to the background shear as alsoshown in local simulations of Riols & Latter (2017).This large scale dynamo is fundamentally differentfrom MRI dynamo and leads to fast accretion. It is re-silient to magnetic diffusivity that kills the MRI com-pletely. We identify tentative resonance and dynamocycles that may be related to the accretion bursts seenin young stellar disks.

The critical cooling rate for GI disk fragmenta-tion is not affected by the MHD turbulence. How-ever, light fragments of a few percent Jupiter masscan form from overdense spirals. Upon the onset ofthe fragmentation, the magnetic fields grow quicklyaround the clump and truncate the mass feedingalong the spiral arms. The low mass fragments inself-gravitating MHD disk provide potential chan-nels for ice giant formation which pose challenges forcore accretion models.

324.11 — Migrating Binaries: Friends or Foes to Cir-cumbinary Planet Formation?

Diego Jose Munoz11 CIERA, Northwestern University (Evanston, Illinois, United

States)

Circumbinary accretion disks are a natural byprod-uct of binary star formation via disk fragmentation.They are key actors in the process of binary coa-lescence/migration, and are the birthplaces of cir-cumbinary planets. Circumbinary disks are thoughtto (1) modulate accretion onto the central objects, (2)modify the orbital elements of the binary and (3) beeccentric, thus altering the usual picture of planetgrowth and migration. Some of these processes areconfirmed observationally, while others are in ten-sion with what is known from observations. In par-ticular, whether the disk decreases or increases theangular momentum of the binary is not a settled is-sue. This debate has broad and important implica-tions for the origin of close binaries, some of which

98

host planets. In this talk, I will discuss how cir-cumbinary hydrodynamics are relevant to star andplanet formation as well as to protoplanetary diskevolution, and will show how moving-mesh numer-ical methods can improve our ability to directly sim-ulate such complex hydrodynamic interactions. Iwill present simulation results produced with theAREPO code, showing that the long-term (steady-state) behavior is consistent with binaries gaining an-gular momentum from the gas. As a result, accret-ing binaries soften rather than harden, contradict-ing the generally accepted picture of disk-induced bi-nary migration. In addition, I will discuss the phe-nomena of pulsed accretion, alternating accretion,and disk eccentricity excitation, all of which are rel-evant to current observations of binary T-Tauri stars(e.g., DQ Tau, UZ Tau E, TWA 3A). In particular, un-derstanding disk eccentricity excitation is essentialfor addressing the planet-forming potential of youngbinary systems, and for establishing parallels to mainsequence binaries (like Kepler-16 and Kepler-38) thatare known to host planets.

324.12 — Modelling Infrared Line Spectra of Com-plex Organic Molecules in Protoplanetary Disks

Hideko Nomura1,2; Chen-En Wei2; Catherine Walsh3;Tom J. Millar4

1 Division of Science, National Astronomical Observatory of Japan(Mitaka, Tokyo, Japan)

2 Department of Earth and Planetary Sciences, Tokyo Institute ofTechnology (Tokyo, Japan)

3 School of Physics and Astronomy, University of Leeds (Leeds,United Kingdom)

4 Astrophysics Research Centre, Queen’s University Belfast (Belfast,United Kingdom)

Protoplanetary disks are the natal place of planets.Understanding chemical components of gas, dustand ice in the disks is essential to investigate the ori-gins of materials in our Solar system and other plan-etary systems. We investigate the synthesis of com-plex organic molecules (COMs) in protoplanetarydisks using a large gas-grain chemical network in-cluding COMs together with a 2D steady-state phys-ical model of a disk irradiated by UV and X-raysfrom the central star. COMs are efficiently formedon cold and warm grains in the disk midplane viahydrogen adding as well as radical-radical reactionson grain surface. Radiation processing on ice formsreactive radicals and helps build further complexity.Part of the icy molecules are photodesorbed into gasand their transition lines become observable. Actu-ally, ALMA observations have detected CH3CN and

CH3OH from protoplanetary disks. The line emit-ting region of these molecules are the outer relativelycold disk, which suggests that the molecules are non-thermally desorbed from grains following the for-mation on dust grains. Based on our model calcula-tions, we perform ray-tracing calculations to predictline spectra of complex organic molecules observablewith SPICA, which shows that detectable lines tracewarm inner region of the disk where radical-radicalreactions and thermal desorption take place. Also,we discuss possible connection of COMs in proto-planetary disks to those in the Solar system objects,such as comets.

324.13 — Irradiated Disks Naturally Form Ringsand Gaps

Yoram Lithwick1; Yanqin Wu21 Northwestern University (Evanston, Illinois, United States)2 University of Toronto (Toronto, Ontario, Canada)

Images of protoplanetary disks (e.g., by ALMA) haveshown that typical disks are not the smooth power-laws beloved by theorists. Rather, bright rings anddark gaps are ubiquitous. Much work on explainingthese surprising features has blamed unseen planets.Here we show that the features can be accounted forby a different possibility: that irradiated disks nat-urally form rings and gaps when the self-consistentmigration of dust is properly accounted for. Planetformation likely proceeds differently in such a struc-tured disk.

325 — Debris Disk Observationsand Modeling, Poster Session325.01 — Dust Spreading in Debris Discs: DoSmall Grains Cling on to Their Birth Environment?

Nicole Pawellek1,2; Attila Moór2; Ilaria Pascucci1,3;Alexander Krivov4

1 Max-Planck-Institut für Astronomie (Heidelberg, Germany)2 Konkoly Observatory (Budapest, Hungary)3 Lunar and Planetary Laboratory, University of Arizona (Tuscon,

Arizona, United States)4 Astrophysikalisches Institut und Universitätssternwarte (Jena,

Germany)

Debris discs are dusty belts of planetesimals aroundmain-sequence stars, similar to the asteroid andKuiper belts in our solar system. The planetesimalscannot be observed directly, yet they produce de-tectable dust in mutual collisions. Observing the

99

dust, we can try to infer properties of invisible plan-etesimals. Here we address the question of what isthe best way to measure the location of outer plan-etesimal belts that encompass extrasolar planetarysystems. A standard method is using resolved im-ages at mm-wavelengths, which reveal dust grainswith sizes comparable to the observational wave-length. Smaller grains seen in the infrared (IR) aresubject to several non-gravitational forces that dragthem away from their birth rings, and so may notclosely trace the parent bodies. In this study, weexamine whether imaging of debris discs at shorterwavelengths might enable determining the spatiallocation of the exo-Kuiper belts with sufficient accu-racy. We find that around M-type stars the dust bestvisible in the mid-IR is efficiently displaced inwardfrom their birth location by stellar winds, causing thediscs to look more compact in mid-IR images thanthey actually are. However, around earlier-type starswhere the majority of debris discs is found, discs arestill the brightest at the birth ring location in the mid-IR regime. Thus, sensitive IR facilities with good an-gular resolution, such as MIRI on JWST, will enabletracing exo-Kuiper belts in nearby debris disc sys-tems.

325.02 — Testing the Paradigm of Asteroidal Dustaround White Dwarfs with the Prototype

Ted von Hippel1; Judi Provencal3; Jay Farihi4; ScotKleinman2; Gilles Fontaine5; Jim Pringle6

1 Embry Riddle Aeronautical University (Daytona Beach, Florida,United States)

2 Gemini Observatory (Hilo, Hawaii, United States)3 University of Delaware (Newark, Delaware, United States)4 University College London (London, United Kingdom)5 University of Montreal (Montreal, Quebec, Canada)6 University of Cambridge (Cambridge, United Kingdom)

At least one quarter of all white dwarfs are ac-tively accreting debris from planetesimals or plane-tary fragments. The prototype system G29-38 wasdiscovered at the IRTF in 1987. Yet despite the in-tervening decades and a complete paradigm shift inthe explanation from interstellar material to exoplan-etary debris, there remain fundamental questions.The common assumption now is that the dust debrisis in a circumstellar disk, yet if so the geometry andvertical optical depth are observationally degenerate.Optically thin and thick cases vary in disk mass —and hence parent body mass — by orders of mag-nitude. The parent body masses have far-reachingimplications for planetary system architecture andlong-term dynamics. We report on our work to breakthis degeneracy and even to test whether the dust is

in a disk at all or some other geometrical distributionbased on the fact that the prototype system containsa well-studied, pulsating star. Using MORIS andSpeX at the NASA 3-meter Infrared Telescope Facil-ity, we simultaneously monitored the optical stellarpulsations and the ensuing infrared dust response.These observations can distinguish among a range ofdust configurations, based on the observed infraredresponse to the known geometry of the optical pul-sations.

325.04 — A survey for resolved debris disks in theSco-Cen association with the Gemini Planet Im-ager

Jenny Patience1; Justin Hom1; Thomas M. Esposito2;Paul Kalas3; Marshall D. Perrin7; Elisabeth Matthews4;Pauline Arriaga6; Christine Chen7; Johan Mazoyer8;Maxwell A. Millar-Blanchaer8; Stanimir Metchev5;Brenda Matthews9; Michael P. Fitzgerald6; SchuylerWolff10; Gaspard Duchene2

1 Earth and Space Exploration, Arizona State University (Tempe,Arizona, United States)

2 University of Leiden (Leiden, Netherlands)3 Astronomy, UC Berkeley (Berkeley, California, United States)4 University of California, Berkeley (Berkeley, California, United

States)5 MIT (Somerville, Massachusetts, United States)6 Physics & Astronomy, University of Western Ontario (London,

Ontario, Canada)7 UCLA (Los Angeles, California, United States)8 Space Telescope Science Institute (Baltimore, Maryland, United

States)9 JPL (Pasadena, California, United States)10 HIA (Victoria, British Columbia, Canada)

With the Gemini Planet Imager (GPI), we are con-ducting a survey of debris disk systems in theScorpious-Centaurus OB association. Each targetwas observed in the polarimetry mode of GPI whichis specifically designed for spatially-resolved, high-contrast observations of debris disks. A subset ofthe targets was also observed in the spectroscopicmode of GPI, covering the same H-band as the po-larimetry mode data. The target sample consistsof a complete set of 28 early-type A and F stars lo-cated within the Upper Centaurus Lupus and LowerCentaurus Crux regions of Sco-Cen that are brightenough for GPI observations and exhibit the high-est Spitzer-detected infrared excesses, with values ofLIR/Lstar above 2.5×10−4. Overall goals of the ongo-ing survey include exploring the range of debris diskproperties around stars of similar age and formationenvironment, placing constraints on disk geometricproperties and the potential dynamical signatures of

100

planets. Of the systems observed thus far, 80 per-cent have been spatially resolved. Among the newly-resolved systems, the majority have close to an edge-on geometry, one system shows a ring structure andanother disk exhibits significant asymmetry, possi-bly indicating the presence of a substellar companiongravitationally interacting with the debris disk.

326 — Planetary Atmospheres —Hot Jupiters326.01 — Exploring the Internal Structures of hotJupiters using the GCM DYNAMICO: Deep, Hot,Adiabats as a Possible Solution to the Radius Infla-tion Problem

Felix Sainsbury-Martinez1; Pascal Wang2,1; SebastianFromang3; Pascal Tremblin1; Thomas Dubos4; YannMeurdesoif5; Jermey Leconte6; Aymeric Spiga4; Is-abelle Baraffe7; Nathan Mayne7; Florian Debras2; GillesChabrier2,7; Ben Drummond7

1 MDLS, CEA Paris Saclay (Gif-sur-Yvette cedex, France)2 ENS Lyon (Lyon, France)3 DAP CEA Paris-Saclay (Saclay, France)4 LMD/IPSL, Ecole Polytechnique (Saclay, France)5 LSCE/IPSL, Universite Paris-Saclay (Saclay, France)6 Universite de Bordeaux (Bordeaux, France)7 Astrophysics, University of Exeter (Exeter, United Kingdom)

The anomalously large radii of highly irradiated ex-oplanets have long remained a mystery to the Exo-planetary community, with many different solutionssuggested and tested. These solutions have includedtidal heating of the atmosphere, or ohmic heatingfrom a strong magnetic field. Another solution wasalso suggested by Tremblin et Al. (2017): The in-flated radii of highly irradiated exoplanets can be ex-plained by the advection of potential temperature,via mass and longitudinal momentum conservation,leads to the deep atmosphere attaching to a hotteradiabat than would be suggested by 1D models, thusimplying an inflated radius. In that paper this mech-anism was tested using 2D steady-state models, andsuccessfully reproduced an inflated HD209458b sce-nario. Here we extend this work to both the time-dependent and 3D regimes using the GCM Dynam-ico (Itself developed as a new dynamical core forLMD-Z, and verified against Hot Jupiter benchmarksas part of this work), exploring the evolution of thedeep P-T profile, and the stability of a deep adiabat asthe steady state solution. As a result of these calcula-tions we confirm that a deep, hot, adiabat is both thetarget of long term evolution of the deep atmosphere,and is stable against typical forcing expected at deep

pressures — we also note that this deep adiabat takesa very significant time to form from an isothermalinitial condition (hence why it has not previouslybeen seen in GCM simulations beyond a kink in thedeep profile), and suggest that future GCM modelsshould use an adiabatic profile to initialise the deepatmosphere. Taken as a whole, our results confirmthe theory of Tremblin et Al. (2017): the inflated radiiof highly irradiated exoplanets can be explained byconnecting the atmosphere with a deep, hot, internaladiabat.

326.02 — Admissible types of magnetospheres ofhot Jupiters

Dmitry Bisikalo11 Institute of astronomy of the Russian Academy of Sciences

(Moscow, Russian Federation)

The orbits of exoplanets of the hot Jupiters type, i.e.,exoplanets with masses comparable to the mass ofJupiter and orbital semi-major axes less than 0.1 AU,as a rule, are located close to the Alfven point ofthe stellar wind of the parent star. At this, manyhot Jupiters can be located in the sub-Alfven zonein which the magnetic pressure of the stellar windexceeds its dynamic pressure. Therefore, magneticfield of the wind must play an extremely importantrole for the flow of the stellar wind around the atmo-spheres of the hot Jupiters. This factor must be con-sidered both in theoretical models and in the inter-pretation of observational data. The analysis showsthat many typical hot Jupiters should have shock-less intrinsic magnetospheres, which, apparently, donot have counterparts in the Solar System. We con-firmed this inference by the three-dimensional nu-merical simulation of the flow of the parent star stel-lar wind around the hot Jupiter HD 209458b in whichwe took into account both proper magnetic field ofthe planet and magnetic field of the wind.

326.03 — Dynamical effects of hydrogen disso-ciation on atmospheric circulation of ultra-hotJupiters

Xianyu Tan1; Thaddeus Komacek21 Department of Physics, University of Oxford (Oxford, England,

United Kingdom)2 University of Chicago (Chicago, Illinois, United States)

Recent secondary eclipse spectral and phase curveobservations of ultra-hot Jupiters with dayside tem-peratures in excess of 2500 K have found evidence fornew physical processes at play in their atmospheres.

101

Here, we investigate the dynamical effects of the dis-sociation of molecular hydrogen and recombinationof atomic hydrogen on the atmospheric circulation ofultra-hot Jupiters. To do so, we incorporate these ef-fects into a general circulation model (GCM) for hotJupiter atmospheres, and run a large suite of mod-els varying the incident stellar flux and strength offrictional drag. We find that including hydrogendissociation and recombination reduces the day-to-night temperature contrast of ultra-hot Jupiter atmo-spheres and causes the speed of the equatorial jet todecrease. This is because the large energy input re-quired for hydrogen dissociation cools the dayside ofthe planet, and the energy released due to hydrogenrecombination warms the nightside. The associatedlarge heating/cooling rate and the mean molecularweight change modify the wave-mean-flow interac-tions, which likely results in the weaker equatorialjet. The results from our GCM experiments qualita-tively agree with previous theory that the day-nighttemperature contrast of ultra-hot Jupiters should de-crease due to hydrogen dissociation and recombina-tion. Lastly we compute full-phase light curves fromour suite of GCMs, finding that the reduced day-to-night temperature contrast in ultra-hot Jupiter atmo-spheres causes a smaller phase curve amplitude. Thereduction in phase curve amplitude due to hydrogendissociation and recombination could explain therelatively small phase curve amplitudes of observedultra-hot Jupiters WASP-33b, WASP-103b and KELT-9b. Our work would first provide valuable under-standing on the basic dynamical processes of ultra-hot Jupiters, helping to interpret future observationsof their atmospheres. Secondly our work wouldstimulate further theoretical investigations, reveal-ing complex interplays between different physicalprocesses in atmospheres of ultra-hot Jupiters.

326.04 — Time Variability in Hot Jupiter Atmo-spheres

Thaddeus Komacek1; Adam Showman21 University of Chicago (Chicago, Illinois, United States)2 Lunar and Planetary Laboratory (Tucson, Arizona, United States)

Hot Jupiter atmospheres are expected to be dynamicenvironments with time-variable large scale circula-tion patterns. However, to date there has been nodetection of variability of hot Jupiter atmospheresin the infrared, while the visible light phase curveof HAT-P-7b has been observed to show a periodicoscillation in its phase offset. In this work, we per-form the first study of the expected infrared time-variability of a broad range of hot Jupiter atmo-spheres in preparation for JWST. To do so, we per-

form a large suite of atmospheric circulation mod-els, varying the incident stellar flux and atmosphericdrag strength. We place lower limits on the atmo-spheric variability, as we do not include the effectsof Lorentz forces, large-scale shear instabilities, andclouds. In general, we find that the amplitude ofvariability is largest in the equatorial regions and in-creases with increasing incident stellar flux and fric-tional drag strength, both in terms of temperatureand emergent flux. We show that JWST will be ableto detect variability due to atmospheric circulationin the secondary eclipse depth and may be able todetect variability in the phase curve amplitude andoffset. The periodicity and amplitude of detectedvariability will provide a first-order understandingof the interaction of large-scale standing waves in hotJupiter atmospheres. Additionally, if variability sig-nificantly larger than expected from our simulationsis found, that may provide evidence for additionalphysical processes (e.g., magnetic effects, clouds) af-fecting the emergent flux of hot Jupiters.

326.05 — Leaking Exoplanets: Understanding howStars Affect Atmospheric Escape in Exoplanets

Andrew Cleary1; Aline Vidotto11 Trinity College Dublin (Dublin, Leinster, Ireland)

The atmospheres of highly irradiated exoplanets areobserved to undergo hydrodynamic escape, result-ing in planetary mass loss. However, stellar windscan shape and even prevent atmospheric escape,affecting observable signatures of escape such asLyman-α and H-α line profiles. In this work, wesimulate atmospheric escape of close-in exoplanetsand investigate whether they are affected by stellarwinds. We show that, although younger hot-Jupitersexperience higher levels of atmospheric escape, ow-ing to a favourable combination of higher irradiationlevels and weaker planetary gravity, stellar windsare also stronger at this young age, which act to re-duce/inhibit escape rates of young exoplanets.

326.06 — Revisiting the NUV Transmission Spec-trum of HD 209458b: Signs of Ionized Iron Beyondthe Roche Lobe

Patricio Cubillos1; Luca Fossati1; Tommi Koskinen3;Mitchell Young1; Kevin France2; Michael Salz4; AikaraSreejith1; Carole Haswell5

1 Space Research Institute, Austria (Graz, Austria)2 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)3 Lunar and Planetary Laboratory, University of Arizona (Tucson,

Arizona, United States)

102

4 Hamburger Sternwarte, Universitaet Hamburg (Hamburg, Ger-many)

5 Department of Physical Sciences, The Open University (MiltonKeynes, United Kingdom)

Ultraviolet transit observations probe the upper at-mosphere of exoplanets, where mass loss occurs.Our analysis of the archival HST/STIS NUV trans-mission observations of HD 209458b shows evidencefor ionized iron, but no evidence for neutral iron,neutral magnesium, nor ionized magnesium. Whileour non-detection of neutral magnesium resolvesthe tension with theoretical models from previousresults, our results are at still odds with lower-atmosphere models resulting from optical and in-frared observations. These upper-atmosphere obser-vations indicate that hydrodynamic escape is strongenough to carry heavy atoms like iron beyond theplanetary Roche lobe; however, lower-atmosphereobservations suggest the presence of cloud conden-sates. With iron-bearing aerosols condensating morestrongly than magnesium-bearing aerosols, if mag-nesium is trapped in the lower atmosphere, ironshould be as well. The intricate relationship betweenlower- and upper-atmosphere properties makes thecombination UV and optical/infrared observationsmore valuable than the sum of its individual parts.The unique properties of the HD 209458 system placeits transiting hot Jupiter in a pivotal role in ourunderstanding of planetary atmospheres, few otherplanets will ever enable such precise measurementsof both their upper- and lower-atmosphere proper-ties. Here, I will present the analysis and theoreticalinterpretation of the HD 209458b NUV observations.Then I will discuss the prospects of future observa-tions to elucidate the puzzling nature of this planet’satmosphere as a whole, which is of particular valuebefore the the imminent launch of the James WebbSpace Telescope.

326.07 — Supervised machine learning for inter-preting ground-based, high-resolution transmis-sion spectra of exoplanets

Chloe Fisher1; H. Jens Hoeijmakers1,2; DanielKitzmann1; Simon Grimm1; Pablo Márquez-Neila3;David Ehrenreich2; Raphael Sznitman3; Kevin Heng1

1 Center for Space and Habitability, University of Bern (Bern,Switzerland)

2 Observatoire astronomique de l’Université de Genève (Geneva,Switzerland)

3 ARTORG Center for Biomedical Engineering, University of Bern(Bern, Switzerland)

We present a novel, unpublished approach to per-

forming atmospheric retrieval on high-resolutionground-based data for exoplanets using supervisedmachine learning. We have developed a techniquethat combines the well-established method of cross-correlation with our random forest retrieval algo-rithm.

High-resolution spectroscopy using meter-class,ground-based telescopes has revolutionized ourability to identify atoms and molecules in the at-mospheres of exoplanets. However, the high lev-els of noise and large number of spectral points pro-vide a challenge for traditional methods of retrieval.Currently, detections of molecules are made usingthe technique of cross-correlation, which matchesline positions of atomic and molecular species withthe high-resolution absorption spectra. But used onits own, cross-correlation does not yield the pos-terior distribution of the abundance of an atom ormolecule, or the properties of the atmosphere beingobserved.

We introduce a hybrid retrieval method that com-bines the cross-correlation method with a supervisedmachine learning method using the random forest.It leverages the statistical content of the spectrum toovercome the high level of noise and uses featureengineering to reduce the size of the training set.We use the hybrid method to interpret the HARPS-N transmission spectrum of KELT-9b, deriving pos-terior distributions for the metallicity and temper-ature and demonstrating that it is able to diagnosemissing physics in the retrieval. The hybrid methodwill be decisive for performing retrieval on suites ofhigh-resolution spectra with broad wavelength cov-erage as the next generation of ground-based spec-trographs come online.

326.08 — Modeling disequilibrium chemistry ofexoplanet atmospheres using a sequence of post-processed forward models

Robin Baeyens1; Leen Decin1; Ludmila Carone2; OliviaVenot3

1 KU Leuven (Leuven, Belgium)2 Max-Planck-Institut für Astronomie (Heidelberg, Germany)3 Laboratoire Interuniversitaire des Systèmes Atmosphériques

(Créteil, France)

In anticipation of the next era of space telescopesfor exoplanet characterization (James Webb, ARIEL)it is essential that sophisticated modeling tools forthe atmospheres of transiting planets are developed.However, the associated effects of strong stellar ir-radiation and tidal locking make these objects in-herently three-dimensional (3D) in nature, and mul-tidimensional forward models are thus required to

103

accurately simulate the multitude of processes thatcomprise an atmospheric system. This is especiallythe case for the out-of-equilibrium chemical compo-sition, which is tightly coupled to the planetary cli-mate through dynamical quenching and can showlarge longitudinal variations due to day-night tem-perature differences and photochemical reactions.Despite fast developments in the field, coupling 3Dgeneral circulation models (GCM) with radiativetransfer and chemistry, computation times are a ma-jor bottleneck of these models and thus they are onlyapplied to a small number of planets.

In an effort to remedy this limitation, we employ arange of post-processed forward models in sequence:a 3D GCM (MITgcm, Adcroft+2004) with simpli-fied, Newtonian radiative transfer (based on petit-CODE, Mollière+ 2015), a post-processed pseudo-2Dchemical network solver (Agundez+ 2014) and a ray-tracing code (petitRADTRANS, Mollière+ 2019), tocompute an extensive grid of planetary atmospheresand synthetic transmission spectra. This allows us tostudy the mechanisms of disequilibrium chemistryand their effect on the observables in a systematicway for a large range of planets. More specifically,we report on the change in synthetic transmissionspectra due to longitudinally-varying vertical mixingand photochemistry. This enables us to derive gen-eral parametrizations for these processes for imple-mentation in 1D retrieval codes, a necessary step inpreparing for the interpretation of high-quality datacoming from the James Webb Space Telescope andARIEL.

326.09 — Exoplanet atmosphere characterizationwith SPIRou: first results for HD 189733b

Anne Boucher1; Antoine Darveau-Bernier1; DavidLafrenière1; Romain Allart2; Stefan Pelletier1; NeilCook1; Étienne Artigau1; Christophe Lovis2; BjörnBenneke1; René Doyon1; Claire Moutou3

1 Institute for Research on Exoplanets, Université de Montréal(Montreal, Quebec, Canada)

2 Geneva observatory, University of Geneva (Versoix, Switzerland)3 Canada-France-Hawaii Telescope (Waimea, Hawaii, United States)

SPIRou is the new high-resolution, near-infraredspectro-polarimeter at the Canada-France-HawaiiTelescope. Primarily built to detect earth-like exo-planets around M-dwarfs through precise radial ve-locity measurements, it is also, thanks to its largespectral range (Y to K band) and its high R∼70,000resolving power, an excellent instrument for exo-planet atmospheric characterization via transit andemission spectroscopy. SPIRou observed its first ex-oplanet transit — of the planet HD 189733b — in

September 2018. I will present preliminary resultsof this first observation, and of other transits thathave been observed since then as part of the SPIRouLegacy Survey. Thus far, we can re-confirm the de-tections of water and metastable helium signals inthe transmission spectrum of HD 189733b.

326.10 — Atmospheric Characterization ofHD209548 and HD189733 with High ResolutionCross Correlation Spectroscopy

Joseph Zalesky1; Rebecca Webb2; Michael Line1; MatteoBrogi2

1 Arizona State University (Tempe, Arizona, United States)2 University of Warwick (Warwick, United Kingdom)

High Resolution Cross Correlation Spectroscopy(HRCCS) has become a powerful tool to constrainboth the physical characteristics and abundances ofatomic/molecular constituents in exoplanetary at-mospheres. Brogi & Line (2019) recently introduceda novel Bayesian atmospheric retrieval methodol-ogy that can combine observations from both longerwavelength (2-4 micron), ground-based, HRCCS andshorter wavelength (1-2 micron) space-based obser-vatories such as the Hubble Space Telescope (HST).Here we present results from the first application ofthis technique to both new and previously publishedobservations of HD209458b and HD189733b fromVLT/CRIRES, HST, and Spitzer. The more com-plete wavelength coverage provides a more com-prehensive assessment of the atmosphere by way ofstronger constraints on the thermal profiles, atmo-spheric metallicity, and carbon/oxygen inventoryfor these two benchmark planets. We also inves-tigate the impact of possible model-induced biasesincluding assumptions regarding molecular cross-sections, cloud model prescriptions, and thermalprofile parameterizations. Finally, we present whatconstraints may be possible in the future by per-forming retrievals of synthetic observations fromthe next generation of high-resolution spectrographslike CRIRES+. This work has laid a foundationaldataset that combines both space and ground-basedobservations to comprehensively characterize exo-planetary atmospheres and will be a useful bench-mark in comparison to future efforts for both transit-ing and non-transiting atmospheric characterization.

326.11 — Three-Dimensonal Mixing of Photochem-ical Hazes in the Atmospheres of Hot Jupiters

Maria Elisabeth Steinrueck1; Adam Showman1; TommiKoskinen1; Panayotis Lavvas2

104

1 Lunar and Planetary Laboratory, University of Arizona (Tucson,Arizona, United States)

2 Groupe de Spectrométrie Moleculaire et Atmosphérique, Univer-sité de Reims Champagne Ardenne, Reims (Reims, France)

The transmission spectra of many hot Jupiters showsignatures of high-altitude aerosols. The natureand composition of these aerosols is unknown—one possible explanation is that they are producedthrough photochemical processes. Previous studiesof photochemical hazes on tidally locked exoplan-ets used one-dimensional models. These 1D mod-els have to make strongly simplifying assumptionsabout the strength of vertical mixing. Furthermore,they ignore that the strong day-night contrast onhot Jupiters, the spatially varying production rate ofphotochemical species acting as haze precursors andthe interaction with the atmospheric circulation canlead to inhomogeneous aerosol distributions. Gen-eral circulation models (GCMs) are needed to bet-ter understand how photochemical hazes are mixedby the atmospheric circulation, what the resulting3D aerosol distributions are, and how 1D modelerscan make more informed choices to represent 3Dprocesses as accurately as possible in a 1D frame-work. As photochemical hazes are produced atmuch higher altitudes than condensate clouds andare expected to have small particle sizes, they mayinteract differently with the atmospheric circulationthan condensate clouds.

We present results from GCM simulations of hotJupiter HD 189733b to explore the mixing of pho-tochemical hazes. With an equilibrium tempera-ture near 1,200 K, this well-studied planet is coolenough that photochemical hazes and condensateclouds with a variety of compositions are both ex-pected to exist. We use passive tracers represent-ing photochemical hazes to study how hazes aretransported by the atmospheric circulation. We alsoinclude tracers representing condensate clouds forcomparison. We present the resulting 3D distribu-tions of each species for different constant particlesizes. Furthermore, we analyze the efficiency of ver-tical mixing and derive effective eddy diffusion co-efficients that describe resulting global-mean parti-cle distribution, to be used in 1D models. Our studyalso addresses sensitivity to initial conditions of thetracer abundance and the long-term behavior of thesimulations.

326.12 — Transit Timing Variations and Transmis-sion Spectroscopic Studies of Seven Exoplanetswith Thai Telescopes

Supachai Awiphan1; Eamonn Kerins2; Jake Morgan2;Josh Hayes2; Iain Mcdonald2; Ekburus Boonsoy3;Patcharawee Munsaket3; Sutthawee Yodmongkol3;Prangsutip Cherdwongsung4; Tanagodcha-porn Inyanya5; Premtanut Tundee6; PongpichitChuanraksasat1; Suphakorn Suphapolthaworn7; SiramasKomonjinda6; Nuanwan Sanguansak3; Phichet Kittara4

1 National Astronomical Research Institute of Thailand (ChiangMai, Chiang Mai, Thailand)

2 Jodrell Bank Centre of Astrophysics, University of Manchester(Manchester, United Kingdom)

3 School of Physics, Institute of Science, Suranaree University ofTechnology (Nakhon Ratchasima, Thailand)

4 Department of Physics, Faculty of Science, Mahidol University(Bangkok, Thailand)

5 Yupparaj Wittayalai School (Chiang Mai, Thailand)6 Department of Physics and Materials Science, Faculty of Science,

Chiang Mai University (Chiang Mai, Thailand)7 Department of Physics, School of Science, Hokkaido University

(Sapporo, Japan)

Since 2012, National Astronomical Research Instituteof Thailand (NARIT) has operated the 2.4 metre ThaiNational Telescope and seven 0.5-0.7 metre class tele-scopes of the Thai Robotic Telescopes network, lo-cated in Thailand, Chile, China, USA and Australia.As a part of collaborations with the Spectroscopy andPhotometry of Exoplanetary Atmospheres ResearchNetwork (SPEARNET), seven transitting exoplan-ets: HAT-P-26b, HAT-P-36b, HAT-P-43b, KELT-3b,WASP-11b/HAT-P-10b, WASP-43b and WASP-127b,were monitored with Thai telescopes from 2015 to2019. Transit Timing Variations (TTV) and transmis-sion spectroscopic analyses are performed on multi-wavelength optical photometric observational data,yielding results which indicate interesting proper-ties of monitered exoplanets. TTV analysis of WASP-43b shows that the case of orbital decay can be ruledout, while in WASP-127b a significant Rayleigh scat-tering effect is confirmed through transmission spec-troscopic analysis of u’-band light curves. Moreover,TTV analyses of two planets with large TTVs: HAT-P-26b and HAT-P-43b, have been performed in or-der to verify results published in the previous works.The first optical transmission spectroscopic analysesof two hot-Jupiters HAT-P-36b and WASP-11b/HAT-P-10b are also carried out.

105

326.13 — Helium in extended exoplanet atmo-spheres observed at high spectral resolution

Romain Allart1; Vincent Bourrier1; Christophe Lovis11 Geneva Observatory, University of Geneva (Versoix, Switzerland)

Since the early age of exoplanetology, the near in-frared helium triplet was seen as one of the mostpromising tracers of exoplanet atmospheres. How-ever, it is only very recently, with the help of high-resolution spectrographs, that it could be detectedand its full potential exploited.

I will show that high-resolution spectrographs canretrieve a fully resolved helium signature. I willpresent two helium detections obtained on the warmNeptune HAT-P-11b and the warm Saturn WASP-107b, and I will show how the helium triplet canprobe the atmospheric expansion and dynamics. Us-ing the CARMENES spectrograph, the helium fea-tures have been spectrally and temporally resolved,at an overall significance of more than 20 σ. Thehelium profile of HAT-P-11b is clearly symmetricbut slightly blueshifted, while the helium profile ofWASP-107b is asymmetric with a clear excess absorp-tion in its blue wing. We model these profiles witha 3D code allowing us to characterize the structureand dynamics of their extended atmospheres (ther-mosphere and exosphere). The HAT-P-11b modelingshows that helium is present in its thermosphere, isnegligible in its exosphere and reveals the presenceof zonal wind from the day to the night side. On thecontrary, the modeling of WASP-107b shows that he-lium fills the thermosphere but the majority of it ispresent in the exosphere in the form of a cometary-like tail.

In general, the near infrared helium triplet offersstrong synergies with space-based UV observations.It will help us understand the formation and evolu-tion of the hot Neptune evaporation desert by study-ing planets from Jupiters to super-Earths in differentconditions of irradiation and stellar environments.

326.14 — Remote sensing of exoplanetary atmo-spheres with ground-based high resolution near-infrared spectroscopy

Denis Shulyak1; Miriam Rengel1; Ansgar Reiners2; UlfSeemann2; Fei Yan2

1 Max Planck Institute for Solar System Research (Goettingen,Germany)

2 Institute for Astrophysics (Goettingen, Germany)

Thanks to the advances in modern instrumentationwe learned about many exoplanets that spawn awide range of masses and composition. Studying

their atmospheres provides insight into planetary di-versity, origin, evolution, dynamics, and habitabil-ity. Present and future observing facilities will ad-dress these important topics in very detail by us-ing more precise observations, high-resolution spec-troscopy, improved analysis methods, etc. In thiscontribution we investigate the feasibility to retrievethe vertical temperature distribution and molecularabundances from expected exoplanet spectra at veryhigh spectral resolution from ground based obser-vations. We focus our research on Cryogenic high-resolution IR Echelle Spectrometer (CRIRES+) on theVery Large Telescope scheduled by the end of 2019for the European Southern Observatory (ESO). Theinstrument will operate between 0.9-5.3 micron witha highest achievable resolving power of R=100k. Westudied several cases of simulated observations atdifferent spectral bands, their combinations, spectralresolving powers, and signal-to-noise ratios (S/N).Our simulations show that we can retrieve accu-rate temperatures in very wide range of atmosphericpressures between 1 bar and 10−6 bar dependingon the chosen spectral region. Retrieving molecularmixing ratios is very challenging, but a simultaneousobservations in two separate infrared bands will helpto obtain accurate estimates. For that, the exoplane-tary spectra must be of relatively high S/N>10. Wealso present determinations of optimal observationalstrategies to increase accuracy in the retrievals andlist potential targets. We show that high resolutionnear infrared spectroscopy is a powerful tool to studyexoplanet atmospheres because profiles of numer-ous lines of different molecules could be analysed si-multaneously. Instruments similar to CRIRES+ willprovide data for detailed retrievals and bring newimportant constraints for the atmospheric chemistryand physics.

326.15 — Theoretical Spectra of Young Giant Plan-ets at Optical Wavelengths

Brianna Lacy1; Adam Burrows11 Astrophysical Sciences, Princeton University (Princeton, New

Jersey, United States)

Optical characterization of young planets is a natu-ral intermediate step as the community strives to de-velop the technology necessary to characterize ma-ture exoplanets. While young planets are dimmerin the optical than in the infrared, they are gener-ally still much brighter in the optical than a matureplanet of similar mass. An instrument just barelyable to characterize a mature planet, should easilyattain high precision photometric imaging and spec-troscopy of young self-luminous planets. In this

106

work we present spectra for a selection of knownand hypothetical self-luminous substellar compan-ions extending into optical wavelengths. For thehypothetical objects we compute spectra at severalpoints along their evolution for a grid of masses andplanet-star separations. For each mass and planet-star separation we determine the age at which boththe reflected-light and the self-luminous portion ofthe planet spectrum are contributing significantly tothe optical spectrum, because this regime presentsexciting possibilities for breaking the degeneracy be-tween a planet’s phase function and radius. Thespectra computed for known systems are selected tobe possible observing targets for WFIRST-CGI.

326.16 — Uniformly hot nightside temperatures onshort-period gas giants

Dylan Keating1; Nick B. Cowan1; Lisa Dang11 Physics, McGill University (Montréal, Quebec, Canada)

I will present results of a meta analysis of the full-orbit phase curves of twelve hot Jupiters, rangingfrom HD 189733b to the ultra-hot Jupiter WASP-33b. Using published phase curves at multiple wave-lengths for each planet, we inferred the wavelengthdependent day side and night side brightness tem-peratures, and used Gaussian process regression toestimate effective temperatures for the day and nightsides of each planet. We found that even thoughdayside temperatures on these planets increase lin-early with increasing amounts of stellar irradiation,their nightside effective temperatures are all clus-tered around 1100K, with a slight upward trend.Neither atmospheric model we attempted to fit couldexplain the nightside trend — our favoured explana-tion is that these planets all have very similar cloudspecies on their nightsides, which condense slightlyabove ∼1100K, and radiate at a similar temperature.

326.17 — Irradiation of exoplanetary atmosphere

Lalitha Sairam11 Institute for Astrophysics, University of Goettingen (Goettingen,

Germany)

Nearly 75% of the stellar population in our galaxyis made up of low mass/ red stars, making them themost common stars in our neighborhood, in turn, themost frequent planet hosts. Due to intrinsically lowluminosities of red stars, the habitable zone lies closeto the host stars making the orbiting world extremelyvulnerable. The high-energy radiation from the hoststars strongly determine the amount of gas lost fromthe atmospheres of close-in exoplanets. In this talk,

I will address the consequence of high energy radi-ation absorbed by the upper planetary atmosphere.I will present my ongoing work on a systematic anddetailed characterization of irradiation-induced exo-planetary mass-loss.

326.18 — Multi-color transit observations of thewarm Jupiter WASP-80b with MuSCAT/MuSCAT2

Yuka Terada1; Akihiko Fukui1; Norio Narita2; Moto-hide Tamura1,2; John Livingston1; Jerome Pitogo DeLeon1; Mayuko Mori1; Nobuhiko Kusakabe2; NoriharuWatanabe3; Taku Nishiumi4

1 The University of Tokyo (Tokyo, Japan)2 Astrobiology Center (Tokyo, Japan)3 Astronomical Science, SOKENDAI (Graduate University for

Advanced Studies) (Mitaka, Tokyo, Japan)4 Kyoto Sangyo University (Kyoto, Japan)

Transit depths of a true planet have a weak wave-length dependence caused by the nature of the plan-etary atmosphere. To measure the weak wavelengthdependence in transit depths is useful to study thecomposition of a planetary atmosphere. WASP-80bis a warm Jupiter in a 3-day orbit around a late-K/early-M dwarf. The number of giant planetsaround a late-K/early-M dwarf which have been dis-covered so far is small. As the planet has an equi-librium temperature of ∼800 K or less, the existenceof haze is theoretically suggested. The transmissionspectrum observed by Kirk et al. (2018) indicatesa weak negative slope, suggesting the possible ex-istence of haze in the atmosphere. On the otherhand, Parviainen et al. (2017) reported a flat spec-trum, indicating a negative result on the haze exis-tence. However, in observations made so far, theplanet-to-star radius ratio in g-band has not yet beenconstrained enough. If haze exists, the possibilitythat the spectrum in the g band becomes larger thanother bands due to the effect of Rayleigh scattering.Therefore, we conducted multi-color transit observa-tions including the g band using the Okayama 188cm telescope / MuSCAT and the Canary Islands 1.5m telescope / MuSCAT 2. In this study, we usetwo approaches to remove systematics, linear modeland Gaussian Process (GP). As a result, we foundthat GP could remove systematics better than linearmodel. As a final result, neither the flat model northe Rayleigh slope model could be rejected with theresult of both MuSCAT and MuSCAT2, but the re-sult of the only MuSCAT2 supported the flat model.In the future, we plan to investigate the characteris-tics of warm planets’ atmosphere by observing simi-lar planets.

107

326.19 — Wind of Change: Revealing thermo-spheres of exoplanets from high-resolution spec-troscopy

Julia Victoria Seidel1; David Ehrenreich1; VincentBourrier1; Lorenzo Pino2; Baptiste Lavie1; AurelienWyttenbach3

1 Geneva Observatory, University of Geneva (Versoix, Switzerland)2 Anton Pannekoek Institute for Astronomy, University of Amster-

dam (Amsterdam, Netherlands)3 University of Leiden (Leiden, Netherlands)

The sodium doublet in the optical is one of the mostpowerful probes of exoplanet atmospheric proper-ties, when observed in transmission spectroscopyduring transits. Recent high-spectral resolution ob-servations of the sodium doublet in hot gas giantsallowed us to resolve the line shape, opening theway for extracting thermospheric properties usingline-profile fitting. I will present the latest resultsfrom the HEARTS survey for hot exoplanetary atmo-spheres at high-spectral resolution with HARPS andHARPS-N (Seidel et al. 2019a, A&A) which founda strongly broadened sodium doublet in the ultra-hot Jupiter WASP-76b. I interpret the findings viaa retrieval method exploiting the resolution of theline profile to determine the temperature-pressureprofile and the velocity of high-altitude winds inthe thermosphere of WASP-76b and HD189733b.The method could be applied to the whole sampleof planets from the on-going HEARTS & SPADESsurvey for hot exoplanetary atmospheres at high-spectral resolution with HARPS and HARPS-N, andto the observations with next generation spectro-graphs like ESPRESSO. With the thus determinedtemperature-pressure profile and velocity of high-altitude winds I will create a clearer picture of theimpact of winds and stellar irradiation on planetaryupper atmospheres (Seidel et al. 2019b, in prep.).

326.20 — Probing Transiting Exoplanet Atmo-spheres With Multi-Object Spectrophotometry

Kamen O. Todorov1; Jean-Michel Desert2; CatherineM. Huitson3; Jacob L. Bean4; Vatsal Panwar1; Filipe deMatos1; Kevin Stevenson5; Jonathan Fortney6

1 Anton Pannekoek Institute for Astronomy, University of Amster-dam (Amsterdam, Netherlands)

2 Anton Pannekoek Institute for Astronomy (API), University ofAmsterdam (UvA) (Amsterdam, Netherlands, Netherlands)

3 University of Colorado, Boulder (Boulder, Colorado, UnitedStates)

4 University of Chicago (Chicago, Illinois, United States)5 STScI (Baltimore, Maryland, United States)6 UC Santa Cruz (Santa Cruz, California, United States)

Observations of exoplanets during transit andeclipse have revealed the atmospheric compositionsand thermal properties of some of these objects.In this context, ground-based facilities provide theopportunity to study close in transiting exoplanetsthrough spectral characterization studies. We showvisible-light and near-infrared spectra of hot Jupitersobtained with ground-based multi-object spectro-graphs (MOS). We compare the spectra to atmo-spheric emission and transmission models and inter-pret the results in terms of atmospheric properties,constraining the physics of our targets in the broadercontext of comparative exoplanetology. We presentour findings in the context of the upcoming ELT andJWST observatories.

326.21 — Metals in the emission spectrum of Kelt-9b

Lorenzo Pino1; Jean-Michel Desert1; Luca Malavolta2;Francesco Borsa3

1 Anton Pannekoek Institute for Astronomy (API), University ofAmsterdam (UvA) (Amsterdam, Netherlands, Netherlands)

2 INAF – Astrophysical Observatory of Catania (Catania, Italy)3 INAF — Astrophysical Observatory of Brera (Merate, Italy)

Ultra-hot Jupiters are planets hotter than 2,200 K,that thus offer the opportunity to study the physicsand chemistry of planetary atmospheres in a regimethat is not accessible in the solar system or in otherexoplanets. At such temperatures, molecules beginto dissociate and atoms become ionized, which givesrise to opacity sources such as H- or metal lines.Indeed, recent transmission spectroscopy observa-tions at high resolution (R>100,000) revealed for thefirst time lines from metals in the atmosphere ofthe hottest planet of this category: KELT-9b (4,000K). These observations probe the terminator of theplanet, at the transition between its permanent day-side and its permanent nightside. While they offerthe first glimpse in the refractory component of thecomposition of a planet, these observations are diffi-cult to interprete due to the uncertainty in the ther-mal conditions at the day-night transition region. Inthis talk, we present a set of complementary obser-vations of Kelt-9b that target the light directly emit-ted from its dayside. We report a detection of metallines in emission, and discuss the techniques usedand our interpretation of the observations. In com-bination with the existing transmission spectroscopydata, our observations provide the chance to map the3D structure of the planet.

108

326.22 — ACCESS: the Arizona-CfA-Catolica-Carnegie Exoplanet Spectroscopy Survey

Chima McGruder1; Mercedes Lopez-Morales2; DánielApai4; Andrés Jordán6; David Osip5; Alex Bixel4; Nés-tor Espinoza3; Jonathan Fortney7; James Kirk2; NikoleLewis8; Benjamin Rackham4; Florian Rodler9; IanWeaver1

1 Astronomy, Harvard University (Cambridge, Massachusetts,United States)

2 Harvard, Center for Astrophysics (Cambridge, Massachusetts,United States)

3 Max-Panck-Institut für Astronomie (Heidelberg, Baden-Württemberg, Germany)

4 Astronomy, University of Arizona, Steward Observatory (Tucson,Arizona, United States)

5 Carnegie Institution for Science (Washington, District ofColumbia, United States)

6 Pontificia Universidad Catolica de Chile (Santiago, Chile)7 Astronomy & Astrophysics, University of California, Santa Cruz

(Santa Cruz, California, United States)8 Astronomy, Cornell University (Ithaca, New York, United States)9 European Southern Observatory: Santiago de Chile (Santiago,

Chile)

Transmission spectroscopy provides a powerful toolto study the atmospheric properties of exoplanets.Optical transmission spectra are particularly impor-tant, since they provide the spectral baselines forclear, cloudy, and hazy atmospheres needed to cor-rectly interpret infrared transmission spectra ob-served with HST and the upcoming JWST. For 6years the Arizona-CfA-Católica-Carnegie ExoplanetSpectroscopy Survey (ACCESS) has been observ-ing and analyzing the atmospheres of over a dozenplanets, ranging from hot Jupiters to super-Earths.These observations have been collected with IMACSmounted on the 6.5-m Magellan telescope with a ho-mogeneous setup designed for measurements from0.4-0.9 microns. The homogeneity of ACCESS’s datais extremely important because it allows for mini-mization of systematic differences between instru-ments and better optimization of data analysis tech-niques. This dataset has allowed us to study 1)how stellar heterogeneities can masquerade as at-mospheric features and some methods to correct forsuch effects, 2) optimal detrending algorithms thatproduce the highest accuracy and precision, 3) ob-serving techniques with ACCESS’s telescopes thatproduce the greatest scientific yield, and more. Wegive an overview of the status of the survey and sum-marize the main findings from our published stud-ies. We also describe the expansion of ACCESS to thenorthern hemisphere with the new BINOSPEC spec-trograph on the 6.5-m MMT and our plans to start

characterizing the atmospheres of transiting planetsdiscovered by TESS.

326.23 — Comparing low- and high-resolutiontransmission spectra of hot Jupiters: What do wereally know?

Neale Gibson1; Stevanus Kristianto Nugroho2; StephanieMerrit2; Jamie Wilson2

1 School of Physics, Trinity College Dublin (Dublin, Ireland)2 School of Mathematics and Physics, Queen’s University Belfast

(Belfast, Northern Ireland, United Kingdom)

The analysis and interpretation of exoplanet spec-tra from time-series observations remains a signifi-cant challenge to our current understanding of exo-planet atmospheres, due to the complexities in un-derstanding instrumental systematics. Observationswith ground- and space-based telescopes often re-sult in conflicting data. The relatively recent devel-opment of high-resolution time-series spectroscopyoffers a new avenue to confirm or refute atmosphericsignals detected at low-resolution. Here we presentlow- and high-resolution transit observations of thehot Jupiters WASP-31b and WASP-121b with UVESand FORS2 on the VLT and GMOS on Gemini-South.Both planets have been extensively observed fromground and space, and have large atomic and molec-ular signals detected at optical wavelengths, mak-ing them excellent targets for high-resolution fol-lowup to enable comparison of signals across instru-ments and methods, and to verify observational tech-niques. Our results demonstrate significant limita-tions to our understanding of instrumental system-atics even with our most stable space-based instru-mentation, but also the power of combining low- andhigh-resolution techniques to gain a complete pic-ture of exoplanet atmospheres.

326.25 — Atmospheric Characterization of Ex-tremely Inflated Exoplanets: The Curious Case ofKELT-11b

Knicole Colon1; Laura Kreidberg2; Michael Line8;Nikku Madhusudhan9; Thomas Beatty3; PatrickTamburo12; Kevin Stevenson11; Avi Mandell1; LuisWelbanks9; Joseph Rodriguez2; Thomas Barclay1,5;Daniel Angerhausen4; Jonathan Fortney6; David James7;Eric Lopez1; Keivan Stassun10

1 NASA Goddard Space Flight Center (Greenbelt, Maryland,United States)

2 Vanderbilt University (Nashville, Tennessee, United States)3 Space Telescope Science Institute (Baltimore, Maryland, United

States)4 Boston University (Boston, Massachusetts, United States)

109

5 Harvard University (Cambridge, Massachusetts, United States)6 University of Arizona (Tucson, Arizona, United States)7 University of Bern (Bern, Switzerland)8 University of Maryland Baltimore County (Baltimore, Maryland,

United States)9 University of California Santa Cruz (Santa Cruz, California,

United States)10 Smithsonian Institution Astrophysical Observatory (Boston,

Massachusetts, United States)11 Arizona State University (Phoenix, Arizona, United States)12 University of Cambridge (Cambridge, United Kingdom)

In recent years, a population of extremely inflatedsub-Saturn-mass exoplanets orbiting bright stars hasemerged from the thousands of exoplanets known.These exoplanets are phenomenal targets for atmo-spheric characterization due to their bright host starsand large atmospheric scale heights. Subsequently,they are being studied intensively to measure theabundances of water and other metals in their at-mospheres in order to place constraints on the lo-cation of their origin within a disk. Such inflatedsub-Saturn-mass planets bridge the divide betweensuper-Earths and Jupiters, so constraints on their at-mospheric composition and metallicities are particu-larly important for informing planet formation mod-els.

We present new results for the atmospheric char-acterization of the hottest planet in this population:KELT-11b. We have an ultra-precise near-infraredtransmission spectrum collected with the HubbleSpace Telescope as well as infrared transit and eclipsemeasurements from the Spitzer Space Telescope.The Transiting Exoplanet Survey Satellite also re-cently provided an extremely precise measurementof the optical transit. We combined these data toprobe the atmospheric constituents and metallicityof the hot, inflated planet KELT-11b for the first time.The data surprisingly reveal that KELT-11b likely hasan unusually metal-poor atmosphere, yet it orbitsa metal-rich star. This combined with the fact thatKELT-11b’s host star is also slightly evolved enablesus to perform a unique comparison between KELT-11b and the other planets in the inflated sub-Saturn-mass population that all orbit main-sequence stars.We present what the various observations of KELT-11b and the other planets in this population have re-vealed about their origins so far.

326.26 — Search for clouds on the planetary day-sides by ground-based wavelength-resolved albe-dos

Matthias Mallonn1

1 Leibniz Institute for Astrophysics Potsdam (AIP) (Potsdam, Ger-many)

One decade of transmission spectroscopy of extraso-lar planets has shown the general presence of cloudsat the exoplanet terminator region. At the day sideof the same planets, however, we are still ignorantof the presence of clouds. The planets and moonsin the solar system, which are massive enough tohold a significant atmosphere, all show a substan-tial reflectivity, i.e. a high geometric albedo, whichis caused by clouds high in their atmospheres. Alsofor extrasolar planets, clouds of condensates are pre-dicted to cause geometric albedos of 0.4 at short op-tical wavelengths. The measurement of the reflectiv-ity of an exoplanet, i.e. its geometric albedo, has sofar been the realm of space-based telescopes. Theirdata, however, rarely delivered wavelength-resolvedalbedo information, although theory showed it to bea very useful tool to investigate the chemical compo-sition of the atmospheres on the day side of the exo-planet. In this presentation, I will describe my obser-vational effort to measure the wavelength-resolvedalbedo for a sample of selected exoplanets. Oneapproach includes the use of the well-proven spec-trophotometric capabilities of low-resolution spec-trographs at 8m-class telescopes. The other methodemployed to obtain multi-color albedos is the co-addition of broadband secondary eclipse light curvesof small telescopes taken in different filters. I willpresent first results from 400 to 900 nm showingdark worlds without significant cloud reflection, andI will outline the near future for the measurements ofground-based exoplanet albedo spectra.

326.28 — The role of mid-sized telescopes for thecharacterisation of exoplanets

Petr Kabath1; Jiri Zak2,1; Henri Boffin3; Valentin D.Ivanov3; Marek Skarka2,1

1 Astronomicai Institute of Czech Academy of Sciences (Ondrejov,Czechia)

2 Masaryk University (Brno, Czechia)3 ESO Garching (Munich, Germany)

The TESS space mission and, later, PLATO, will findseveral thousand of exoplanetary candidates aroundbright host stars, which will be suitable targets forfollow-up with 2m-class telescopes. We will be ableto characterise and study atmospheres of a few hun-dred gas giants via transmission spectroscopy. Herewe discuss the synergy of mid-sized telescopes withTESS and PLATO space missions. Also, we presentsome new exoplanetary results from the Perek 2-m

110

telescope located at the Ondrejov observatory andoperated by the Czech Academy of Sciences.

326.29 — The convective blueshift center-to-limband line depth variations using Solar eclipse obser-vation

Mahmoudreza Oshagh11 Institute for Astrophysics, Georg-August-University of Göttingen

(Göttingen, Germany)

High spectral resolution transmission spectroscopyhas been proven to be the most powerful techniqueto study transiting planet atmospheres. The trans-mission spectra retrieval relys on an accurate knowl-edge of the host star spectra and its variation dur-ing the planetary transit, which are mostly based onthe theoretical models of stellar atmosphere. There-fore, high resolution spectroscopic observations ofthe disk integrated Sun (Sun-as-a-star) during aneclipse can help us to obtain better understanding ofthose variation and also define observational strate-gies to mitigate its impact in transmission spectraretrieval. We will present a new analysis of spec-troscopic observation which were taken during par-tial solar eclipse with our Fourier Transform Spec-trograph at the Institute for Astrophysics Göttingenwith using I2 as its wavelength reference. We useda two dimensional cross-correlation function (CCF)using I2 and G2 star mask for estimating RV of Sunduring the eclipse. In order to re-investigate the con-vective blueshift and its dependence on line strengthand also limb angles, we also estimated the RV asa function of line depths during the solar eclipse.Our result show a steady decrease in the convectiveblueshift from shallow to deep lines and also fromcenter to limb of the Sun. Our results can be imple-mented for more realistic and robust stellar contam-ination correction for studying transmission spectraof planets transiting a solar-like stars with ongoingand upcoming missions such as ESPRESSO, ARIELand JWST.

326.30 — New Empirical Trends From Two LargeSpitzer Phase Curve Surveys

Kevin Stevenson1; Jacob Bean2; Jonathan Fraine9; Dy-lan Keating3; Megan Mansfield2; Nick B. Cowan3; LisaDang3; Drake Deming6; Jean-Michel Desert4; JonathanFortney10; Tiffany Kataria5; Eliza Kempton6; LauraKreidberg7; Nikole Lewis11; Michael Line12; CarolineMorley13; Vivien Parmentier14; Emily Rauscher8; AdamShowman15; Taylor Bell3

1 STScI (Baltimore, Maryland, United States)2 UC Santa Cruz (Santa Cruz, California, United States)

3 Cornell University (Cornell, New York, United States)4 Arizona State University (Tempe, Arizona, United States)5 UT Austin (Austin, Texas, United States)6 University of Oxford (Oxford, United Kingdom)7 University of Arizona (Tucson, Arizona, United States)8 University of Chicago (Chicago, Illinois, United States)9 Physics, McGill University (Montréal, Quebec, Canada)10 Anton Pannekoek Institute for Astronomy (API), University of

Amsterdam (UvA) (Amsterdam, Netherlands, Netherlands)11 JPL/Caltech (Pasadena, California, United States)12 University of Maryland (College Park, Maryland, United States)13 Harvard University (Cambridge, Massachusetts, United States)14 Astronomy, University of Michigan (Ann Arbor, Michigan,

United States)15 Space Science Institute (Boulder, Colorado, United States)

A long-standing mystery about close-in exoplanetsis what parameters control the transport of energyin their atmospheres. Thermal phase curve mea-surements hold the key to resolving this mystery,but previous work on this topic has been limitedby the available sample size. We have recently ob-tained phase curves of 15 previously-unobserved hotJupiters with a range of physical parameters as partof observing programs totaling ∼1,300 hours withthe Spitzer Space Telescope. This sample doubles thenumber of planets with phase curve data, thus en-abling unprecedented statistical analyses and com-parative studies. We present results from this survey,including a definitive test of the fundamental predic-tion that the irradiation level is the primary factorcontrolling heat transport for close-in planets. Fur-thermore, we identify empirical trends in the datathat help to inform us on the roles that planet grav-ity and rotation rate play in atmospheric dynamicsand cloud formation. Finally, we highlight resultsfrom compelling individual systems, including theinference of nightside clouds in the WASP-43b-twinQatar-2b and the detection of molecular dissocia-tion/recombination energy transport in the extremeplanet KELT-9b.

326.31 — A Complete Optical to Infrared Transmis-sion Spectrum of HAT-P-32Ab: Interpreting Atmo-spheric Properties in the Presence of Clouds

Munazza K. Alam1; Mercedes Lopez-Morales21 Dept of Astronomy, Harvard University (Cambridge, Mas-

sachusetts, United States)2 Center for Astrophysics | Harvard & Smithsonian (Cambridge,

Massachusetts, United States)

We present a 0.29-1.7 micron transmission spectrumof the hot Jupiter HAT-P-32Ab observed with theSpace Telescope Imaging Spectrograph (STIS) and

111

Wide Field Camera 3 (WFC3) instruments mountedon the Hubble Space Telescope. The spectrum iscomposed of 49 spectrophotometric bins measuredto a median precision of 48 ppm, and is character-ized by a steep slope in the optical and a weak wa-ter feature at 1.4 microns. Comparing the observedtransmission spectrum to a grid of 1D radiative-convective forward models indicates the presenceof clouds/hazes, consistent with previous ground-based observations. To provide more robust con-straints on the planet’s atmospheric properties, weperform the first full optical to infrared retrieval anal-ysis for this planet and verify our results using twoindependent retrieval frameworks. This result is partof the HST Panchromatic Comparative ExoplanetaryTreasury (PanCET) Program (GO 14767).

326.32 — Testing Assumptions in Atmospheric Re-trievals of Spectroscopic Phase Curves

Ying Feng1; Jonathan Fortney3; Michael Line2; LauraKreidberg4; Kevin Stevenson5

1 Astronomy & Astrophysics, UC Santa Cruz (Santa Cruz, Califor-nia, United States)

2 Arizona State University (Tempe, Arizona, United States)3 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)4 Harvard University (Cambridge, Massachusetts, United States)5 STScI (Baltimore, Maryland, United States)

From Jupiter’s banded structure and iconic GreatRed Spot, to Earth’s constantly changing cloud pat-terns, planetary atmospheres are complex 3D struc-tures. Can a 1D atmospheric model capture this com-plexity? With the upcoming JWST, we will obtaindata from exoplanet atmospheres of higher precisionthan ever before. Given this opportunity, we mustunderstand the accuracy of our data interpretation.Our model-driven interpretation of an atmosphere’sspectrum fuels our understanding of the day-nightheat redistribution and establishes chemical abun-dance estimates, which impact theories on planetformation.

Without the ability to spatially resolve exoplanetatmospheres, we often approximate them as 1D anduniform to model hemispherically-averaged spectra.Retrieval models have traditionally followed this as-sumption. Could our inferences be biased as a re-sult? In 2016, I pioneered an investigation for hotJupiters, which have large day-night temperaturecontrasts due to close-in, tidally-locked orbits. Ishowed that one can fundamentally mischaracterizeplanetary abundances when relying on a 1D modelfor atmospheres made up of half-hot and half-coolregions.

In my poster, I will discuss my latest work inpreparation for the era of high-precision spectra.How does the 1D assumption bias our estimates ofplanetary composition as a function of orbital phase?HST has obtained “spectroscopic phase curves” for asmall number of planets, but JWST will study moretargets with higher quality observations. What reli-able inferences can be made across the orbit? I con-sider synthetic data taken both with HST and fu-ture data expected from JWST. I also demonstratethe effect on actual Hubble phase curve data of a hotJupiter WASP-43b. With JWST-quality data, our sim-plifying 1D assumptions always lead to incorrect es-timates of all molecules, compared to a more realistic2D temperature structure. The need for more accu-rate models is clear. The framework I will present isalso ideal for incorporating patchy clouds, hot spots,and chemical gradients. This invites input from thecommunity as we develop the appropriate modelgiven a data set.

326.33 — Unveiling hazes and energy balance in theexoplanets atmospheres with STIS

RAISSA ESTRELA1,2; Mark Swain1; Gael Roudier11 Jet Propulsion Laboratory, California Institute of Technology

(Pasadena, California, United States)2 Center for Radioastronomy and Astrophysics Mackenzie (Sao

Paulo, Brazil)

The discovery of thousands of transiting exoplan-ets in the last decade and the upcoming planets tobe discovered with TESS makes the characterizationof exoplanets atmospheres one of the most promis-ing fields. By analyzing the atmosphere of the plan-ets, we can infer their composition and obtain in-formation about their formation and evolution. Inparticular, by observing in the visible wavelengthsone can probe the presence and size of atmosphericaerosols (haze and clouds), and therefore differenti-ate between haze (a sloped spectrum) and clouds (aflat spectrum). In this work, we analyse the visiblespectra of various targets observed by the STIS/HSTinstrument (0.3—1.0 μm) using our Bayesian data-reduction pipeline to obtain the transmission spec-tra. Our analysis also incorporates spectral retrievalsof all targets providing the best set of atmosphericparameters using a MCMC sampling.

326.34 — Gemini/GMOS Transmission Spectro-scopic Survey of Gas Giant Exoplanets

Vatsal Panwar1; Jean-Michel Desert2; Kamen O.Todorov1; Catherine M. Huitson3; Jacob Bean4; JonathanFortney5; Marcel Bergmann6

112

1 Anton Pannekoek Institute for Astronomy, University of Amster-dam (Amsterdam, Noord Holland, Netherlands)

2 Anton Pannekoek Institute for Astronomy (API), University ofAmsterdam (UvA) (Amsterdam, Netherlands, Netherlands)

3 University of Colorado (Boulder, Colorado, United States)4 University of Chicago (Chicago, Illinois, United States)5 University of California, Santa Cruz (Santa Cruz, California,

United States)6 NOAO (Tucson, Arizona, United States)

Estimating the nature and abundances of chemi-cal species and clouds in exoplanetary atmospheresforms the backbone of comparative exoplanetology.We present a long-term ground-based survey of adozen transiting hot Jupiters observed in the visi-ble bandpass using the Gemini Multi-Object Spectro-graph (GMOS). By observing transits of an ensem-ble of hot Jupiters spanning a range of masses, radii,and host star types, and using a consistent methodol-ogy for extracting their optical transmission spectraacross the sample, we derive common properties fortheir atmospheres. We present the results of this sur-vey, the challenges faced by such an experiment, andthe main lessons learned for future MOS observa-tions and instrument designs. Ultimately, our surveyimproves our understanding of the diversity of phys-ical processes at play in exoplanetary atmospheres.We also introduce novel techniques to analyze thedata that lead to improved precision and expand thecapabilities of the MOS technique towards the obser-vations of exoplanets transiting bright stars, which isrelevant for compelling targets discovered by TESS.

326.35 — Atmospheric Rossiter-McLaughlin effectof KELT-9b

Monica Rainer1; Francesco Borsa2; Aldo Bonomo3;Domenico Barbato4,3; Luca Fossati5; Luca Malavolta6;Valerio Nascimbeni7; Antonino Lanza6; MassimilianoEsposito8; Laura Affer9; Gloria Andreuzzi10,11; SerenaBenatti9; Katia Biazzo6; Andrea Bignamini12; MatteoBrogi13,14; Ilaria Carleo15; Riccardo Claudi7; RosarioCosentino10; Elvira Covino16; Mario Damasso3; SilvanoDesidera7; Antonio Garrido Rubio9,17; Paolo Giacobbe3;Esther González-Álvarez18; Avet Harutyunyan10;Cristina Knapic12; Giuseppe Leto6; Roxanne Ligi2; An-tonio Maggio9; Jesus Maldonado9; Luigi Mancini19,20;Aldo Fiorenzano10; Sabrina Masiero9,21; GiuseppinaMicela9; Emilio Molinari22; Isabella Pagano6; MarcoPedani10; Giampaolo Piotto23; Lorenzo Pino24; EnnioPoretti2,10; Gaetano Scandariato6; Riccardo Smareglia12;Alessandro Sozzetti3

1 OAA, INAF (Florence, Florence, Italy)2 TNG, FGG (Brena Baja, Spain)3 OAR, INAF (Rome, Italy)

4 OATs, INAF (Trieste, Italy)5 Department of Physics, University of Warwick (Coventry, United

Kingdom)6 Centre for Exoplanets and Habitability, University of Warwick

(Coventry, United Kingdom)7 Astronomy, Wesleyan University (Middletown, Connecticut,

United States)8 OACn, INAF (Naples, Italy)9 Department of Physics and Chemistry Emilio Segré, University of

Palermo (Palermo, Italy)10 CSIC-INTA (Madrid, Spain)11 Department of Physics, University of Rome Tor Vergata (Rome,

Italy)12 OAB, INAF (Milan, Italy)13 Max Planck Institute for Astronomy (Heidelberg, Germany)14 Fondazione Gal Hassin (Isnello, Italy)15 OAC, INAF (Cagliari, Italy)16 Department of Physics and Astronomy Galileo Galilei, Univer-

sity of Padua (Padua, Italy)17 Anton Pannekoek Insitute for Astronomy, Universiteit van Ams-

terdam (Amsterdam, Netherlands)18 OATo, INAF (Turin, Italy)19 Department of Physics, University of Turin (Turin, Italy)20 Space Research Institute, Austrian Academy of Sciences (Graz,

Austria)21 OACt, INAF (Catania, Italy)22 OAPd, INAF (Padua, Italy)23 Thüringer Landessternwarte Tautenburg (Tautenburg, Germany)24 OAPa, INAF (Palermo, Italy)

We observed the planet-hosting star KELT-9 (A-typestar, vsini ∼110 km/s) in the framework of theGAPS project (Global Architecture of Planetary Sys-tems). GAPS is a long-term italian program using theHARPS-N and GIANO-B spectrographs at the TNGtelescope in order to detect and characterize plan-etary systems, with a strong focus on the study ofthe exoplanets’ atmospheres. We extracted from thehigh-resolution optical HARPS-N spectra of KELT-9the mean stellar line profiles with a custom analysisbased on the Least Square Deconvolution technique.Then, we computed the stellar radial velocities witha method developed for fast rotators, by fitting themean stellar line profile with a purely rotational pro-file instead of using a Gaussian function. The newspectra and analysis led us to update the orbital andphysical parameters of the system. The former de-pends on the improvement on the Kstar value, thatwe measured with an 8-σ significance, a remarkableresult for such a fast rotating star. We discovered ananomalous in-transit radial velocity deviation fromthe theoretical Rossiter-McLaughlin effect solution,previously calculated from the known spin-orbit an-gle obtained from the line profile tomography. We

113

prove that this deviation is caused by the planetaryatmosphere of KELT-9b, thus we name this effect At-mospheric Rossiter-McLaughlin effect. By analysingthe magnitude of the radial velocity anomaly, weobtained information on the extension of the plane-tary atmosphere as weighted by the model used toretrieve the stellar mean line profiles. The Atmo-spheric Rossiter-McLaughlin effect will be observ-able for other exoplanets whose atmosphere has non-negligible correlation with the stellar mask used toretrieve the radial velocities, in particular ultra-hotJupiters with iron in their atmospheres. The dura-tion and amplitude of the effect will depend not onlyon the extension of the atmosphere, but also on thein-transit planetary radial velocities and on the pro-jected rotational velocity of the parent star, and thiscould led to strange in-transit RV variations.

326.36 — Characterizing Exoplanet Atmospheresusing the all new SPIRou High-Resolution Spec-trograph

Stefan Pelletier1; Björn Benneke11 Université de Montréal (Montreal, Quebec, Canada)

Over the last decade, high-resolution near-infraredspectroscopy has led to a multitude of unambiguousmolecular detections in transiting and non-transitingexoplanet atmospheres by identifying the distinctsignature of their unique spectral features. Oneshortcoming of many of the previous observationsis, however, that they only cover a very narrowwavelength range thus greatly hindering the infer-ence of precise molecular abundance ratios. Thismajor limitation can be overcome using the brand-new ultra-stable SPIRou spectrograph recently in-stalled at CFHT. Its unique combination of wave-length stability, high spectral resolution (R∼75000),and wide wavelength coverage (0.98-2.5um) enablesus to detect multiple molecules simultaneously andconstrain their relative abundances. Here we presenthow SPIRou observations of hot Jupiters can be com-bined with state-of-the-art modelling frameworks toconstrain molecular abundances in exoplanet atmo-spheres and provide robust carbon to oxygen ratiomeasurements that would greatly help shed lightonto long-withstanding questions about planetaryformation.

326.37 — The strange chemistry of the Dr. Daysideand Mr. Nightside

Giuseppe Morello1; Pascal Tremblin2; CamillaDanielski1; Marine Martin-Lagarde3; René Gastaud1;Daniel Dicken1; Pierre-Olivier Lagage1

1 AIM, CEA-Saclay (Paris, France)2 Maison de la Simulation, CEA (Gif-Sur-Yvette, France)3 Department of Astrophysics, CEA-Saclay (Gif-sur-Yvette, France)

We discuss phase-curve spectroscopy of threehot Jupiters with ultra-short periods (P<1 day):WASP43b, WASP18b and WASP103b. To date,these are the only exoplanets with spectroscopicphase-curve observations. Multiple phase-curveshave been observed with HST/WFC3 at 1.1-1.7 μmand with Spitzer/IRAC at 3.6 and 4.5 μm. Becauseof their short orbital periods, these exoplanets aretidally-locked, therefore exhibiting a hotter daysideand a cooler nightside. Phase-curve spectroscopyenables unique access to horizontal distribution anddynamics of the exoplanet atmospheres. However,state-of-the-art atmospheric modelling tools fail toprovide a consistent model for the observations atall wavelengths, ofter invoking peculiar chemistryto explain the excess of infrared emission fromtheir daysides. We present here revised data anal-yses and comparisons with a grid of atmosphericmodels for the aforementioned exoplanets. Weselect those models from the grid that give the mostconsistent results at all wavelengths and discusstheir likelihoods. Finally, we comment about thepossible causes for the remaining (but reduced)discrepancies, and whether such causes could beastrophysical or instrumental in nature.

326.38 — New models and thermal emission spec-tra for hot Jupiters, and a look at population-leveltrends

Megan Mansfield1; Jacob Bean2; Michael Line3;Jonathan Fortney4; Vivien Parmentier5; JacobArcangeli6; Jean-Michel Desert7; Eliza Kempton8; BrianKilpatrick9; Laura Kreidberg10; Matej Malik11; KevinStevenson12

1 Department of Geophysical Sciences, University of Chicago(Chicago, Illinois, United States)

2 Harvard University (Cambridge, Massachusetts, United States)3 Astronomy, University of Maryland (College Park, Maryland,

United States)4 STScI (Baltimore, Maryland, United States)5 University of Chicago (Chicago, Illinois, United States)6 Arizona State University (Tempe, Arizona, United States)7 University of California - Santa Cruz (Santa Cruz, California,

United States)8 University of Oxford (Oxford, United Kingdom)9 Universiteit van Amsterdam (Amsterdam, Netherlands)10 Anton Pannekoek Institute for Astronomy (API), University of

Amsterdam (UvA) (Amsterdam, Netherlands, Netherlands)11 University of Maryland (College Park, Maryland, United States)12 Brown University (Providence, Rhode Island, United States)

114

Hot Jupiters are compelling targets for thermal emis-sion observations because their high signal-to-noiseallows precise atmospheric characterization. The-ory predicts a continuum of thermal structuresand resulting emission spectra for these objects:Planets below ∼1800 K are expected to have un-inverted atmospheres and display absorption fea-tures in their emergent spectra, while those warmerthan ∼2100 K should have inverted atmospheresand emission features. Ultra-hot planets with tem-peratures above ∼2500 K are predicted to displayblackbody-like spectra because of a combination ofhigh-temperature effects including water dissocia-tion and H− opacity. We present new high signal-to-noise spectra for three hot Jupiters observed withthe Hubble Space Telescope Wide Field Camera 3(HST/WFC3) from 1.1 to 1.8 microns. We alsopresent an update to the Fortney et al. (2008)grid of self-consistent 1D hot Jupiter model thermalstructures and emission spectra for comparison tothese data. This new grid extends to higher tem-peratures and includes all the physics and chem-istry for ultra-hot planets that has been discussedrecently in the literature. We combine our newlymeasured hot Jupiter emission spectra with 9 pre-viously published HST/WFC3 observations to per-form a population study of the 12 planets with thehighest signal-to-noise WFC3 spectra. We computethe degree of absorption or emission observed intheir emergent spectra by quantifying their devia-tion from a blackbody in a new color-magnitude-type plot. While cooler planets do show absorp-tion features as expected, the ensemble of planets attemperatures above ∼2100 K do not show the emis-sion features predicted by our models. When ana-lyzed together as a statistical sample, these planetsinstead show spectra consistent with blackbodies totemperatures much lower than expected from cur-rent theory. This suggests that existing models maystill be missing important physical effects that couldact to mute emission features at lower temperatures.We present some possible explanations for the dis-crepancy and observations that could be done to testthese hypotheses.

326.39 — A comparison of retrieved water abun-dances using the clear-atmosphere, hot-SaturnWASP-39b as a test case

James Kirk1; Mercedes Lopez-Morales1; Peter Wheatley2;Ian Weaver1; Ian Skillen3; Tom Louden2; JamesMcCormac2; Nestor Espinoza4

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 University of Warwick (Coventry, United Kingdom)3 Isaac Newton Group of Telescopes (Santa Cruz de La Palma,

Spain)4 Max-Planck-Institut fur Astronomie (Heidelberg, Germany)

WASP-39b provides a benchmark test case for com-paring different retrieval approaches and assump-tions. It has a deep and precisely measured waterfeature, however, the water abundances retrieved inthe literature differ by over four orders of magni-tude. We will present a new analysis of all availabletransmission spectra of this planet in addition to anew WHT/ACAM transmission spectrum obtainedthrough the LRG-BEASTS survey. We have run re-trievals on various combinations of literature datasets and a combined transmission spectrum cover-ing a wavelength range of 0.29 to 5.06 microns. Wewill highlight how different assumptions made dur-ing retrieval analyses can lead to order of magni-tude differences in the retrieved water abundances,which in turn lead to different conclusions about theplanet’s formation. WASP-39b is one of the very besttargets for transmission spectroscopy, with a transitdepth per atmospheric scale height of 450 ppm. Wemust understand and be aware of the impact of ourassumptions before we can begin to understand thetransmission spectra of targets with smaller signals.

326.40 — Investigating Trends in AtmosphericComposition of Gas Giant Planets Using SpitzerSecondary Eclipses

Nicole Lisa Wallack11 Division of Geological and Planetary Sciences, Caltech (Pasadena,

California, United States)

It is well-established that the magnitude of the in-cident stellar flux is the single most important fac-tor in determining the day-night temperature gra-dients and atmospheric chemistries of short-periodgas giant planets. However it is likely that otherfactors, such as planet-to-planet variations in atmo-spheric metallicity, C/O ratio, and cloud proper-ties, also contribute to the observed diversity of in-frared spectra for this population of planets. In thisstudy we present new 3.6 and 4.5 micron secondaryeclipse measurements for five cool (T < ∼1000K)transiting gas giant planets: HAT-P-15b, HAT-P-17b,HAT-P-18b, HAT-P-26b, and WASP-69b, as well asfive hotter transiting gas giant planets: HAT-P-5b,HAT-P-38b, WASP-7b, WASP-72b, and WASP-127b.For the cooler planets, we expect that the ratio ofmethane to CO and CO2 should vary as a func-tion of atmospheric metallicity and C/O ratio. Weuse our measured eclipse depths for these planets

115

to search for trends in CH4/(CO+CO2) ratio as afunction of planet mass, temperature, bulk metallic-ity, and host star metallicity. We see no evidencefor a solar system-like correlation between planetmass and atmospheric metallicity, but instead iden-tify a potential (1.9 σ) correlation between the in-ferred CH4/(CO+CO2) ratio and stellar metallicity.Building on the work of Garhart et al. (2019), weplace these new planets into a broader context bycomparing them with the sample of all planets withmeasured Spitzer secondary eclipses (over 80 plan-ets). This allows us to search for empirical trends inthe spectral slopes of these planets that cannot be ex-plained purely by changes in incident flux.

326.41 — Detecting analogues of the L-T transitionin eccentric warm Jupiters.

Jason Dittmann1; Jacqueline Faherty2; Daniella BardalezGagliuffi2; Elena Manjavacas3; Johanna M. Vos2

1 Earth and Planetary Science, Massachusetts Institute of Technol-ogy (Boston, Massachusetts, United States)

2 Astrophysics, American Museum of Natural History (New YorkCity, New York, United States)

3 W. M. Keck Observatory (Kamuela, Hawaii, United States)

One of the main goals of current exoplanet researchis in studying the atmospheric composition and dy-namics of exoplanet atmospheres. Recent successhas been made in identifying the spectral featuresof carbon monoxide and titanium oxide in the atmo-spheres of highly irradiated planets as well as iden-tification of broad metallic ionization lines. How-ever, these studies have generally been performedfor tidally locked planets in circular orbits, and arelikely in a steady state. In this talk, I will discuss theoptical phase curve of KOI 614.01, a warm-Jupiter inan eccentric orbit around a Sun-like star. KOI 614.01exhibits smooth brightness variations in its phasecurve, but resides at an orbital distance where theamplitudes of doppler beaming, tidal distortion, andreflection effects are greatly diminished. When com-bined with extant radial velocity data, these bright-ness changes are correlated with the planet’s dis-tance from the host star. The system’s total bright-ness increases by 30 ppm when the planet resides ata distance with an equilibrium temperature of 1000Kelvin, and takes approximately 12 hours to changestates. In this talk I will discuss the possible originof this effect and its similarities to the L-T transi-tion in brown dwarfs as well as future prospects forstudying the effects of time-variant irradiation on ex-oplanet atmospheres.

326.42 — Hubble spectroscopic phase curve obser-vations for the ultrahot Jupiter WASP-121b

Thomas Evans1; Tiffany Kataria2; David Sing5; NikoleLewis3; Hannah Wakeford6; Joanna Barstow4; JessicaSpake7

1 MIT (Cambridge, Massachusetts, United States)2 JPL/Caltech (Pasadena, California, United States)3 Astronomy, Cornell (Ithaca, New York, United States)4 UCL (London, United Kingdom)5 JHU (Baltimore, Maryland, United States)6 STScI (Baltimore, Maryland, United States)7 University of Exeter (Exeter, United Kingdom)

WASP-121b is an extreme system, even by exoplanetstandards. With a radius 1.8× that of Jupiter, itis one of the most inflated planets known, and or-bits a bright F6 star only ∼15% beyond its Rochelimit, where it is subjected to intense tidal forcesand the atmosphere is heated to ∼2700K. Radiationfrom the dayside hemisphere has been measured us-ing Hubble and Spitzer, revealing spectral featuresdue to H- ions, H2O, and CO. Notably, these fea-tures are observed in emission rather than absorp-tion, providing the first definitive evidence for a ther-mal inversion in an exoplanet atmosphere. This day-side thermal inversion is likely caused by strong ab-sorption of incident stellar radiation at optical wave-lengths, a picture supported by the Hubble trans-mission spectrum, which probes the atmosphericopacity at the day-night boundary. It shows evi-dence for significant absorption across the optical,possibly due to VO gas, as well as a dramatic near-UV absorption signal, not seen before for any otherplanet. We will present new observations of WASP-121b, in the form of a full-orbit spectroscopic phasecurve acquired with Hubble at near-infrared wave-lengths. Phase curves provide information about thecirculation and 3D properties of the atmosphere, un-obtainable through primary transits and secondaryeclipses alone. In particular, phase curves allow theday-night brightness contrast to be determined andcan reveal evidence for advective heat transport inthe form of the dayside hot spot being offset fromthe substellar point, both constraining atmosphericdynamics. The new phase curve data will also trackhow the H2O band within the Hubble bandpasstransitions from being in emission on the daysideto absorption on the cooler nightside, producing alongitudinally-resolved map of WASP-121b’s verti-cal thermal structure. Through detailed compari-son with 3D circulation models, these measurementswill provide valuable new insights into the efficiencywith which optical absorbers such as TiO and VO areremoved from the upper atmosphere via condensa-

116

tion cold-trapping.

326.43 — The atmospheric dynamics of WASP-49b

Tom Louden1; Peter Wheatley11 University of Warwick (Kenilworth, United Kingdom)

I will present an analysis of the spectrum of WASP-49b using Terminator, a code developed to spatiallyresolve the atmospheres of exoplanets. Simulationsof hot Jupiters with GCM’s predict the presence of astrong equatorial jet, as well as a day-to-night flow.Without spatial resolution, it is only possible to mea-sure an average velocity of the planet, and hence it isnot possible to disentangle these contributions. Tran-sit Limb Scanning makes it possible to spatially re-solve the atmosphere of an exoplanet during transit.This technique was first used in Louden & Wheatley2015 to spatially resolve the atmosphere of the hotJupiter HD 189733b. I build upon this technique andshow that it is also possible to separate out contribu-tions from polar and equatorial regions of the planet.I will present an analysis of WASP-49b showing 3 dis-tinct velocity regions in the atmosphere; eastern andwestern equatorial regions with velocities of -3 +- 1and 1+-1 km/s, and a polar region with an averagevelocity of -1 +- 1 km/s. This is only the second timethat an equatorial jet has been directly measured onexoplanet, and is the first evidence of a distinct polarregion.

326.44 — Consistent Cloud Models in Retrieval

Jasmina Blecic11 Physics Department, New York University Abu Dhabi (Abu

Dhabi, United Arab Emirates)

To date, there have been two theoretical approachesused to model clouds and understand cloud forma-tion: (1) the phase-non-equilibrium concept of mi-crophysical dust formation, Helling et al., and (2)the phase-equilibrium concept of thermal stability,Marley et al. While the first approach describesthe end state of the cloud formation, assuming thatthe gas is well mixed, transported above the cloudbase and grains fall under the influence of gravity;the second approach explores the kinetic processesof cloud formation, where the dust particles form,fall down and grow, until they evaporate below thecloud base. Understanding the nature of clouds inexoplanetary atmospheres is crucial for interpretingtransit observations, particularly those made withJames Web Space Telescope (JWST), as aerosols areextremely common in planetary atmospheres and

dramatically change the appearance of spectral fea-tures. Within the retrieval community, until veryrecently, only simple cloud parametrizations (grayor opaque cloud decks) were considered when an-alyzing current observations, as the data are not ofsufficient quality to respond to the complexity ofmore advanced cloud models. With the imminentapproach of the JWST era, a need for more consis-tent cloud models in retrieval have become a neces-sity. Here, we present the first implementation ofthe microphysical kinetic cloud model in 1D retrievalthat explores the aggregate cloud formation in a self-consistent way; and a parametrized thermal stabilitycloud model, based on the phase-equilibrium con-cept, that utilizes a parametrized cloud shape and ef-fective particle size, and wraps the atmospheric mix-ing and settling inside the cloud extent and numberdensity of the particles as free parameters.

326.45 — ACCESS: A Flat Visual Spectrum of theHot Jupiter WASP-43b without evidence for Na orK

Ian C. Weaver11 Astronomy, Center for Astrophysics | Harvard & Smithsonian

(Somerville, Massachusetts, United States)

We present a new ground-based visual transmissionspectrum of the hot Jupiter WASP-43b, obtained aspart of the ACCESS Survey. The spectrum was de-rived by combining four transits observed between2015 and 2018, with combined wavelength coveragebetween 5,300 Å - 9,000 Å and an average photomet-ric precision of 708 ppm in 230 Å bins. We performa full atmospheric retrieval of our transmission spec-trum combined with literature HST/WFC3 observa-tions to search for the presence of clouds/hazes aswell as Na, K, Hα, and water planetary absorptionand stellar spot contamination over a combined spec-tral range of 5,317.90 Å - 16,420 Å. We do not de-tect a statistically significant presence of Na I or KI alkali lines in the atmosphere of WASP-43b. Wedo find, however, that the transmission spectrum isbest explained, although only marginally, by includ-ing the effect of heterogeneities on the photosphereof the host star, yielding a log-evidence of 3.875 ±0.20 higher than a flat (featureless) spectrum. In par-ticular, the models favor the presence of large, low-contrast stellar heterogeneity over the four transitepochs with an average covering fraction fhet = 0.47± (0.30, 0.23), temperature contrast ΔT = 113 K ± 65K. Within the planet’s atmosphere, we recover a totallog water volume mixing ratio of -5.61 ± (2.47,-2.58)from the combined visual/NIR data.

117

326.46 — Optical Transmission Spectra for the TwoHot Jupiters WASP-79b and WASP-101b

Alexander Rathcke11 DTU Space, Technical University of Denmark (Copenhagen N,

Denmark)

Several advancements in the characterization of ex-oplanetary atmospheres, such as the improvementand refinement of analysis techniques with theimplementation of more flexible noise-modellingframeworks, has led to progress at a rapidly increas-ing pace. This progress implies that the near fu-ture will allow us to go from atmospheric charac-terization on an individual basis to a statistical sig-nificant sample of characterized atmospheres, whichwill help explain the observed planetary diversityas a result of system architectures, planetary com-positions and planetary environments. In this talk,I present optical transmission spectra of two hotJupiters, WASP-79b and WASP-101b, where in partic-ular the determining factor(s) between a cloudy anda clear atmosphere is still not understood. These ob-servations were obtained with HST/STIS as part ofthe Panchromatic Comparative Exoplanetary Trea-sury program (PanCET), which aims at providingthe community with the first large-scale simultane-ous UVOIR comparative study of atmospheres. Us-ing a Gaussian process approach to model system-atics, we produce a high precision, low resolutionspectrum of these two planets between 2900 & 10200Å. The resulting transmission spectrum of WASP-101b shows a flat, featureless spectrum, indicatinga high-altitude cloud cover muting spectral features.In contrast, the transmission spectrum for WASP-79bis varying with wavelengths, which hints at an at-mospheric opacity that is wavelength dependent. Iwill discuss possible explanations for this and relateit to what JWST might find when it probes the planetin transmission as part of the Transiting ExoplanetCommunity Early Release Science Program in cycle1.

326.47 — Storms or systematics? The search for at-mospheric variability in hot Jupiters

Matthew John Hooton1,2; Ernst J.W. De Mooij3; ChrisA. Watson2; Neale Gibson4; Francisco Galindo-Guil5;Rosa Clavero6; Stephanie Merritt2

1 Center for Space and Habitability, University of Bern (Bern,Switzerland)

2 Astrophysics Research Centre, Queen’s University Belfast (Belfast,Northern Ireland, United Kingdom)

3 School of Physical Sciences and Centre for Astrophysics and Rela-tivity, Dublin City University (Dublin, Ireland)

4 School of Physics, Trinity College Dublin (Dublin, Ireland)5 Departamento de Astrofísica, Centro de Astrobiología (INTA-

CSIC) (Madrid, Spain)6 Instituto de Astrofísica de Canarias (San Cristóbal de La Laguna,

Santa Cruz de Tenerife, Spain)

The observation of exoplanets during their primarytransits and secondary eclipses is a powerful tool tocharacterise their atmospheres. These observationsprovide important information about their temper-ature structures and circulation efficiencies, as wellas the ability to detect the presence of clouds, hazesand specific molecular features. However, repeat ob-servations of hot Jupiters routinely yield significantlydifferent results. Without understanding the sourceof each of these disagreements, which could arisedue to systematic errors or genuine atmosphericvariability in the exoplanets themselves, it is difficultto reliably constrain the atmospheric properties ofthese exoplanets. I will present a summary of thesediscrepancies—including my own secondary eclipseobservations of WASP-12b—along with a discussionof their various causes. I will also describe futureground-based and space-based observational strate-gies to discriminate between explanations involvingstorms and systematics.

326.48 — Search for alkali metals in the atmo-spheres of low-density exoplanets

Guo Chen1; Enric Palle2; Nuria Casasayas-Barris2; Fe-lipe Murgas2; Lisa Nortmann2; Hannu Parviainen2;Jorge Prieto-Arranz2; Nikku Madhusudhan3; LuisWelbanks3; Siddharth Gandhi3

1 Purple Mountain Observatory, CAS (Nanjing, China)2 Investigacion, Instituto de Astrofisica de Canarias (La Laguna,

Spain)3 Institute of Astronomy, University of Cambridge (Cambridge,

United Kingdom)

Exoplanets with relatively clear atmospheres areprime targets for detailed studies of chemical com-positions and abundances. Alkali metals havelong been suggested to exhibit broad wings dueto pressure broadening, but most of the alkali de-tections only show very narrow absorption cores,which probably hints the presence of clouds. Wehave been conducting an optical transmission spec-troscopy survey using the low-resolution OSIRISspectrograph at the 10.4 m GTC. Our survey hasrevealed very diverse alien worlds, from cloudy orhazy atmospheres to partly cloudy atmospheres, andto clear skies. We will present a comparative studyof low-density exoplanets with similar equilibrium

118

temperatures but different masses. We detected al-kali metals in these planets, which show distinct ab-sorption profiles. In particular, we detected Li inone of them, which, if confirmed, could play a crit-ical role in understanding the planet formation his-tories. When pressure-broadening is resolved, weperformed spectral retrieval to constrain the chem-ical abundances and cloud properties. The absorp-tion profiles of alkali metals, when acquired at low-and high-resolution, can probe a wide range of at-mospheric layers. This poses a bright future for us tocomprehensively understand the atmospheric struc-tures, from troposphere to thermosphere, as well asthe underlying physics, chemistry and dynamics.

327 — Planetary Atmospheres— Terrestrial Planets and Mini-Neptunes, Poster Session327.01 — Accretion of Nebula-originated Proto-atmospheres on Planets with Eccentric Orbits

Chuhong Mai1; Steven J. Desch1; Rolf Kuiper2; Gabriel-Dominique Marleau2; Cornelis Dullemond3

1 School of Earth and Space Exploration, Arizona State University(Phoenix, Arizona, United States)

2 Institut für Astronomie and Astrophysik, Universität Tübingen(Tübingen, Germany)

3 Institut für Theoretische Astrophysik (ITA), Zentrum für As-tronomie (ZAH), Universität Heidelberg (Heidelberg, Germany)

Protoplanets are believed to form before gas dissi-pates in the protoplanetary disk and are likely to cap-ture proto-atmospheres from the nebula gas. Suchhydrogen-rich atmospheres have been detected andcharacterized in exoplanetary systems (e.g. low-density super Earths and mini Neptunes). Theaccretion process and the structure of the proto-atmosphere is subject to the disk environment suchas the evaporation of nebula gas, the eccentricityof the planet’s orbit and the planet mass, etc. Inthis study, we used the hydrodynamic code PLUTOand the radiation transport module MAKEMAKE tomodel the accretion event of H2-dominated atmo-spheres. We established a 2-D radiative accretionmodel with sophisticated opacity treatment to sim-ulate low-mass protoplanets on eccentric orbits cap-turing proto-atmospheres. We revealed astonishingrecycling behaviors of gas flow around the planet,forming an asymmetric but stable atmosphere insidethe bow shock structure. We also quantitatively ex-plored how such primordial atmospheres are sen-sitive to the relative velocity of the planet and disk

gas, the planet mass, disk gas density/pressure andthe overall orbital evolution. A supersonic environ-ment turns out to be favorable for planets to keepan early stable atmosphere, rather than harmful. Ingeneral, the bigger the planet and the smaller theMach number (still supersonic), the thicker the at-mosphere can be retained. The orbital evolution ofthe planet can also insert a forced oscillation on theatmosphere properties. Our study provides impor-tant insights in understanding how planet migrationand orbital eccentricity affect the formation and evo-lution of proto-atmospheres for Earth-size planets.The problem of proto-atmosphere accretion is fun-damental to science topics including the nebula ori-gin of terrestrial water and other volatiles, the supplyof noble gases, etc. Understanding how the proto-atmospheres could influence or even create the cur-rent terrestrial planetary environments has generalsignificance to the study of exoplanet geochemistryand habitability.

327.02 — Thermo-compositional diabatic convec-tion in the atmospheres of brown dwarfs, exoplan-ets, and in Earth’s atmosphere and oceans

Pascal Tremblin11 Maison de la Simulation, CEA (Gif-Sur-Yvette, France)

By generalizing the theory of convection to any typeof thermal and compositional source terms (diabaticprocesses), we show that thermohaline convectionin Earth’s oceans, fingering convection in stellar at-mospheres, and moist convection in Earth’s atmo-sphere are derived from the same general diabaticconvective instability. We also show that “radia-tive convection” triggered by the CO/CH4 transitionwith radiative transfer in the atmospheres of browndwarfs is analogous to moist and thermohaline con-vection. We derive a generalization of the mixing-length theory to include the effect of source terms in1D codes. We show that CO/CH4 “radiative” con-vection could significantly reduce the temperaturegradient in the atmospheres of brown dwarfs simi-larly to moist convection in Earth’s atmosphere, thuspossibly explaining the reddening in brown dwarfspectra. By using idealized 2D hydrodynamic simu-lations in the Ledoux unstable regime, we show thatcompositional source terms can indeed provoke a re-duction of the temperature gradient. The L/T tran-sition could be explained by a bifurcation betweenthe adiabatic and diabatic convective transports andseen as a giant cooling crisis: an analog of the boilingcrisis in liquid/steam-water convective flows.

The study of the impact of different parameters(effective temperature, compositional changes) on

119

CO/CH4 radiative convection and the analogy withEarth moist and thermohaline convection is openingthe possibility of using brown dwarfs to better un-derstand some aspects of the physics at play in theclimate of our own planet.

This mechanism, with other chemical transitions,could be present in many giant and Earth-like exo-planets. We present first results towards the simula-tion of this process for the CO/CO2 transition in sec-ondary atmospheres of hot rocky exoplanets (youngor irradiated) which could be applied to the primitivestages of the atmospheres of Earth, Mars or Venus.

327.03 — Near-Infrared Transit Spectroscopy of 55Cancri e

Emily Deibert1; Ernst J.W. De Mooij2; Andrew Ridden-Harper3; Ray Jayawardhana4; Suresh Sivanandam1;Marie Karjalainen5; Raine Karjalainen5

1 University of Toronto (Toronto, Ontario, Canada)2 School of Physical Sciences and Centre for Astrophysics and Rela-

tivity, Dublin City University (Dublin, Ireland)3 Department of Astronomy, Cornell University (Ithaca, New York,

United States)4 Cornell University (Ithaca, New York, United States)5 Isaac Newton Group of Telescopes (La Palma, Canary Islands,

Spain)

Thousands of transiting exoplanets have been dis-covered, but the extreme brightness contrast be-tween these planets and their host stars makes char-acterizing their atmospheres particularly challeng-ing. Recent work has focused on transmission spec-troscopy during transits, when the light from thehost star passes through the planet’s atmosphere andallows for the detection of any atomic or molecu-lar species present. While this method has beenused to make atmospheric detections around severalhot Jupiters, the atmospheres of cooler, lower-massplanets remain elusive, requiring extremely high-resolution observations that are only now becom-ing available with the latest generation of spectro-graphs. In this poster, I will present our recent analy-sis of high-resolution ground-based observations ofthe hot super-Earth 55 Cancri e, using near-infrareddata from the CARMENES instrument at the CalarAlto Observatory and the newly-operational SPIRouinstrument at the Canada-France-Hawaii Telescope.Through use of the Doppler cross-correlation tech-nique pioneered by Snellen et al. (2010), which takesadvantage of the extreme changes in radial velocityduring close-in exoplanets’ transits, we investigateand place constraints on the presence of CO, CO2,H2O, and HCN in the atmosphere of 55 Cancri e.

Our high-resolution observations allow for an un-precedented look at this super-Earth’s atmosphere inthe infrared regime, not only allowing us to inves-tigate the potential HCN detection by Tsiaras et al.(2016), but also resulting in some of the most strin-gent constraints to date on the atmospheric compo-sition of this enigmatic exoplanet.

327.04 — Do Habitable Worlds Require MagneticFields?

David Brain1; Yaxue Dong1; Robin Ramstad1; HilaryEgan1; Tristan Weber1; Rebecca Jolitz1; Kanako Seki2;Janet Luhmann3; Jim McFadden3; David Mitchell3;Laila Andersson1; Jared Espley4; Jasper Halekas5; BruceJakosky1

1 University of Colorado (Boulder, Colorado, United States)2 University of Tokyo (Tokyo, Japan)3 University of California Berkeley (Berkeley, California, United

States)4 NASA GSFC (Greenbelt, Maryland, United States)5 University of Iowa (Iowa City, Iowa, United States)

In recent years the idea that a planet’s magnetic fieldshields its atmosphere from being removed to spacethrough interaction with a stellar wind has beenincreasingly questioned, based largely on observa-tions of atmospheric escape from Venus, Earth, andMars. This has motivated new thinking about therole that a magnetic field plays in determining atmo-spheric escape rates — whether it shields the atmo-sphere by deflecting incident charged particles fromthe planet’s host star, or whether it collects more en-ergy from the stellar wind to drive escape by increas-ing the cross-sectional area of the planet. Since anatmosphere is required to keep water stable in liquidform at a planet’s surface, this question is relevant toour understanding of the habitability of planets.

Various approaches are currently being used tomake progress on this question. Basic theoretical ar-guments have been advanced, but require observa-tions to confirm them. Computer simulations of at-mospheric escape are being pursued with promisingresults, but are difficult to validate. Observations of-fer opportunity for ground truth, but comparisonsbetween planets in our own solar system are compli-cated by the fact that planets differ in many ways, sothat isolating the influence of a global magnetic fieldis non-trivial.

Mars may offer an opportunity for a planetary-scale control experiment on the influence of mag-netic fields on atmospheric escape. This is becauseMars possesses regions of strongly magnetized crust,as well as regions of non-magnetized crust. Thus

120

it is possible to study atmospheric escape from dif-ferent regions of Mars while controlling for many ofthe other variables that could influence escape. TheMAVEN spacecraft currently orbiting Mars is mak-ing measurements that allow us to address this ques-tion.

In this presentation we will summarize the cur-rent understanding about the role of magnetic fieldsin atmospheric retention, and present an analysis ofMAVEN measurements of charged particle escapefrom the atmosphere from magnetized and unmag-netized regions. The analysis consists of both globalstatistics of the spacecraft observations and detailedcase studies above individual regions.

327.05 — Exploring the Climate of Exoplanets withOASIS

Joao M. Mendonca1; Lars Buchhave11 National Space Institute, Technical University of Denmark (Kon-

gens Lyngby, Denmark)

In the last two decades, the astrophysical commu-nity discovered a multitude of planets orbiting otherstars. The variety of planetary environments thatthese exoplanets may harbour is still unknown. Mostimportantly it propels the fundamental scientific andphilosophical quest of searching for the first detec-tion of life beyond our own planet. As more obser-vational data become available, models of exoplane-tary atmospheres are essential, at a first level to in-terpret the data and more importantly to reproduceand explain the physical and chemical processes thatgenerate the climate of planets.

My poster presents the new planet simulator, OA-SIS, that I have developed from scratch to studyplanet’s habitability and search for life in exoplanets.Our new planetary virtual lab includes a new state-of-the-art 3D atmospheric model (THOR) coupledself-consistently with other modules that representthe main physical and chemical processes that shapeplanetary climates and their evolution. We have re-cently submitted a manuscript presenting our firstresults using OASIS. Using our new platform, wesuccessfully performed challenging simulations ofVenus-like environments and compared the model’sresults to Venus observations. We will present ournew results and describe the main advantages of us-ing OASIS compared with other models. The new re-sults show that OASIS is a robust and efficient tool tosimulate a large diversity of planet’s environments.

327.06 — Clouds of fluffy mineral aggregates inwarm mini-Neptunes

Kazumasa Ohno1; Satoshi Okuzumi1; Ryo Tazaki21 Earth and Planetary Sciences, Tokyo Institute of Technology

(Tokyo, Japan)2 Astronomical Institute, Tohoku University (Sendai, Japan)

Recent observational efforts have revealed that mini-Neptunes are often covered by clouds that exist atextremely high altitudes. Although a number of the-oretical studies have examined cloud formation inexoplanetary atmospheres, it is still highly uncertainwhat processes lead to the formation of clouds at ex-tremely high altitude. One of the promising mech-anisms of the high-altitude cloud formation is theevolution of particle porosity. Solid mineral con-densates, likely formed in warm exoplanetary at-mospheres, can grow into fluffy aggregates that areeasily lofted to upper atmospheres; however, therehas not been a study on microphysical modeling ofsuch fluffy-aggregate clouds to date. Here, we in-vestigate how the porosity of cloud particles evolvesin exoplanetary atmospheres and influences obser-vations of transmission spectra. We first constructa porosity evolution model that takes into accountthe fractal aggregation and compression of aggre-gates caused by the gas drag and high-energy colli-sion. Our cloud microphysical model coupled withthe porosity evolution model demonstrates that thecloud particle aggregates can be highly porous with-out significant compression. As a result, fluffy-aggregate clouds ascend to altitude much higherthan classically assumed compact-sphere clouds. Wealso calculate synthetic transmission spectra andfind that the fluffy-aggregate clouds largely obscurethe molecular signatures, while it produces spectralslopes originated by the scattering properties of ag-gregates. Comparing the synthetic spectra with theobservations of GJ1214b, a mini-Neptune suggestedto be covered by the high-altitude clouds, we findthat its flat spectrum could be explained by the fluffy-aggregate clouds if the atmosphere is highly metal-rich (≥100×solar). The high-metallicity atmospherepotentially encapsulates the information of past gasaccretion processes onto GJ1214b.

327.07 — An extensive search for metallic ions inthe upper atmosphere of the warm Neptune GJ 436b

Leonardo Dos Santos1; David Ehrenreich1; VincentBourrier1; Mercedes Lopez-Morales2; David Sing3,4;Alain Lecavelier des Etangs5

121

1 Department of Astronomy, University of Geneva (Versoix, GE,Switzerland)

2 Center for Astrophysics, Harvard & Smithsonian (Cambridge,Massachusetts, United States)

3 Department of Earth & Planetary Sciences, Johns Hopkins Uni-versity (Baltimore, Maryland, United States)

4 Department of Physics & Astronomy, Johns Hopkins University(Baltimore, Maryland, United States)

5 Institut d’Astrophysique de Paris (Paris, France)

The M2.5 star GJ 436 hosts a warm Neptune that dis-plays an extended atmosphere that dwarfs its ownhost star. Predictions of atmospheric escape in suchplanets state that H atoms escape from the upper at-mosphere in a collisional regime and that the flowcan drag heavier atoms to the upper atmosphere. Weleveraged the extensive coverage of HST/COS ob-servations of GJ 436 from the Hubble PanCET pro-gram to search for signals of metallic ions in the up-per atmosphere of GJ 436 b, as well as to study theactivity-induced variability of the star at short wave-lengths. Despite being a relatively quiet M dwarf,we found that the FUV an unexpectedly high flarerate of GJ 436 is ∼10 events per day. Furthermore,the host star seems to possess either a long-lived ac-tive region or an active longitude that modulates theFUV fluxes coherently for several years. We repro-duced the Lyman-α transit of GJ 436 b with COS, de-spite the strong geocoronal contamination, indicat-ing that the large atmospheric loss rate is stable overthe timescale of a few years. Several metallic lines ofions in the transition region of GJ 436 are present inour datasets, but the in-transit light curves for eachspecies do not reveal any evidence for the presence ofsuch ions to be escaping from the planet. The previ-ously claimed in-transit absorption in the Si III lineis likely an artifact resulting from the stellar mag-netic cycle. Given its featureless optical transmissionspectrum, it is still not completely clear if it has ahigh metallicity or a cloudy atmosphere. The non-detection of metallic species in its exosphere, in par-ticular Si, suggests that, if GJ 436 b possesses a cloudyatmosphere, then mixing is not efficient in draggingthe Si-rich clouds high enough for sublimation andallow for a significant escape rate of metallic ions.

327.08 — The Transit Spectrum of the Super-Earth55 Cnc e at Low and High Resolution

Michael Zhang11 Caltech (Pasadena, California, United States)

The ultra-hot (∼2000 K) exoplanet 55 Cnc e is themost favorable super Earth for atmospheric charac-terization, as well as the only super Earth for which

there have been convincing claims of atmosphericdetections. Notably, Tsiaras et al 2016 report an up-ward sloping spectral slope in the HST/WFC3 1.1–1.7 μm bandpass, which they attribute to HCN ina hydrogen dominated atmosphere. We jointly an-alyze the low resolution transit spectrum of 55 Cnce taken with multiple instruments at multiple wave-lengths, namely HST/WFC3 at 1.1–1.7 μm, Spitzerat 3.6 μm, and Spitzer at 4.5 μm. In addition, wesearch for the molecular signatures of HCN, CH4,and H2O using high-resolution L band (2.8–3.7 μm)transit spectra collected with the upgraded NIRSPECon Keck. Our 14-transit dataset is the largest andmost diverse yet assembled for this planet. We con-firm the upward sloping spectral shape reported byTsiaras et al 2016 in the WFC3 bandpass. Using aBayesian retrieval analysis, we also confirm that itcan be explained by HCN in a hydrogen dominatedatmosphere under conditions of super-solar metal-licity and high carbon-to-oxygen ratio. However, wedo not see evidence of the excess absorption due toHCN expected in the Spitzer 3.6 μm bandpass, norof the forest of absorption lines expected at high res-olution in L band. We discuss possible instrumentaland physical reasons for the discrepancy.

327.09 — Characterizing exoplanetary atmospheresin the TESS era

Ilaria Carleo1; John Livingston3; Terra Ganey1; PrajwalNiraula2; Seth Redfield1

1 Astronomy, Wesleyan University (Middletown, Connecticut,United States)

2 MIT (Cambridge, Massachusetts, United States)3 University of Tokyo (Tokyo, Japan)

The characterization of exoplanet atmospheres willbe a major scientific endeavor in the coming decades,in particular the search for biosignatures in the at-mospheres of temperate, rocky exoplanets. It is aprimary science driver for upcoming space-based(e.g., JWST) and ground-based (e.g., GMT) facili-ties. There are currently several different researchdirections in exoplanet atmosphere characterizationthat probe fundamental questions in exoplanet for-mation and evolution, as well as provide steppingstones to developing the facilities and techniques toultimately detect biosignatures in the atmospheresof small planets. Observations of extended atmo-spheres provide an opportunity to not only measurethe current conditions in the planetary atmosphere,but also put constraints on formation history and in-terior structure (Owen et al. 1999), interactions withhost star (Cauley et al. 2017), and atmospheric and

122

planetary evolution (Oberg et al. 2011). One such re-search direction is the characterization of extendedexoplanetary atmospheres with different spectral ab-sorption lines, namely Lyman-α, H-α, or HeI at dif-ferent wavelengths. Here, we present an analysisaimed to estimate the planetary mass-loss rate andthe signal of these lines starting from the propertiesof the TESS candidates sample. This allows to selectthe more promising targets for a follow-up with ra-dial velocities and possibly atmospheric characteri-zation. We also present a summary of our follow-upobservations to characterize the planets in the sys-tem GJ9827, the nearest planetary system that Kepleror K2 has ever found, with three super-Earth plan-ets in 1:3:5 commensurability. GJ9827b has a rela-tively hot atmosphere, which makes it an ideal tar-get for measuring atmospheric escape, particularlygiven the high activity of its host star.

327.10 — On the observational assessment of M-Dwarf Rocky Planet Atmospheres

Matej Malik1; Daniel D. B. Koll3; Megan Mansfield2;Edwin Kite2; Eliza Kempton1; Jacob Bean2; DorianAbbot2; Renyu Hu4

1 University of Maryland (College Park, Maryland, United States)2 University of Chicago (Chicago, Illinois, United States)3 MIT (Cambridge, Massachusetts, United States)4 Jet Propulsion Laboratory (Pasadena, California, United States)

Preface: This abstract is about three new studies,forming a collaborative effort about rocky exoplanetatmospheres. In agreement with the email instruc-tions of the SOC these studies are not yet publishedand thus “new”, and also “exciting” because theytreat the question of observability and characteristicsof rocky exoplanet atmospheres, a class of exoplan-ets that will undoubtedly be in the focus of futureobservations.

Abstract: Maybe the most promising candidatesfor near-future characterization attempts with JWSTare super-Earths orbiting small M-dwarfs, as theyprovide a unique insight into smaller, rocky exoplan-ets while offering a comparatively high signal con-trast. In order to understand these objects better, wesimulate model atmospheres in radiative-convectiveequilibrium including the effects of a solid plane-tary surface of the three super-Earths, TRAPPIST-1b,GJ 1132b, and LHS 3844b. Our self-consistent mod-eling shows that the atmospheric radiative transferis significantly influenced by the cool M-star irradi-ation. Specifically, we find temperature inversionsin the upper atmosphere and a mitigation of con-vection in the lower atmosphere due near-infrared

non-gray gas opacity. Since the strength of spec-tral features directly depends on the temperaturegradients in the visible atmosphere, we stress thatparametrized atmospheric models, often used in ob-servability studies, are not sufficient to accuratelypredict spectral emission of exoplanets. The ques-tion remains whether these planets retain atmo-spheres at all, given the high-energy flux of their hoststar. In this context, we explore the opportunities ofsecondary eclipse photometry with JWST as a test forthe presence of atmospheres. We find that secondaryeclipse measurements are well suited to distinguishbetween rocky planets that do and do not possess at-mospheres. The underlying idea is that measuring arelatively cool dayside or high Bond albedo is indica-tive of an atmospheric day-night heat transport or thepresence of reflective clouds. Our findings allow usto optimize the scientific return of future JWST ob-servations and are thus interesting to the broad exo-planet community.

327.11 — Living on the Edge: The Effects of a Sur-face on Atmospheric Circulation and Emission Fea-tures for 1.5 REarth Planets

Erin M. May1; Emily Rauscher11 University of Michigan (Ann Arbor, Michigan, United States)

It is well known that planets between the radius ofEarth and Neptune have been the most commonlydetected to-date. Without a direct comparison in ourSolar System, we are left without an immediate un-derstanding of what the compositions of these typesof planets are. To classify them, we have tradition-ally turned to mass-radius relations and compositioncurves in order to determine the likelihood of such aplanet being rocky or gaseous. While previous workhas determined that at a radius of approximately 1.5times that of Earth a planet is equally likely to beeither terrestrial or gaseous, we cannot expect thatany transition between these two compositions willbe a sharp cut-off due to the wide range of possiblecomposition curves that can agree with a given den-sity measurement for a planet of this size. There-fore, we must turn to alternative methods to clas-sify these planets when they fall in this regime andwhen aerosols limit our ability to break this degener-acy by measuring atmospheric composition directlythrough transmission and emission spectroscopy. Inthis talk we present one such new method using 3Dgeneral circulation models of 1.5 Earth radii planets.With this method, we study the effects of a surfaceon observable quantities such as equator-to-pole dif-ferences in emitted and reflected light with an eye

123

on detecting the presence of a surface through sec-ondary eclipse mapping with future instrumenta-tion. In this talk we will present our updated GCM,verified on the circulations of Earth and Neptune, be-fore discussing our modeling choices and results forour transition planets. Finally, we discuss the com-parison of models with surfaces at various pressuresand without surfaces while exploring the prospectsof detecting a surface through secondary eclipse ob-servations in the near future.

327.14 — Atmospheric Models and Transit Spectrafor Hot Rocky Planets

Roxana Lupu1,21 BAER Institute (Mountain View, California, United States)2 NASA Ames (Moffett Field, California, United States)

We are building a set of atmospheric models to cal-culate the structure, composition, and both transitand eclipse spectra for hot rocky exoplanets in shortperiod orbits. These hot worlds offer the best op-portunity to characterize and understand the for-mation and evolution rocky exoplanets with cur-rent and future instruments. We are using a fullynon-grey radiative-convective atmospheric structurecode with cloud formation combined with a self-consistent treatment of gas chemistry above themagma ocean. Being in equilibrium with the surface,the vaporized rock material can be a good tracer ofthe bulk composition of the planet. We are investi-gating both volatile-poor and volatile-rich composi-tions, with the volatile poor ranging from completelydepleted, to water-free (Venus-like), to containingonly sulfur and halogens (Io-like). To properly ac-count for these exotic compositions and thermody-namic regimes, including vertical mixing, condensa-tion, and photochemistry. We present our modelsfor the atmospheric structure of hot rocky planets, aswell as new predicted transit and eclipse spectra withthe identification of specific spectral signatures in theoptical through mid-IR. Our models will inform thedesign and interpretations of follow-up observationswith JWST and ground-based instruments, and pro-vide a better understanding of volatile behavior inthese atmospheres.

327.15 — Limits on the atmosphere of a habitable-zone terrestrial planet from ground-based spec-troscopy

Hannah Diamond-Lowe1; David Charbonneau1; JasonDittmann2; Eliza Kempton3

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 Earth and Planetary Science, Massachusetts Institute of Technol-ogy (Boston, Massachusetts, United States)

3 University of Maryland (College Park, Maryland, United States)

LHS 1140b is a terrestrial exoplanet orbiting in thehabitable zone of an M4.5 star (Dittmann et al. 2017b;Ment et al. 2019). From radial velocity measure-ments of its mass and transit measurements of its ra-dius, we know that LHS 1140b has a rocky composi-tion, though its radius of 1.7 Earth radii puts it at theupper limit of the population of rocky worlds (Ful-ton et al. 2017). It is an open question whether ornot temperate terrestrial planets orbiting low-massstars can retain atmospheres. The exploratory in-vestigation we present here aims to put upper limitson what we can achieve with ground-based obser-vatories to answer this question. This investigationfaced challenges: the 25-day orbital period and 2-hour transit duration of LHS 1140b means that thereare few observable transits per year from any sin-gle observing location. Over the course of 2017 and2018 we observed two transits of LHS1140b, eachtime using both 6.5 m Magellan Telescopes at LasCampanas Observatory simultaneously. We usedthe multi-object spectrographs LDSS3C and IMACSto produce a transmission spectrum from 610-1030nm. The residuals in our data are within 1.5× thephoton-noise limit. This data set is not sufficientfor us to test realistic models of the atmosphere ofLHS 1140b, owing to its cool equilibrium tempera-ture (235 K), high surface gravity (23.7 m/s2), andlikely high mean molecular weight composition. Al-ternatively, if we can identify a system similar to LHS1140 (i.e., having a terrestrial planet orbiting in thehabitable zone of a mid-M star) but at a distance of5 pc and with the Earth’s surface gravity, an atmo-spheric detection could be achieved in as few as 8transits, depending on the atmospheric composition.In the era of extremely large telescopes (ELTs), wemay be able to access the atmospheres of hard-to-reach planets like LHS 1140b.

HDL is supported by an NSF Graduate ResearchFellowship. This work is supported by a grant fromthe John Templeton Foundation.

327.16 — The evaporating atmosphere of the youngexoplanet K2-25b

Keighley Rockcliffe1; Elisabeth Newton11 Physics and Astronomy, Dartmouth College (Lebanon, New

Hampshire, United States)

Host stars create environments that shape the atmo-spheric structure and composition of their exoplan-ets, the effects of which we observe today. The stel-

124

lar wind and high energy radiation present withinastrospheres can erode planetary atmospheres, lead-ing to the detections of atmospheric evaporation forshort-period planets Gl 436b and GJ 3470b. We haveobtained 2 transit observations of the Neptune-sizedexoplanet K2-25b in Lyman α with HST/STIS. Theyoung age of the K2-25 system and K2-25b’s sizeand similarities with Gl 436b suggest the possibil-ity of a large amount of atmospheric escape. Study-ing this planet will further our understanding of theevolution of small planets by probing how planetaryproperties, like size and youth, impact atmosphericescape. By characterizing the total intrinsic Lymanα and X-ray flux from K2-25, we will present con-straints on the EUV environment the exoplanet re-sides in and the mass loss it is experiencing.

327.17 — Iceland, Iceline, IcePlanets

Li Zeng11 Earth and Planetary Sciences, Harvard University (Cambridge,

Massachusetts, United States)

Exoplanet radii show a bimodal distribution, withtwo peaks corresponding to smaller planets (likelyrocky) and larger intermediate-size planets, re-spectively. We apply interior structure model,growth model, as well as atmospheric escape model,and conduct Monte Carlo simulations, to demon-strate that many intermediate-size planets are wa-ter worlds. This result has profound implicationson planet formation theory and origins of life.(https://arxiv.org/abs/1906.04253)

327.18 — The transition from primary to secondaryatmospheres

Edwin S. Kite11 University of Chicago (Chicago, Illinois, United States)

I will address two questions: (1) Do TESS/Keplersuper-Earths have atmospheres? (2) For sub-Neptunes, how does magma-atmosphere exchangeset the composition of the atmosphere? Both involvecoupling between magma oceans and H2-rich atmo-spheres.

(1) After stellar XUV flux declines, Super-Earthvolcanism continues for Gyr. Late volcanism mightrejuvenate atmospheres, but only if the solid man-tle has enough volatiles. Volatiles that have high-molecular-mass (high-μ) are delivered in the first∼108 yr. On this timescale, for short-period exoplan-ets, early-accreted nebular atmospheres are ablated.Thus, rocky exoplanet volatiles might be entrainedby escaping H2. The extent of high-μ loss depends

on magma ocean crystallization timescales. If this isslow, then because the atmosphere and magma areequilibrated, all of the high-μ volatiles can escape.However, early crystallization favors later existenceof an atmosphere. The (small) fraction of volatileswhich are trapped within the crystals are shieldedduring the era of high XUV. Crystallization timescaledepends on stellar luminosity, so this model sug-gests a period threshold for atmosphere absence onSuper-Earths.

(2) Sub-Neptunes are mostly magma by mass, andare mostly atmosphere by volume. I explore theeffect of the magma on the atmosphere. For theFe-Mg-Si-O-H system, the fate of H (and the sub-Neptune atmosphere’s composition) is set by magmaoxidation state. In turn this oxidation state is setby the details of planet assembly. Depending onmagma redox state, a nebular-derived atmospherecan masquerade as an outgassed atmosphere andvice versa. Fortunately, planet radius and atmo-sphere scale height can be used to tell these scenar-ios apart. H2O dissolved in the magma is a majorH sink for sub-Neptunes: the total volatile budgetcan be mostly dissolved H2O even when μatm ∼2.Because of these sinks, the amount of H2 needed toexplain the radius of sub-Neptunes is much greaterthan usually assumed.

These results point the way to future modelingwork, combining magma-ocean chemistry with frac-tionating escape during the transition from primaryto secondary atmospheres.

327.19 — Testing Atmospheric Retrieval Assump-tions

Luke Tremblay1; Jacob Lustig-Yaeger21 School of Earth and Space Exploration, Arizona State University

(Phoenix, Arizona, United States)2 Astronomy & Astrophysics, University of Washington (Seattle,

Washington, United States)

Transmission spectroscopy is a powerful probe of thephysical properties of exoplanet atmospheres and,with the upcoming launch of the James Webb SpaceTelescope (JWST), will soon be used to study theEarth-sized planets orbiting TRAPPIST-1. However,the method by which we interpret transmission spec-tra, known as atmospheric retrievals, often employssimplified physics to make as a means to improve sta-tistical rigor. Here, we attempt to validate numer-ous physical assumptions, common in atmosphericretrievals, by inferring the atmospheric compositionfrom synthetic transmission spectra generated witha line-by-line full physics radiative transfer code.

125

Specifically, we investigate non-uniform vertical at-mospheric thermal structures and molecular mix-ing ratios, aerosol parameterizations, and ray trac-ing with refraction versus geometric optics. Thesetests help to validate the information that may beretrieved from observed terrestrial exoplanet trans-mission spectra, but also expose potential biases thatemerge from using simplified physical models. Ourresults will elucidate a more complete picture of thefeasibility of JWST to accurately retrieve informationon the atmospheres, habitability, and biosignaturesof small rocky planets.

328 — Planetary Atmospheres —Directly Imaged Planets and BrownDwarfs, Poster Session328.01 — First Light of the Keck Planet Imager andCharacterizer

Jason Wang1; Dimitri Mawet1; Nemanja Jovanovic1;Jacques Robert Delorme1; Charlotte Bond2; JacklynPezzato1; Sylvain Cetre3; Daniel Echeverri1; J. KentWallace4; Randall Bartos4; Scott Lilley3; Sam Ragland3;Garreth Ruane4; Peter Wizinowich3; Mark Chun2; JiWang5; Michael P. Fitzgerald7; Andy Skemer6; EmilyMartin6; Ed Wetherell3; Eric Wang7; Shane Jacobson2;Eric Warmbier2; Charles Lockhart2; Donald Hall2

1 Astronomy, California Institute of Technology (Pasadena, Califor-nia, United States)

2 Institute for Astronomy, University of Hawaii (Hilo, Hawaii,United States)

3 Keck Observatory (Waimea, Hawaii, United States)4 Jet Propulsion Laboratory, California Institute of Technology

(Pasadena, California, United States)5 Astronomy, The Ohio State University (Comlubus, Ohio, United

States)6 Astronomy, University of California, Santa Cruz (Santa Cruz,

California, United States)7 Astronomy, University of California, Los Angeles (Santa Monica,

California, United States)

High-resolution spectroscopy has enabled detailedcharacterization of the composition and dynamicsof exoplanets through measurements of individualspectral lines. The Keck Planet Imager and Char-acterizer (KPIC) consists of upgrades to the Keck IIadaptive optics (AO) system and instrument suite.This presentation will focus on the first light scienceresults from the novel fiber injection unit (FIU) ofKPIC that isolates and injects planet light from theAO system into the NIRSPEC spectrograph via a sin-gle mode fiber. By combining high-contrast imag-

ing and high-resolution spectroscopy, the KPIC FIUis the first instrument to implement high dispersioncoronagraphy to study faint imaged exoplanets closeto their host stars at high spectral resolution. We areobtaining K-band R∼35,000 exoplanet spectra sen-sitive to spectral lines of water, methane, and car-bon monoxide in their atmospheres, as well as plane-tary radial velocities and spin rates. These measure-ments provide insights into the composition, forma-tion, and accretion history of these young gas giants.I will present early science results from data we haveobtained this year. I will conclude with upcomingupgrades to KPIC that will enhance its science capa-bilities such as detecting radial velocity exoplanetsand discuss the science case for high dispersion coro-nagraphy on the Thirty Meter Telescope based on theperformance of KPIC.

328.02 — Spectral and atmospheric characterisationof a new benchmark brown dwarf

Emily Rickman1; Damien Ségransan1; AnthonyCheetham2

1 University of Geneva (Versoix, Switzerland)2 Institute for Astronomy (MPIA) (Heidelberg, Germany)

Evolutionary models of brown dwarfs are plaguedby a lack of observational constraints. The complexmolecular chemistry of their atmospheres leaves arelatively wide parameter space for models to span.Placing accurate mass and luminosity data to ob-servationally populate the mass-luminosity relation-ship provides a major contribution to an understand-ing of the interplay between gravitational contractionand nuclear burning.

To date, individual dynamical masses are knownfor only a handful of brown dwarfs, therefore anynew detections contributes greatly to brown dwarfmodels as they provide important analogues for thecharacterisation of exoplanets.

Radial velocity measurements provide only alower limit on the measured masses due to the un-known orbital inclination. Therefore directly imag-ing these candidates is needed to break that degener-acy and provide constraints on the dynamical massof the companion. With over 20 years worth of ra-dial velocity measurements from the CORALIE sur-vey for extrasolar planets, we have identified sev-eral promising candidates that we have directly ob-served.

In this talk we will announce the detection of a newbenchmark ∼30 MJup brown dwarf with radial ve-locity and direct detections allowing us to constrainits dynamical mass, luminosity, spectral type and weobtain a low resolution spectrum of the companion.

126

The discovery of benchmark sources provides apowerful and critical tool of advanced evolutionarymodels. As we move toward imaging smaller andsmaller objects it is important to use these objects asa laboratory to test theoretical atmospheric models.

328.03 — Detection of New Strongly VariableBrown Dwarfs in the L/T Transition

Simon Christoffer Eriksson1; Markus Janson1; PerCalissendorff1

1 Astronomy, Stockholm University (Vallentuna, Stockholm, Swe-den)

Brown dwarfs in the spectral range L9 – T3.5, withinthe so called L/T transition, have been shown tobe photometrically variable at higher amplitudesand with greater frequency than other field dwarfs.This strong variability allows for the probing oftheir atmospheric structure in 3D through multi-wavelength observations for studying the underly-ing physical mechanisms responsible for the vari-ability. The few known strongly variable dwarfsin this range have been extensively studied. Now,more such variables need to be discovered and stud-ied to better constrain atmospheric models, also crit-ical for the understanding of giant exoplanets, andto shed light on a number of possible correlationsbetween brown dwarf characteristics and variabil-ity. The largest ground survey to date (Radigan etal. 2014) suggest an occurrence rate for strong vari-ability (peak-to-peak amplitudes > 2 %) of up to ∼39% among brown dwarfs within the L/T transition.

Here we present the results from a survey with the2.5 m Nordic Optical Telescope, where we have ob-served 10 L9.5 - T3.5 brown dwarfs, alongside theknown strongly variable 2MASS J01365662+0933473,in the J-band during 2017/2018 and find 4/10 to bestrongly variable at a high level of significance, yield-ing a variability occurrence rate of 40+32

−19 %. Wedetect variablity amplitudes up to 10.7 % for the T1dwarf 2MASS J22153705+2110554, for which we alsodetect significant light curve evolution between the2017 and 2018 epochs, possibly similar to the evo-lution seen in other brown dwarfs, e.g. T2 2MASSJ13243553+6358281 (Apai et al. 2017). We furtherconfirm the continued strong variability of J0136 andfind marginally significant but strong variability inanother target.

This survey almost doubles the number of knownstrong variables in the L/T transition, provid-ing much needed targets for the continued multi-wavelength study of brown dwarf atmospheres us-ing the SST or HST. Combining our findings with

those in the literature, we’ve also created an exten-sive multi-epoch catalogue over all known stronglyvariable brown dwarfs and discuss possible correla-tions that can be identified from this overview.

328.04 — Spectral characterization of newly de-tected young substellar binaries with SINFONI

Per Calissendorff1; Markus Janson1; Ruben AsensioTorres2; Rainer Köhler3,4

1 Department of Astronomy, Stockholm University (Stockholm,Stockholm, Sweden)

2 Stockholm University (Stockholm, Sweden)3 Sterrewacht Leiden (Leiden, Netherlands)4 Department of Astrophysics, University of Vienna (Vienna, Aus-

tria)

Multiplicity studies of stars and brown dwarfs haveshown a decrease in multiplicity frequency as a func-tion of primary mass and spectral type, with fewermultiple systems for primaries of lower mass andlater spectral types, which stretches down all theway to the substellar mass-regime. However, at thevery bottom of the stellar and substellar initial massfunction, binarity and multiplicity is not very well-constrained. Recent efforts in associating low-massmembers to young moving groups have providedmuch sought means to constrain ages for substel-lar objects, which otherwise prove difficult to do.As such, we are now at a point where we can com-pare the multiplicity rates for both older and youngersamples of substellar brown dwarfs, which provideinsight to their formation channels and dynamicalevolution. This is of particular interest as very low-mass brown dwarfs are analogous to directly imagedexoplanets, which when resolved can be studied inhigh detail. We present the results from observationsof 14 young low-mass substellar objects using theVLT/SINFONI integral field spectrograph with laserguide star adaptive optics, which we employ to de-tect and characterize 3 new binary systems. These re-sults indicate for higher multiplicity frequencies forthe younger populations of brown dwarfs, and thatolder systems may have undergone dynamical inter-actions disrupting the primordial binaries. By uti-lizing substellar theoretical models and the resolvedbrightnesses we obtain from our astrometric tech-niques, we discover some of these companions to beof planetary-mass. Furthermore, we find that the bi-nary systems have small separations, which trans-lates to orbital periods of just a few decades. As such,dynamical masses can be obtained within just a fewyears of astrometric monitoring, making these sys-tems excellent benchmarks for calibrating evolution-

127

ary models in an otherwise scarcely probed mass-regime.

328.05 — Revealing the Turbulent, Cloudy Natureof Young Giant Exoplanet Analogs

Johanna M. Vos1; Jacqueline Faherty1; Kelle Cruz2;Emily Rice3; Jonathan Gagné4; Mark Marley5; JohnGizis6

1 Astrophysics, American Museum of Natural History (New YorkCity, New York, United States)

2 CUNY, Hunter College (Manhattan, New York, United States)3 CUNY, College of Staten Island (Staten Island, New York, United

States)4 University of Montreal (Montreal, Quebec, Canada)5 NASA Ames (Mountain View, California, United States)6 University of Delaware (Delaware, Delaware, United States)

An apparent correlation between enhanced cloudsand youth has emerged as an important insight intocool-temperature atmospheres. Spitzer variabilitymonitoring allows us to trace the presence of inho-mogeneous clouds in brown dwarf and exoplanet at-mospheres. This method is particularly sensitive tothe high-altitude clouds that are suspected in low-gravity atmospheres. In this talk I will present pre-liminary results from our ongoing survey for Spitzervariability in an age-calibrated sample of young (20-130 Myr), low-gravity objects that share a striking re-semblance with the directly-imaged planets. Com-bining our sample with similar variability studiesof old brown dwarfs (3-5 Gyrs) will unequivocallyreveal the extent of turbulent clouds in the atmo-spheres of young, exoplanet analogs, while also pro-viding crucial information on the angular momen-tum evolution and atmospheric dynamics of young,isolated exoplanet analogs.

328.06 — Atmosphere and Evolutionary Models forBrown Dwarf and Giant Exoplanets

Mark Phillips1; Pascal Tremblin2,1; Isabelle Baraffe1,3;Gilles Chabrier3,1; Nicole Allard4,5

1 University of Exeter (Exeter, United Kingdom)2 Maison de la Simulation, CEA (Gif-Sur-Yvette, France)3 Université de Lyon, (Lyon, France)4 Observatoire de Paris (Paris, France)5 Sorbonne Université (Paris, France)

The study of brown dwarfs and giant exoplanets israpidly evolving as ever-improving instrumentationbecomes sensitive to cooler objects. Accurate and re-liable atmosphere and evolutionary models are im-portant for placing mass and age constraints on thesenewly discovered objects, and understanding the

rich chemistry and physics taking place in their at-mospheres. We are expanding on the widely usedCOND evolutionary models by developing a gridof model atmospheres (Teff=200-2000K, log(g)=2.5-5.5) with our state-of-the-art 1D radiative-convectiveequilibrium code ATMO. ATMO includes the latestopacities for important molecular absorbers such asH2O, CH4 and NH3, and the latest line shapes for thecollisionally broadened Na and K resonance lines.These model improvements allow us to follow theevolution of 1-75MJup objects down to the coolest ef-fective temperatures (Teff=200K). We present com-parisons of this new model set to those previouslypublished, illustrating how the evolutionary tracksand the substellar boundary have changed due to im-proved opacities and the usage of a new equationof state. We also compare our model grid to ob-servational datasets in colour-magnitude diagramsand investigate the impact of our new Na and K lineshapes in reproducing brown dwarf spectra. Ourfuture work will involve expanding on this initialgrid, to investigate the effects of metallicity, C/Oratio and non-equilibrium chemistry in cool browndwarfs and giant exoplanets.

328.07 — Moderate Resolution Spectroscopy of Di-rectly Imaged Planets

Kielan Kathryn Wilcomb1; Quinn Konopacky1; TravisBarman2; Christopher Theissen1; Laci S. Brock2; BruceMacintosh3; Jean-Baptiste Ruffio3; Christian Marois4

1 Physics, UC San Diego (San Diego, California, United States)2 Lunar and Planetary Laboratory, University of Arizona (Tucson,

Arizona, United States)3 Stanford University (Stanford, California, United States)4 NRC-Herzberg (Victoria, British Columbia, Canada)

Recent direct imaging of exoplanets has revealeda population of Jupiter-like objects that orbit atlarge separations (∼10-100 AU) from their host stars.These planets, with masses of ∼2-14 MJup and tem-peratures of ∼500-2000 K, remain a problem for thetwo main planet formation models—core accretionand gravitational instability. We present results fromour ongoing survey of directly imaged planets withmoderate (R∼4000) spectral resolution. We are mak-ing use of OSIRIS on the W.M. Keck I 10 meter tele-scope, which offers some of the best spectra to-datefor directly imaged substellar companions. Thus far,we have observed eight companions in the K band(∼2.2 µm), including the “super-Jupiter” Kappa An-dromeda b. Our spectra reveal resolved molecularlines from water and CO, allowing for the derivationof atmospheric properties such as temperature, sur-face gravity, metallicity, and C/O ratio. In partic-

128

ular, we confirm that Kappa And b has a low sur-face gravity, consistent with a young age and massnear the deuterium burning limit. We also find thatKappa And b potentially has a sub-solar metallic-ity. We compare our spectra of the companions inK band to those of other brown dwarfs and gas gi-ant planets, and to each other. Our survey will im-prove our knowledge of the intricate atmospheres ofyoung, substellar objects.

328.08 — Detecting a substellar high-contrast com-panion to an M-dwarf with CARMENES

Maritza Soto11 School of Physics and Astronomy, Queen Mary University (Lon-

don, United Kingdom)

M-dwarf stars are the most abundant type of starin the galaxy, but not very studied due to theirlow temperatures and radiation mostly in the in-frared. The CARMENES spectrograph was createdwith the aim of studying these objects, by observ-ing both in the visible and infrared wavelengths,and delivering high resolution spectra. Here westudy the spectra obtained for one M dwarf ob-served with CARMENES, and attempt to detect thehigh contrast brown dwarf companion using cross-correlation techniques. We combine the spectra ob-tained to create a template, which is then removedfrom each individual spectrum. The spectra areshifted to the companion’s restframe, and the residu-als are cross-correlated with model spectra of differ-ent temperatures. We show that a companion with acontrast level of 0.001 at 1 micron is detectable. Thismakes this target a good candidate for future directimaging efforts.

328.09 — Directly-imaged atmospheric characteri-sation with TauREx retrievals.

Niall Whiteford1; Alistair Glasse2; Beth Biller1; KenRice1; Paul Palmer1

1 University of Edinburgh (Edinburgh, United Kingdom)2 UKATC (Edinburgh, United Kingdom)

Inverse methods have become a fundamental anal-ysis technique for modelling exoplanetary atmo-spheres. This technique explores a variety of po-tential bulk and atmospheric model parameters thatcombine to best-fit an observed spectrum. TauREx(Tau Retrieval of Exoplanets) is a Bayesian retrievalsuite designed to be applied to spectroscopic ob-servations of extrasolar planetary atmospheres. Wehave adapted TauREx for analysis of near-infraredspectrophotometry from a variety of directly-imaged

gas giant exoplanets and brown dwarfs. This in-cludes the HR 8799 system and 51 Eri b observed us-ing instruments such as SPHERE and GPI. This anal-ysis returns estimates for target mass, radius, surfacegravity and temperature-pressure structure, as wellas confirming and constraining the presence of a va-riety of molecular species including H2O and CO.This project aims to help bridge directly-imaged exo-planet observations with robust, efficient and precisecharacterisation. The development and adaptationof this retrieval tool is very timely and relevant giventhe upcoming launch of JWST.

328.10 — High Spatial Resolution Thermal InfraredSpectroscopy with ALES

Zack Briesemeister1; Andy Skemer1; Jordan Stone2;Travis Barman3

1 Astronomy and Astrophysics, University of California, SantaCruz (Santa Cruz, California, United States)

2 Steward Observatory, University of Arizona (Tucson, Arizona,United States)

3 Lunary and Planetary Laboratory, University of Arizona (Tucson,Arizona, United States)

The Arizona Lenslets for Exoplanet Spectroscopy(ALES) mode of the Large Binocular Telescope In-terferometer is the first adaptive-optics-fed integralfield spectrograph capable of delivering 3-5 micronspectral data cubes of directly imaged exoplanetsand substellar companions. By extending spectro-scopic wavelength coverage of the spectral energydistributions of companions beyond near-infraredspectrophotometry, we become more sensitive to theensemble of atmospheric processes that shape theirthermal emission. ALES exists as a precursor instru-ment to the proposed third generation Keck instru-ment, SCALES, and the red channel of the proposedsecond generation TMT instrument, PSI. Throughdetailed instrument simulation, we evaluate the sen-sitivity of the integral field spectrograph modes ofthe three instruments for LBTI, Keck, and TMT, re-spectively, for various science cases.

328.11 — A new look at the solar neighborhood:distances and atmospheres of the coldest planetarymass brown dwarfs

Emily Martin1; Davy Kirkpatrick2; Richard Smart3;Charles Beichman2; Alfred Cayago4; Michael Cushing6;Jacqueline Faherty5; Christopher Gelino2; SarahLogsdon7; Federico Marocco2; Ian McLean8; BrittanyMiles1; Adam Schneider11; Andy Skemer9; ChristopherTinney10; Edward Wright8

1 UC Santa Cruz (Santa Cruz, California, United States)

129

2 UNSW (Sydney, New South Wales, Australia)3 Arizona State University (Tempe, Arizona, United States)4 IPAC/Caltech (Pasadena, California, United States)5 INAF/ OATO (Torino, Italy)6 UC Riverside (Riverside, California, United States)7 AMNH (New York, New York, United States)8 University of Toledo (Toledo, Ohio, United States)9 NASA Goddard (Greenbelt, Maryland, United States)10 UC Los Angeles (Los Angeles, California, United States)11 Astronomy, UC Santa Cruz (Santa Cruz, California, United

States)

The coldest brown dwarfs, late-T and Y dwarfs, aretestbeds for atmospheric studies of the extremelycold atmospheres of free-floating planetary-mass ob-jects and provide insights into the lowest mass endof the star formation process. Using Spitzer/IRACimaging, we present new and updated parallaxes for>140 late-T and Y dwarfs in the nearest ∼20 pc (Mar-tin et al. 2018, Kirkpatrick et al. 2019). In this talk, Iwill present an updated understanding of the phys-ical properties of the coldest brown dwarfs based onthese parallaxes. I will also present a new look at thesubstellar field mass function, which shows that thestar formation process does not have a physical cut-off at the deuterium burning limit, and if a low-masscut-off exists, it is likely below 5 Jupiter masses. Fi-nally, I will present the first-ever Keck/NIRES spec-tra of Y dwarfs; the highest resolution near-infraredspectra of Y dwarfs to date (Martin et al, in prep).These spectra combined with our new parallaxesgive us a detailed look at the coldest, free-floating,planetary mass objects.

328.12 — Gemini Planet Imager Spectroscopy of theReddest Known Substellar Companion HD206893B

Kimberly Ward-Duong2,1; Jennifer Patience3; RobertDe Rosa4; Julien Rameau5; Katherine B. Follette2; MarkMarley6; Abhijith Rajan7; Alexandra Greenbaum8

1 Five College Astronomy Department (Amherst, Massachusetts,United States)

2 Amherst College (Amherst, Massachusetts, United States)3 School of Earth and Space Exploration, Arizona State University

(Tempe, Arizona, United States)4 Stanford University (Stanford, California, United States)5 IPAG/Univ. Grenoble Alpes (Grenoble, France)6 NASA Ames (Mountain View, California, United States)7 STScI (Baltimore, Maryland, United States)8 University of Michigan (Ann Arbor, Michigan, United States)

From the Gemini Planet Imager Exoplanet Sur-vey, we present new near-infrared spectroscopy ofHD206893 B, a substellar companion orbiting within

the debris disk of an F5V star. New J, H, K1, andK2 spectra with GPI demonstrate the extraordinaryred color of the object, presenting a challenging at-mosphere to model with existing model grids. Wepresent comparison with field and young L-dwarfsto assess whether the NIR spectra are consistent withupper atmosphere sub-micron hazes. Multi-epochastrometric monitoring of the system suggests aprobable semimajor axis of 10 au, well within the es-timated disk inner radius of ∼50 au. As only the sec-ond brown dwarf imaged within the innermost re-gion of a debris disk, the properties of this system of-fer important dynamical constraints for companion-disk interaction and a useful benchmark for browndwarf and giant planet atmospheric study.

329 — Planetary Atmospheres —Theory, Poster Session329.01 — Towards a more complex description ofchemical profiles in exoplanets retrievals: A 2-layerparameterisation

Quentin Changeat1; Billy Edwards1; Ingo Waldmann1;Giovanna Tinetti1

1 Physics and Astronomy, University College London (London,United Kingdom)

State of the art spectral retrieval models of exoplanetatmospheres assume chemical profiles which areconstant with altitude/pressure. This assumptionis justified by the information content of currentlyavailable datasets which do not allow, in most cases,for the molecular/atomic abundances as a functionof atmospheric pressure/altitude to be constrained.In the context of the next generation of space tele-scopes, a more accurate description of chemical pro-files with additional levels of flexibility may becomecrucial to interpret observations and gain new in-sights into atmospheric physics. We explore here thepossibility of retrieving pressure-dependent chemi-cal profiles from transit spectra as recorded by fu-ture space observatories, without injecting any pri-ors from theoretical chemical models in the retrievalalgorithms. The “2-layer” retrieval parameterisationpresented here allows for the independent extractionof molecular/atomic abundances above and belowa certain atmospheric pressure. By simulating var-ious cases, we demonstrate that this evolution fromthe assumption of constant chemical abundances isjustified by the information content of transit spec-tra provided by future space instruments. Compar-isons with traditional retrieval models show that as-sumptions made on chemical profiles may signifi-

130

cantly impact retrieved parameters, such as the at-mospheric temperature, and justify the attention wegive here to this issue. We find that the 2-layer re-trieval is able to accurately capture discontinuitiesin the vertical chemical profiles, which could becaused by disequilibrium processes — such as ver-tical mixing or photo-chemistry — or the presence ofclouds/hazes. The 2-layer retrieval could also helpto constrain the composition of clouds and hazesby exploring the correlation between the chemicalchanges in the gaseous phase and the pressure atwhich the condensed/solid phase occurs. The 2-layer retrieval presented here therefore represents animportant step forward in our ability to constraintheoretical chemical models and cloud/haze compo-sition from the analysis of future observations.

329.02 — The Formation of Zonal Flow and a Hot-spot Shift on Tidally Locked Planets

Mark Hammond1; Raymond Pierrehumbert11 Physics, University of Oxford (Oxford, United Kingdom)

The global atmospheric circulation and temperaturedistribution of tidally locked exoplanets is key to in-terpreting observations such as phase curves, eclipsemaps, and atmospheric retrievals. The dominantdynamical process in their atmospheres is a super-rotating zonal equatorial jet. The meridional cir-culation on tidally locked planets is rarely investi-gated, as it is assumed to be secondary to the zonalday-night circulation. This poster will show that themeridional circulation is in fact vital to the formationof the zonal flow, via the same ”Gierasch-Rossow-Williams” mechanism that operates on Venus. It willshow how the meridional circulation adds eastwardangular momentum to the atmosphere and trans-ports this angular momentum upwards. This mech-anism will then be used to explain the scaling be-haviour of the zonal jet position and speed in a suiteof GCM simulations, by predicting how the variousterms in the zonal momentum equation scale withstellar flux and rotation rate.

Finally, I will link this to our recently publishedstudy using a rotating shallow-water model lin-earized about this eastward zonal flow. I will showthat the response to day-night forcing is modified bythis jet, producing the distinctive eastward hot-spotshift that appears in observations. The key result isthat the shift is not produced by advection of heat,but rather is a result of interactions between forcedstationary waves and the mean flow. This theoreticalprediction of the equilibrium zonal flow and the re-sulting temperature distribution could be useful forinterpreting observations of tidally locked planets.

329.03 — Atmospheric Mass Loss due to Giant Im-pacts

Almog Yalinewich11 CITA (Toronto, Ontario, Canada)

Exoplanet systems (especially Kepler 36b/c and Ke-pler 107b/c) exhibit large density variations betweenneighbouring planets. This difference is attributedto atmospheric content. One mechanism that is usu-ally implicated in atmospheric mass loss is photoe-vaporation, but this mechanism cannot explain theKepler 107b/c dichotomy, for which the planet withthe atmosphere is closer to the host star and less mas-sive. Another explanation is giant impacts that occurshortly after the dispersal of the protoplanetary disc.In this talk I will present new results from a state ofthe art, moving mesh, hydrodynamic simulations ofsuch collisions. These simulations capture the prop-agation of the shock wave through the interior of theplanet. Using these results it is possible to calculatethe amount of atmospheric mass loss from a widerange of parameters. In contrast to other atmosphericloss processes, giant impacts can remove both theprimordial and secondary atmospheres with a largermolecular weight. In this talk I will present new,yet unpublished results that include the effects ofan oblique collision and take into account the finitespeed of sound in the interior of the target planet.With these more realistic simulations I will show thatunder the right circumstances, giant collisions canaccount for the differences in atmospheric content inKepler 36b/c and Kepler 107b/c.

329.04 — Do Magnetic Fields Prevent AtmosphericEscape?

Hilary Egan1,5; Riku jarvinen2,3; Yingjuan Ma4; DavidBrain5,1

1 Astrophysical and Planetary Science, University of ColoradoBoulder (Boulder, Colorado, United States)

2 Department of Electronics and Nanoengineering, Aalto Univer-sity (Espoo, Finland)

3 Finnish Meteorological Institute (Helsinki, Finland)4 Department of Earth Planetary and Space Science, University of

California Los Angeles (Los Angeles, California, United States)5 Laboratory for Atmospheric and Space Physics, University of

Colorado (Boulder, Colorado, United States)

Atmospheric escape is capable of shaping a planet’satmospheric composition and total mass, and thusthe planet’s long-term habitability. Loss to spaceof atmospheric particles has played a key role inthe atmospheric evolution of both Mars and Venus.Intrinsic planetary magnetic fields like the Earth’s

131

have long been thought to shield planets from at-mospheric erosion via stellar winds; however, re-cent arguments have suggested that a magnetic fieldwill increase the interaction area with the solar wind,collecting correspondingly more energy that can beused to drive increased escape.

Using a set of global three-dimensional hybridplasma simulations validated via observations atMars and Venus, we find that neither of theseparadigms are complete descriptions. Rather thansolely inhibiting or driving ion escape, there is avalue of magnetic field strength associated withmaximum ion outflow. For weaker magnetic fields,ion escape is enhanced due to shielding of thesouthern hemisphere from “misaligned” ion pickupforces. For stronger magnetic fields ion escape de-creases due to trapping associated with closed mag-netic field lines. The peak escape rate occurs wherethe intrinsic magnetosphere (caused by the planetarymagnetic field) reaches the induced magnetosphere(caused by ionospheric conductivity). As the size ofthe intrinsic magnetosphere is determined by pres-sure balance between the incoming solar wind andthe planetary magnetic field, the magnetic field as-sociated with peak escape is critically dependent onthe solar wind pressure.

Where possible we have fit power laws for the vari-ation of fundamental parameters (escape rate, es-cape power, polar cap opening angle and effectiveinteraction area) with magnetic field, and assessedupper and lower limits for the relationships. Suchpower laws can be used in generalized studies of at-mospheric escape and potential habitability to bettercharacterize a wide variety of systems.

329.05 — THOR version 2: a GPU-enabled, non-hydrostatic general circulation model for extra-solar planets

Russell Deitrick1; Joao M. Mendonca2; UrsSchroffenegger1; Shang-Min Tsai3; Simon Grimm1;Kevin Heng1

1 Center for Space and Habitability, University of Bern (Bern, Bern,Switzerland)

2 National Space Institute, Technical University of Denmark (Kon-gens Lyngby, Denmark)

3 University of Oxford (Oxford, United Kingdom)

We present the first major update to THOR, a non-hydrostatic, GPU enabled 3D general circulationmodel, which is the culmination of 8 years of workin the Heng group in Bern. THOR is the first GCMthat has been built from the ground up for the studyof exoplanets. Thus, it is entirely free of tunings to-ward solar system planets and contains as few as-

sumptions as possible. It is also publicly availableand we actively encourage the community to becomeinvolved in further development. With this model,we have the capability to model atmospheres withor without the hydrostatic approximation, indepen-dent of the additional approximations that lead tothe primitive equations of meteorology. We use themodel to study whether the climate structures of ex-oplanets are robust to the assumption of hydrostaticequilibrium. We demonstrate that the hydrostaticapproximation alone is sufficient to significantly al-ter the zonal and vertical winds of hot jupiters. Thisimplies that aerosol sizes derived from spectra maybe miscalculated, if the wind velocities are basedupon hydrostatic GCMs. The divergence betweenhydrostatic and non-hydrostatic simulations appearsto be a function of temperature. We further dis-cuss improvements and additions to the model thathave been implemented since the release of version1.0, including grey radiative transfer, chemical trac-ers, and an insolation scheme that allows for arbi-trary orbits and rotation parameters. We have be-gun adapting the model for terrestrial planets withthe goal of studying atmospheric collapse on tidally-locked worlds. Additionally, we reproduce a num-ber of benchmark tests for dynamical cores.

329.06 — Effect of disequilibrium chemistry on thespectra of exoplanet atmospheres

Yui Kawashima1; Michiel Min11 SRON Netherlands Institute for Space Research (Utrecht, Nether-

lands)

Recently, transmission and/or emission spectra ofexoplanet atmospheres have been observed by bothspace- and ground-based telescopes. Forthcomingspace missions such as JWST and ARIEL are ex-pected to enable high-precision observation of thesespectra. Most of the current retrieval models usedto derive the atmospheric properties from observedspectra assume the abundance profiles of chemicalspecies in the atmospheres to be thermochemical-equilibrium or constant ones throughout the atmo-sphere for simplicity. However, in reality, the abun-dance profiles can depart from the thermochemi-cal equilibrium ones by the disequilibrium processessuch as quenching effect. In this study, we exam-ine how the quenching process affects the spectra ofexoplanet atmospheres for some atmospheric prop-erties such as temperature and eddy diffusion co-efficient. For this purpose, we have developed 1-Dmodel to simulate the abundance profiles consider-ing both thermochemical reactions and eddy diffu-sion transport with the use of the chemical timescale

132

for each species derived by Tsai et al. (2018). We dis-cuss the conditions in which we can assume equilib-rium chemistry or disequilibrium chemistry.

329.07 — A coherent disentanglement of the finger-prints of physics, chemistry and dynamics of exo-planets on their atmospheric spectra

Karan Molaverdikhani1; Thomas Henning1; PaulMollière2

1 Max Planck Institute for Astronomy (Heidelberg, Germany)2 Sterrewacht Leiden, Huygens Laboratory (Leiden, Netherlands)

Characterization of planetary atmospheres has beenalways a challenge. While the next generation of fa-cilities, such as E-ELT, JWST, and ARIEL, will help toimprove the status, the number of well-characterizedexoplanet atmospheres will still be limited. Large-scale simulations could assist us by predicting thediversity of the planetary atmospheres, and point-ing toward the regions on the parameter space wherewe have a higher chance of finding interesting targetswith desired properties.

We present the results of an extensive investiga-tion, with a three-step strategy, to understand thefingerprints of physics, chemistry and dynamics ofexoplanets on their atmospheric spectra. In thefirst step, we study the synthetic spectra of 28,224self-consistent cloud-free models; assuming effectivetemperature, surface gravity, metallicity, C/O ratioof the planet, and host star’s stellar type as the freeparameters. We propose a new classification schemeand find a region (Methane Valley) between 800 and1500 K, where a greater chance of CH4 detection isexpected. The first robust CH4 detection on an irra-diated planet places HD102195b within this region;supporting our prediction.

We then investigate the fingerprints of disequilib-rium chemistry on the atmospheric spectra by per-forming 84,672 full chemical network kinetic simu-lations with ChemKM. We find that the quenchingpressure decreases with the effective temperature ofplanets, but it also varies with other atmosphericparameters. We show that the atmospheric mixingdoes not change the shape of the two main color-populations in the Spitzer color-maps and thus anydeviation of observational points from these popula-tions are likely due to the presence of clouds and notdisequilibrium processes. However, we find somecolder planets (Teff<900 K) with very low C/O ra-tios (<0.25) that show significant deviations; makingthese planets interesting cases for further investiga-tions.

We further present the results of 38,500 self-consistent cloudy models to demonstrate how this

picture changes when the radiative feedback ofclouds is included in the models.

329.08 — Retrieving Exoplanet Spectra using deeplearning

Ingo Waldmann11 Physics & Astronomy, University College London (London,

United Kingdom)

The field of exoplanetary spectroscopy is as fast mov-ing as it is new. Analysing currently available ob-servations of exoplanetary atmospheres often invokelarge and correlated parameter spaces that can be dif-ficult to map or constrain. This is particularly true forthe theoretical modelling of their atmospheres andthe atmospheric parameter retrieval from observeddata. Issues of low signal-to-noise data and large,non-linear parameter spaces are nothing new andcommonly found in many fields of engineering andthe physical sciences. Recent years have seen vast im-provements in statistical data analysis and machinelearning that have revolutionised fields as diverse astelecommunication, pattern recognition and medicalphysics. In this talk, I will discuss the use of ma-chine and deep learning in inverse retrievals of ex-oplanetary atmospheres. I will present the ExoGANretrieval framework, which learns the intrinsic like-lihood surface of a radiative transfer retrieval usinggenerative adversarial networks (GANs) and com-pare this approach to other classical and machinelearning solutions in the recent literature. As wefirmly move into the era of ‘big data’ and increas-ing model complexities in the era of JWST, ELTs andARIEL, intelligent algorithms will play an importantpart in facilitating the analysis of these rich data setsin the future.

329.09 — Nucleation of TiO2molecular clusters inthe context of cloud formation on hot Jupiters

Jan Philip Sindel1; David Gobrecht11 Instituut voor Sterrenkunde, KU Leuven (Leuven, Belgium)

Clouds form when a super-saturated gas condenseson small dust grains, leading to an optically thickgas-liquid mixture. On rocky planets like earth theprocesses are well-understood, as small dust grainsare easily transferred from the surface to the up-per atmosphere by winds. On hot Jupiters however,there is no solid surface that supplies dust grains toact as condensation cores. Yet, clouds have been ob-served in the atmospheres of hot Jupiters. The cur-rent understanding of the formation process is thatthe required condensation cores are formed through

133

nucleation of molecular clusters of highly refractorymolecules, similar to the nucleation of dust in stellaroutflows.

In this work we investigate the cluster-formationof Titanium dioxide (TiO2) as a potential candidatefor the seed-nucleus. In order to establish a selfcon-sistent nucleation pathway, it is crucial to obtain thebond-energies for all isomers and geometries of theclusters with high accuracy. We achieve this by usingquantum chemical density functional theory (DFT)calculations. We found that the B3LYP/cc-PVTZlevel of theory comes closest to the experimental val-ues from the Janaf-Nist tables. Since the DFT cal-culations quickly get too computationally expensivewith cluster size, we employ a force-field approachto approximate the energies of larger clusters as ac-curate as possible. Using an interatomic bucking-ham pair potential we find new cluster-geometriesand their respective energies using a simulated an-nealing technique. We use the new clusters and theirenergies as input for a kintetic nucleation model thatserves as the basis of a cloud formation code. Finally,we investigate the impact of the cluster geometriesand energies on the nucleation process and therebyon the cloud formation on hot Jupiters.

329.10 — The effect of internal gravity waves oncloud evolution in sub-stellar atmospheres

Amy Parent1; Ruth Falconer1; Karen Meyer1; Craig R.Stark1

1 Division of Computing and Mathematics, University of AbertayDundee (Dundee, United Kingdom)

Substellar objects exhibit photometric variabilitywhich is believed to be caused by a number of pro-cesses such as magnetically-driven spots or inhomo-geneous cloud coverage. Recent substellar modelshave shown that turbulent flows and waves, includ-ing internal gravity waves, may play an importantrole in dust cloud evolution. The aim of this paper isto investigate the effect of internal gravity waves ondust cloud nucleation and dust growth, and whetherobservations of the resulting cloud structures couldbe used to recover atmospheric density information.For a simplified atmosphere in two dimensions, wenumerically solve the governing fluid equations tosimulate the effect on dust nucleation and mantlegrowth as a result of the passage of an internal grav-ity wave. Furthermore, we derive an expressionthat relates the properties of the wave-induced cloudstructures to observable parameters in order to de-duce the atmosphere density. Numerical simula-tions show that the density, pressure and tempera-ture variations caused by gravity waves lead to an

up to 600-fold increase of the dust nucleation rateand an up to 80% increase of the dust growth rate inthe linear regime. These variations lead to bandedareas in which dust formation is much more pro-nounced. We show that internal gravity waves insubstellar atmospheres lead to banded cloud struc-tures similar to those observed on Earth. Using theproposed method, potential observations of bandedclouds could be used to estimate the atmosphericdensity of substellar objects.

329.11 — Atmospheric escape: new windows,longer baselines and demographic influences

John McCann2,1; Ruth Murray-Clay1; Mark Krumholz4;Kaitlin M. Kratter3

1 Astronomy and Astrophysics, UC Santa Cruz (Santa Cruz, Cali-fornia, United States)

2 Physics, UC Santa Barbara (Santa Barbara, California, UnitedStates)

3 Astronomy and Steward Observatory, Univ. of Arizona (Tucson,Arizona, United States)

4 Astronomy and Astrophysics, Australian National University(Canberra, Australian Capital Territory, Australia)

We present new quasi-global 3-D radiative-hydrodynamic simulations of close-in exoplan-ets undergoing atmospheric escape. By trackingthe ionization state of outflows driven by ioniza-tion heating, we produce self-consistent syntheticobservations. The resulting synthetic Lyman-α ob-servations find several distinct large-scale featuresof atmospheric escape, which we classify into threeregimes dependent upon the properties of the inter-planetary medium (e.g., stellar wind, ionizing flux,orbital separation). Several of these new featuresproduce substantial obscuration of the star manyhours outside of transit. We therefore demonstratethat long-baseline transit observations in Lyman-αand other non-optical lines are needed to constrainmass loss mechanisms. We compare to the severalobservations of known systems undergoing atmo-spheric escape, and discuss which aspects of thetheory are still missing. By incorporating recentdevelopments in transit observations of escapingexospheres, such as ground-based observations inH-α and Helium 10830, we expand our modelsand probe the core of hydrodynamic escape missedin Lyman-α. We conclude by discussing how ourresults inform our understanding of the evaporationvalley evident in super-Earth demographics.

134

329.12 — Planetary Interior Physics from Near In-frared Spectroscopy

Jonathan Fortney11 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)

The atmosphere can often be a window into pro-cesses happening within a planetary interior. In-deed, in some cases for rocky planets, and in mostcases for giant planets, the properties of the radiativeatmosphere entirely regulate the cooling of deeperplanetary layers, including any surface or deep in-terior. Often this atmosphere/interior interface ishidden from view. However, at near-infrared wave-lengths, where atmospheric opacity sources (such aswater vapor) are often at their minima, one can seerelatively deeply into a planet’s atmosphere. In someparticular cases, this gives us access to the physics ofthe deeper atmosphere or interior. The first example:In hot Jupiters, the mechanism that causes inflatedradii, a long-standing problem, is still open. Throughthe use of 1D models we show that in the NIR onecan sometimes observe flux that emerges from be-low the radiative-convective boundary, which is a di-rect probe of the interior entropy of these planets,which will place new constraints on the radius infla-tion mechanism. The second example: Planets on ec-centric orbits have this eccentricity damped, in whichorbital energy is converted to thermal energy in theplanetary interior. This thermal flux can heat inte-riors well in excess of that expected for simple cool-ing models, by orders of magnitude, leading to de-tectable fluxes from the deep interior. Inferring thesefluxes from spectra can lead to a direct constraint ona planet’s tidal quality factor, Q, which is typicallynot a measurable quantity. We show an examplefor well-studied transiting Neptune-class planet GJ436b, and generalize this work for other planets, gi-ant and rocky, on eccentric orbits.

329.13 — Investigating the Mineralogy of Clouds inSubstellar Atmospheres

Jessica L. Luna1; Caroline Morley11 Astronomy, University of Texas at Austin (Austin, Texas, United

States)

Brown dwarfs, substellar objects not massive enoughto fuse hydrogen, have cool atmospheres with thetemperatures of giant planets. Their atmospheresare cool enough to form clouds and their tempera-ture determines which species condense. Thick lay-ers of dust, likely made of silicates, blanket L dwarfatmospheres, limiting the depths probed by spectra;

these clouds clear dramatically at the L/T transition.Cloud chemistry and microphysics is challengingto model from first principles; clouds clearly form,but the specific species that condense are not well-constrained from theory. This uncertainty is a ma-jor barrier to understanding exoplanet atmospheres.The next key step in understanding these clouds is toempirically determine which clouds form using mid-infrared spectroscopy to identify mineral species.Currently there is tentative evidence from Spitzerthat silicate features are present in L dwarf spectra.JWST could allow us to measure these features inmany L dwarfs. Before these observations are made,we need to understand in detail at what wavelengthsthe strongest cloud absorption features will be, andpredict which objects will have the largest amplitudesignals. We present our results exploring the impactof individual cloud species, including how particlesizes and cloud mineralogy change spectral features.We investigate which objects are most ideal to ob-serve, exploring a range of temperatures and surfacegravities. We find that silicate, corundum and per-ovskite clouds have a strong cloud absorption fea-ture for small particle sizes (<1 um). Silicate cloudsstrongly absorb at 10 um while corundum and per-ovskite absorb at 11.5 um and 14 um respectively. Wesimulate time-series observations with the MIRI in-strument on JWST for a range of nearby, cloudy, andphotometrically variable brown dwarfs. Our predic-tions suggest that with JWST, by measuring spectro-scopic variability inside and outside a mineral fea-ture (eg. the silicate feature), we can uniquely iden-tify a range of clouds species. Mid-infrared time-series spectroscopy can therefore be used to empir-ically constrain the complex cloud condensation se-quence in substellar atmospheres

329.15 — Help or Hinder? Assessing the Role ofClouds on the Spectra of Exoplanet Atmospheres

Zafar Rustamkulov1,2; Jonathan Fortney11 Astronomy and Astrophysics, University of California, Santa

Cruz (Oceanside, California, United States)2 Earth and Planetary Sciences, Johns Hopkins University (Balti-

more, Maryland, United States)

As we move toward a more detailed reconnaissanceof exoplanet atmospheres, the role of clouds will be-come more apparent. Self-consistent treatments ofclouds in the reflection, emission, and transmissiongeometries will be necessary to interpret multiwave-length observations for a more complete understand-ing of a breadth of planetary atmospheres. Here wepresent analytic toy models for understanding howclouds can alter the information within a planetary

135

spectrum. This “information content” is character-ized by the column density of a given absorber in re-flection and emission spectra, and the transit depthamplitudes for features in transmission spectra. Weproduce parameterizations for the information con-tent of a range of planet types in the three main ob-serving modes as a function of bulk cloud properties.The effects of cloud optical depth, height, vertical ex-tent, patchiness, coverage, and albedo, are importantin shaping the information content and are often in-herently degenerate. We show how clouds can in-crease the atmospheric signal in reflection spectra,decrease the signal in transmission spectra, and pro-duce a wider range of behavior in emission spectra.This work provides an approachable introduction forunderstanding the results of more sophisticated nu-merical models, and highlights how cloud geometrycan add nuance to exoplanet atmosphere characteri-zation.

329.16 — Extreme Storms on Synchronized Exo-planets

James Cho1; Jack William Skinner2; Heidar Thrastarson31 CCA, Flatiron Institute (New York, New York, United States)2 Astronomy Unit, Queen Mary University of London (London,

United Kingdom)3 JPL, Callifornia Institute of Technology (Pasadena, California,

United States)

We propose to present the latest results from anextensive set of numerically converged and vali-dated simulations of global atmospheric dynamicson tidally-synchronized exoplanets. Significantly,the simulations are the highest resolution, best ‘con-serving’ (e.g., angular momentum, energy, and en-strophy) simulations to date. These characteris-tics permit highly energetic, long-lived storms thatchaotically stir and redistribute hot and cold tem-peratures on a wide range of scales, including theplanetary scale (roughly planetary radius). Here,the emergence, morphology, and life-cycle of thestorms are carefully studied in a hot-Jupiter context,as a generic example, with a major focus on howthe storms affect the three-dimensional temperaturestructure and observable temporal variability. Forexample, tropical modons (cyclone-pairs) transportheat (cold) from day (night) side to night (day) sideand produce discernable quasi-periodic signaturesin the time series of disk-integrated temperature fluxover many thousands of orbits. Modeling and char-acterizing such behaviors with fidelity are criticalfor understanding the nature of extreme planetaryatmospheres, in which the large-scale flow speeds(characterized by the Mach number) are high, as well

as for interpreting and guiding current and futureobservations.

329.17 — Salt Clouds, Metal Rain and Rock Stormsin the Deep Atmosphere of Jupiter and Implica-tions on Exoplanets and Brown Dwarfs

Xi Zhang1; Huazhi Ge1; Diana Powell1; Cheng Li2; Pe-ter Gao2

1 University of California Santa Cruz (Santa Cruz, California,United States)

2 University of California Berkeley (Berkeley, California, UnitedStates)

Salts, silicates and metals, such as KCl, ZnS, TiO2,Mg2SiO4, MnS, Fe, and Al2O3, condense as clouds inthe deep atmosphere of Jupiter and greatly affect itschemical and dynamical structure. This new studyhas three motivations: (1) convective (rock) stormsmight trigger the observed internal mode oscillationsof Jupiter (e.g., Gaulme et al. 2011; Markham andStevenson 2018). (2) Salt grains could chemicallymodify the local plasma environment and producepossible signals for the JUNO microwave observa-tions (Jansen et al. 2017). (3) Those high-temperaturecondensates are direct analogs of the clouds detectedon hot Jupiters (e.g., Sing et al. 2016), directly im-aged exoplanets (e.g., Zhou et al. 2016) and browndwarfs (e.g., Apai et al. 2013), where the clouds formhigh in their photospheres. We first use a 1D mi-crophysics model, CARMA (Gao et al. 2018; Pow-ell et al. 2018), to simulate multiple cloud layers ofsalts, silicates and metals above 104 bar. TiO2 andKCl self-nucleates to form condensation nuclei forthe other vapors to condense. Iron droplets can growup to sub-milimeter size and rain down and evap-orate at ∼104 bar. Mg2SiO4 is the most abundantcloud species, extending from 103 bar to ∼1 bar. Themicron-sized silicate and salt grains might serve asthe seeds for the water cloud formation. We thenuse a dynamical model, SNAP (Li and Chen 2018),to simulate the silicate thunderstorm. Latent heat re-lease from the silicate condensation and deep con-vective flux drive the cloud-level moist convection.Jupiter’s silicate “hydrological cycle” is found to beseveral orders of magnitude larger than the water hy-drological cycle on Earth. We consider the convec-tive storm as a heat engine and use the CAPE (con-vective available potential energy) to estimate themechanical work done by the storm. The total avail-able energy of a single rock storm can reach ∼1026

erg but it depends on the area fraction of the con-vection. Our ongoing 3D convection simulations willshed more light on the power from the storm to ex-cite possible seismic modes on Jupiter.

136

329.18 — Assessment of Machine Learning forAnalysing Exoplanet Atmospheres

Matthew Conor Nixon1; Nikku Madhusudhan11 Institute of Astronomy, University of Cambridge (Cambridge,

Cambridgeshire, United Kingdom)

Recently several attempts have been made to use ma-chine learning algorithms either to complement orreplace existing techniques used to study the atmo-spheres of exoplanets. However, these attempts haveoften led to predictions that do not capture the in-herent uncertainties and degeneracies that we expectto be present in both observed data and in forwardmodels. In this project we independently repro-duce previous results and discuss the differences be-tween results of machine learning retrieval and tradi-tional methods. We subsequently exploer a new su-pervised machine learning approach and apply thismethod both to a previously examined case, as wellas to the atmosphere of another hot giant planet, HD209458b. In the latter case we use a fully numericalforward model. Our method leads to a closer matchto traditional retrievals than other machine learning-based approaches. Finally we discuss the benefitsand drawbacks of incorporating machine learninginto the analysis of exoplanetary atmospheres.

329.19 — The Accuracy of Mass Measurements Re-quired for Robust Atmospheric Characterization

Natasha Batalha1; Taylor Lewis2; Jonathan Fortney2;Natalie Batalha2; Eliza Kempton3

1 University of California Santa Cruz (Santa Cruz, California,United States)

2 Astronomy and Astrophysics, University of California, SantaCruz (Santa Cruz, California, United States)

3 University of Maryland (College Park, Maryland, United States)

TESS discoveries have already become fruitful tar-gets for HST follow up, and this will undoubtedlycontinue through the JWST era. In addition toproviding targets for atmospheric characterization,TESS’ Level One Science Requirement is to measuremasses for 50 transiting planets smaller than 4 Earthradii. A full suite of ground based facilities will beworking together to optimize the TESS science yield.Somewhat surprisingly though, no study has quan-tified the accuracy of mass constraints required toyield robust atmospheric properties of small plan-ets. Previous work showed that the mass of a transit-ing exoplanet could be inferred from its transmissionspectrum alone. The method leverages the effect ofthe planet’s surface gravity on the atmospheric scaleheight, which in turn influences the transmission

spectrum. However, significant degeneracies existbetween transmission spectra of planets with differ-ent masses and compositions, making difficult to un-ambiguously determine the planet’s mass and com-position in many cases. I will present the first quan-titative answer to this pressing question. Our anal-ysis places definitive limits on how accurate massconstraints need to be in order to unambiguously de-termine atmospheric composition for a diverse arrayof planets ranging from terrestrial-size (TRAPPIST-1-like) to mini-Neptunes and hot Jupiters. Theseresults broadly impact the community of scientistsworking on exoplanets — from the full breadth ofground based observers conducting TESS follow-up,to those studying planet populations, and finally tothose planning atmospheric investigations. This isparticularly timely as the STScI Director charged theHST-TESS Advisory committee to report to the SpaceTelescope Users Committee on how HST can bestsupport follow-up observations of TESS exoplanetdiscoveries. The community needs to determine op-timal strategies for maximizing the rapid scientificreturn from TESS targets.

330 — Future Missions and Instru-mentation, Poster Session330.01 — GLINT, a pathfinder instrument for ex-oplanets characterization through nulling interfer-ometry

Tiphaine Lagadec1; Peter Tuthill1; Barnaby Norris1; Si-mon Gross2; Alex Arriola2; Thomas Gretzinger2; Nickcvetojevic3; Jon Lawrence4; Michael Withford2; Marc-Antoince Martinod1

1 Physics, University of Sydney (Camperdown, New South Wales,Australia)

2 Macquarie University (Sydney, New South Wales, Australia)3 Observatoire de la Cote d’Azur (Nice, France)4 Australian Astronomical Observatory (Sydney, New South Wales,

Australia)

High angular resolution (of order milliarcseconds) athigh contrast (fractional flux of order 10−6 or fainter)is required to directly detect exoplanets light, re-vealing their intrinsic atmosphere and surface prop-erties. Here we present the GLINT (Guided LightInterferometric Nulling Technologies) instrument, apathfinder developed at the University of Sydney inAustralia to tackle this task. It employs nulling in-terferometry to actively reject the light of a host starthrough destructive interference. Such advancedcontrol and processing of starlight is accomplished

137

by way of photonic technologies fabricated into inte-grated optical chips. This platform immediately of-fers versatile and complex optical layouts all minia-turized onto highly stable and low-loss components.The first GLINT prototype was demonstrated on skyat the 4m Anglo Australian Telescope. Our currenttestbed on the 8m Subaru telescope boasts signifi-cantly expanded capabilities including multiple in-put array elements feeding a spectrally-dispersed in-terferometer delivering multi-channel nulling andcomplex visibility data.

330.02 — An Updated Study of Potential Targets forAriel

Billy Edwards1; Lorenzo Mugnai2; Giovanna Tinetti1;Enzo Pascale2,3; Subhajit Sarkar3

1 University College London (London, United Kingdom)2 La Sapienza Universita di Roma (Rome, Italy)3 Cardiff University (Cardiff, United Kingdom)

Thousands of exoplanets have now been discoveredwith a huge range of masses, sizes and orbits. How-ever, the essential nature of these exoplanets remainslargely mysterious: there is no known, discerniblepattern linking the presence, size, or orbital parame-ters of a planet to the nature of its parent star. Wehave little idea whether the chemistry of a planetis linked to its formation environment, or whetherthe type of host star drives the processes controllingthe planet’s birth and evolution. Progress with thesescience questions demands a large, unbiased spec-troscopic survey of exoplanets and Ariel has beenselected as ESA’s M4 mission for launch in 2028.By studying a large and diverse population of exo-planetary atmospheres, Ariel will provide insightsinto planetary formation and evolution within ourgalaxy. I will present the latest study of poten-tial targets for Ariel in which we assessed the suit-ability of currently-known exoplanets and predictedTESS yields. This list of planets has been utilised toform an example Mission Reference Sample (MRS) todemonstrate that Ariel’s mission goals could be metfrom this planetary population. I will also presentthe results from the latest studies into the expectedscientific capability of Ariel.

330.03 — Measuring the Stellar Drivers of Exo-planet Habitability with the ESCAPE Mission (-or-Why are those solar systems so extreme???)

Kevin France11 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)

The long-term stability of exoplanetary atmospheresdepends critically on the extreme-ultraviolet (EUV)flux from the host star. The EUV flux likely drivesthe demographics of the short-period planet popu-lation as well the ability for rocky planets to main-tain habitable environments long enough for theemergence of life. In this talk, I will present theExtreme-ultraviolet Stellar Characterization for At-mospheric Physics and Evolution (ESCAPE) mission,an astrophysics Small Explorer proposed to NASA.ESCAPE employs extreme- and far-ultraviolet spec-troscopy (70 - 1800 Angstroms) to characterize thehigh-energy radiation environment in the habitablezones (HZs) around nearby stars. ESCAPE providesthe first comprehensive study of the stellar EUV en-vironments that control atmospheric mass-loss anddetermine the habitability of rocky exoplanets. ES-CAPE’s prime mission is driven by two spectroscopicsurveys: 1) a broad survey of EUV and FUV flux from200 nearby (d < 100 pc) A, F, G, K, and M stars, pro-viding direct input into atmospheric evolution mod-els. The mission targets stars with a range of agesand activity levels, and places an emphasis on starswith known exoplanets. 2) A deep monitoring sur-vey (∼2 weeks per star) of 24 targets-of-interest tomeasure the stellar flare frequency distribution andconstrain the CME rate and high-energy particle flu-ence from these objects. Together, these surveys pro-vide the crucial stellar drivers that regulate habitableenvironments on planets targeted by upcoming at-mospheric characterization missions, from JWST toLUVOIR.

330.04 — A Fabry Perot Instrument for OxygenSearches in Exoplanet Atmospheres

Surangkhana Rukdee1; Sagi Ben-Ami1; Mercedes Lopez-Morales1; Juliana Garcia-Mejia1; David Charbonneau1;Andrew Szentgyorgyi1

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

The large light-gathering capacity of upcoming Gi-ant Segmented Mirror Telescopes are expected tohave enough collecting area to start searching forpotential biosignature gases, notably molecular oxy-gen O2, in the atmospheres of small planets aroundnearby stars. Recent simulations have shown thatspectral resolutions of 300,000 - 400,000 are optimalto detect O2 in the atmosphere of an earth analog.In order to increase detection capabilities, we havedeveloped a Fabry perot Interferometer for OxygenSearches (FIOS), an interferometer array coupled toa high resolution spectrograph. FIOS can achieve aspectral resolution of 500,000 at the O2 A-band (760

138

nm). It is optimal for O2 detection, while maintain-ing a higher throughput compared to instrumentswith similar spectral resolving power. We describethe instrument concept, a simulation of its sensitiv-ity, and preliminary results from our lab prototype.

330.05 — The Tierras Observatory: An Ultra-Precise Photometer to Characterize Nearby Terres-trial Exoplanets

Juliana Garcia-Mejia1; David Charbonneau1; DanFabricant1; Jonathan Irwin1

1 Astronomy, CfA | Harvard & Smithsonian (Cambridge, Mas-sachusetts, United States)

We are currently building the Tierras Observatory, a1.3m ultra-precise fully-automated photometric ob-servatory located atop Mt. Hopkins, Arizona dedi-cated to following up nearby transiting planets dis-covered by TESS and other surveys, refine esti-mates of their radii, and find the longer period (andhence more temperate) worlds. Tierras will regularlyachieve a photometric precision of 700 ppm, enoughto measure the transit of Earth-sized planets orbiting0.1− 0.3R⊙ stars with 3σ significance. I will providean overview of the current state of the observatory, aswell as the design choices that will enable its sciencegoals. These include: (i) a custom designed four-lensfocal reducer and field-flattener to increase the field-of-view (FOV) of the telescope from a 12’’ to a 0.5° di-agonal; (ii) a 4K × 4K pixel deep-depletion low readnoise CCD operated in fast frame transfer (shutter-less) mode with 80% quantum efficiency at the wave-length of observation (compared to 35% for regu-lar CCDs); (iii) a custom narrow (50 nm) bandpassfilter centered at 865 nm to minimize precipitablewater vapor errors, known to limit ground-basedphotometry of red dwarfs; and (iv) a custom-madenano-fabricated beam-shaping diffuser to mold thefocal plane image star into a broad and stable top-hat shape, increasing the dynamic range of our ob-servations while minimizing flat-fielding, guiding,and phase-induced errors due to seeing. Tierras willbe on-sky by January of 2020, in time to carry outplenty of follow-up observations of TESS targets inthe northern hemisphere. This work is supported bythe National Science Foundation, the Ford Founda-tion, the John Templeton Foundation, and the Har-vard Origins of Life Initiative.

330.06 — Correcting instrumental systematics inSpitzer transit light curves using probabilisticLong Short Term Memory networks

Mario Morvan1; Ingo Waldmann1; Angelos Tsiaras1

1 Department of Physics & Astronomy, UCL (London, UnitedKingdom)

The instrument systematic errors associated with thevarious space-based transit detectors remain onlypartly understood. Yet, they condition the wholesubsequent analysis of transit light curves, from sig-nal detection up to the planet characterization. Inparticular, the retrieval of atmospheres from Spitzer’sIRAC or Hubble’s WFC3 data require precise mea-surements of the transit depths at several wave-lengths, and it is widely believed that a better cor-rection algorithm might allow to improve the qual-ity, robustness and complexity of the retrieval. HereI present a machine learning approach to correctingthe instrument systematics and photon noise, aimingat disentangling the instrument signal from the tran-sit signal. The algorithm makes use of the individ-ual pixel light curves as well as the centroid positiontime series, and try to learn the instrument behaviourfrom the out-of transit parts while assuming an ana-lytical form for the transit shape. While the first re-sults are shown on Spitzer transit light curves, suchan approach is widely adaptable to most of presentor future space-based detectors.

330.07 — Correcting Transiting Exoplanet LightCurves for Stellar Spots: A Machine LearningChallenge for the ESA Ariel Space Mission

Nikolaos Nikolaou1; Ingo Waldmann1; Subhajit Sarkar1;Angelos Tsiaras1; Billy Edwards1; Mario Morvan1; KaiHou Yip1; Giovanna Tinetti1

1 Department of Physics and Astronomy, University College Lon-don (London, United Kingdom)

The field of exoplanet discovery and characterisationhas been growing rapidly in the last decade. How-ever, several big challenges remain, many of whichcould be addressed using machine learning method-ology. For instance, the most successful method fordetecting exoplanets, transit photometry, is very sen-sitive to the presence of stellar spots. The currentapproach is to identify the effects of spots visuallyand correct for them manually or discard the data.As a first step to automate this process, we are or-ganising a competition for the 2019 European Con-ference of Machine Learning (ECML) on data gen-erated by ArielSim, the simulator of the EuropeanSpace Agency’s upcoming Ariel mission, whose ob-jective is to characterise the atmosphere of 1000 exo-planets. The data consists of pairs of light curves cor-rupted by stellar spots and the corresponding cleanones, along with auxiliary observation information.The goal is to correct light curves for the presence

139

of stellar spots (multiple signal denoising). This is ayet unsolved problem in the community. In this talkwe will discuss the problem, the impact of a solution,introduce the basics of machine learning and presentthe outline of the competition as well as initial base-line solutions.

330.08 — Exoplanet direct detection and characteri-zation with the ELT/HARMONI integral field spec-trograph

Mathis Houllé1; Arthur Vigan1; Alexis Carlotti2; ElodieChoquet1; Mickaël Bonnefoy2; Niranjan Thatte3; BenoitNeichel1

1 Laboratoire d’Astrophysique de Marseille (LAM) (Marseille,France)

2 Institut de Planétologie et d’Astrophysique de Grenoble (IPAG)(Grenoble, France)

3 Department of Physics, University of Oxford (Oxford, UnitedKingdom)

High-contrast imaging is a unique method to probethe outer regions of young exoplanetary systems,and thus gives insight into the formation, com-position and evolution of young giant exoplan-ets. Current high-contrast instruments, such asVLT/SPHERE, Gemini/GPI or Subaru/SCExAO,provide spectro-imaging capabilities at low spec-tral resolution (R = 30-100) that allow the detectionand first-order characterization (temperature, sur-face gravity) of giant planets, but significantly higherresolutions would be required for more detailed es-timations (abundances, orbital and rotational veloc-ities). The next generation of instruments on ex-tremely large telescopes (ELTs) will be ideal for thistask. HARMONI will be one of the first-light in-struments mounted on ESO’s ELT, currently plannedfor 2025. It is a medium-resolution (up to R =17,000) integral field spectrograph in the optical andthe near-infrared, which will be equipped with asingle-conjugated adaptive optics system to reachthe diffraction limit of the ELT in the H- and K-bands.For direct exoplanet detection and characterization,HARMONI will include a high-contrast module thatwill provide unprecedented contrast limits at sepa-rations between 50 and 400 mas, i.e. down to 1 AUfor a star at 20 pc. Exoplanet detection will be fur-ther facilitated by the medium spectral resolution ofHARMONI, which allows detection based on the ex-pected planetary spectral features. In this poster, wewill present an estimation of HARMONI capabili-ties for exoplanet detection, based on realistic simu-lated data for the high-contrast mode and new anal-ysis tools exploiting the spectral information, such

as “molecule mapping” techniques (e.g. Hoeijmak-ers et al. 2018). We will also show preliminary esti-mations of HARMONI performances for orbital andatmospheric characterization of exoplanets.

330.09 — Following up TESS’s temperate terrestri-als with MAROON-X

Jacob Bean11 University of Chicago (Chicago, Illinois, United States)

MAROON-X is a new high precision radial veloc-ity instrument that is currently being commissionedon the Gemini North telescope. MAROON-X isoptimized in terms of its wavelength coverage, ef-ficiency, stability, and pairing with a large tele-scope for following up TESS’s habitable zone planetcandidates. I will give an update on the statusof MAROON-X, including commissioning progress,performance of the instrument, and how the com-munity can use it for their own science. I will alsodiscuss my group’s plans for using the instrumentfor both radial velocity and atmospheric characteri-zation work.

330.10 — The Search for Biosignatures and Exo-Earths with the LUVOIR Mission Concept

Giada Nicole Arney11 Planetary Systems Laboratory, NASA Goddard Space Flight Cen-

ter (Silver Spring, Maryland, United States)

The Large Ultra Violet-Optical-Infrared (LUVOIR)Surveyor is one of four mission concepts being stud-ied by NASA in preparation for the 2020 Astro-physics Decadal Survey. LUVOIR is a general-purpose space-based observatory with a large aper-ture (8-15 m) and a total bandpass spanning fromthe far-UV to the near-infrared. One of LUVOIR’smain science objectives is to directly image temperateEarth-sized planets in the habitable zones of Sun-likestars, measure their spectra, analyze the chemistryof their atmospheres, and obtain information abouttheir surfaces. LUVOIR can also observing poten-tially habitable exoplanets transiting nearby M dwarfstars. Such observations will allow us to evaluatethese worlds’ habitability and search for the pres-ence of remotely detectable signs of life known as“biosignatures.” We will discuss the strategies forExo-Earth detection and characterization, includingspecific observational requirements for astrobiologi-cal assessments of exoplanetary environments withLUVOIR. The survey of the atmospheric composi-tion of dozens of potentially habitable worlds wouldbring about a revolution in our understanding of

140

planetary formation and evolution, and may usherin a new era of comparative astrobiology.

330.11 — Lessons Learned for JWST TransitingPlanet Time Series from Ground-based Studies ofthe NIRCam Detectors and Control Electronics

Everett Schlawin2; Karl Misselt2; Tom Greene1; JarronLeisenring2; Thomas Beatty2; Marcia Rieke2

1 NASA Ames (Mountain View, California, United States)2 University of Arizona (Tucson, Arizona, United States)

JWST transmission and emission spectra of transitingexoplanets will provide invaluable glimpses at exo-planet atmospheres. These spectra will reveal atmo-spheric compositions and temperature structures ata level never achieved before. This promising sciencefrom JWST, however, will require exquisite precisionand understanding of systematic errors that can im-pact the time series of planets crossing in front of andbehind their host stars. This is especially true if JWSTis used to probe the atmospheres of temperate Earth-sized planets. Here, we discuss the lessons learnedfrom ground-based characterization of the NIRCamH2RG detectors and ASIC control electronics, whichwill be used in the slitless grism time series mode forexoplanet spectra. These detectors are the same typeused in JWST’s NIRISS and NIRSpec spectrographsas well as ground-based near-infrared instruments.Ground-based tests provide new understandings ofthe crosshatching on the detectors that trace to thecrystallographic structure of the HgCdTe. We showthat the crosshatching behavior extends to the sub-pixel level and can be modeled in the Fourier do-main. We provide models that estimate the impactof subpixel crosshatching when subject to pointingjitter of the telescope. Ground-based tests also re-veal that correlations in the read noise (1/f noise)can strongly affect time series apertures. If uncor-rected, these correlations can rival photon noise onshort integrations. 1/f noise correction algorithmsare of special importance to targets where only a fewsamples may be read per integration and exquisiteprecision (sub 50 ppm) is desired. We discuss severalstrategies for removing correlated read noise. Wesummarize the lessons learned from ground-basedtesting of flight detectors and electronics at NASAGoddard, NASA Johnson as well as flight-like elec-tronics and detectors at the University of Arizona de-tector labs.

330.12 — Design and Performance of NEID Ultra-Stable Environmental Control System

Emily Lubar2; Paul Robertson1; Frederick Hearty2; Gud-mundur Kari Stefansson2; Andrew Monson2; SuvrathMahadevan2; Jason Wright2; Shubham Kanodia2; ChadBender3; Joe Ninan2; Arpita Roy4; Sam Halverson4

1 University of California, Irvine (Irvine, California, United States)2 The Pennsylvania State University (University Park, Pennsylva-

nia, United States)3 University of Arizona (Tucson, Arizona, United States)4 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)

NEID is a NASA & US NSF funded ultra-stable, op-tical spectrometer designed to achieve Radial Veloc-ity (RV) precision on the order of 10cm/s. Achievingthis level of measurement precision requires extremethermo-mechanical stability within the instrumentwhich we achieve by maintaining a vacuum on theorder of microTorr as well as sub-milliKelvin temper-ature stability. In this poster, we will outline NEID’sEnvironmental Control System (ECS) and Tempera-ture Monitoring and Control (TMC) System, whichwere both inherited and improved upon from thatof the Habitable-zone Planet Finder (HPF) infraredspectrograph. We have achieved our target stabilityby demonstrating < 0.4mK RMS temperature vari-ability over the course of a 30 day stability run inthe lab. We expect our stability to improve at theobservatory as the WIYN instrument room is morestable than our instrument development lab at PennState. NEID will be commissioned in Fall 2019 at KittPeak National Observatory on the 3.5m WIYN Tele-scope. It will serve the exoplanet community as avital resource for the detection and confirmation oflow-mass exoplanets.

330.13 — Creating realistic synthetic observationsof transiting exoplanets with JWST introducing in-strumental systematics

Marine Martin-Lagarde1; Pierre-Olivier Lagage1; RenéGastaud1; Alain Coulais2,1; Christophe Cossou3; DanielDicken1

1 AIM, CEA, CNRS, Université Paris-Saclay, Université de Paris(F-91191 Gif-sur-Yvette, France)

2 LERMA, Observatoire de Paris, CNRS (F-75014, Paris, France)3 Institut d’Astrophysique Spatiale, CNRS/Université Paris-Sud,

Université Paris-Saclay, bâtiment 121, Université Paris-Sud (91405Orsay Cedex, France)

The launch and commissioning of the James WebbSpace Telescope (JWST) in 2021 will provide game-changing astronomical observations, in particular

141

for exoplanets. The Mid-InfraRed Instrument (MIRI)with its Low-Resolution Spectrometer (LRS) willcarry out transit spectroscopy of the exoplanet at-mospheres with unprecedented precision. The EarlyRelease Science (ERS) program of JWST includes acomplete light-curve spectroscopy of WASP43-b hot-Jupiter. In order to prepare the analysis of the obser-vations, several pieces of software have been devel-oped to create synthetic data of such observations.The instrument consortium has created MIRISim, asimulator reproducing as accurately as possible theinstrument behaviour, including its noise and sys-tematics, for imaging and spectroscopy modes ofMIRI. The consortium made it publicly available.Complementary to MIRISim, we have developed Ex-oNoodle, a Python tool to generate time-series spec-tra of star-planet systems, varying over time as theplanet orbits around the star. It aims at providingMIRISim with Time-Series Observations (TSO) inputfiles. MIRISim does not include a TSO mode. Chang-ing the source spectrum over time means to restarta new computation, therefore getting rid of the sys-tematics evolution. Here, we present how we ap-plied post-processing to MIRISim data to generaterealistic MIRI-LRS observation of the WASP43 sys-tem, including the instrument systematics. Thesesimulated data are used to test and improve the datareduction and retrieval techniques the communityis building. They will be used in the framework ofMIRI exoplanet data challenge to be conducted toprepare the exoplanet ERS program.

330.14 — Connection between free-floating and ice-giant exoplanets

Radoslaw Poleski11 Department of Astronomy, Ohio State University (Columbus,

Ohio, United States)

Gravitational microlensing technique has alreadyprovided first-order estimate of the occurrence rateof free-floating planets and detections of a few ice-giant planets. These objects are detected as, re-spectively, very short single-lens single-source mi-crolensing events and short anomalies that are well-separated from otherwise single-lens events. How-ever, some very short events that look like single-lens can be caused by bound planets which host starswere not closely approached by the sources. To con-strain the existence of host we can use three lines ofevidence: microlensing signal by the host, anoma-lous shape of the planetary sub-event, or host lightdetection (using adaptive optics or Hubble SpaceTelescope imaging). In some cases, none of these willgive definite answer and this fact leads to possible

misidentification of free-floating and bound planets.The chances of this misidentification are higher if thepopulation of the wide-orbit planets is large. Thus,full inference of occurrence rate of free-floating plan-ets has to take into account the false-positive boundplanets. In a few years, large samples of both free-floating and ice-giant exoplanets will be found by theNASA flagship WFIRST mission and these sampleshave to be analyzed jointly. Full understanding ofthe origin of free-floating planets requires measur-ing their masses, but this can be efficiently done onlyif Euclid satellite conducts microlensing survey con-currently with WFIRST. In contrast, for the boundplanets the mass can be measured in two additionalways. First, the microlensing parallax detection is en-abled by timing of the planet and host events. Sec-ond, the host light detection constrains the mass-distance relation. Either of these can be combinedwith source size measurement to give direct mea-surement of the lens mass and distance.

330.15 — ‘Alopeke, Zorro, and NESSI: Three dual-channel speckle imaging instruments at Gemini-North, Gemini-South, and the WIYN telescopes.

Nicholas Jon Scott11 SSA, NASA ARC/BAERI (Burlingame, California, United States)

Zorro is the newest in a line of three speckle imagersbuilt at NASA’s Ames Research Center for commu-nity use at the WIYN and Gemini telescopes. Thethree instruments are functionally similar and in-clude the capability for wide-field imaging in ad-ditional to speckle interferometry. The diffraction-limited imaging available through speckle effectivelyeliminates distortions due to the presence of Earth’satmosphere by ‘freezing out’ changes in the atmo-sphere by taking extremely short exposures andcombining the resultant speckles in Fourier space.This technique enables angular resolutions equal tothe theoretical best possible for a given telescope,effectively giving space-based resolution from theground. Our instruments provide the highest spa-tial resolution available today on any single aperturetelescope. The instruments have been installed andcommissioned and are in regular use at these obser-vatories. In addition to diffraction-limited speckleimaging, they are equipped with standard Sloan fil-ters and can perform extremely high cadence pho-tometry in either a narrow speckle field-of-view ofless than 10 arcseconds or a one-arcminute widefield-of-view.

A primary role of these instruments is exoplanetvalidation for the Kepler, K2, TESS, and many RVprograms. Contrast ratios of 6 or more magnitudes

142

are easily obtained in speckle mode. The instru-ments use two emCCD cameras providing simulta-neous dual-color observations help to characterizedetected companions. High resolution imaging en-ables the identification of blended binaries that con-taminate many exoplanet detections, leading to in-correctly measured radii. In this way small, rockysystems, such as Kepler-186b and the TRAPPIST-1planet family, may be validated and thus the de-tected planets radii are correctly measured.

330.16 — Small Satellites for Extreme Systems:Upcoming Space Missions and Concepts for Exo-planet Exploration

Brian Fleming1; Kevin France11 Laboratory for Atmospheric and Space Physics, University of

Colorado (Boulder, Colorado, United States)

Extreme solar systems are often dynamic, orbit ac-tive stars, and require dedicated facilities for mon-itoring and characterization. The ultraviolet (UV;100 – 300 nm) bandpass offers sensitive probes of at-mospheric mass loss, protoplanetary disks, and theinterplay between exoplanet atmospheres and thestellar radiation field, yet resources remain elusive.The UV can only be accessed from large, shared-use,space observatories. Small satellites and cubesats of-fer a new paradigm for studying transient systems.We present three missions under development at theUniversity of Colorado that will demonstrate the po-tential of dedicated, low cost space instruments. TheColorado Ultraviolet Transit Experiment (CUTE) is anear-UV (250 - 330 nm) spectrograph in a 6U cube-sat set to launch in mid-2020 to study atmosphericmass loss in close-orbiting giant exoplanets. TheSPRITE cubesat is also a 6U spectrograph, but op-erates in the far-UV (100 - 200 nm) in a bandpasscurrently only accessible to Hubble and UVIT. TheSPRITE instrument is a pathfinder for future far-UVcube- and smallsats. Finally, we present a concept forsuch a smallsat, COMPASS; an ESPA-class missionsthat could deliver unprecedented sensitivity in thedeep FUV (100 - 115 nm) for a fraction of the cost ofan Explorer-class mission. COMPASS measures theFUV irradiance of host stars and proxies for impul-sive particle outbursts to assess the habitability of or-biting exoplanets. These missions represent part ofthe first step towards a new era of exoplanet surveysand characterization from space-based platforms.

330.17 — Exoplanet Host Star Characterization withQWSSI

Gerard van Belle1; Catherine Clark2,1; Elliott Horch3,1;David Trilling2

1 Lowell Observatory (Flagstaff, Arizona, United States)2 Northern Arizona University (Flagstaff, Arizona, United States)3 Southern Connecticut State University (New Haven, Connecticut,

United States)

QWSSI, the Quad-camera Wavefront-SensingSpeckle Imager, is a next-generation speckle imagerthat is being developed for Lowell Observatory’s4.3-meter Discovery Channel Telescopes. The prin-ciple behind QWSSI is to extend the capabilities ofthe speckle camera currently resident at Lowell, theDifferential Speckle Survey Instrument (DSSI), intwo ways. First, while DSSI currently observes intwo visible channels, QWSSI will simultaneouslyobserve in six narrow-band channels: four in thevisible (0.5-0.9um), and one each in J- and H-band(1.2 and 1.6um). Second, the visible light unusedfor speckle imaging is carefully preserved and feedsa wavefront sensor (WFS), which is also run simul-taneously with the speckle imaging. Simulationsby Löbb (2016) indicate WFS data will providesignificant gains in exploring stellar multiplicity,with marked improvements in primary-secondarycontrast ratios and inner working angle (Horch etal. 2018). QWSSI will also be mountable on one ofthe three 1-meter telescopes being installed on theNPOI Array for engineering tests and preliminaryscience observations. QWSSI will expand on thealready considerable exoplanetary work of thespeckle imagers DSSI, NESSI (@ WIYN), Alopeke(Gemini-N), and Zorro (Gemini-S), improving thediscovery space for existing targets, as well openingup new regions of that discovery space with its NIRchannels.

330.18 — Progress on Starlight Suppression Tech-nologies for NASA Direct Imaging Missions

Gary Blackwood1; Nicholas Siegler1; Brendan Crill1;Karl Stapelfeldt1; Eric Mamajek1; Kendra Short1; PhillipWillems1

1 NASA Jet Propulsion Laboratory, Caltech (Pasadena, California,United States)

Extensive interest in exoplanet research is motivatedin part by the search for evidence of life on a subsetof these worlds. This scientific search relies on mea-surements of star spectra, and upon the planet spec-tra, photometry, and mass. The ability to measurethese planet properties at the required sensitivity has

143

been the focus of sustained technology investmentsby NASA for the past decade, investments which arenow achieving milestones and sensitivity thresholdsthat for the first time enable mission concepts capableof a search for evidence of life on exoplanets. Mostnotably, broadband contrast levels for coronagraphsand starshades have recently been demonstrated tomeet better than 4×10−10 and 10−11, respectively —thresholds required for spectral characterization ofexo-earths in the habitable zones of sun-like stars.

NASA’s Exoplanet Exploration Program (ExEP)identifies technology gaps pertaining to possible ex-oplanet missions and works with the communityto identify and track technologies to prioritize forinvestment, and ultimately to close the gaps. TheDecadal Survey Testbed, a NASA ExEP facility fortesting next-generation coronagraphs, has recentlydemonstrated contrast performance at better than4×10−10. A sub-scale starshade validation demon-stration has met the 10% broadband visible contrastthreshold of better than 10−10 contrast. A technologyreadiness level 5 milestone was recently met with alaboratory demonstration of a novel formation flyingsensor that demonstrates the required lateral posi-tion sensitivity.

This poster describes the recent technology break-throughs and the enabling impact on mission con-cept studies being submitted to the 2020 AstroDecadal Survey for direct imaging of exoplanets andspectral characterization to search for evidence oflife.

330.19 — The WFIRST Coronagraphic Instrument’sRole in the Direct Imaging of Planetary Systems

John Debes1; Vanessa Bailey3; Jeremy Kasdin5; NikoleLewis2; Bruce Macintosh4; Bertrand Mennesson3; JasonRhodes3; Aki Roberge6; Margaret Turnbull7; MargaretA. Ferking3; Feng Zhao3

1 AURA for ESA, Space Telescope Science Institute (Baltimore,Maryland, United States)

2 Astronomy, Cornell (Ithaca, New York, United States)3 Jet Propulsion Laboratory (Pasadena, California, United States)4 Stanford University (Stanford, California, United States)5 Princeton University (Princeton, New Jersey, United States)6 Goddard Space Flight Center (Greenbelt, Maryland, United

States)7 SETI Institute (Mountain View, California, United States)

The Wide Field Infrared Survey Telescope (WFIRST)coronagraphic instrument (CGI) is predicted to becapable of high contrast imaging and spectroscopyof exoplanets and circumstellar dust in reflected vis-ible light. CGI will be a technology demonstra-tion of five main areas that aid future direct imag-

ing missions such as LUVOIR and HabEX: exquisitewavefront control through a pair of deformable mir-rors, suppression of an on-axis star’s diffraction pat-tern through occulting masks or shaped pupils, theuse of photon counting visible detectors and post-processing techniques at high contrast in space, andhigh contrast spectroscopy. We will present anoverview of CGI’s design and of its predicted sci-entific capabilities. The technology demonstrationphase of CGI operations will include the spectro-scopic characterization of at least one known giantplanet, as well as a photometric characterization ofseveral additional known giant exoplanets orbitingnearby stars. Additionally, CGI will be sensitive toa wide range of circumstellar disks around main se-quence stars, from protoplanetary disks to tenuousexozodiacal disks that may interfere with the detec-tion of Earth analogs. These observations will pro-vide constraints on the atmospheric and cloud chem-istry of giant planets and will characterize the phys-ical properties of circumstellar disks over several or-ders of magnitude in dust mass. CGI will retire manyof the risks currently associated with attempting todirectly image and characterize Earth-like planetsduring its technology demonstration phase. If theCGI performance during the technology demonstra-tion phase warrants it, additional science campaignsmay be undertaken by a competitively selected Par-ticipating Scientist Program team during later phasesof the WFIRST mission.

330.21 — The Next decade (or two) of X-ray Exo-planet Studies.

Scott Wolk11 High Energy, Harvard-Smithsonian Center for Astrophysics

(Cambridge, Massachusetts, United States)

High energy photons and particles create some of themost extreme environments for exoplanets. I high-light the contributions to exoplanet science made byXMM-Newton and Chandra X-ray observatory overthe last two decades. This research has includedmany individual exceptional system such as CoRoT-2, HD17156 and GJ436. Conversely, So many X-raybright exoplanet hosts have been identified that sta-tistical analysis, such the distribution of exoplanetmasses in the presence of X-ray irradiation, are pos-sible. As an example recent results that suggeststhat planet atmospheres have been eroded by theirhost star coronal emissions. I discussion of prospectsfor observations with future X-ray missions XRISM,SEEJ, Athena and Lynx.

144

330.22 — The NASA Exoplanet Exploration Pro-gram: Update and Prospects for the 2020’s and Be-yond

Eric E. Mamajek1; Gary Blackwood1; Kendra Short1;Karl Stapelfeldt1; Nicholas Siegler1; Brendan Crill1;Keith Warfield1; John L. Callas1; Phillip Willems1; AnyaBiferno1; Douglas Hudgins2; Martin Still2; ShahidHabib2; Chas Beichman3

1 NASA Jet Propulsion Laboratory, Caltech (Pasadena, California,United States)

2 Astrophysics Division, NASA Headquarters (Washington, Dis-trict of Columbia, United States)

3 NASA Exoplanet Science Institute, IPAC/Caltech (Pasadena,California, United States)

The NASA Exoplanet Exploration Program (ExEP) isresponsible for implementing NASA’s plans for dis-covering and characterizing exoplanets, and identi-fying candidates that could harbor life. ExEP man-ages concept studies, technology development pro-grams, precursor and follow-up ground-based sci-ence programs that aim towards achieving the sci-ence goals of current and future NASA missions (andthat enable the design of next generation exoplanetmissions), and communicates the excitement of exo-planet research to the public. We will review recentactivities in the NASA-NSF Exoplanet ObservationalResearch (NN-EXPLORE) partnership, progress inthe characterization of exozodiacal light, the statusof ongoing studies of future exoplanet flagship mis-sions, and recent technology milestones — includ-ing updates on the progress of starshade and coro-nagraph technology capable of imaging and charac-terizing nearby habitable exoplanets.

330.23 — The Oxyometer: A Novel Instrument Con-cept for Characterizing Exoplanet Atmospheres

Ashley Baker1; Cullen Blake1; Sam Halverson21 Physics & Astronomy, University of Pennsylvania (Philadelphia,

Pennsylvania, United States)2 MIT (Boston, Massachusetts, United States)

With TESS and ground-based surveys searching forrocky exoplanets around cooler, nearby stars, thenumber of Earth-sized exoplanets that are well-suited for atmospheric follow-up studies will in-crease significantly. For atmospheric characteriza-tion, the James Webb Space Telescope will only beable to target a small fraction of the most interest-ing systems, and the usefulness of ground-based ob-servatories will remain limited by a range of effectsrelated to Earth’s atmosphere. Here, we explorea new method for ground-based exoplanet atmo-spheric characterization that relies on simultaneous,

differential, ultra-narrow-band photometry. The in-strument uses a narrow-band interference filter andan optical design that enables simultaneous observ-ing over two 0.3 nm wide bands spaced 1 nm apart.We consider the capabilities of this instrument in thecase where one band is centered on an oxygen-freecontinuum region while the other band overlaps the760 nm oxygen band head in the transmission spec-trum of the exoplanet, which can be accessible fromEarth in systems with large negative line-of-sight ve-locities. We find that M9 and M4 dwarfs that meetthis radial velocity requirement will be the easiesttargets but must be nearby (<8 pc) and will requirethe largest upcoming Extremely Large Telescopes.The oxyometer instrument design is simple and ver-satile and could be adapted to enable the study ofa wide range of atmospheric species. We demon-strate this by building a prototype oxyometer andpresent its design and a detection of a 50 ppm simu-lated transit signal in the laboratory. We also presentdata from an on-sky test of a prototype oxyometer,demonstrating the ease of use of the compact instru-ment design.

331 — Habitability and Biosigna-tures, Poster Sessopm331.01 — Dying to Live: Post-Main Sequence Hab-itability

Thea Kozakis1; Lisa Kaltenegger11 Carl Sagan Institute, Cornell Univeristy (Ithaca, New York,

United States)

During the post-main sequence phase of stellar evo-lution the orbital distance of the habitable zone,which allows for liquid surface water on terres-trial planets, moves out past the system’s originalfrost line, providing an opportunity for outer plan-etary system surface habitability. We use a 1D cou-pled climate/photochemistry code to study the im-pact of the stellar environment on the planetary at-mospheres of Earth-like planets/moons throughoutits time in the post-main sequence habitable zone.We also explore the ground UV environments ofsuch planets/moons and compare them to Earth’s.We model the evolution of star-planet systems withhost stars ranging from 1.0 to 3.5 MSun through-out the post-main sequence, calculating stellar massloss and its effects on planetary orbital evolutionand atmospheric erosion. The maximum amountof time a rocky planet can spend continuously inthe evolving post-MS habitable zone ranges between56 and 257 Myr for our grid stars. Thus, during

145

the post-MS evolution of their host star, subsurfacelife on cold planets and moons could become re-motely detectable once the initially frozen surfacemelts. Frozen planets or moons, like Europa in ourSolar System, experience a relatively stable environ-ment on the horizontal branch of their host stars’ evo-lution for millions of years.

331.02 — Exploring Giant Planets & Exomoons inthe Habitable Zone.

Michelle Hill1; Stephen Kane11 Earth & Planetary Science, UC Riverside (West Hollywood, Cali-

fornia, United States)

While the search for exoplanets has been focused pri-marily on trying to find Earth like planets, there havebeen discoveries of many different worlds that havecaused us to revise our ideas as to what could bea potentially habitable planet. Interestingly a sig-nificant number of giant exoplanets (>3 earth radii)have been detected in the habitable zone of their star.These giant planets are likely gas giants and thus arenot considered habitable on their own, but they eachcould potentially be host to large rocky exomoonswhich would also exist in the habitable zone. Thesemoons, should they exist, will offer new ways to un-derstand the formation and evolution of planetarysystems, and widen the search for signs of life outin the universe. The occurrence rates of these moonsare related directly to the occurrence rates of giantplanets in the habitable zone of their star, thus weestimated the frequency with which we expect giantplanets to occur in the habitable zones. We then com-piled a proposed exoplanet target list to be used inthe search for detectable exomoons and for perform-ing more detailed follow-up studies. We identified121 giant planets whose orbits lie within either theoptimistic habitable zone (OHZ) or the conservativehabitable zone (CHZ). As well as the potential exis-tence of exomoons, these giant planets in the HZ oftheir star raise questions as to the formation and evo-lution of these systems. As giant planets are thoughtto form beyond the snow line these planets likely mi-grated inwards during their formation in order forthem to reside in the HZ today. One possible ex-planation as to why these planets stopped migratingonce they reached the HZ is that of orbital resonancewith other planets. Thus we fit each of the 121 HZ gi-ant planets radial velocity (RV) data to confirm theirorbital solution and look for linear trends to deter-mine if there were indications for additional plan-etary companions. Of the 121 giant planets tested,51 showed indications of additional companions (>3σ). Highlights of our calculations will be presented

along with up to date results from ongoing RV ob-servations of the most promising of these systems.

331.03 — Predicting the UV Emission of M dwarfswith Exoplanets from Ca II and H-α

Katherine Melbourne1; Allison Youngblood2; AkiRoberge2; Sarbani Basu1; Kevin France3; CynthiaFroning4; J. Sebastian Pineda3; Evgenya L. Shkolnik6;Travis Barman5; R. O. Parke Loyd6; Elizabeth Newton7;Isabella Pagano8; Sarah Peacock5; Joshua Schlieder2;Adam Schneider6; David John Wilson4

1 Yale University (New Haven, Connecticut, United States)2 NASA Goddard Space Flight Center (Greenbelt, Maryland,

United States)3 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)4 University of Texas at Austin (Austin, Texas, United States)5 University of Arizona (Tucson, Arizona, United States)6 Arizona State University (Tempe, Arizona, United States)7 Dartmouth College (Hanover, New Hampshire, United States)8 National Institute of Astrophysics (Cantania, Italy)

Given the current capabilities of exoplanet detec-tion methods, M dwarf stars are excellent candi-dates around which to search for temperate, Earth-sized planets. The UV wavelength regime is impor-tant to evaluate the photochemistry of the planetaryatmosphere because many molecules have highlywavelength dependent absorption cross sections thatpeak in the UV (900-3200 Å). M dwarfs are highlyactive stars with unique spectra that can drive theabiotic production of key planetary biosignatures.We seek to provide a broadly applicable method ofestimating the UV emission of an M dwarf, with-out direct UV data, by identifying a relationship be-tween non-simultaneous optical and UV observa-tions. Our work uses the largest sample of low-massstar UV observations yet assembled, including datafrom the MUSCLES and Mega-MUSCLES TreasurySurveys (Measurements of the Ultraviolet SpectralCharacteristics of Low-mass Exoplanetary Systems),the FUMES survey (Far Ultraviolet M-dwarf Evolu-tion Survey), and the HAZMAT survey (HAbitableZones and M dwarf Activity across Time). We mea-sure Hα equivalent widths and the Mount WilsonCaII H&K S and R’HK indices using ground-basedoptical spectra from the HARPS, UVES, and HIRESarchives and new HIRES spectra. Archival and newHubble Space Telescope COS and STIS spectra areused to measure line fluxes for the brightest chro-mospheric and transition region emission lines be-tween 1200-2800 Å. Our results show a correlationbetween UV line luminosity and CaII R’HK with stan-dard deviations in the range of 0.25-0.54 dex about

146

the best-fit lines. Correlations between UV luminos-ity and Hα or the S index are weak. The results pre-sented in this talk will be important for near-futureallocations of competitive Hubble time as well as thepost-Hubble era. We demonstrate that with a pre-cise R’HK measurement obtained from the ground(e.g., 5-10% precision), the estimate of the intrinsicLyα luminosity is ∼12-20%, which is typically bet-ter than what can be achieved with direct, low-to-medium S/N Hubble spectra. After we gather moredata this summer, we will also be able to detail de-pendencies on age and spectral type.

331.04 — 3-D Climate Models For CharacterizingHabitable Terrestrial Extrasolar Planets

Ravikumar Kopparapu1; Anthony Del Genio2; MichaelWay2; Eric Wolf3; Thomas Fauchez1; Nancy Kiang2;Linda Sohl2; Jacob Haqq-Misra4; Scott Guzewich1;Stephen Kane5; John Armstrong6; Chester Harman2;Kostas Tsigardis2; Daria Pidhorodetska1; ShawnDomagal-Goldman1; Mark Marley7

1 NASA Goddard Space Flight Center (Greenbelt, Maryland,United States)

2 NASA GISS (New York, New York, United States)3 University of Colorado-Boulder (Boulder, Colorado, United States)4 Blue Marble Space Institute of Science (Seattle, Washington,

United States)5 U. California Riverside (Riverside, California, United States)6 Weber State University (Ogden, Utah, United States)7 NASA Ames (Moffet Field, California, United States)

While the recently discovered extrasolar planetshave both challenged our imaginations and broad-ened our knowledge of planetary systems, per-haps the most compelling objective of exoplanetscience is to detect and characterize habitable andpossibly inhabited terrestrial worlds around otherstars. In our quest to characterize habitable worlds,three-dimensional (3-D) general circulation models(GCMs) should be used to evaluate the potentialclimate states and their associated observable sig-nals. 3-D models allow for self-consistent and re-alistic simulations of the climates of terrestrial ex-trasolar planets around a variety of stellar spectraltypes. Future observatories have the capability to de-duce transmission, thermal emission, and reflectancespectra as a function of orbital phase. A complete un-derstanding of terrestrial exoplanetary atmospheres,gained through comprehensive 3-D modeling, is crit-ical for interpreting spectra of exoplanets. In thispresentation, we will highlight the recent advancesin 3-D climate model studies of habitable climates,and their impact on observables. We show that thecurrent assumption of a (modern) Earth-analog as

a template for a habitable planet around other stel-lar spectral types is not a representative model forimportant features in the observed spectrum. Im-proving upon our models of habitability is particu-larly relevant for planets within the habitable zonesof low-mass stars (late-K and all M-dwarfs), whichmight be amenable for characterization in the nearfuture either with JWST or large ground-based tele-scopes or the mission study concept OST. Such im-provements will also be equally important for plan-ets around Sun-like stars due to different evolution-ary and planetary system history, which can be stud-ied by mission studies like LUVOIR and HabEX. Thepresentation will include potential synergies that canbe fostered between theorists and observers, with acommon goal of finding an inhabited exoplanet.

331.05 — Geochemistry of Carbon Cycles on RockyExoplanets

Kaustubh Hakim1; Pierre Auclair-Desrotour1; RussellDeitrick1; Daniel Kitzmann1; Dan J. Bower1; CarolineDorn2; Kevin Heng1

1 Center for Space and Habitability, University of Bern (Bern,Switzerland)

2 University of Zurich (Zürich, Switzerland)

The long-term carbon cycle (also known as thesilicate-carbonate cycle) acting on a timescale of theorder of hundreds of thousands of years providesthe essential negative feedback to maintain temper-ate climates on Earth. With the discovery of almost athousand rocky exoplanets and ongoing hunts for anEarth-twin, it is imperative to understand the work-ing of the carbon cycle on such planets. The aim isto investigate the factors of the Earth’s carbon cy-cle that are critical to stabilize and destabilize carboncycles on rocky exoplanets. These factors could bedependent on the orbital, planetary and stellar pa-rameters as well as planet-specific properties such asrock composition, land and ocean fractions, amongother factors. In this study, we focus on modelingthe chemical kinetics of rock-water interaction fordifferent rock types (depending on the planet’s sur-face composition), as well as pH. We incorporate aset of silicate weathering reactions leading to the for-mation of carbonates. In addition to continental sil-icate weathering, we explore the effects of seafloorweathering especially in the context of varying land-mass fractions, and shallow and deep ocean frac-tions. Other components of the carbon cycle such assubduction, ridge and arc volcanism are parameter-ized based on previous studies. The effects of planetsize, oxidation states, and tidal locking are also in-vestigated.

147

331.06 — Stellar Flares and Habitable (?) M-dwarfWorlds: Exploring a New Sample with TESS

Maximilian N. Günther11 MIT (Cambridge, Massachusetts, United States)

Finding and characterizing small exoplanets tran-siting small stars naturally poses the question oftheir habitability. A major contributing factor to thismight be stellar flares, originating from powerfulmagnetic reconnection events on the star. While toopowerful flaring can erode or sterilize exoplanets’ at-mospheres and diminish their habitability, a mini-mum flare frequency and energy might be requiredfor the genesis of life around M-dwarfs in first place.Here, I will highlight our study of stellar flares fromthe Transiting Exoplanet Survey Satellite (TESS). Inthe first year of TESS data, we already identifiedthousands of flaring M-dwarfs, most of which arerapidly rotating and of late spectral types. Our sam-ple includes superflares that showed over 30× bright-ness increase in white light. I will link our resultsto criteria for prebiotic chemistry, atmospheric lossthrough coronal mass ejections, and ozone steriliza-tion. Expanding this with upcoming TESS sectors,stellar flare studies will ultimately aid in defining cri-teria for exoplanet habitability.

331.07 — Abiotic Oxygen on Venus-Like Exoplan-ets Around M-Dwarfs

Michael L. Wong1,3; Victoria S. Meadows1,3; Peter Gao2;Carver Bierson4; Xi Zhang4

1 Astronomy & Astrobiology, University of Washington (Seattle,Washington, United States)

2 University of California, Berkeley (Berkeley, California, UnitedStates)

3 Virtual Planetary Laboratory (Seattle, Washington, United States)4 University of California, Santa Cruz (Santa Cruz, California,

United States)

Terrestrial exoplanets in the habitable zones ofnearby M dwarfs represent the first targets for thesearch for life outside of the Solar System. It has beenlong thought that one of the most obvious biosig-natures on alien worlds would be the spectroscopicdetection of O2 and/or O3, created by a global bio-sphere of photosynthetic life forms (Meadows et al.,2018). However, modeling has suggested that largeamounts of O2 can be created abiotically—especiallyon terrestrial planets around M dwarfs. In particu-lar, Gao et al. (2015) showed that desiccated worldswith CO2-rich atmospheres can build up ∼15% O2via CO2 photolysis. Venus, nonetheless, has little at-mospheric O2, despite ongoing CO2 photolysis. This

has been attributed to catalytic cycles involving ClOxand SOx that regenerate CO2 from CO and O (Millset al., 2007; Yung & DeMore, 1999). We seek to ascer-tain how these cycles behave on Venus-like planetsaround different types of stars.

We have constructed a 1-D photochemical modelbased on Zhang et al. (2012) to study the atmosphericchemistry of Venus-like exoplanets. The model sim-ulates an atmosphere primarily composed of CO2(∼90 bars) and N2 (∼3 bars) with trace amounts ofH2O, SO2, HCl, and other constituents composed ofH, C, O, N, S, and Cl, which contribute the HOx,ClOx, SOx, and NOx catalysts that can recombinephotochemically generated CO and O into CO2. Wecompare the effect of G- and M-dwarf spectral en-ergy distributions on Venus-like worlds, placing theplanets at orbital distances with the same total inci-dent flux as Venus.

We find that Venus-like worlds are rich in catalyststhat can recombine CO and O into CO2. Around bothG and early M dwarfs, as the catalytic ClOx chem-istry is sufficient recombine CO and O. We identifycatalytic cycles involving Cl–S molecules that con-trol the buildup of large amounts of photochemi-cal O2 around late M dwarfs. Specifically, aroundTRAPPIST-1, low SO2 mixing ratios significantly re-duces the action of Cl–S catalysts that scrub O2and reconstitute CO2. This implies that Venus-likeplanets around late M dwarfs must maintain someamount of active SO2 outgassing to be robust againstabiotic O2 production.

331.08 — A laboratory-to-model approach to under-standing exoplanet biosignatures

Tiffany Kataria1; Scott Perl1; Pin Chen1; Laura M.Barge1; Yuk L. Yung2

1 JPL/Caltech (Pasadena, California, United States)2 California Institute of Technology (Pasadena, California, United

States)

The assessment of exoplanet habitability is predom-inantly based on the measurement of biosignaturegases, usually in the form of triplicate sets of CH4,O2, CO2, and O3, among other molecules indicativeof life. Because exoplanets are distant, this is predi-cated on the ability to characterize atmospheres thatwould contain these gases at detectable limits for re-mote telescopes. However, this methodology oftenlacks a mechanism relating atmospheric detectionsto the potentially biogenic sources that emit thesegases at the planetary surface. This can lead to mis-interpretations between abiotic signatures and trulybiotic sources. Here we present a study to quantita-tively link surface processes and atmospheric chem-

148

istry for potentially habitable exoplanets using actualmicrobiological experiments. We will measure gasoutputs from actual field samples of microbial com-munities that are from various ecosystems in which amultitude of major biogenic gases can be quantified.These measurements will serve as exoplanet surfaceinputs to the Atmsopheric-Rock-Ocean-Chemistry(AROC) model, which couples an aqueous geochem-istry code and KINETICS (a photochemistry code) totrace surface-to-atmosphere chemistry. In this way,we can to bridge the gap between exoplanet biosig-natures and the very biology and metabolisms foundin nature. The results from this study can help guidethe design of future ground- and space-based tele-scopes (e.g., JWST, ARIEL, ELTs, HabEx, LUVOIR,Origins) by identifying additional biosignatures thatwould help distinguish between atmospheric condi-tions that may or may not be conducive to life.

331.09 — Stellar Energetic Particle-induced radia-tion dose as a constraint on the habitability of ter-restrial exoplanets

Dimitra Atri11 New York University Abu Dhabi (Abu Dhabi, United Arab Emi-

rates)

Space weather has a profound impact on plane-tary atmospheres and has the potential to disrupthospitable environments on exoplanets. The ef-fect is more significant in case of close-in exoplan-ets around active stars. In addition to X-rays andEUV from stellar flares, energetic charged particlescan ionize the atmosphere leading to photochemi-cal changes, result in atmospheric erosion and canenhance radiation dose on the planetary surface.Charged particles of GeV energies undergo hadronicinteractions in the atmosphere producing secondaryparticles, a fraction of which traverse down to thesurface with harmful biological effects. Using datafrom 70 major SPEs (Solar Particle Events) over thepast century as a proxy, and using GEANT4 (CERN)Monte Carlo model, we simulate radiation dose in-duced by Stellar Energetic Particles on presentlyknown habitable exoplanets for various atmosphericand magnetospheric conditions. This is the first com-prehensive study to quantify the effects of SEPs onexoplanets. We compare the results with experimen-tal radiobiology data and discuss its implications onconstraining the habitability of terrestrial exoplanets.

331.10 — Transition from Eyeball to SnowballDriven by Sea-ice Drift on Tidally Locked Terres-trial Planets

Jun Yang11 Dept. of Atmospheric and Oceanic Sciences, Peking University

(Beijing, Beijing, China)

Background: Among the ≈4000 confirmed exoplan-ets, most of them are orbiting around M dwarfs be-cause they are relatively easy to detect and M dwarfsare the most common type of star in the galaxy.About 15 exoplanets are most likely to have rockycompositions and meanwhile in the habitable zonewithin which the surface is temperate to maintainliquid water. These planets are the prime targets forfuture atmospheric characterizations of potentiallyhabitable systems, especially the three nearby ones–Proxima b, TRAPPIST-1e, and LHS 1140b. Previ-ous studies suggest that if these planets have surfaceocean they would be in an eyeball-like climate state:ice-free in the vicinity of the substellar point and ice-covered in the rest regions. However, an importantcomponent of the climate system–sea ice dynamicshas not been well studied in their work.

Fundamental question: Would the open ocean ofthe eyeball-like climate be stable against a globallyice-covered snowball state? Or, could sea-ice driftclose the open ocean?

Methods: Through a series of climate experiments,we investigate the effects of oceanic heat transportand sea-ice drift on the surface ice coverage of tidallylocked terrestrial planets. Hundreds of numeri-cal experiments were performed in order to clearlyknow the robustness of the results.

Conclusion: Ocean dynamics trend to expand theopen-ocean area, but more importantly wind-drivensea-ice drift toward the substellar point shrinks theopen-ocean area and even drives some of them to asnowball state. This works for both a synchronousorbit and a resonance orbit. The dominated mecha-nism is that sea-ice drift cools the sea surface throughabsorbing heat during ice melting when the ice flowsto the warmer substellar region.

Implication: Previous studies have shown thatstellar radiation, atmospheric composition and evo-lution, atmosphere dynamics, clouds, and ocean dy-namics are important for the climate and habitabil-ity of tidally locked planets. Here, we show that an-other critical factor—sea ice dynamics, which is ableto drive an eyeball-like climate state to a globally ice-covered snowball state.

149

331.11 — TETH: Towards Extra-Terrestrial Habitats

Mitchell Young1; Luca Fossati1; Colin Johnston2;Michael Salz3; Herbert Lichtenegger1; Patricio Cubillos1;Kevin France4; Helmut Lammer1

1 Institut Fur Weltraumforschung (Graz, Austria)2 University of Vienna (Vienna, Austria)3 Hamburg Sternwarte (Hamburg, Germany)4 Astrophysical and Planetary Science, University of Colorado

(Boulder, Colorado, United States)

The discovery and characterisation of new extra-solar planets (exoplanets) is ongoing, but to date onlya handful of low-mass planets have been found orbit-ing in the habitable zone of sun-like stars. The nextgeneration of major facilities (e.g. TESS and PLATO)aimed at the systematic search for Earth-like plan-ets orbiting solar-like stars will be operational in thecoming years, and some of the planets they will findmay orbit stars close enough for atmospheric char-acterisation, including the possible detection of bio-signature gases. Studies on the formation and evo-lution of the Earth reveal that an Earth-like habi-tat is characterised by a N-dominated atmosphereand could be detected by measuring the relative at-mospheric abundances of N, O, C, and H (NOCH).However, N, which is the main fingerprint of anEarth-like habitat, is extremely difficult to detect andmay be possible only in the ultraviolet, a wavelengthrange that has not been studied for low-mass exo-planets. Before starting the search for bio-signatureswith future facilities (e.g. ELTs, LUVOIR), we needto explore our capabilities to detect Earth-like habi-tats. Here, we present several synthetic transmis-sion spectra for the Earth’s atmosphere, for the wave-length range 915 to 11000 Å, at a spectral resolutionof R = 100,000. We focus on both atomic and molec-ular features, and discuss the detectability of N.

331.12 — Assessing the Potential of Volatile Or-ganic Compounds as Exoplanet Biosignatures

Yuka Fujii1; Yui Kawashima2; Jade Checlair3; AlexisGilbert4,1; Sebastian Danielache5,1

1 Earth-Life Science Institute (Tokyo, Japan)2 SRON Netherlands Institute for Space Research (Utrecht, Nether-

lands)3 University of Chicago (Chicago, Michigan, United States)4 Tokyo Institute of Technology (Tokyo, Japan)5 Sophia University (Tokyo, Japan)

Remotely detectable signs of life, or biosignatures,are being widely studied for future exoplanet ob-servations. While the majority of the studies fo-cus on simplest molecules such as Oxygen, Ozone,

or Methane, Earth’s biosphere emits a significantlybroader range of gaseous species into the atmo-sphere, including what is called Volatile OrganicCompounds (VOCs). The specific inventory of theserelatively complex molecules would have more lim-ited possibilities to be produced by abiotic processeson terrestrial planets, making them potentially use-ful, complementary biosignatures. A drawback isthat they are photochemically vulnerable and lesslikely to be accumulated in the atmosphere. In ourpresentation, we make a general assessment of VOCsas exoplanet biosignatures by studying the charac-teristics of their spectral features and the conditionsin which VOCs may accumulate in planetary atmo-spheres to a detectable level. We show that for anEarth-like atmosphere the mixing ratio of larger than∼1 ppm would be necessary to be able to identify theinfrared spectral features of VOCs. This level of accu-mulation could be possible around late M-type starsdue to the reduced OH and O3 production. We alsodiscuss how to distinguish VOC worlds from abiotichydrocarbon worlds.

331.13 — Identifying Potential Venus Analogs fromExoplanet Discoveries

Colby Ostberg1; Stephen Kane11 Earth and Planetary Sciences, University of California, Riverside

(Riverside, California, United States)

With a radius of 0.95 Rearth and a mass of 0.85 Mearth,Venus is the most analogous planet to Earth in the so-lar system. Study of Venus and Venus-like exoplan-ets is invaluable in understanding factors that deter-mine a planet’s habitability throughout its evolution.Fortunately, many Venus-analogs are expected tosoon be discovered as the recently launched Transit-ing Exoplanet Survey Satellite (TESS) mission is sen-sitive to planets in close proximity to their host stars.TESS is predicted to discover hundreds of terrestrialplanets within the inner boundary of their host star’sHabitable Zone (HZ), placing them in the ‘VenusZone’ (VZ), defined by Kane et al. (2014). TESS intandem with the launch of the James Webb SpaceTelescope in the coming years will allow for the char-acterization of these planets’ atmospheres, provid-ing a better understanding of atmospheric compo-sitions of planets inside the VZ. This will help de-lineate the primary factors that determine whether aplanet develops sustainable temperate surface condi-tions, or if it would be pushed into a runaway green-house state, leading to a more well-defined outerboundary for the VZ. Here we provide a progressreport on discoveries from the TESS mission, iden-tification of planets in the VZ, and methods used to

150

determine runaway greenhouse scenarios. The ob-served properties of these planets will be applied toGlobal Climate Models, such as ROCKE-3D, to bet-ter constrain the boundaries of the HZ and VZ, andstudy the atmospheric demographics of terrestrialplanets.

331.14 — A Matter of Time: The Coupled Roleof Stellar Abundances, Exoplanet Radiogenic HeatBudgets and Climatic Evolution

Cayman Unterborn1; Bradford Foley2; Patrick Young1;Greg Vance1; Lee Chieffle1; Steve Desch1

1 Arizona State University (Tempe, Arizona, United States)2 Penn State University (State College, Pennsylvania, United

States)

A planet’s heat budget is a complex combination ofthe heat of formation, the energy released duringcore formation, and decay of the long-lived radionu-clides U, Th and 40K. A planet’s radiogenic heat bud-get is solely a function of the total amount of theseelements present. Observations of Solar twins showa range of Th abundances between 60 and 250% ofthe Sun’s (Unterborn et al., 2015). We show simi-lar ranges are expected for U and K. If this range isindicative of the span of exoplanet radiogenic heatbudgets, an exoplanet’s thermal and chemical evolu-tion may be quite different from the Earth’s.

We present results of geodynamical models fromFoley & Smye, 2018 for stagnant lid planets withvarying radiogenic heat budget. We compile a rangeof observed radionuclide abundances reported fromthe literature, adopting a Monte-Carlo approach fordetermining input abundances in individual mod-els. We focus on stagnant lid planets, as stagnant lidconvection is likely to be more common than platetectonics. We show that changes in a planet’s radio-genic heat budget affect the rates of volcanism, sur-face weathering, and volatile degassing from the in-terior. This allows us, through the measurement ofthe host star’s abundances of radionuclides, to con-strain a planet’s thermal history and quantify thetimescale over which it can maintain a temperate cli-mate.

In general we find those stagnant lid planets witha greater starting abundance of radionuclides aretemperate for a longer period of time. We calcu-late a conservative estimate of degassing lifetimesfor a 1 Earth mass stagnant-lid planet of 1.6±0.6 Gyracross our observationally constrained range of ra-dionuclide abundances. We argue those non-tidally-locked planets orbiting stars older than this are un-likely to be actively degassing in the absence of platetectonics, including both TRAPPIST-1 (7.6±2.2 Gyr;

Burgasser & Mamajek, 2017) and Kepler 444 (10±1.5Gyr; Mack et al., 2018). These planets thereforeare unlikely to have temperate climates, and insteadlikely lie in a snowball climatic regime, limiting theirpotential to be habitable and ”Earth-like.”

331.15 — Detecting Pre-Biosignatures in the Atmo-spheres of Earth-like Planets Around Other Stars

Sarah Rugheimer1; Paul Rimmer21 University of Oxford (Oxford, United Kingdom)2 University of Cambridge (Cambridge, United Kingdom)

When we observe the first terrestrial exoplanet atmo-spheres, we expect to find planets at a wide range ofgeological conditions and evolution including plan-ets that may be in the early stages of biological de-velopment or failed biospheres that reached only acertain point of pre-biotic chemistry.

Understanding the UV environment of the hoststar is particularly important for contextualizing thehabitability of an exoplanet. Depending on the in-tensity, UV radiation can be both useful and harm-ful to life as we know it. UV radiation from 180 - 300nm can inhibit photosynthesis and cause damage toDNA and other macromolecule damage (e.g. Ker-win & Remmele 2007). However, these same wave-lengths also drive several reactions thought neces-sary for the origin of life (e.g. Ritson & Sutherland2012; Patel et al. 2015).

Molecules such as HCN, NH3, CH4, and C2H6would be interesting to detect in an exoplanet at-mosphere since they are known to be useful for keyprebiotic chemical pathways. We find that someof these molecules could be produced abiotically ina CO2/CH4/H2 rich atmosphere with lighting andphotochemistry. HCN, for example, is present ateach of the initial photochemical reactions that pro-duce lipids, amino acids and nucleosides, the threebuilding blocks of life. Reactions to form HCNcan be accomplished via photochemistry, lightning,impacts, or volcanism. As well, the C/O ratio ofthe planet will greatly influence the likely domi-nate reactions in that planet’s atmosphere. We dis-cuss the chemical mechanisms by which HCN canbe formed and destroyed on rocky exoplanets withEarth-like N2 content and surface water inventories,varying the oxidation state of the dominant carbon-containing atmospheric species.

We finally examine the plausibility of detectingprebiotically interesting molecules, such as HCN,NH3, CH4, and C2H6 in an early-Earth type atmo-sphere around stars with different UV environmentsusing early Earth, a flaring M dwarf, and a quiescent

151

M dwarf as a range of host stellar types and UV en-vironments.

331.16 — Near-UV transmission spectrum of Earth-as-an-exoplanet obtained during a lunar eclipse

Allison Youngblood1; Giada Nicole Arney1; JohnStocke3; Kevin France2; Aki Roberge1

1 NASA Goddard Space Flight Center (Greenbelt, Maryland,United States)

2 Astrophysical and Planetary Science, University of Colorado(Boulder, Colorado, United States)

3 Center for Astrophysics and Space Astronomy (Boulder, Colorado,United States)

We performed UV spectroscopy with the HubbleSpace Telescope of the January 2019 lunar eclipseto obtain the first UV observation of Earth as atransiting exoplanet. The observatories and in-struments that will be able to perform transmis-sion spectroscopy of exo-Earths are beginning to beplanned, and characterizing the transmission spec-trum of Earth is key to ensuring that key spec-tral features (e.g., ozone) are appropriately capturedin mission concept studies. Ozone is photochemi-cally produced from O2, a product of the dominantmetabolism on Earth today, and it will be soughtin future observations as critical evidence for lifeon exoplanets. Ground-based lunar eclipse observa-tions have provided the Earth’s transmission spec-trum at optical and near-IR wavelengths, but thestrongest ozone signatures are in the near-UV. We de-scribe the observing strategy and the methods usedto extract a transmission spectrum from Hubble lu-nar eclipse spectra. Finally, we identify spectral fea-tures in the transmission spectra and compare themto Earth models to determine if current models accu-rately capture key transmission features of the Earthin the near-UV.

331.17 — Prospects for detecting extraterrestrial O2with the ELTs

Dilovan Serindag1; Ignas Snellen11 Leiden Observatory, Leiden University (Leiden, Netherlands)

High-dispersion transmission spectroscopy usingthe upcoming extremely large telescopes (ELTs) hasthe potential to probe the atmospheres of temper-ate, Earth-like exoplanets transiting nearby, late-typestars. Such observations may reveal the presence ofbiomarkers like molecular oxygen (O2), which couldindicate the presence of extraterrestrial life. Previ-ous studies performed simulations with purely syn-thetic noise distributions to estimate the detection

feasibility of the O2 A-band at 7600 Å. We improveon these simulations by incorporating real data withreal white and red noise, specifically, archival, time-series spectra taken of the M-dwarf Proxima Cen-tauri with the Ultraviolet and Visual Echelle Spec-trograph (UVES) on the Very Large Telescope (VLT).Since the expected flux difference between Proximaand the brightest transiting M-dwarf systems is sim-ilar to the difference in collecting area between theVLT and the European ELT (E-ELT), these UVES datashould have noise characteristics similar to future E-ELT observations. By injecting oxygen transmissionsignals into the UVES data of Proxima, we determinethat an A-band detection with the E-ELT will require20-50 transits for an Earth-twin transiting a nearby(d≈7 pc) M5V star. This result from simulations witha more realistic noise distribution is similar to previ-ous estimates.

331.18 — Evryscope flares as probes of the habit-ability of Proxima b and the nearest rocky exoplan-ets

Nicholas Law1; Ward Howard1; Henry Corbett1; AmyGlazier1; Matt Tilley2

1 Physics and Astronomy, UNC Chapel Hill (Chapel Hill, NorthCarolina, United States)

2 University of Washington (Seattle, Washington, United States)

In March 2016, the Evryscope observed the first su-perflare from Proxima Centauri. The Evryscope ar-ray of small optical telescopes recorded the super-flare as part of an ongoing survey of all bright south-ern stars, monitored simultaneously at 2 minute ca-dence since 2015. Evryscope flares act as probes ofthe space weather environment and potential habit-ability of nearby exoplanets in three ways: by con-straining their UV surface environments, by lookingfor planetary magnetic fields via star-planet interac-tion and flares that phase up with planet orbits, andby monitoring optical counterparts to radio flare ob-servations. We will illustrate each of these probes forProxima Centauri, and go on to discuss the first re-sults from our ongoing program to measure the long-term flare behavior of all TESS planet-search targets.By modeling the photochemical effects of particleevents accompanying large flares, we find Proxima’srepeated flaring is sufficient to reduce the ozone col-umn of an Earth-like atmosphere at the orbit of Prox-ima b by 90% within five years. Surface UV-C levelsduring the Evryscope superflare reached ∼100× theintensity required to kill simple UV-hardy microor-ganisms without ozone, suggesting that life wouldstruggle to survive in the areas of Proxima b exposedto these flares.

152

Across the sky, we report ∼2× the previous largestnumber of 1034 erg high-cadence flares from nearbycool stars. We find 8 flares with amplitudes of 3+magnitudes, with the largest reaching 5.6 magni-tudes and releasing 1036 erg. We measure the super-flare rate per flare-star and quantify the superflareproperties of TESS-planet-search stars as a functionof spectral type. We observe 14.6±2% of the starsaround which TESS may discover temperate rockyplanets emit flares large enough to significantly af-fect the potential habitability of those planets. Wefound 17 stars that may deplete an Earth-like at-mosphere via repeated flaring, and we will discussobservations of a superflare with sufficient energyto photo-dissociate all ozone in an Earth-like atmo-sphere in a single event.

331.19 — Investigating the Habitability of Exoplan-ets Orbiting High-Mass Stars

Johnathon Ahlers11 NASA/GSFC & USRA (Silver Spring, Maryland, United States)

High-mass stars (M∗ ≥ 1.3MSun) provide fundamen-tally different environments for planets than Sun-like and smaller stars. Their outer-most layers areradiative rather than convective, and therefore typi-cally do not emit flares that make habitability aroundM-dwarfs questionable. High-mass also stars com-monly rotate near their breakup speeds, resultingin pole-to-equator temperature gradients that canvary by more than a thousand Kelvin. Additionally,planets orbiting high-mass stars commonly misaligninto highly inclined or even retrograde orbital con-figurations, causing planets to vary in exposure be-tween their host stars’ hotter poles and cooler equa-tors throughout their orbits. In this presentationI will show how the stellar and orbital propertiescommonly seen in A/F-type systems can combineto produce insolation patterns unlike anything seenfor planets orbiting Sun-like stars, and I will discusstheir implications for climate and habitability.

331.20 — How surface albedo shapes a planet — in-side our Solar System and out.

Jack Madden1,2; Lisa Kaltenegger1,21 Astronomy and Space Sciences, Cornell University (Ithaca, New

York, United States)2 Carl Sagan Institute (Ithaca, New York, United States)

Different planetary surfaces can strongly influencethe climate, atmospheric composition, and remotelydetectable spectra of planets and exoplanets. Fromhighly reflective ice, cooling a planet’s surface, to

dark oceans, absorbing most of the incident lighteach surface type needs careful study. Our own So-lar System contains a diverse set of bodies (planetsand moons) with a wide range of rocky, icy, andgaseous surfaces that can be used as a reference cat-alog for comparison against exoplanet observation.We show how this spectral reference catalog of 19 So-lar System objects, can be used to initially character-ize different surface types using color filters, whichcan assist in the prioritization of exoplanets for time-intensive follow-up with next-generation ExtremelyLarge Telescopes (ELTs) and space-based direct ob-servation missions. The feedback between surfaceand climate is a large factor when modeling terres-trial exoplanets. We further explore the changes to aplanetary environment for different surfaces arounda wide range of host stars using a wavelength de-pendent surface albedo. This allows us to create adatabase of high-resolution spectra (R>100,000) ofremotely detectable atmospheric features, includingbiosignatures, for exoplanet observations with up-coming telescopes. The feedback between planetarysurfaces and climate is critical to understanding theenvironment of potentially habitable worlds.

331.21 — Comparitive exo-Planetology with Po-larimetry for Detecting Habitability Markers

Kimberly Bott1,2; Lyan Guez2; Victoria Meadows2,1; An-drew Lincowski1,2; Jeremy Bailey3,4; Lucyna Kedziora-Chudczer3,4

1 VPL, University of Washington (Seattle, Washington, UnitedStates)

2 Astronomy, University of Washington (Seattle, Washington,United States)

3 Astronomy, UNSW Australia (Sydney, New South Wales, Aus-tralia)

4 Australian Centre for Astrobiology (Sydney, New South Wales,Australia)

We present results from research on the use of po-larimetry to distinguish between planet types anddetect key features of habitability. This work ex-plores the signatures of planets with realisticallymodelled surfaces and atmospheres in orbit aroundG and M dwarfs to determine which features in a po-larized light curve are easiest to detect. We modelterrestrial type planets as well as mini-Neptuneswith volatile envelopes. For terrestrial planets,our models include different ground types, oceans,species dependent Rayleigh scattering (based on en-vironmentally dependent chemical evolution), andclouds. In some key areas such as distinguish-ing glint from cloud and disentangling the effectof cloud heights polarimetry provides a powerful

153

tool. With upcoming larger aperture ground basedtelescopes polarimetry could enable observations ofplanets that would be within the inner working an-gle of a coronagraph, complementing space basedtelescopes. We find that, although the signaturesof some planet types are distinct, considerations of“secondary effects” (subtle changes in the reflectiv-ity of contaminated ices for example) need to be con-sidered. The presentation will portray this work inthe context of the broader work of the Virtual Plan-etary Laboratory to understand exoplanet habitabil-ity and biosignatures, showcasing work by our stu-dents. We examine observational strategies with cur-rent and future instruments to assess the likelihoodof detection and compare the utility of polarimetryto other observation techniques.

331.22 — Prospects for Biosignature Detection withJWST

Victoria Meadows1; Andrew Lincowski1; Jacob Lustig-Yaeger1; Guadalupe Tovar Mendoza1

1 Astronomy, University of Washington (Seattle, Washington,United States)

The James Webb Space Telescope (JWST) may pro-vide the first opportunity to characterize the com-position of atmospheres of M dwarf terrestrial plan-ets. Due to the character of M dwarf host stars’UV spectra, atmospheric biosignatures may build upto higher abundances than for planets orbiting Gdwarfs, potentially making them more detectable.Krissansen-Totton et al., (2018) argue that the si-multaneous detection of CO2 and CH4 could con-stitute a disequilibrium biosignature in an anoxicenvironment. However, whether JWST will havethe sensitivity to detect biosignatures from oxygenicphotosynthesis, or other plausible metabolisms, re-mains an open question. To provide a comprehen-sive exploration of the potential detectability of dif-ferent types of biosignatures with JWST, we haveused coupled 1-D climate-photochemical and radia-tive transfer models to generate synthetic spectra ofsimulated planetary environments for TRAPPIST-1d and e that support a range of different biospheres.We then simulate JWST observations for these en-vironments, and identify optimal observing modes,exposure times, and retrieval methods for detect-ing biosignatures and environmental context. Wesimulate Archean-Earth-like environments with ei-ther a dominant sulfur- or methane-producing bio-sphere for clear, cloudy and hazy cases, as well asmodern Earth analogs with photosynthetic oxygen-producing biospheres for clear and cloudy cases, andalso assess the habitability of these environments.

We quantify the detectability of the CO2and CH4biosignature. We find that other biosignatures, in-cluding the ethane signature of sulfur biospheres,and oxygen and ozone for photosynthetic biosigna-tures, will be extremely challenging to detect. Wealso explore the detectability of other possible biosig-natures, including methyl chloride, which is pre-dominantly produced by tropical plants on Earth,and could serve as an alternative indicator of oxy-genic photosynthesis.

331.23 — Low-Cost Inference of Terrestrial Cli-mates With Broadband Photometry

Adiv Paradise1; Kristen Menou2,1; Christopher Lee3; BoLin Fan1

1 Astronomy & Astrophysics, University of Toronto (Toronto, On-tario, Canada)

2 Physical & Environmental Sciences, University of Toronto Scar-borough (Toronto, Ontario, Canada)

3 Physics, University of Toronto (Toronto, Ontario, Canada)

A universal aspect of exoplanet science is that the in-formation available is very limited, and the sampleof planets we can study in situ is very small. De-tailed, informative observations of exoplanets are ob-servationally resource-intensive and technically dif-ficult, and a growing body of numerical simula-tions of planet formation, evolution, and climateshows that inference from these limited observationsis fraught with numerous observational and physicaldegeneracies. At a time when the number of obser-vational targets is large but the resources availableto study them are limited, we need low-cost tech-niques which nonetheless permit robust classifica-tion. Critically, these techniques must be developedin advance to prevent wasted resources. The par-ticular challenge of classifying rocky exoplanet cli-mates in the habitable zones of Sun-like stars hasin the past been frustrated by many potential de-generacies, particularly the possibility of widespreadclouds. Using hundreds of full 3D climate mod-els, each with detailed simulated reflectance spec-tra, we present a new technique to uniquely identifycold, ice-covered ‘snowball’ planets through broad-band visible and infrared photometric colors. Theoccurrence rate of snowball planets in the habitablezone is sensitive to planets’ geophysical properties,so this low-cost technique for identifying snowballplanets in observed populations represents a possi-bly unique way to test geophysical theories of planetformation and evolution. While in our models seri-ous degeneracies prevent the ultimate goal of easyand robust identification of temperate, Earth-like cli-mates, this technique could permit pre-screening of

154

potential follow-up targets. We identify several use-ful observing bands that will be accessible to plannedinstruments on upcoming large ground-based tele-scopes, and which may guide future instrument de-sign. Our work shows low-SNR broadband photom-etry can distinguish between different climates onEarth-like planets, which permits informative surveyscience and potentially provides a roadmap for im-proving other exoplanet inference problems.

331.24 — Constraining the Water Loss Histories ofthe TRAPPIST-1 Exoplanets

David Fleming1,2; Rory Barnes1,2; Rodrigo Luger3,21 Astronomy, University of Washington (Seattle, Washington,

United States)2 Virtual Planetary Laboratory (Seattle, Washington, United States)3 Center for Computational Astrophysics, Flatiron Institute (New

York City, New York, United States)

JWST is poised to detect and characterize terres-trial exoplanet atmospheres in the search for biosig-natures (Morley et al., 2017, Lustig-Yaeger et al.,2019), but the correct interpretation of those ob-servations is predicated on understanding the sys-tem’s long-term evolution. A well-known example,TRAPPIST-1, harbors 7 planets that received signifi-cant high-energy fluxes during the 1 Gyr stellar pre-main sequence, likely driving water loss (Luger &Barnes 2015, Wheatley et al., 2017). We describeand employ two new software tools, VPLanet andapproxposterior, to derive probabilistic constraintsfor TRAPPIST-1’s XUV luminosity evolution andestimate the probability that these planets couldhave water today. We use VPLanet, a general pur-pose planetary system evolutionary code (Barnes etal., 2019), to simulate stellar evolution and waterloss and apply approxposterior, a machine learningPython package for Bayesian inference (Fleming &VanderPlas, 2018), to compute accurate approxima-tions of posterior distributions for how much waterthe planets could have lost, accounting for observa-tional uncertainties and correlations between param-eters. approxposterior obtains nearly identical re-sults as traditional Markov Chain Monte Carlo meth-ods (e.g. Foreman-Mackey et al., 2013), but requires500× less computational time. We find that there isa 46% chance that TRAPPIST-1 is still in the satu-rated phase today, potentially causing TRAPPIST-1eto lose about 8 Earth oceans, releasing ∼1400 barsof O2, a false biosignature. We define the ExoplanetHabitability Index (EHI) to quantify the probabilitythat a planet possesses water, given our model, andtherefore may be a viable target for JWST biosigna-ture observations. TRAPPIST-1e is likely a good can-

didate with an EHI ∼1, depending on the assumedinitial water distribution. As new nearby transitingplanets are discovered, our framework can be ap-plied to efficiently identify those that could possessliquid water today and can be readily generalizedto account for additional physical processes to gaininsights and generate predictions for potentially-habitable and uninhabitable worlds.

331.25 — Planetary Magnetism as a Parameter inExoplanet Habitability

Sarah R.N. McIntyre1; Charley H. Lineweaver1,2;Michael Ireland1

1 Research School of Astronomy & Astrophysiscs, Australian Na-tional University (Weston Creek, Australian Capital Territory, Aus-tralia)

2 Research School of Earth Sciences, Australian National University(Canberra, Australian Capital Territory, Australia)

Evidence from the Solar system suggests that, un-like Venus and Mars, the presence of a strong mag-netic dipole moment on Earth has helped main-tain liquid water on its surface. Therefore, plane-tary magnetism could have a significant effect onthe long-term maintenance of atmosphere and liquidwater on rocky exoplanets. We use Olson & Chris-tensen’s (2006) model to estimate magnetic dipolemoments of rocky exoplanets with radii Rp ≤ 1.23Rearth. Even when modelling maximum magneticdipole moments, only Kepler-186 f has a magneticdipole moment larger than the Earth’s, while ap-proximately half of rocky exoplanets detected in thecircumstellar habitable zone have a negligible mag-netic dipole moment. This suggests that planetarymagnetism is an important factor when prioritizingobservations of potentially habitable planets.

331.26 — An Astroecological Model for Character-izing Exoplanet Habitability

Alma Yesenia Ceja1; Stephen Kane11 Earth and Planetary Sciences, University of California, Riverside

(Riverside, California, United States)

A primary objective of astrobiology is to identifyworlds outside of our own which are capable of sup-porting life. Here, an integrative approach is ap-plied to characterize the habitability of rocky ex-oplanets in the habitable zone of their host stars.We explore the relationship between modeled alienenvironments and terrestrial life with a novel as-troecology model which can be used as a tool toasses habitability on exoplanet surfaces. In thismodel, simulated exoplanet environments from real

155

derived exoplanetary parameters are convolved witha biological layer derived from laboratory experi-ments. The general circularization model (GCM),ROCKE-3D (Way et al. 2018), a fully-coupled 3-dimensional oceanic-atmospheric model is used togenerate exoplanet environmental parameters. TheGCM output is coupled in the astroecology model inthe agent-based modeling software, NetLogo, withempirically-derived thermal performance curves of1,627 cell strains representing extremophiles fromall six Kingdoms. This dataset, termed the Bioki-netic Spectrum for Temperature (Corkrey et al.,2016), arises from a meta-analysis of cellular growthrate as a function of temperature. In this agent-based model, the survivability of a biological organ-ism is determined by its thermal response to simu-lated local and global exoplanet temperatures. Thiswork produces a list of exoplanets with the high-est probability of having temperate surface condi-tions compatible with terrestrial-based thermophys-iology. Life, however, is dependent upon multiplevariables including the presence of liquid water, nu-trient content, and an energy source. Caveats of themethodology and application of our results are dis-cussed with implications for extraterrestrial evolu-tion.

331.27 — Habitability of the Teegarden’s Star Plan-ets

Amri Wandel1; Lev Tal-Or21 Physics, Hebrew University of Jerusalem (Jerusalem, Israel)2 Geophysics, Tel Aviv University (Tel Aviv, Israel)

We study the habitability of the two 1.5±0.4 Earth-mass planets, recently detected by the CARMENEScollaboration, around the ultra-cool nearby M dwarfTeegarden’s Star. With orbital periods of 4.9 and 11.4days, both planets are likely to be within the Habit-able Zone and tidally locked. They are among themost Earth-like exoplanets yet discovered. We find(Wandel and Tal-Or 2019, ApJL) that one or bothplanets are likely to support liquid water on at leastpart of their surface for a wide range of possible at-mospheres, characterized by their atmospheric heat-ing factor (i.a. the greenhouse effect) and global cir-culation. At least one of the TG planets may be habit-able for atmospheric heating in the range 0.3-15 thatof Earth. As demonstrated by detailed numerical cal-culations of similar planets, Teegarden’s Star presentcalmness and old age favor the retaining or repro-duction of a sufficiently massive atmosphere, withinthe habitability range. The demonstrated habitabil-ity of the Teegarden’s Star planets for a wide rangeof atmospheres, combined with their small distance

(3.85 pc) and highest similarity to Earth (ESI=0.94)makes them most attractive targets for bio-signaturesearches.

332 — Other — Theory, Poster Ses-sion332.03 — Machine Learning and Big Data for Ex-oplanets and Astrobiology: Results from NASAFrontier Development Lab

Daniel Angerhausen1,21 Center for Space and Habitability, Bern University (Bern,

Switzerland)2 Blue Marble Space Institute of Science (Seattle, Washington,

United States)

We present results from NASA’s Frontier Develop-ment Lab 2018, an Artificial Intelligence/MachineLearning incubator tackling challenges in variousfields of space sciences. Herw we focus on the resultsof the Exoplanet and Astrobiology teams: planetcandidate classification in survey data and modelingand retrieval of exoplanet atmospheres and spectrain the context of life detection. A particular focuswill be on two data sets produced: a set of 3 millionexoplanet spectra calculated with the GSFC Plane-tary Spectrum Generator (PSG) and a set of 150.000exoplanet atmospheres computed with ATMOS. TheExoplanet team used state-of-the-art deep learningmodels to automatically classify Kepler and TESStransit signals as either exoplanets or false positives(Ansdell et al. 2018, Osborn et al. 2019). Their As-tronet code expanded upon work of Shallue & Van-derburg 2018 by including additional scientific do-main knowledge into the models to significantly in-crease overall performance . The Astrobiology team1 project demonstrated how cloud computing capa-bilities can accelerate existing technologies and mapout previously neglected parameter spaces (Bell etal., 2019). They succeeded in modelling tens-of-thousands of atmospheres over a few days, usingthe software ATMOS that was originally intendedfor use in single run applications. In Soboczenski etal., 2018 and Cobb/Himes at al., 2019 the Astrobi-ology 2 team presented a ML-based retrieval frame-work called INARA that consists of the first Bayesiandeep learning model for retrieval and a data set of 3,million synthetic rocky exoplanetary spectra gener-ated using PSG (Zorzan et al, in prep.; Himes et al.,in prep.). References: Bell, A., et al. 2018, NIPS 2018CiML workshop; Soboczenski, F., et al. 2018, NeurIPSWorkshop on Bayesian Deep Learning, arXiv:1811.03390;Ansdell, M., et al. 2018, ApJL, 869 (1), L7; Shallue, C. J.,

156

& Vanderburg, A. 2018, AJ, 155, 94 Cobb A., Himes M.et al. 2019, AJ, in press, arXiv:1905.10659 ; Osborn H. etal. 2019, A&A, in press, arXiv:1902.08544

332.04 — Magnetic field effects on the motion ofcharged dust in rings and discs, motivated by Sat-urn’s spokes.

Mia Mace11 School of Physics, University of Bristol (Bristol, United Kingdom)

Spokes are a transient feature in Saturn’s rings.Widely accepted to be charged dust levitated awayfrom the main ring plane, the exact mechanismsand physical environment leading to their formationand propagation is still shrouded in mystery. Sev-eral theories have been proposed, including micro-meteoroid impacts and lightning. However, nonecan fully explain the appearance of spokes.

This study is an ongoing exploration into chargeddust motions in rotating tilted dipole magnetic fields.A systematic investigation of the orbital dynam-ics and resonances in Saturn’s rings has been per-formed with simulations. The numerical methodused is flexible and allows a comparison to the othernon-diffuse ring systems of our solar system, wherespokes are not observed. The aim is a greater un-derstanding of why spokes are only observed in Sat-urn’s B ring. Extending the work beyond planetaryrings, predictions can be made about the existence ofspoke-like features in more extreme solar systems,which are analogous in structure — a central bodyand disc, such as white dwarfs.

332.05 — Persistent Homology of Flows on Extraso-lar Planets

Jack William Skinner1; James Cho2; Heidar Thrastarson31 Astronomy Unit, Queen Mary University of London (London,

United Kingdom)2 Flatiron Institute (New York, New York, United States)3 Jet Propulsion Laboratory, California Institute of Technology

(Pasadena, California, United States)

High-resolution simulations and observations gen-erate copious amounts of high dimensional, largevolume, heterogeneous datasets, which are increas-ingly difficult (if not prohibitive) for analysis by tra-ditional (statistical, spectral, or graphical) methodsalone. Persistent homology is a novel computa-tional method for practically ascertaining the topo-logical ‘shape’ of such data. Here the shape ischaracterized by tallying the number of connectedelements and n-dimensional ‘holes’ (e.g., closedloops, three and higher dimensional voids, etc.), as

well as ‘coves’ (depressions or protrusions on theholes), in the data. An example is the recent high-resolution, long-duration simulations of hot-Jupiteratmospheres that produce highly complex flow andtemperature fields, containing up to many thou-sands of storms across a wide range of spatial andtemporal scales. To clearly demonstrate the efficacyof the homology analysis method, we use it to ana-lyze an idealized vortex model of these storms, fo-cusing on the nonlinear evolution of such stormsat the extremely high Reynolds number associatedwith planetary flows. Features, such as the num-ber of storms and filaments around their periphery,their ‘tubular’ or ‘blobby’ morphologies, and peri-odic bursts of instability are captured and quantified.Understanding such features is crucial for validatingtheory and numerical models, as well as for inter-preting and guiding observations. Broadly, homo-logical analysis is a widely applicable tool that canhelp to directly address the large data problem facedin many areas.

332.06 — Extreme Solar Systems and the Fermiparadox : limits to growth?

Aurelien Crida1,21 Lagrange, Université Cote d’Azur (Nice, France)2 Institut Universitaire de France (Paris, France)

The amazing data we now have on exoplanets allowsus to estimate that our Galaxy hosts about 50 billionterrestrial exoplanets. On the other hand, on ourown planet, life appeared very quickly, then devel-oped slowly, until a remarquable acceleration in thelast 500 million years that lead to mankind. In thedevelopement of our civilisation, a similar exponen-tial acceleration is observed, with the distance fled byman made objects multiplied roughly by 10 every tenyears since the first plane flew accross a field at thedawn of the XXth century. At this pace, we shouldconquer the whole Milky Way by the end of this cen-tury. But although our planet formed nine billionyears after the first stars, none of the many exoplan-ets seems to be just a century ahead of us : they havenot sent their flying saucers to Earth.

This paradox, first coined by E. Fermi, shows thatsomething must be wrong in the above reasoning. Infact, it is obvious to every participant of this confer-ence that the laws of physics forbid to cross a hun-dred thousand light-years in just a hundred years,so that exploring the whole Galaxy can not be doneso fast. This illustrates that any exponential processreaches its limits quickly : the distance we explorecan not keep being multiplied by ten every ten years.This is also true for the economic growth, which is

157

also exponential, and based on the use of ressourcesand the emission of pollution in a finite world ...

The CO2 emissions already change the Earth’s cli-mate, in such a way that they represent a threat forour future developement. Have all the other extrater-restrial lives destroyed their environment by tryingto conquer the Galaxy? Have they spontaneously de-cided to stop their exponential growth in order topreserve their planet? The fact that none of themsucceeded to come here forces us to think about ourown behaviour. In particular, this conference willcertainly be fantastic, and I will be enjoying it a lot.But is it reasonable that we all fly here? Should ourcommunity start thinking about how to do sciencewith less greenhouse gasses emission?

332.07 — Formation and evolution of short-periodplanets around magnetized host stars

Douglas NC Lin1,21 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)2 Institute of Advanced Studies, Tsinghua University (Beijing,

China)

During the formation and early evolution of short-period planets, kiloguass fields persist on their hoststars. Their interaction with planets’ evolving nataldisks determines not only the amount of buildingblock grains and the orbital destiny of protoplanets,but also emerging stars’ spin rate. After the disk de-pletion, the relative motion between the stellar spinand the planets’ orbit leads to unipolar induction andLorentz force which can cause significant orbital evo-lution, Ohmic dissipation, and low-frequency radioemission. I show how this effect may enable remotesensing of super Earths’ surface composition. I alsoshow how to generate analogous magnetic field inproto-Jupiters and how these process may have de-termined the spin rate of Jupiter and the orbits of itsGalilean moons.

332.08 — Forget Limb Darkening Laws: TransitModeling Using Stellar Atmosphere Intensities

Jerome Orosz1; Donald R. Short1; William F. Welsh1;Gur Windmiller1

1 Astronomy, San Diego State University (San Diego, California,United States)

Limb darkening (LD) laws are ubiquitous in their usein exoplanet transit modeling. They provide a fastand easy way to parameterize the changing intensi-ties across the disk of the star, allowing transit mod-els to be quickly computed and compared with ob-servations. However, we know that LD laws are a

poor representation of the stellar intensities, partic-ularly at the limb where the intensities drop off ex-tremely quickly. This is especially important whenconsidering subtle effects such as planet oblateness,since these effects are most pronounced at the limb.In spite of the shortcomings, researchers often do notuse the LD laws that match stellar models. Instead,they fit for the coefficients of a LD law. This fur-ther removes the transit modeling from actual stel-lar physics. While the fit to the transit may be betterwhen solving for LD coefficients, it is no more thana consequence of (i) allowing more freedom in themodels; (ii) a re-statement that LD laws are not real-istic representations of the stellar intensities.

It would be far more advantageous to use an ac-tual model atmosphere to give the specific intensi-ties, and we provide a fast method to do so, us-ing tabulated model intensities from ATLAS mod-els to compute the transit. The code is only slightlyslower than the one using ad hoc LD laws. By us-ing actual model atmosphere intensities, there are nofree parameters to solve for (given the stellar prop-erties). Comparison of this method with the tradi-tional method using ad hoc LD laws shows the de-rived planet radius is systematically wrong by 0.1%or more, depending on the impact parameter. In ad-dition to improving the accuracy (rather than pre-cision) of transit modeling, we note an area whereour method may be extremely fruitful: transit spec-troscopy. The wavelength-dependent transit depthis a function of both the planet’s atmosphere and thestellar LD variation. By eliminating the use of param-eterized LD laws we (i) completely remove any de-generacy between transit depth and LD coefficients;(ii) include astrophysical knowledge of the star’s in-tensity distribution, which varies strongly as a func-tion of both position and wavelength.

333 — Other — Observational,Poster Session333.01 — Endeavours towards precise M-dwarfproperties: Activity robust multi-line modeling inthe visual and near-infrared

Vera Maria Passegger1; Andreas Schweitzer1; DenisShulyak2; Evangelos Nagel1; Peter H. Hauschildt1;Ansgar Reiners3; Pedro J. Amado4; José A. Caballero5;Miriam Cortés-Contreras5; Alejandro J. Domínguez-Fernández6; David Montes6; Andreas Quirrenbach7;Ignasi Ribas8,9

1 Hamburg Observatory (Hamburg, Germany)2 Max Planck Institute for Solar System Research (Goettingen,

158

Germany)3 Institute for Astrophysics, Georg-August-University Goettingen

(Goettingen, Germany)4 Instituto de Astrofísica de Andalucía (Granada, Spain)5 Centro de Astrobiología (Madrid, Spain)6 Universidad Complutense de Madrid (Madrid, Spain)7 Landessternwarte, U Heidelberg (Heidelberg, Germany)8 Institut de Ciències de l’Espai (Barcelona, Spain)9 Institut d’Estudis Espacials de Catalunya (Barcelona, Spain)

A precise characterisation of planet-hosting stars isvery important to derive and constrain the physicalproperties of orbiting planets. The CARMENES in-strument, which is searching for habitable planetsaround M dwarfs, provides us with high-resolutionspectra in the visual (0.52–0.96 μm) and near-infrared wavelength range (0.96–1.71 μm). We fit themost recent PHOENIX-SESAM stellar atmospheremodels simultaneously to both wavelength rangesto determine effective temperature, surface gravity,and metallicity for 282 M dwarfs. With these tem-peratures we also derive stellar masses and radii us-ing luminosities and Gaia DR2 parallaxes. Althoughstellar activity is widely unconsidered in stellar pa-rameter determination, we show the importance oftaking into account this property by carefully select-ing magnetically insensitive lines, especially for thenear-infrared wavelength range. For the first time,we directly compare stellar parameters such as ef-fective temperature, surface gravity, and metallic-ity derived from multiple wavelength ranges for thesame spectra. We recommend using a combinationof the visual and near-infrared wavelength ranges forparameter determination in order to maximise theamount of spectral information and minimise possi-ble effects due to model imperfections.

333.02 — Optical/Near-IR Microwave Kinetic In-ductance Detector-based Integral Field Spectro-graphs for High-Contrast Observations

Isabel Lipartito1; Benjamin A. Mazin1; Alexander B.Walter1; Clinton Bockstiegel1; Neelay Fruitwala1; SethMeeker2; Paul Szypryt4; Nicholas Zobrist1; GregoireCoiffard1; Sarah Steiger1; Noah Swimmer1; JenniferSmith1; John I. Bailey1; Kristina Davis1; Henry Ru-pert Dodkins1; Olivier Guyon3,5; Nemanja Jovanovic6;Julien Lozi3; Ananya Sahoo3; Sebastien Vievard3; Dim-itri Mawet6; Michael Bottom2; Clarissa Rizzo1

1 Physics, University of California, Santa Barbara (Santa Barbara,California, United States)

2 Jet Propulsion Laboratory (Pasadena, California, United States)3 National Astronomical Observatory of Japan (Hilo, Hawaii,

United States)

4 National Institute of Standards and Technology (Boulder, Col-orado, United States)

5 College of Optical Sciences, University of Arizona (Tucson, Ari-zona, United States)

6 California Institute of Technology (Pasadena, California, UnitedStates)

I will present an overview of the high-contrastimaging techniques achievable with Microwave Ki-netic Inductance Detector (MKID)-based instru-ments. Optical/Near-IR MKIDs are low noise detec-tors which can resolve both the energy and arrivaltime of individual photons, returning microsecond-accurate time-tagged photon lists. These lists allowfor post-processing techniques that take full advan-tage of photon spectral and arrival time information,such as photon-counting Stochastic Speckle Discrim-ination (SSD), a technique which enables direct ob-servations of exoplanets at small separations fromtheir host star, and Spectral Differential Imaging(SDI), a technique which exploits the chromatic be-havior of speckles to model the stellar Point-SpreadFunction (PSF). These techniques enable the subtrac-tion of the speckle background approaching the pho-ton noise limit. Active feedback techniques using theMKIDs as a focal plane wavefront sensor promiseto further improve instrumental sensitivity. I willalso present results from the MKID Exoplanet Cam-era (MEC). MEC is a 20,440 pixel MKID infraredcamera that interfaces with the SCExAO planet find-ing instrument on the Subaru Telescope at MaunaKea. MEC received first light in June 2018 and hasbeen used to image young massive planets and de-bris disks.

333.03 — Capturing the Now — the AAS Oral His-tory Project

Jarita Holbrook11 Physics & Astronomy, University of the Western Cape (Belliville,

Western Cape, South Africa)

The Oral History Project of the American Astronom-ical Society is in its sixth year. It is part of the ac-tivities of the Historical Astronomy Division. Whatis an oral history? Oral histories are interviews withindividuals that are meant to capture some aspect oftheir, if not their entire, life. For the AAS project,we spend up to two hours with each person gath-ering background information, noting career moves,highlighting mentors, but also touching on currentissues relevant to our community such as diversity,sexual harassment, data science, queue observing,tenure, and getting individual recognition for collab-orative research. We have interviewed over 150 sci-

159

entists, technicians, family members, and STEM sup-port staff. Some of these interviews can be foundon the AIP archive of oral history interviews. TheAAS Oral History Project is unique in that we in-terview anyone who volunteers to be interviewed:You do not need to be old and famous (like DougLin, but we do want to interview him, too!). Ourinterviews include undergraduates to emerita. Theposter shares the full list of our interviewees andsome of our preliminary analysis of trends. Finally, ifyou would like to be interviewed contact Jarita (jhol-brook@uwc.ac.za).

333.04 — A Fresh Look at Red Giant Planet HostsUsing TESS: A Study of Stellar Mass and SurfaceGravity

Joleen Carlberg1; Doug Branton1; Jeff Valenti11 Space Telescope Science Institute (Baltimore, Maryland, United

States)

A detailed understanding of even the most basicparameters of planetary systems requires accurateknowledge of the host star’s bulk properties. His-torically, measuring accurate masses for field red gi-ant stars has been especially difficult because starsof a wide range of masses evolve through similarparameter space on the Hertzsprung Russell dia-gram, sometimes passing through the same phasespace multiple times. For a given luminosity andtemperature, the star’s surface gravity must also beknown. Until recently, surface gravity was predom-inately measured spectroscopically, requiring theabundances of metals inferred from both neutral andsingly-ionized species to agree (ionization balance).Such measurements tend to suffer from systematicuncertainties that are difficult to fully quantify, re-sulting in independent gravities measured for thesame star that differ by much more than the quoted(random) uncertainties. Asteroseismology has pro-vided the key to solving this problem. High preci-sion, long baseline photometric monitoring yield de-tailed oscillation spectra, and the frequency of max-imum oscillation power is directly proportional tosurface gravity. TESS is collecting such data over thewhole sky, and here we present preliminary mea-surements of asteroseismic log g from a sample ofred giant planet hosting stars in the Southern Hemi-sphere. Our sample has been observed by TESS in2-minute cadence, and we have spectroscopic log gmeasurements both from heterogeneous sources inthe literature and measured homogenously by us.We discuss the frequency with which asteroseismiclog g’s differ substantially from the asteroseismicones (even when spectroscopic measurements agree)

and we explore the implications of revised stellar logg and masses on the interpretation of the planetarycompanions around these stars.

333.05 — Machine Learning for the Identificationof Exomoon Candidates in Kepler

Alex Teachey1; David Kipping11 Astronomy, Columbia University (New York, New York, United

States)

Convolutional neural networks (CNNs) are wellsuited for image classification problems, particularlywhen the data volume is too large for by-eye clas-sification. They have recently been applied withgreat success to the problem of distinguishing gen-uine planets from false positives in the Kepler, K2,and TESS datasets. In this work we apply a CNNto the Kepler data for the purpose of identifyingcandidate exomoon signals. We train the CNN on∼200,000 artificial light curves — real Kepler lightcurve segments with injected planet/moon signals— to achieve ∼95% accuracy for moon signals of suf-ficient SNR in the validation set. We apply the CNNto every transit of every KOI to identify potentialexomoons in the data, after which we will vet themost promising candidates with a full photodynam-ical moon fit and Bayesian model selection.

333.06 — Eclipse Mapping: Creating Two- or Three-Dimensional Images of Exoplanets

Emily Rauscher1; Nick B. Cowan2; Megan Mansfield6;Jacob Arcangeli13; Arthur D. Adams11; Ying Feng5;Prashansa Gupta10; Dylan Keating7; Jacob Lustig-Yaeger9,12; Everett Schlawin3; Kevin Stevenson4;Thomas Beatty8

1 Astronomy, University of Michigan (Ann Arbor, Michigan,United States)

2 Département de Physique, Université de Montréal (Montréal,Quebec, Canada)

3 Yale University (New Haven, Connecticut, United States)4 NExSS Virtual Planetary Laboratory (Seattle, Washington,

United States)5 University of Amsterdam (Amsterdam, Netherlands)6 McGill University (Montreal, Quebec, Canada)7 Astronomy, University of Arizona (Tucson, Arizona, United

States)8 STScI (Baltimore, Maryland, United States)9 Astronomy & Astrophysics, UC Santa Cruz (Santa Cruz, Califor-

nia, United States)10 Geophysical Sciences, University of Chicago (Chicago, Illinois,

United States)11 Physics, McGill University (Montréal, Quebec, Canada)12 University of Arizona (Tucson, Arizona, United States)

160

13 University of Washington (Seattle, Washington, United States)

Eclipse mapping is a method that can be used to re-solve an image of the day side of an exoplanet. Asthe planet passes into secondary eclipse, the stellarlimb blocks successive slices of the planet disk, whilea different set of slices are revealed as the planetcomes back out of eclipse. If we measure the de-tailed change in flux during these partially obscuredtimes of ingress and egress, we can reconstruct thebrightness distribution across the planet disk. If wetake spectra during these times, then different wave-lengths are probing different depths into the atmo-sphere and so these data contain three-dimensionalinformation about the planet’s atmosphere. In prac-tice, eclipse mapping is a nuanced technique, sub-ject to various sources of uncertainty and degener-acy. Here we present a mathematically optimizedmethod for retrieving the maximum possible spatialinformation from an eclipse mapping dataset. Wecan also use this method to evaluate whether signifi-cant orbital-mapping degeneracies exist, for any par-ticular system and observational set-up. We then in-troduce a way to extend this framework to includemulti-wavelength observations, by using machinelearning methods to identify dominant spectral com-ponents and group them into spatial regions. Weshow applications of this new spectral-spatial map-ping method to test-case toy models, with various ar-tificial atmospheric states, demonstrating its abilityto retrieve correct information. Finally, we estimatethe scientific return when this method is applied toupcoming JWST data.

333.07 — The Starchive – An Extreme Open AccessArchive

Angelle Tanner1; Demitri Muna21 Physics and Astronomy, Mississippi State University (Mississippi

State, Mississippi, United States)2 Eureka Scientific (New York, New York, United States)

The Starchive (starchive.org) is an open access, opensource stellar database and web application like noother. We have designed an interface which is intu-itive, comprehensive and adaptable. Currently, thedatabase contains multiple stellar samples includ-ing all stars within 30pc, all known brown dwarfsand white dwarfs, stars with planets and circumstel-lar disks and stars in young stellar associations. Wehave plans to incorporate the WDS, Gaia and TESScatalogs. The web app allows users to search thedatabase using coordinates, names or an ADS ref-erence code. A search can include a single star or

multiple stars (batch mode). If users search for infor-mation on a single star, the result page contains allpublished measurements and derived physical pa-rameters on that star, a Vizier image as well as anyavailable high contrast images via js9. If the star isin a multiple system, there is a clear hierarchical treewith live links to the other members of the system.If available, users will have access to wavelength cal-ibrated spectra and time series of that star all in onelocation. If a user submits a list of stars or utilizesthe rank list search option, the web app provides adynamic table of multiple stars with links to eachindividual star page. Users are able to download atext, .csv or latex file of that table. Directly from themulti-star web page application users will be able touse adaptable plotting tools to visualize the resultingdata set. It is the goal of the Starchive that the plotsbe publication quality thus eliminating the need todownload and then replot data for presentations andpapers. Registered users will be able to upload datainto the database. To ensure the fidelity of the data,we will highly regulate and constantly validate anyuploaded data sets. There will be an API available forusers to access the database directly from their owncode. The front-end scripts will be placed on githuband users will be encouraged to contribute new plot-ting tools.

333.08 — Sonifying Solar Systems: New Tools forResearch and Outreach

Deborah Kala Perkins11 AstroBioethics, AUSN (Woodside, California, United States)

Sonifying planetary transits and the possible chem-ical contents of their atmospheres; ALMA’s publicSoundbank of molecular data from new star birthingregions; discerning the harmonic resonances be-tween planets in distant solar systems; and Ke-pler Space Telescopes’ “Stellar Choir”, demonstrat-ing with a unique soundscape, playable by the in-quirer, the location of known exoplanetary systems:All of these are now offering sonified data for aunique encounter with exo-planetary science both bythe public and research scientists, seeking insightsabout the universe beyond the visible spectrum. Wehave found new insights about the solar corona fromsonified data which the ear could discern though in-visible to the eye. Sonification is opening an entirelynew window for communication with the public,students, and understanding our universe. It is pro-viding those who are visually impaired the abilityto explore the universe, experience it, and as well tobecome astrophysicists. This presentation explores

161

some of the unique things being done at this fron-tier for outreach, blending culture and science, andresearch in the realm of Extreme Solar Systems andthe quest for other life in the cosmos.

333.09 — Exoplanets in the Antarctic Sky: fromSearching to Characterizing

Yu Zhouyi1; Zhang Hui11 Nanjing University (Nanjing, China)

Thanks for the wide-field exoplanet surveys on theground and in the space, thousands of exoplanetsamples have been found in the last two decades.From Dome A, the highest point of the Antarcticplateau, we have also contributed over 100 candi-dates using the AST3 telescopes in 2018. Now, be-sides searching, we are progressing forward to studyspecial exoplanet systems in details to reveal theirdynamics and physics properties. I will first in-troduce our recent works on searching exoplanetswith the help from deep learning methods. And I’llpresent some results on characterizing Proxima Centb, the nearest potential habitable world, using AST3-II. We find a temporary solar-like oscillation in Prox-ima. We know that M dwarf stars are dominated byadvection layer, there should be some kinds of Solar-like oscillation. But the oscillation won’t be stable soalthough believed, no positive detection was made.This may be the first observation proof. This is anexcellent example to show the advantages of moni-toring high-value targets from Dome A. To furtherutilize these advantages I’ll also introduce the KISS(Kunlun Infrared Sky Survey) project and its usageon exoplanet characterization. We also monitoredBeta Pictoris a couple of hours each day during thetwilights (when the weather was permitting), usingAST3 II telescope, in 2017. At the end of the polarwinter, we had acquired around 70,000 frames onthis target at a cadence of 3.5 sec. Although no ob-vious eclipse was found, we’ve found some new pul-sating frequencies, e.g. around 14.3, 20.6, 58.98/day ,and some ultra-high-freqency signals, which are notmentioned before. We think this phenomena can re-veal some properties of Beta Pictoris b’s circumplan-etary environment.

333.10 — The Terrascope: Turning Planets intoTelescopes

David Kipping11 Columbia University (New York, New York, United States)

As our knowledge of exoplanets grows, so to doesour thirst for ever more data. Photometric precision

has been gradually improving from percent levelwith photographic plates, to mmag with ground-based surveys and now in the parts per million withspace-based facilities. It is timely to consider wherethe next order of magnitude gain might come fromthen, to reach parts per billion precision. In this talk,I’ll discuss the idea of using the Earth as a giant lensof distant starlight exploiting atmospheric refraction— a ”terrascope”. By placing a one-meter apertureat the Earth-Sun L2 point, distant point sources areshown to be amplified by a factor of ∼45,000 witha lensing timescale of ∼20 hours, effectively turninga one-meter telescope into a ∼200 meter class tele-scope. I show how the effects of atmospheric extinc-tion and clouds are small provided the terrascope isplaced at a large separation from the Earth, such asL2. The terrascope concept is certainly challenged byseeing, turbulence and airglow and I’ll discuss somepossible strategies to mitigate these effects.

333.11 — The Mysterious Activity of TRAPPIST-1

Brett Morris1; Eric Agol2; James Davenport2; SuzanneHawley2

1 Center for Space and Habitability, University of Bern (Bern, Bern,Switzerland)

2 Astronomy Department, University of Washington (Seattle,Washington, United States)

TRAPPIST-1 is one of the most tantalizing exoplanetsystems discovered to date, with seven Earth-sizedtransiting exoplanets in a resonant chain orbitingan ultra-cool dwarf star. To make robust infer-ences about the properties of the exoplanets orbitingTRAPPIST-1, we must first identify any stellar sur-face inhomogeneities which will confound exoplanettransmission spectroscopy (Morris et al. 2018, Rack-ham et al. 2018, Ducrot et al. 2018). TRAPPIST-1 isthe first M8V star to be scrutinized with long-term∼1% precision photometry in multiple wavebands,and preliminary analyses of the surface features ofthe host star are full of surprises. There is no defini-tive evidence for coverage of the stellar surface bydark starspots, but there is photometric and spectro-scopic evidence for bright, hot regions on the sur-face of the star. Furthermore, the occurence of flaresseems to be correlated with the optical flux of thestar, perhaps suggesting that the apparent rotationalmodulation of the star could instead be evolution ofbright active regions. We will discuss the availableevidence for activity and rotation of the host star,and conclude with a discussion of the implicationsfor transmission spectroscopy of the exoplanets.

162

333.12 — The Transiting Exoplanet Survey Satellite(TESS): Mission Operations and Instrument Per-formance

Roland Vanderspek11 MIT Kavli Institute for Astrophysics and Space Science, Mas-

sachusetts Institute of Technology (Cambridge, Massachusetts, UnitedStates)

The Transiting Exoplanet Survey Satellite (TESS) isperforming a near-all-sky survey to detect exoplan-ets around the nearest and brightest stars in the sky.TESS began performing science observations of thesouthern ecliptic hemisphere in July of 2018. AsTESS begins the survey of the northern ecliptic hemi-sphere, we review the performance of the instrumentand spacecraft during the first year of the sciencemission. We also discuss flight operations in Year 2,including instrument pointing and target selection.

333.13 — Extreme Stellar Systems: Towards Im-proved Stellar Parameters with Single-lined Eclips-ing Binaries

Daniel Joseph Stevens2,1; George Zhou3; Caleb Cañas2,11 Astronomy & Astrophysics, The Pennsylvania State University

(University Park, Pennsylvania, United States)2 Center for Exoplanets and Habitable Worlds (University Park,

Pennsylvania, United States)3 Center for Astrophysics, Harvard & Smithsonian (Cambridge,

Massachusetts, United States)

The precision and accuracy with which we character-ize exoplanets discovered via transit and radial ve-locity (RV) techniques are no better than the preci-sion and accuracy with which we know their hoststars. This relationship is most stark for planetsaround M dwarfs. The most precisely and accu-rately determined stellar parameters have come fromdouble-lined eclipsing binaries (EBs), and M-M EBstudies suggest that low-mass stellar models under-predict radii and over-predict effective temperaturesby 5-10%, — thus determining how precisely andaccurately we can characterize planets around Mdwarfs.

While studies of the ”M dwarf discrepancy” arelimited both by the small sample of well-studied M-M EB and their typically short orbital periods, single-lined EBs — and the advent of precise Gaia par-allaxes — provide one avenue to circumvent theseproblems. We will present results from our ongo-ing campaign to measure few-percent masses, radii,and temperatures for single-lined EBs — specifically,the M dwarf secondaries. our total sample of ∼200single-lined EBs extends the parameter space out to

longer orbital periods (dozens of days) and differentbinary configurations. As such, we can better distin-guish between radius and temperature discrepanciesinherent to low-mass stellar models and the effects ofincreased stellar activity caused by a nearby stellarcompanion.

We will highlight a few key characterizations of A-M and F-M EBs from joint analyses of primary andsecondary eclipses, optical radial velocities (RVs),and spectral energy distributions. We will also shareour exciting discovery of unique EB consisting ofa fully convective, likely pre-main sequence M starand a late-B dwarf. We will discuss prospects forimproving the precision of the inferred parametersand compare parameters inferred from these ”globalanalyses” to those inferred from modeling the out-of-transit flux variations.

333.14 — Exoplanetary Atmospheric Retrieval viaBayesian Machine Learning

Michael Himes1; Adam Cobb2; Frank Soboczenski3; Si-mone Zorzan4; Molly O’Beirne5; Atilim Gunes Baydin2;Yarin Gal9; Shawn Domagal-Goldman7; Giada NicoleArney6; Daniel Angerhausen8

1 Planetary Sciences Group, Department of Physics, University ofCentral Florida (Orlando, Florida, United States)

2 Department of Engineering Science, University of Oxford (Ox-ford, United Kingdom)

3 SPHES, King’s College London (London, United Kingdom)4 ERIN Department, Luxembourg Institute of Science and Technol-

ogy (Esch-sur-Alzette, Luxembourg)5 Department of Geology and Environmental Science, University of

Pittsburgh (Pittsburgh, Pennsylvania, United States)6 Planetary Systems Laboratory, NASA Goddard Space Flight Cen-

ter (Silver Spring, Maryland, United States)7 Planetary Environments Laboratory, NASA Goddard Space Flight

Center (Greenbelt, Maryland, United States)8 Center for Space and Habitability, Bern University (Bern,

Switzerland)9 Department of Computer Science, University of Oxford (Oxford,

United Kingdom)

Atmospheric retrieval, the inverse modeling tech-nique whereby atmospheric properties are inferredfrom observations, is computationally expensiveand time consuming. Recently, machine learning(ML) approaches to atmospheric retrieval have beenshown to provide results consistent with traditionalapproaches in just seconds to minutes. We introduceplan-net, the first ensemble of Bayesian neural net-works for atmospheric retrieval. Our novel likeli-hood function captures parameter correlations, im-proving uncertainty estimations over standard like-lihood functions common in ML. We replicate the re-

163

sults of Marquez-Neila et al. (2018), and we demon-strate plan-net’s improvement in accuracy over theirrandom forest regression tree when applied to theirsynthetic data set of hot Jupiter WFC3 transmis-sion spectra. We apply a trained plan-net ensem-ble to the transmission spectrum of WASP-12b andfind results generally consistent with the literature.We also apply plan-net to our data set of over 3million synthetic terrestrial exoplanet spectra gen-erated using the NASA Planetary Spectrum Gen-erator. AC is sponsored by the AIMS-CDT andEPSRC. FS and MH acknowledge the support ofNVIDIA Corporation for the donation of the TitanXp GPUs used for this research. AGB is funded byLawrence Berkeley National Lab and EPSRC/MURIgrant EP/N019474/1.

333.15 — Planets are Shaped by their Past: Re-constructing the Early XUV Emission of ExoplanetHost Stars

Parke Loyd1; Evgenya L. Shkolnik1; Adam Schneider1;Tyler Richey-Yowell1; Travis Barman2; Sarah Peacock2;Isabella Pagano4; Victoria Meadows3

1 School of Earth and Space Exploration, Arizona State Univeristy(Tempe, Arizona, United States)

2 Lunar and Planetary Laboratory, University of Arizona (Tucson,Arizona, United States)

3 Astronomy, University of Washington (Seattle, Washington,United States)

4 Osservatorio Astrofisico di Catania (Catania, Italy)

X-ray and extreme ultraviolet (XUV) radiation islikely shaping the observed population of short pe-riod exoplanets by powering blow-off of their atmo-spheres. The amount of atmosphere an individualplanet might lose to XUV erosion depends on the his-tory of its XUV irradiation, which could be 10-1000×higher when it formed than at present day. Thiscan explain two classes of ”missing” planets: sub-Jovians with short orbital periods (<5 d) and planetswith radii between super-Earths and sub-Neptunesout to longer periods (< 25 d), colloquially known asthe ”sub-Jovian desert” and the ”radius gap,” respec-tively. However, a star’s present XUV flux is likelynot a good tracer of its XUV past. In consequence, wefound correlating the properties of planets with theirpresent-day XUV irradiation yielded no clear depen-dence in the location of the desert or the gap. We willpresent this test, along with simulations determiningwhether such correlations could be detected in yet-to-be-discovered TESS planets in this talk. Accurateknowledge of a individual stars’ XUV histories couldpermit a robust test of the photoevaporation hypoth-esis for the missing planets and constrain the times-

pan over which photoevaporation is most effective.We will synthesize the current state of knowledge re-garding the evolution of stellar XUV emission fromthe time of planet formation onwards for F-M stars,outlining the obstacles and opportunities for reveal-ing the past lives of highly irradiated planets.

333.16 — Results from the W-Band survey: Search-ing for planetary-mass brown dwarfs

Sophie Dubber11 University of Edinburgh (Edinburgh, United Kingdom)

The W-Band survey has found multiple young, M/L-type dwarfs in the Taurus and Serpens star-formingregions, spectroscopically confirmed using a custom1.45 micron filter to identify water absorption fea-tures. Here, I present examples of the spectra of theseobjects, and details of the 27 sq. degree survey. I alsodiscuss the difficulties in determining spectral typesfor objects in young, star-forming regions, due to thedegeneracy caused by interstellar reddening.

333.17 — The search for radio emission from exo-planets using LOFAR beam-formed observations

Jake D. Turner1,2; Jean-Mathias Griessmeier3,4; PhilippeZarka4,5

1 Astronomy, Cornell University (Ithaca, New York, United States)2 Astronomy, University of Virginia (Charlottesville, Virginia,

United States)3 LPC2E, Université d’Orléans/CNRS (Orleans, France)4 Station de Radioastronomie de Nançay, Observatoire de Paris

(Nançay, France)5 LESIA, Observatoire de Paris (Meudon, France)

The detection of exoplanetary magnetic fields is oneof the most elusive hunts in exoplanet science today.Observing the magnetic field of an exoplanet willgive valuable information to constrain their interiorstructure, atmospheric escape, and the nature of anystar-planet interactions. Additionally, the magneticfields on Earth-like exoplanets might help contributeto their sustained habitability by deflecting energeticstellar wind particles.

The most promising method to detect exoplanetmagnetic fields is radio emission observations sincethis method is not susceptible to false positives. Allthe magnetized planets and moons in our Solar Sys-tem emit in the radio using the Cyclotron Maser In-stability (CMI) mechanism. To date, many ground-based observations conducted to find exoplanet ra-dio emission have resulted in non-detections.

In this talk, we discuss our ongoing observationalcampaign searching for exoplanetary radio emis-sions using beam-formed observations with the Low

164

Band of the Low-Frequency Array (LOFAR). To datewe have observed three exoplanetary systems: 55Cnc, Upsilon Andromedae, and Tau Boötis. Theseplanets were selected according to theoretical predic-tions, which indicated them as among the best can-didates for an observation. Data analysis is currentlyongoing.

In order to test, validate, and quantify the sensi-tivity reached with our LOFAR pipeline, we apply itto a LOFAR observation of Jupiter’s magnetosphericradio emission in which the signal from Jupiter is at-tenuated. The idea is simple: we observe Jupiter,divide its signal by a fixed factor before adding itto an observation of sky background, thereby creat-ing an artificial dataset best described as “Jupiter asan exoplanet”. We then run our pipeline and checkwhether the (attenuated) radio signal from Jupiteris detected. The maximum factor by which we candivide Jupiter’s signal and still achieve a detectiongives the sensitivity of our setup. We find that circu-larly polarized exoplanetary radio bursts can be de-tected up to a distance of 20 pc assuming the level ofemission is 105 times stronger than the peak flux ofJupiter’s decametric burst emission.

400 — Atmospheres – I400.01 — The Galilean Satellites Observed byCassini: a testbed for icy terrestrial exoplanets

Laura Mayorga1; David Charbonneau1; DanielThorngren2

1 Center for Astrophysics | Harvard & Smithsonian (Cambridge,Massachusetts, United States)

2 University of California, Santa Cruz (Santa Cruz, California,United States)

For terrestrial exoplanets with thin atmospheres orno atmospheres, the surface will dominate to the re-flected light signal of the planet. Direct observationof the disk-integrated brightness of bodies in the So-lar System, and the variation with illumination an-gle, wavelength, and planetary longitude, is essentialfor both planning imaging observations of exoplan-ets and interpreting the eventual datasets. We willpresent our analysis of approximately 5,000 Cassiniobservations of the Galilean satellites through an ex-oplanet lens and show their longitudinal and illumi-nation variations. The data span a range of wave-lengths from 400-950 nm and predominantly phaseangles from 0-25 degrees with some constraining ob-servations near 120 degrees. Restricted to observa-tions at the same illumination angle, we show thatwe can clearly detect the spin period of each of the

four moons. We invert these light curves to recon-struct maps of the surfaces and we present compar-isons of these maps to direct images. In the case ofIo, we detect a clear color variation that can be tracedto geologic features with varying quantities of sul-fur compounds and silicates across the surface. De-spite the similarity in size and density between themoons, surface inhomogeneities result in significantchanges in the disk-integrated reflectivity with plan-etocentric longitude and phase angle. This impliesthat future exoplanet observations could exploit thiseffect to deduce surface variations, determine rota-tion periods, and potentially infer surface composi-tion. Furthermore, the Galilean satellites are all dis-tinctly non-Lambertian with steep phase functions,implying that icy exoplanets will be fainter than ex-pected at quadrature and more demanding to char-acterize by direct imaging.

L.C.M. is supported by the Harvard Future FacultyLeaders Postdoctoral fellowship.

400.02 — Prospects for Using H-α Transits to ProbeEscaping Atmospheres

Ruth Murray-Clay1; Mark Dijkstra11 Astronomy and Astrophysics, UC Santa Cruz (Santa Cruz, Cali-

fornia, United States)

The recently-observed dearth of super-Earths withradii ∼1.8 times that of Earth is a member of a rareclass of discoveries: observational confirmation ofa clear theoretical prediction. Two separate groupspredicted this feature using models of photoevapo-rative atmospheric loss. Since the discovery of thesuper-Earth radius gap, these models have been usedto constrain properties of this planetary populationsuch as the core mass distribution, with exciting re-sults. As the quantitative results of photoevapora-tion models become more important (and as photoe-vaporation is compared to core-powered mass loss,a new competing theory for the source of energydriving escape), improved observational constraintson photoevaporation models are sorely needed. Iwill present new results showing that, though tran-sits in hydrogen’s H-α line have thus far providedlimited information about escaping atmospheres, forthoughtfully-chosen planetary samples, this line hasexciting potential. Most direct observations of pho-toevaporation in action have been conducted usingtransits in hydrogen’s Lyman-α line. Because thisline’s center is obscured by ISM absorption, these ob-servations primarily provide information about out-flowing gas far from the planet, making mass lossrate calculations model dependent. For most planetscurrently observed to have escaping gas, the fraction

165

of escaping hydrogen in the n=2 state is too small forsignificant H-α absorption. I will present new the-oretical calculations showing that for planets expe-riencing a larger XUV flux, recombination cascadescan populate the n=2 state at an observable level. Iwill comment on these results in the context of A-stars such as Kelt 9, flaring M-stars, and young stars,and attempt to convince conference attendees thatadditional observational campaigns are warrantedin the conveniently accessible from the ground H-αline.

400.03 — A Sub-Neptune Exoplanet with a Low-Metallicity Methane-Depleted Atmosphere andMie-Scattering Clouds

Björn Benneke1; Heather Knutson2; Joshua Lothringer3;Ian Crossfield4; Julianne I. Moses5; Caroline Morley6;Laura Kreidberg7; Benjamin Fulton8; Diana Dragomir4;Andrew Howard2; Ian Wong4; Jean-Michel Desert9;Peter McCullough10; Eliza Kempton10; JonathanFortney11; Joshua Kammer12; Drake Deming10

1 University of Montreal (Montreal, Quebec, Canada)2 University of Maryland (College Park, Maryland, United States)3 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)4 Southwest Research Institute (San Antonio, Texas, United States)5 California Insitute of Technology (Pasadena, California, United

States)6 University of Arizona (Tucson, Arizona, United States)7 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)8 Space Science Institute (Boulder, Colorado, United States)9 Astronomy, University of Texas at Austin (Austin, Texas, United

States)10 Harvard University (Cambridge, Massachusetts, United States)11 NASA Exoplanet Science Institute / Caltech-IPAC (Pasadena,

California, United States)12 Anton Pannekoek Institute for Astronomy (API), University of

Amsterdam (UvA) (Amsterdam, Netherlands, Netherlands)

The discovery of thousands of exoplanets withmasses and radii intermediate between Earth andNeptune was one of the biggest surprises of ex-oplanet science. These super-Earths and sub-Neptunes likely represent the most common out-come of planet formation. Mass and radius measure-ments indicate a diversity in bulk composition muchwider than for gas giants; however, direct spectro-scopic detections of molecular absorption and con-straints on the gas mixing ratios have largely re-mained limited to planets more massive than Nep-tune. In this talk, we present the main results froman unprecedented HST/Spitzer data set (12 transitsand 20 eclipses) of a sub-Neptune exoplanet, whose

mass of 12.6 Earth masses places it near the half-way point between previously studied exo-Neptunes(22-23 Earth masses) and exoplanets known to haverocky densities (7 Earth masses). Obtained overmany years, our data set provides a robust detectionof water absorption (> 5 σ) and a thermal emissiondetection from the lowest irradiated planet to date.We reveal a low-metallicity, hydrogen-dominated at-mosphere similar to a gas giant, but strongly de-pleted in methane gas. The low, near-solar metal-licity (O/H=0.2-18) sets important constraints on thepotential planet formation processes at low massesas well as the subsequent accretion of solids. The lowmethane abundance indicates that methane is de-stroyed much more efficiently than previously pre-dicted, suggesting that the CH4/CO transition curvehas to be revisited for close-in planets. Finally, wealso find a sharp drop in the cloud opacity at 2-3 µmcharacteristic of Mie scattering, which enables nar-row constraints on the cloud particle size and makesthe planet a keystone target for mid-IR characteriza-tion with JWST.

400.04 — Meteorite Outgassing Experiments to In-form Chemical Abundances of Super-Earth Atmo-spheres

Maggie Thompson1; Myriam Telus2; Jonathan Fortney1;Toyanath Joshi3; David Lederman3

1 Department of Astronomy & Astrophysics, University of Califor-nia, Santa Cruz (Santa Cruz, California, United States)

2 Department of Earth & Planetary Sciences, University of Califor-nia, Santa Cruz (Santa Cruz, California, United States)

3 Department of Physics, University of California, Santa Cruz(Santa Cruz, California, United States)

At present, there is no first-principles understandingof how to connect a terrestrial planet’s bulk compo-sition to its atmospheric properties. Since terrestrialexoplanets likely form their atmospheres throughdegassing (Elkins-Tanton & Seager 2008), a logicalfirst step to build such a theory for super-Earths isto assay meteorites, the left-over building blocks ofplanets, by heating them to measure the outgassedvolatiles. Our Solar System presents a wide vari-ety of meteorite types, including chondrites whichare primitive unaltered rocks believed to be repre-sentative of the material that formed the rocky plan-ets. We present the current results of our meteoriteoutgassing experiments in which we heated a vari-ety of chondritic meteorite samples, at carefully con-trolled rates to temperatures from 200 to 1200 °C andmeasured the partial pressures and relative abun-dances of the outgassed volatile species (e.g., CO2,

166

H2O, CH4, H2, O2, S, Na) as a function of temper-ature and time. Our experimental set-up consistedof a residual gas analyzer connected to a furnace toheat samples at specified rates. We compare the re-sults of these experiments to Schaefer and Fegley’sprior theoretical chemical equilibrium and kineticscalculations which modeled thermal outgassing fora wide variety of chondrites to predict the composi-tion of terrestrial atmospheres formed via outgassingof specific types of meteorites (Schaefer & Fegley2007, Schaefer & Fegley 2010). In addition to testingand validating Schaefer and Fegley’s models, the re-sults from our experiments inform the phase space ofchemical abundances used in atmospheric models ofsuper-Earth exoplanets.

400.06 — Helium-Enhanced Planets at the Edge ofthe Radius Gap

Leslie Rogers1; Isaac Malsky11 Astronomy & Astrophysics, University of Chicago (Chicago,

Illinois, United States)

Primordial hydrogen-helium envelopes surround-ing sub-Neptune-sized planets are susceptible tomass loss driven by ionizing radiation from theirhost star. The effect of mass loss is imprinted onobserved exoplanet populations in the form of a”photo-evaporation desert” and a ”gap” at 1.6 EarthRadii in the planet radius distribution. To date, mod-els of the mass-loss evolution of exoplanets haveassumed that the planetary envelope compositionstays constant over time. However, after an initial∼0.1 Gyr phase of rapid hydrodynamic mass loss,sub-Neptunes may experience a subsequent phaseof thermal escape modulated by diffusive separationbetween hydrogen and helium wherein they gradu-ally become enhanced in helium and metals (relativeto hydrogen) over billions of years. We predict thatplanets on the large radius edge of the ”radius gap”in planet occurrence rates could be significantly en-hanced in helium (or depleted in hydrogen) relativeto solar composition. We have performed the firstself-consistent calculations of the coupled thermal,mass-loss, and compositional evolution of hydrogen-helium envelopes surrounding sub-Neptune massplanets. Our simulations consistently produce plan-ets with envelope helium mass fractions in excess ofY=0.5 (at planet ages of 5 Gyr) near the upper edgeof the radius gap. Our results have important impli-cations for the interpretation of atmospheric trans-mission and emission spectra of low-density sub-Neptune-size planets, which are prime targets foratmospheric characterization with HST and eventu-ally JWST. Enhancement in helium relative to hydro-

gen will affect both the scale height and equilibriumchemical abundances in the atmosphere (e.g., CO rel-ative to CH4). To date, most atmospheric retrievalanalyses have fixed the ratio of hydrogen and heliumto solar abundances; this assumption must now berelaxed. Our prediction further provides a new ob-servational test for the extent to which the radius gapis caused by atmospheric mass loss versus an intrin-sically bimodal outcome of planet formation.

401 — Population Statistics andMass-Radius Relations401.01 — Frequency of Gaseous Planets Beyond theIce Line

Benjamin Fulton1; Lee Rosenthal2; Andrew Howard2;Lea Hirsch3; Howard Isaacson4

1 NASA Exoplanet Science Institute / Caltech-IPAC (Pasadena,California, United States)

2 Astronomy, Caltech (Pasadena, California, United States)3 Physics, Stanford University (Burlingame, California, United

States)4 Astronomy, UC Berkeley (Berkeley, California, United States)

The occurrence distribution of long-period planetsis a key unknown left in our understanding of ex-oplanetary demographics. The presence of Jupiterand Saturn in our solar system likely facilitated theformation of Earth and the other terrestrial planets,yet we do not have a clear picture of the intrinsic fre-quency of gas giant planets orbiting beyond ∼5-10AU in exoplanetary systems. Here we analyze a mas-sive dataset of over 35,000 archival Keck/HIRES andLick radial velocities collected over the past 30 yearsfor a sample of 785 nearby stars. This groundbreak-ing study provides direct measurements of the mass,period, and eccentricity distributions of Jupiter andSaturn-mass planets beyond 5 AU and Neptune-mass planets out to the location of the ice line (∼3 AUfor sun-like stars). We are finally able to unify, com-pare and contrast planet population statistics fromdirect imaging surveys with those of radial velocityand transit surveys. This will have a profound im-pact on the planning of future direct imaging cam-paigns, potentially providing them with targets toobserve and yield predictions. We examine the pe-riod distribution of gas giant planets beyond 10 AUto look for the fall off in the planet occurrence raterequired to reconcile the tension between radial ve-locity studies and direct imaging studies. We mea-sure the mass distribution of Neptune to Jupiter massplanets orbiting beyond 3 AU to look for an enhance-ment in the frequency of sub-Jovian planets near the

167

ice line as is observed for Jovian and super-Jovianmass planets. We compare the mass-distribution ofplanets beyond 3 AU to those orbiting inside 1 AU tolook for the signatures of planet migration and shep-herding of the multitude of small, low-mass plan-ets discovered by Kepler. The results of this studywill have profound impact on the planning of fu-ture direct imaging campaigns, potentially provid-ing them with targets to observe and yield predic-tions. By connecting the regimes of exoplanetarydemographics explored by all of the various detec-tion techniques we provide a unified view of the endproduct of planet formation across the entire proto-planetary disk.

401.02 — Masses and Radii of Exoplanets from Ke-pler, K2, and TESS

David Winslow Latham1; Samuel N. Quinn11 Center for Astrophysics | Harvard & Smithsonian (Cambridge,

Massachusetts, United States)

The population of planets orbiting solar-type stars isdominated by planets smaller than 4 Earth radii, thesize of Neptune. Mass determinations for transitingplanets identified by Kepler and K2 have suggestedthat most planets smaller than about 1.8 Earth radiihave bulk densities that are consistent with inter-nal structures and compositions similar to the terres-trial planets in the Solar system. One of the primarygoals of the TESS mission is to determine masses andbulk densities for more than 50 planets smaller than4 Earth radii, to improve our understanding of thetransition between rocky planets and those more likeNeptune. We will highlight the progress towardsmeeting that goal.

401.03 — On the hunt for Trappist-1 siblings

Didier Queloz1; Michaël Gillon21 U. Cambridge (Cambridge, United Kingdom)2 U. Liege (Liege, Belgium)

The TRAPPIST-South 60cm telescope at La Silla(ESO) is famously known for its detection of the ex-traordinary TRAPPIST-1 planetary system. A dis-covery made during the prototype phase of ourultra-cool dwarf transit survey SPECULOOS (Searchfor Planets EClipsing ULtra-cOOl Stars). This talkwill first report on the self-consistence transit occur-rence analysis of all observations of 42 bright ultra-cool dwarfs made with TRAPPIST-South during aperiod ranging from 2011 to 2017. On the basis that,with the exception of the discovery of TRAPPIST-1

planets, we didn’t detect any other significant tran-siting event, we concluded on a 10% lower limit forthe occurrence of planets similar to TRAPPIST-1b inthis sample. The outcome is very sensitive to thesize and period of the planet considered. A com-prehensive statistic will be presented. Finally, per-formance obtained with our recently commissionedSPECULOOS Southern facility installed at Paranalwill be presented. The lower occurrence limit mea-sured with TRAPPIST survey will be compared withearly results from 6 months of continue SPECULOOScore survey operations

401.04 — ‘Oumuamua and DSHARP Point to 1011

Hidden Planets

Malena Rice1; Gregory Laughlin11 Astronomy, Yale University (New Haven, Connecticut, United

States)

The discovery of the first interstellar asteroid,‘Oumuamua, strongly suggests an enigmatic abun-dance of free-floating asteroids whose ejection intogalactic space is entirely unexplainable by the cur-rent population of known exoplanets. Remark-ably, it signals the existence of a vast undiscov-ered population of wide-separation (∼5+ AU) plan-ets of Neptune’s size or larger that are capable of di-rectly ejecting debris from their environments. TheALMA Disk Substructures at High Angular Reso-lution Project (DSHARP) recently returned 20 ultrahigh-resolution images of protoplanetary disks, il-lustrating a ubiquity of substructures among the pro-toplanetary disk population. We present new resultsdemonstrating for the first time that the planets sug-gested by radial gaps in the DSHARP sample are con-sistent with the undiscovered exoplanetary popula-tion required to produce a significant flux of inter-stellar objects. There are a number of compellingnear-term prospects for detecting the population ofplanets that we have inferred, with which we con-clude.

401.05 — Peas in a Pod: Planets in Kepler’s Multi-planet Systems are Similar in Size and RegularlySpaced

Lauren Weiss11 Institute for Astronomy, University of Hawaii at Manoa (Hon-

olulu, Hawaii, United States)

As part of the California Kepler Survey, we have es-tablished precise planet radii, semimajor axes, inci-dent stellar fluxes, and stellar masses for 909 plan-ets in 355 multi-planet systems discovered by Ke-

168

pler. In this sample, we find that planets within a sin-gle multi-planet system have correlated sizes: eachplanet is more likely to be the size of its neighbor thana size drawn at random from the distribution of ob-served planet sizes. In systems with three or moreplanets, the planets tend to have a regular spacing:the orbital period ratios of adjacent pairs of planetsare correlated. Furthermore, the orbital period ratiosare smaller in systems with smaller planets, suggest-ing that the patterns in planet sizes and spacing arelinked through formation and/or subsequent orbitaldynamics. The regular sizes and spacing of the Ke-pler planets are among the most common outcomesof planet formation, suggesting that our solar systemis not among the majority of planetary system ar-chitectures. New theories of planet formation mightbe required to reproduce the patterns in the Keplerplanetary systems.

401.06 — Planets do not show intra-system unifor-mity

Wei Zhu11 Canadian Institute for Theoretical Astrophysics (Toronto, Ontario,

Canada)

Recent works claim that planets in the same systemshould have similar properties (namely, mass and ra-dius) and be regular spaced. These patterns, if true,would have significant implications for theories ofplanet formation and evolution. In this talk, I willshow that this so-called intra-system uniformity canbe largely, if not entirely, explained by detection bi-ases. The universal signal-to-noise threshold that isused in Kepler transit detections corresponds to dif-ferent thresholds in planetary parameters for differ-ent stars. I will show that it is this variation in detec-tion threshold that is responsible for the majority ofthe claimed correlation in planet properties. With amore robust statistical approach, I find that the pres-ence and the size of the smaller planets is indepen-dent of the presence and size of their largest sibling,arguing against the so-called “intra-system unifor-mity.” Theoretical implications of these findings willalso be discussed.

401.07 — Kepler planets: a uniform population

Yanqin Wu11 University of Toronto (Toronto, Ontario, Canada)

In this talk, I will discuss recent progress regardingthe masses of Kepler planets, and how they seem tomake up a surprisingly uniform population. Thisuniformity is unpredicted, and challenges theories

of planet formation. As an aside, Kepler planets ap-pear to have interesting connections to the formationof stellar binaries.

402 — Planets in and around Bina-ries402.01 — The Role of Stellar Multiplicity in theFormation of Massive Close-In Giant Planets andBrown Dwarf Desert Members

Clémence Fontanive1; Ken Rice1; Mariangela Bonavita1;Eric Lopez2; Koralijka Muzic3; Beth Biller1

1 University of Edinburgh (Edinburgh, United Kingdom)2 NASA Goddard Space Flight Center (Greenbelt, Maryland,

United States)3 Universidade de Lisboa (Lisboa, Portugal)

Stellar multiplicity is believed to influence planetaryformation and migration, although the precise na-ture and extent of this role remain ambiguous. In thistalk, I will present new results from a survey aimedat testing the impact of stellar multiplicity on the for-mation and/or evolution of the most massive, close-in planetary and substellar companions, which areextremely challenging to explain with current the-oretical models. Using direct imaging observationsand the Gaia DR2 catalogue, we searched for wide bi-nary companions to stars hosting massive giant plan-ets or brown dwarfs (M > 7 MJup) on orbits shorterthan ∼1 AU. From a robust statistical analysis, we de-rived a very high binary fraction of ∼80% on separa-tions of 20–10,000 AU for our sample, twice as highas for field stars with a 3-σ significance. These resultsindicate that stellar companions greatly influence theformation or evolution of these systems. This binaryfrequency was also found to be larger than for lower-mass planets on similar orbits, suggesting that the ef-fects of binary companions become more importantfor higher-mass planets. Our survey thus demon-strates that binarity plays a crucial role in the exis-tence of very massive short-period giant planets andbrown dwarf desert inhabitants, almost exclusivelyobserved in multiple systems. These new resultsprovide vital information and constraints for boththeoretical studies of planet formation, and obser-vational campaigns of exoplanets and brown dwarfcompanions.

402.02 — The Mutual Inclinations of the Proto-Tatooine Disks

Ian Czekala1; Eugene Chiang1; Sean Andrews2; EricJensen3

169

1 Astronomy, UC Berkeley (Berkeley, California, United States)2 Radio and Geoastronomy, Center for Astrophysics | Harvard and

Smithsonian (Cambridge, Massachusetts, United States)3 Physics and Astronomy, Swarthmore College (Swarthmore, Penn-

sylvania, United States)

Efforts to learn about the efficiency of circumbi-nary planet formation from the dozen circumbinarysystems discovered by Kepler are hampered by thefact that the occurrence rate is degenerate with theunderlying mutual inclination distribution (i.e., thetypical misalignment between the binary and plan-etary orbital planes). In this talk, I will discuss oursurvey of 20 spatially-resolved circumbinary proto-planetary and debris disks, which includes severalspectroscopic binaries (P < 40 days) whose disks wehave resolved with ALMA for the first time. Cru-cially, these tight binaries are the only protoplane-tary systems that are comparable in scale to the Ke-pler circumbinary planet hosts. Using a hierarchi-cal Bayesian model, we infer that the disks aroundshort-period binaries are intrinsically coplanar (< 5°mutual inclination), implying that the occurrencerate of Kepler circumbinary planets is similar to thataround single stars (∼10% for planets with radii �[4,10] REarth). In stark contrast, however, the mu-tual inclinations of slightly wider binary systems (P> months) are much more varied: there are severalstrongly misaligned disks (> 40°), with even a fewcircumpolar disks (90°) around highly eccentric (e >0.7) binaries. We will demonstrate that a combina-tion of tight binary star formation mechanisms andhigh eccentricity oscillatory effects can explain theextremely strong trends of mutual inclination withbinary period and eccentricity.

402.03 — Circumbinary Planets at the K/T (Kepler-TESS) Boundary

William F. Welsh11 Astronomy, San Diego State University (San Diego, California,

United States)

Eight years ago at the ESS II conference, we pre-sented the first Kepler circumbinary planets. Sincethen, 11 planets in 9 systems have been discovered.In this talk we present the last two unpublished,unambiguous Kepler transiting circumbinary plan-ets: KOI-3152 and KIC 10753734. Both systems haveplanets with rapidly precessing orbits and signifi-cantly spotted stars which has made their charac-terization difficult. The orbital periods are 28.2 and19.4 days for the binaries, and 171 and 260 days forthe planets, respectively. In both systems three tran-sits were detected, yielding planetary radii of 3.4

and 6.0 Re. Like most of the circumbinary planets,these planets orbit near the binary (in)stability ra-dius. KOI-3152 is a particularly interesting case: itseclipses are extremely grazing (impact parameter >1) and thus its eclipse depths and widths are ex-tremely sensitive to changes in the orbital inclination.A change in inclination is in fact detected, and can bemodeled as the result of precession driven by a ∼140Me planet. But this mass is not consistent with theplanet’s 3.4 Re radius — something is wrong. Aftermuch scrutiny, we have solved the problem: The in-creasing eclipse depth is spurious — it is the resultof increasing starspot coverage that has led to an in-correct out-of-eclipse flux normalization. This well-understood (though ignored) bias is of course notparticular to binary stars. But the change in the biasis pernicious. We will put these two new planets incontext and briefly discuss what the ensemble of Ke-pler transiting circumbinary planets has to offer ex-oplanet science: the most accurately known massesand radii, challenges to understanding planet forma-tion and migration, revising the definition of the hab-itable zone, and why circumbinary planets are of-ten more suitable for life than single-star planets. Fi-nally, we close with a mention of how TESS is ex-pected to reveal hundreds of new circumbinary plan-ets via the ”one-two punch” discovery technique,and how we are on the cusp of transitioning fromdetailed characterization of a handful of planets tobeing able to carry out statistical studies of the cir-cumbinary planet population.

403 — Planets around WhiteDwarfs403.01 — Gas giant planets evaporated by hot whitedwarfs

Mathias Schreiber1,21 Institute for Physics and Astronomy, Universidad de Valparaiso

(Valparaiso, Chile)2 Millennium Nucleus for Planet Formation (NPF), Universidad de

Valparaiso (Valkparaiso, Chile)

All known exo-planet hosts stars will evolve intowhite dwarfs. It is well established that many whitedwarfs are accreting small planetary bodies, includ-ing asteroids and comets, indicating that planetarysystems survive, at least in part, the metamorphosisof their host stars. Gravitationally scattering plan-etesimals towards the white dwarf requires the pres-ence of more massive bodies, yet no planet has sofar been detected at a white dwarf. We discovereda moderately hot white dwarf that is accreting from

170

a circumstellar gaseous disc composed of hydrogen,oxygen, and sulphur. The composition of the disc isunlike all previously detected gaseous disks aroundwhite dwafs but resembles predictions for deeper at-mospheric layers of icy gas giants, with H2O andH2S being major constituents. We therefore suggestthat a gas giant orbiting the white dwarf is evap-orated by the strong EUV irradiation from the hotwhite dwarf. While still indirect, this discovery rep-resents the so far clearest evidence for the expectedexistance of gas giant planets around white dwarfs.We extend on this result by calculating the orbitalseparation at which gas giant planets will be evapo-rated by hot white dwarfs. We find that planets suchas the gas giant planets in our solar system are ex-pected to be located at such separations from the sunwhen it completed its metamorphosis into a whitedwarf.

403.02 — A planetesimal orbiting within the debrisdisc around a white dwarf star

Christopher Manser11 Physics, University of Warwick (Coventry, West Midlands,

United Kingdom)

I will present the first and so far only spectroscop-ically detected planetesimal around a white dwarf(Manser et al. 2019, Science, 364, 66). The bodyorbits within a disc of rocky debris on a period of2.06 hours, making it the closest planetary body to awhite dwarf. To withstand the tidal forces this deepin the gravitational field of the white dwarf this plan-etesimal must have some internal strength and/or ahigh density, and I discuss the possibility of it beingthe core of a differentiated planet that has partiallydisintegrated.

During the talk, I will highlight the methods usedto make the discovery, how I can apply this newlydeveloped method to several other white dwarf sys-tems, and how these extreme bodies could explainthe presence of gaseous debris discs at white dwarfs.

403.03 — The Abundances of Metals in Circumstel-lar Gas around Polluted White Dwarfs

Amy Samantha Steele1; John Debes3; Siyi Xu2; PatrickDufour4; Drake Deming1

1 University of Maryland (College Park, Maryland, United States)2 Gemini Observatory (Hilo, Hawaii, United States)3 Space Telescope Science Institute (Baltimore, Maryland, United

States)4 Université de Montréal (Montreal, Quebec, Canada)

Between 30 - 50% of white dwarfs (WDs) show heavyelements in their atmospheres. This ”pollution”

likely arises from the accretion of planetesimals thatwere perturbed by outer planet(s) into the whitedwarf’s tidal radius. A small fraction of these WDsshow either emission or absorption from circumstel-lar (CS) gas. For example, high resolution spectro-scopic observations of WD1145+017 reveal photo-spheric and CS absorption lines of elements heav-ier than helium in multiple transitions. The pho-tospheric abundances have been measured and aresimilar to the bulk composition of the Earth. The CScomponent arises from a gas disk produced throughthe sublimation of a transiting, disintegrating plan-etesimal. Models (to date) have not yet been ableto link the CS species to the total atomic abundancein gas. Here we present self-consistent models ofCS gas in orbit around various types of WDs anddemonstrate how we determine the abundances ofCS lines arising from planetesimals. Additionally,we build a grid of models to place constraints on thegas masses needed for detection of CS gas aroundvarious WDs with current observatories. We test thegrids using new discoveries and WDs with previ-ously known CS gas, in preparation for constrainingthe frequency of CS gas around statistical samplesof WDs. Knowing the abundances of CS gas aroundpolluted white dwarfs will provide a key to under-standing the instantaneous composition of the mate-rial accreting onto the photosphere and will allow adirect comparison to the composition of rocky bod-ies in the Solar System.

404 — Atmospheres – II404.01 — The Most Extreme Case of AtmosphericEscape Detected on the Warm Neptune GJ 3470bwith HST

Vincent Bourrier11 Geneva Observatory, University of Geneva (Versoix, Switzerland)

Observations of exoplanets during the transit of theirhost star allow probing the structure and composi-tion of their atmosphere. The intense stellar energyinput into exoplanets orbiting close to their star canlead to a dramatic expansion of their upper atmo-sphere, and the ’evaporation’ of large amounts ofgas into space. UV observations of hot Jupiters re-vealed the extended exospheres formed by this es-caping gas, and showed that these planets are toomassive to lose a substantial fraction of their atmo-sphere. Lower-mass planets are expected to be muchmore sensitive to evaporation, which has long beenthought to play a role in forming the desert of hotNeptunes (a deficit of Neptune-size exoplanets on

171

very short orbits). I will present the discovery of agiant hydrogen exosphere around GJ3470b, a warmNeptune located at the border of the desert. Thisis the first UV result of the Panchromatic Compar-ative Exoplanet Treasury (PanCET) survey, a Hub-ble program targeting 20 exoplanets across the entirespectrum. Our numerical simulations of the resolvedexospheric transit show that GJ3470b is subjected tomass losses comparable to that of hot Jupiters, mak-ing it the most extreme case of evaporation observedto date. GJ3470b could already have lost up to 40%of its mass over its 2 Gyr lifetime, bringing directobservational confirmation that evaporation shapedthe population of close-in exoplanets. I will com-pare GJ3470b with other known evaporating plan-ets and discuss the reasons for its dramatic escape.Our results strengthen the interest of observing theupper atmosphere of exoplanets to determine theirproperties and understand how they depend on theirpast evolution. This is particularly important forsuper-Earth and Earth-size planets, whose lower at-mosphere could be hidden by clouds. The devel-opment of new tracers of atmospheric escape at op-tical/infrared wavelengths opens thrilling perspec-tives for the characterization of exoplanets via theirupper atmosphere.

404.02 — A Novel New Method for MeasuringWindspeeds on Exoplanets and Brown Dwarfs

Katelyn Allers1; Johanna Vos2; Peter K. G. Williams3;Beth Biller4

1 Physics and Astronomy, Bucknell University (Lewisburg, Penn-sylvania, United States)

2 American Museum of Natural History (New York, New York,United States)

3 Harvard-Smithsonian Center for Astrophysics (Boston, Mas-sachusetts, United States)

4 Institute for Astronomy, University of Edinburgh (Edinburgh,United Kingdom)

Within our solar system, we can directly observe theeffects of rapid rotation on the atmospheric physicsof the giant planets. Zonal winds, a result of rapid ro-tation and convection, play an important role in thebulk atmospheric flow. This, in turn, can impact at-mospheric chemistry, as evidenced by Jupiter’s dis-equilibrium PH3, which dominates its mid-IR spec-trum. Similar to Jupiter and Saturn, recent studiesreveal that many brown dwarfs and directly-imagedexoplanets are also fast rotators with evolving atmo-spheric inhomogeneities. The effects of rotation andconvection are starting to be included in efforts tomodel the atmospheric dynamics of brown dwarfs

and exoplanets. The resulting predictions of windspeed, however, remain relatively untested.

We present the first results of a novel new methodfor measuring wind speeds on exoplanets and browndwarfs. Utilizing a combination of radio obser-vations and infrared photometric variability, wepresent the first observational constraints on windspeed for a cool, cloudless brown dwarf. We discussthe implications of our measurement for models ofatmospheric circulation. Looking to the future, wediscuss the ways in which new observational facil-ities could extend our method of wind speed mea-surement to other brown dwarfs and exoplanets.

404.03 — Exoplanet atmospheres at high spectralresolution with CARMENES

Enric Palle11 Investigacion, Instituto de Astrofisica de Canarias (La Laguna,

Spain)

Transmission spectroscopy using high-resolutionspectrographs is quickly becoming a major toolto detect and understand planetary atmospheres,from ultra hot Jupiters to Neptunes-size planets.The CARMENES spectrograph started operations in2016, and since then we have been using it for thestudy of planetary atmospheres taking advantage ofit simultaneous wavelength coverage from visible tonear-infrared (0.5-1.7 micron). This has led to sev-eral innovative results, including the first ground-based detections of the He I triplet, allowing thestudy of exoplanetary tales and scape ratios, or thedetection for the first time of the Ca triplet (togetherwith FeII, Na I, and the Balmer series of Hα, Hβ, andHγ) in the atmosphere of the ultra hot Jupiter (UHJs)MASCARA-2b/KELT20-b. In this talk we will up-date the several He detections on a sample of abouta dozen planets, including various levels of stellar ir-radiation and planetary masses. I will also discussCARMENES’s capabilities for the characterization ofUHJs atmospheres, where our results are consistentwith theoretical models, predicting a rich day-sideionosphere.

For the SOC: The CARMENES team as a large sam-ple of exoplanets observed (some published somenot yet), which can for the first time provide correla-tions between stellar irradiation and the detectabilityof He I triplet and Hα lines, in some cases both, use-ful for both observers and modelers. I can review theoverall results, currently in preparation for a numberof publications. For UHJ atmospheres, we can de-tect several atmospheric species with both trasnmis-sion spectroscopy and cross-correlation techniquessimultanoeuly, again for a sample of hot planets. So

172

I think this talk will nicely summaryze several re-sults in two important hot topics of exoplanet atmo-spheres.

404.04 — Unlocking the Hidden Secrets of HotJupiter Atmospheres through Near-UltravioletSpectroscopy: A Case Study of HAT-P-41b

Nikole Lewis1; Hannah Wakeford2; Ishan Mishra1;David Sing3; Natasha Batalha4; Mark Marley5; NorPirzkal2; Thomas Evans6; Jayesh Goyal7; GregoryHenry8; Tiffany Kataria9; Nikolay Nikolov3; JessicaSpake7; Kevin Stevenson2

1 Astronomy, Cornell (Ithaca, New York, United States)2 STScI (Baltimore, Maryland, United States)3 JHU (Baltimore, Maryland, United States)4 UCSC (Santa Cruz, California, United States)5 NASA Ames (Mountain View, California, United States)6 MIT (Cambridge, Massachusetts, United States)7 Exeter (Exeter, United Kingdom)8 Tennessee State (Nashville, Tennessee, United States)9 JPL/Caltech (Pasadena, California, United States)

Near-Ultraviolet (NUV, 200-400 nm) spectra of plan-ets hold rich information about the chemistry andphysics at work in their upper atmospheres. In thesolar system, NUV spectroscopy has been criticalin identifying and measuring the abundances of avariety of hydrocarbon and sulfur-bearing species,produced via photochemical mechanisms, as well asoxygen and ozone. To date, less than 20 exoplanetshave been probed in this critical wavelength range,with mixed results, limited by the wavelength cov-erage and sensitivity of the workhorse instrumentfor such studies, HST’s STIS G430L and E230M grat-ings. In HST Cycle 25, our team embarked on a jour-ney to explore the potential of HST’s WFC3/UVISG280 grism, which offers the highest throughput ofall HST’s instruments in the NUV and is up to 25times more sensitive than its STIS counterparts at 350nm. The WFC3/UVIS G280 grism does offer onechallenge, the presence of overlapping spectral or-ders similar to those of JWST’s NIRISS instrument,which required us to develop new data reductionand analysis techniques. The first target to be ex-plored with this newly unlocked mode on HST wasthe hot Jupiter HAT-P-41b, which had been previ-ously observed with HST’s STIS G430L grism. Ourhigh-precision spectrum of HAT-P-41b, which com-bines information from both the positive and neg-ative spectral orders, has revealed features in theNUV that cannot be explained by standard equilib-rium chemical models, the presence of aerosols, orstellar activity. Drawing on solar system and stel-lar studies, we considered dozens chemical species

that are known to absorb strongly at NUV wave-lengths. Through detailed atmospheric modelingand retrieval analyses we have uncovered not yetconsidered chemistry and physics at work in the at-mosphere of HAT-P-41b, which is likely present inmany exoplanet atmospheres. In this talk I will de-tail the opportunities that have been opened up withthe HST WFC3/UVIS G280 grism in the explorationof exoplanet atmospheres and reveal the once hiddensecrets of HAT-P- 41b’s atmosphere.

404.05 — New Theoretical Models for Cloudy Sub-stellar Atmospheres

Caroline Morley1; Mark Marley2; Didier Saumon31 Astronomy, University of Texas at Austin (Austin, Texas, United

States)2 NASA Ames Research Center (Mountain View, California, United

States)3 Los Alamos National Laboratory (Los Alamos, New Mexico,

United States)

Ample evidence suggests that exoplanets of all kindshave clouds, likely made of many different materi-als from refractory minerals, to silicate dust, to salts,to volatile ices. Clouds are complex to model andchallenging to understand from limited observationsof exoplanets. Fortunately, planet-mass free-floatingobjects provide a key venue for understanding cloudformation in substellar atmospheres. These objectshave the temperatures of planets but, critically, lacka nearby star, making high signal-to-noise, high pre-cision measurements possible. To understand thephysics and chemistry of these atmospheres, weneed to compare these high fidelity observed spec-tra to state-of-the-art models. Recent improvementsto the ingredients in substellar atmosphere modelsinclude new line lists for various important species(methane, alkali metals, water, etc.), as well as up-dated chemistry calculations for a range of metallici-ties and carbon-to-oxygen ratios. Here, we presenta new set of substellar atmosphere models for ob-jects warmer than 1000 K including clouds. Weshow how these models differ from previous cloudybrown dwarf models (Saumon & Marley 2008), anddemonstrate how metallicity affects cloudy substel-lar spectra. We present results comparing thesemodels to field brown dwarfs, free-floating plan-ets, and directly-imaged companions, demonstrat-ing how gravity changes cloud properties and emer-gent spectra. Finally, we present a new technique forunderstanding the compositions and mineralogy ofclouds in brown dwarfs using mid-infrared spectro-scopic time-series measurements with JWST. Thesemodels will publicly available and provide a critical

173

tool for the community in the lead-up to the launchof JWST.

404.06 — Infrared Eclipses and Transits of the BestTESS Planets

Ian Crossfield1; Laura Kreidberg2; Diana Dragomir3;Björn Benneke4; Michael W. Werner5; Drake Deming7;Varoujan Gorjian5; Xueying Guo1; Courtney Dressing8;Liang Yu1; Stephen Kane6; Jessie Christiansen5; DavidBerardo1; Farisa Morales5

1 Physics, MIT (Cambridge, Massachusetts, United States)2 Harvard University (Cambridge, Massachusetts, United States)3 MIT/UNM (Cambridge, Massachusetts, United States)4 U. Montreal (Montreal, Quebec, Canada)5 JPL (Pasadena, California, United States)6 UC Riverside (Riverside, California, United States)7 U. Maryland (College Park, Maryland, United States)8 UC Berkeley (Berkeley, California, United States)

A key TESS goal is to identify the best exoplanet tar-gets for atmospheric study. We will report on initialresults from our large-scale Spitzer program to fol-low up TESS planets with mid-infrared transits andeclipses. Spitzer’s unparalleled infrared sensitivityand photometric stability are allowing us to refinethe properties of these new planets and ensure thattheir transits and eclipses can be recovered for manyyears to come — e.g., with HST and JWST. Our pro-gram focuses on the smaller (i.e., sub-Jovian) planetsfor which ground−based observations are impracti-cal and for which JWST spectroscopy will have a highimpact. Our most exciting results will include theonly secondary eclipse measurements of these sub-Jovian planets until JWST launches. Our program in-cludes eclipse observations of planets from 1–6 Earthradii and equilibrium temperatures from 800-2500K. In addition, we are producing some of the onlythermal phase curves known for such planets. Ourprogram is providing the touchstone sample of TESSplanets that will be studied in great detail for manyyears to come.

404.07 — Peering into the formation history of BetaPic b with long-baseline interferometry

Mathias Nowak1; Sylvestre Lacour2; Paul Mollière4; Ja-son Wang3; Benjamin Charnay2

1 Observatoire de Paris (Meudon, France)2 LESIA, Observatoire de Paris (Meudon, France)3 Astronomy, California Institute of Technology (Pasadena, Califor-

nia, United States)4 Leiden Observatory (Leiden, Netherlands)

Beta Pictoris is arguably the best-known stellar sys-tem outside of our own. 30 years of study have re-vealed a highly structured circumstellar disk withrings, belts, and a giant planet. But very little isknown about how it came into being. In particu-lar, the giant planet beta Pictoris b is known to haveplayed a crucial role in the structuring of the sys-tem, but its formation history remains elusive, de-spite some attempts to settle the cold / hot startquestion. I will present the first interferometric ob-servations of the giant planet Beta Pic b, obtainedwith GRAVITY, on the combined four 8.2 m tele-scopes of the VLTI. These observations resulted inthe cleanest (S/N > 50), medium resolution (R=500),K-band spectrum of a giant planet ever obtained. Iwill show that this spectrum, combined with exist-ing low-resolution data, can be used to estimate theplanetary C/O ratio, which in turn can be used totrace down the formation history of the planet. Inparticular, I will present two interpretations of thelow C/O ratio obtained, one in the gravitational coll-pase formation paradigm for planet formation, andone in the core-accretion paradigm.

500 — Atmospheres – III500.01 — Non-Equilibrium Chemistry of theCoolest Brown Dwarfs: Implications for DirectlyImaged Exoplanets

Brittany Miles1; Andy Skemer5; Caroline Morley2; Kate-lyn Allers3; Jacqueline Faherty6; Thomas Geballe7; MarkMarley8; Jonathan Fortney4; Michael Cushing9; AdamSchneider10; Gordon Bjoraker11

1 Astronomy and Astrophysics, UC Santa Cruz (SANTA CRUZ,California, United States)

2 Arizona State University (Tempe, Arizona, United States)3 NASA Goddard (Greenbelt, Maryland, United States)4 Astronomy, University of Texas at Austin (Austin, Texas, United

States)5 Bucknell University (Lewisburg, Pennsylvania, United States)6 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)7 Astronomy, UC Santa Cruz (Santa Cruz, California, United

States)8 Department of Astrophysics, American Museum of Natural His-

tory (New York, New York, United States)9 Gemini Observatory (Hilo, Hawaii, United States)10 NASA Ames (Mountain View, California, United States)11 University of Toledo (Toledo, Ohio, United States)

The Y-dwarf spectral class is composed of only ∼23brown dwarfs with effective temperatures below 450K and atmospheres rich in gaseous methane, ammo-

174

nia, and water. Y-dwarfs are colder than any currentdirectly imaged exoplanet, offering previews into theatmospheric physics of gas giants that will be dis-covered and characterized in the future with JWSTand SCALES. The coldest brown dwarf WISE 0855(250 K) showed water absorption and evidence ofclouds across its M-band (4.5 – 5.0) spectrum. In ad-dition to this, WISE 0855 is also only ∼100 K hot-ter than Jupiter, yet its lack of phosphine absorp-tion means that it has significantly different atmo-spheric mixing properties than Jupiter. This differ-ence implies that there could be a large degree ofvariation in atmospheric physics at extremely cooltemperatures. In this work, we expand the sam-ple of cool brown dwarf M-band spectra to coverthe temperature range of 250 K – 700 K by tak-ing low resolution Gemini/GNIRS spectra of 4 T/Y-dwarfs (50 hours, WISE 1541, WISE 2056, WISE 0313,UGPS 0722) and placing them in context to previ-ously published spectra (WISE 0855, 2MASS 0415,Gl 570D). With the exception of Gl 570D, cloud-free,solar metallicity brown dwarf models do not accu-rately fit the spectral slopes of our sample and betterfits are achieved when non-equilibrium abundancesof carbon monoxide are added into the atmosphericmodels. Atmospheric mixing can bring up warmercarbon monoxide gas and our sample suggests thatmixing becomes stronger at cooler effective temper-atures. We discuss why these types of atmosphericanalyses are essential for 1) planning and interpret-ing higher quality JWST Y-dwarf observations and 2)predicting what may be observable on cooler, low-gravity gas giant exoplanets.

500.02 — New signatures of Planet Formation Sce-nario in Gas Giant Exoplanet Atmospheres

Jean-Michel Desert1; Lorenzo Pino11 University of Amsterdam (UvA) (Amsterdam, Netherlands)

We present a portfolio of observational projects forwhich we have recently succeeded in measuring theatmospheric composition and metallicities of gas gi-ant exoplanets. We study the atmospheric propertiesof giant exoplanets over a broad range of masses andequilibrium temperatures, and retrieve their theirchemical and dynamical properties. We introducenovel observational techniques and diagnostics toprobe exoplanet atmospheres and climates of tidallylocked planets; we measure their composition at dif-ferent longitudes, from their dayside to their night-side, and with multiple spectral resolution. Theseprojects use diverse and complementary approachesto retrieve atomic and molecular abundances and

the atmospheric metallicity, including new metallic-ity tracers (e.g. alkali metals, iron, hydrides). Wepresent the breakthrough results but also the mainchallenges to overcome while making these mea-surements. When combined together, our findingson the measurements of atmospheric metallicity al-low us to make quantitative comparisons amongstexoplanets, and with the Solar System planets, in or-der to test predictions from planet formation models.The first results from these projects offer the tanta-lizing suggestion that some of the trends seen in theSolar System are also seen in extrasolar systems, butothers are not. Ultimately, I will present the implica-tion of these new results in the context of exoplanets,and shed light on our understanding of exoplanets’formation, evolution and architectures.

500.03 — Heavy and rare-Earth metals in the trans-mission spectra of Ultra Hot Jupiters

Jens Hoeijmakers1,2; David Ehrenreich1; DanielKitzmann2; Romain Allart3; Simon Grimm2; Julia Vic-toria Seidel4; Aurelien Wyttenbach7; Lorenzo Pino5;Louise Nielsen1; Chloe Fisher2; Paul Rimmer8; VincentBourrier1; Heather Cegla1; Baptiste Lavie1; ChristopheLovis1; Beate Patzer9; Joachim Stock6; Francesco Pepe1;Kevin Heng2

1 Geneva Observatory, University of Geneva (Versoix, Switzerland)2 Center for Space and Habitability, University of Bern (Bern,

Switzerland)3 Geneva observatory, University of Geneva (Versoix, Switzerland)4 Astronomy, University of Geneva (Versoix, GENEVA, Switzer-

land)5 Anton Pannekoek Insitute for Astronomy, Universiteit van Ams-

terdam (Amsterdam, Netherlands)6 City University of New York (Brooklyn, New York, United States)7 Leiden Observatory (Leiden, Netherlands)8 Cambridge University (Cambridge, United Kingdom)9 Technische Universität Berlin (Berlin, Germany)

Ultra-hot Jupiters form a new class of exoplanets thattend to orbit hot early type stars in short periods. Thefirst Ultra-hot Jupiter known to exist is KELT-9 b: Amassive gas giant heated to a temperature of over4,000K on the day-side by its 10,000K A-star. Theextreme temperature dissociates all but traces of themost strongly bound molecules (CO and H2O) intotheir constituent atoms. A significant fraction of theatomic gas is thermally ionised. Under these circum-stances, line absorption lines by metals and contin-uum absorption by the hydrogen anion are the dom-inant sources of opacity. Clouds and aerosols are no-tably absent, and the timescales of chemical reactionsare much shorter than those of mixing and photo-ionisation over most of the atmosphere at the day-

175

side and terminator regions. This means that the in-terpretation of the transmission spectrum of KELT-9b should be greatly simplified compared to planetsfor which aerosols and non-equilibrium chemistryare important.

We recently performed a survey of the transmis-sion spectrum of KELT-9 b as observed with the high-resolution HARPS-N spectrograph and discoveredmultiple heavy metals, including iron, chromium,scandium and yttrium. We find that the absorptionlines of neutral iron are almost perfectly predicted bya model of an isothermal atmosphere in hydrostaticand chemical equilibrium, but that all other lines areanomalously strong, indicating strong atmosphericinflation below millibar pressures.

We have since started analyses of high-resolutionobservations of a number of other, cooler hotJupiters. During my talk I will announce new, strongdetections of neutral and ionised metals in these at-mospheres. The discovery of multiple metals in mul-tiple planets at high confidence demonstrates thatdetailed chemical analyses of ultra-hot Jupiter atmo-spheres are indeed possible, and that the communitycan move beyond the notion of metallicity, but in-stead study the chemistry of individual trace metals.In addition, these spectral lines are powerful tools toconstrain atmospheric dynamics and structure.

500.04 — Detection of planetary rotation and astrong East wind on a ultra-hot gas giant withESPRESSO at the Very Large Telescope

David Ehrenreich11 Department of Astronomy, University of Geneva (Versoix, GE,

Switzerland)

The stellar radial velocity anomaly measured dur-ing an exoplanet transit (the Rossiter-McLaughlin ef-fect) allows to constrain the orbital architecture ofa transiting planetary system and resolve the sur-face velocity structure of the transited star. Thanksto the exquisite radial velocity precision of thenew ESPRESSO high-resolution spectrograph at theVery Large Telescope, we show that the Rossiter-McLaughlin effect of a highly-irradiated gas giantcan also be used to diagnostic the existence of aultra-hot atmosphere. Removing the stellar signa-ture (known as the Doppler shadow) reveals anotherDoppler signature moving along with the planet,which we attribute to the planetary atmosphere.Since this absorption signal is obtained throughthe cross-correlation of the spectra with a stellarmask, the exoplanet must contain atomic iron, themain component of stellar cross-correlation masks.The cross-correlation function of the planet appears

slightly redshifted at the beginning of the transit andbecomes strongly blue-shifted as the East (evening)limb enters the stellar disc, indicating a strong asym-metry in the wind velocity between the two limbs.This is seen consistently at two different epochs ofobservation. We measure the average full-width athalf maximum (FWHM) of the planetary signal andobtain the projected rotational velocity of the planet,which we compare to the expected tidally-locked ro-tation of the close-in exoplanet. ESPRESSO startedscience operations at the VLT on September 2018 andthis is the first scientific result from the consortiumthat built the instrument. It reveals ESPRESSO, anESO instrument open to the Community, as an out-standing characterization machine for exoplanetaryatmospheres.

500.05 — The Global Climates, Clouds, and Dy-namics of the Hottest Jupiters

Thomas Beatty1; Taylor James Bell2; Karen Collins5;Nick B. Cowan2; Lisa Dang2; Drake Deming7; JonathanFortney4; B. Scott Gaudi3; Tiffany Kataria6; DylanKeating2; Joshua Lothringer1; Avi Mandell8; TamaraRogers9; Adam Showman1; Keivan Stassun10

1 University of Arizona (Tucson, Arizona, United States)2 Vanderbilt University (Nashville, Tennessee, United States)3 Physics, McGill University (Montreal, Quebec, Canada)4 Ohio State University (Columbus, Ohio, United States)5 Astronomy and Astrophysics, University of California, Santa

Cruz (Santa Cruz, California, United States)6 Harvard CfA (Cambridge, Massachusetts, United States)7 JPL/Caltech (Pasadena, California, United States)8 University of Maryland (College Park, Maryland, United States)9 NASA Goddard (Greenbelt, Maryland, United States)10 Newcastle University (Newcastle, United Kingdom)

The atmospheres of ultra-hot Jupiters (>3000K) ex-ist in an extreme state of day-night disequilibrium,giving us one of the best opportunities to study thedynamic processes in giant planet atmospheres. Byusing orbital phase curve observations of these plan-ets we can construct a global map of their thermalemission, and watch their atmospheres change asgas moves from day to night and back again. Be-sides providing us with a better understanding ofhot Jupiter atmospheres, this also allows us to studywhat would otherwise be observationally inaccessi-ble atmospheric processes. Specifically, we can seethe formation of clouds near planetary dusk, theirdestruction shortly after dawn, and use this to con-strain the physics of cloud formation and dissolu-tion in both exoplanets and brown dwarfs. We canalso use phase curve observations of extremely hot,

176

cloudless, planets to directly trace the underlying dy-namics and mixing in these atmospheres.

We will illustrate this using new results from ournew HST/WFC3 phase curves of KELT-1b and newdual-band Spitzer phase curves of KELT-9b. ForKELT-1b, our spectroscopic phase curves of KELT-1b show — for the first time — the broadband andspectroscopic signatures of the formation and break-up of these nightside clouds. Coupled with pre-vious broadband Spitzer phase curve observations,this gives us new insight into the cloud formationtimescales and cloud compositions on hot Jupiters.

We will also present new 3.6um and 4.5um Spitzerphase curves of KELT-9b, which, at 4600K, is thehottest giant planet known. Unlike all other hotJupiters, KELT-9b shows a strongly non-sinosoidalphase curve in both Spitzer bands. Also unlike allother hot Jupiters, the extreme temperature of KELT-9b’s atmosphere means that it is completely cloud-less, and we will discuss how the phase curve varia-tion we see is driven by atmospheric dynamics. Thehigh temperature also causes molecular hydrogen todissociate on the dayside, and by using the strongopacity difference between H and H2 in the twoIRAC channels, we can use the hydrogen dissocia-tion / recombination reaction as a direct tracer of theatmospheric gas dynamics.

500.06 — A Window into Planetary Magnetismwith Exo-Aurorae

Melodie Kao1; Evgenya L. Shkolnik1; J. SebastianPineda2

1 School of Earth and Space Exploration, Arizona State University(Tempe, Arizona, United States)

2 Laboratory for Atmospheric and Space Physics, University ofColorado Boulder (Boulder, Colorado, United States)

Planetary magnetic fields influence atmosphericevaporation from space weather, yield insights intoplanet interiors, and are essential for producing au-rorae. The most direct way of measuring magneticfields on exoplanets is by observing exo-aurorae atradio frequencies. Our discovery of the first ra-dio exo-aurora on the ∼12.7 MJ brown dwarf SIMPJ01365662+0933473 marks the beginning of an erafor directly probing magnetism at planetary masses.Low-frequency radio arrays such as the Owens Val-ley LWA and the Square Kilometre Array will soonbe sensitive to exoplanet aurorae, providing a newmeans of exoplanet detection and characterization.Now is a critical time to prepare for these up-coming searches by harnessing detailed studies ofexo-aurorae on observationally accessible exoplanet

analogs, planetary-mass brown dwarfs. I will syn-thesize the state of the art for searches of browndwarf exo-aurorae, including our new results froma survey of young, planetary-mass objects and thedeepest study to date of an 11-12 MJ brown dwarf.I will discuss implications for and highlight oppor-tunities to probe exoplanet magnetism with the nextgeneration of ground- and space-based radio facili-ties.

501 — Disks501.01 — Protoplanetary disks and their dynamichost stars

Catherine Espaillat11 Boston University (Boston, Massachusetts, United States)

Protoplanetary disks play a key role not only in un-derstanding planet formation, but also in unlockingthe fundamental physics of transport processes giventhat these are some of the closest astrophysical ac-cretion disks. The young stars hosting these disksare known to be remarkably variable, but it is notclear how the variable high-energy radiation fieldsof a young star impact the disk and hence planetformation and accretion processes. In order to gaininsight to these dynamic systems, multi-epoch andmulti-wavelength datasets are necessary. I presentthe first near-simultaneous multi-epoch X-ray, ultra-violet, optical, infrared, and radio observations of anaccreting, young star. These data reveal the first ob-servational evidence of the star-disk-jet connection.The observations show that an increase in the surfacedensity in the inner disk resulted in more mass load-ing onto the star and therefore a higher accretion rateonto the star, which led to a higher mass-loss rate inthe jet. This suggests a linked origin, presumably thestellar magnetic field, which can both channel mate-rial onto the star as well as eject it in collimated jetsalong twisted field lines. This showcases the pos-sibilities for future progress in time-domain studiesand emphasizes the importance of coordinated mul-tiwavelength work of these dynamic young systems.

501.02 — Large total masses and small amounts ofCO gas in protoplanetary disks

Diana Powell1; Ruth Murray-Clay1; Laura Perez2; HilkeE. Schlichting3,4; Peter Gao5; Michael Rosenthal1

1 UC Santa Cruz (Santa Cruz, California, United States)2 Universidad de Chile (Santiago, Chile)3 UC Los Angeles (Los Angeles, California, United States)4 Massachusetts Institute of Technology (Cambridge, Mas-

sachusetts, United States)

177

5 UC Berkeley (Berkeley, California, United States)

The total mass in protoplanetary disks and the C-to-O ratio of solids and gas are critical initial conditionsfor understanding the speed of planet formation andplanetary composition after formation. Several re-cent studies show that the standard assumptions forboth of these quantities are likely incorrect. I will re-port on our new set of models that reconcile theorywith observations of protoplanetary disks and cre-ate a new set of initial conditions for planet forma-tion models. This modeling makes use of recent re-solved multiwavelength observations of disks in themillimeter to constrain the aerodynamic propertiesof dust grains, allowing us to infer total disk masswithout an assumed dust opacity or tracer-to-H2 ra-tio. The 7 disks modeled using this method thus farare close to the limit of gravitational stability at cer-tain radii and raise the possibility that all disks aremore massive than has been previously appreciated.This qualitative change to the initial conditions ofplanet formation has sweeping implications. I willpresent new, unpublished work that combines themicrophysics of cloud formation in planetary atmo-spheres and our new models of protoplanetary disksto show that the observed depletion of CO in TW Hyais consistent with freeze-out processes and that thevariable CO depletion observed in disks can be ex-plained by the processes of freeze-out and particledrift. This work both solves an outstanding problemin observations of protoplanetary disks and robustlyconstrains the C-to-O ratio in gas and solids availablefor planet formation.

501.03 — Kinematic detection of embedded proto-planets

Christophe Pinte1,2; Gerrit van der Plas2; FrancoisMenard2; Daniel Price1; Valentin Christiaens1; TraceyHill3; Gaspard Duchene4; Daniel Mentiplay1; Chris-tian Ginski5; Elodie Choquet6; Yann Boehler2; SebastianPerez7; Simon Casassus7; Bill Dent3

1 Monash university (Clayton, Victoria, Australia)2 IPAG, Grenoble (Grenoble, France)3 ALMA (Santiago, Chile)4 Berkeley (Berkeley, California, United States)5 UVA (Amsterdam, Netherlands)6 LAM (Marseille, France)7 Universidad de Chile (Santiago, Chile)

We still do not understand how planets form, or whyextra-solar planetary systems are so different fromour own solar system. Recent observations of pro-toplanetary discs have revealed rings and gaps, spi-rals and asymmetries. These features have been in-

terpreted as signatures of newborn protoplanets, butthe exact origin is unknown, and remains poorly con-strained by direct observation. In this talk, we showhow high spatial and spectral resolution ALMA ob-servations can be used to detect embedded planet intheir discs. We report the kinematic detections ofJupiter-mass planets in the discs of HD 163296 andHD 97048. For HD 97048, the planet is located ina gas and dust gap. An embedded planet can ex-plain both the disturbed Keplerian flow of the gas,detected in CO lines, and the gap detected in the dustdisc at the same radius. While gaps appear to be acommon feature in protoplanetary discs, we presenta direct correspondence between a planet and a dustgap, indicating that at least some gaps are the resultof planet-disc interactions.

501.04 — ALMA Reveals a Misaligned Inner GasDisk inside the Large Cavity of a Transitional Disk

Satoshi Mayama1; eiji akiyama3; olja panic4; JamesMiley4; Takashi Tsukagoshi5; takayuki muto6; RuobingDong2; Jerome Pitogo De Leon7; toshiyuki mizuki8; ohdaehyeon9; jun hashimoto10; jinshi sai7; thayne currie5;michihiro takami11; carol A. Grady12; masahiko hayashi5;Motohide Tamura7; Shu-ichiro Inutsuka13

1 SOKENDAI(The Graduate University for Advanced Studies)(Kanagawa, Japan)

2 Astrobiology Center (Tokyo, Japan)3 Academia Sinica (Taipei, Taiwan)4 nasa (Greenbelt, Maryland, United States)5 Nagoya University (Nagoya, Japan)6 University of Victoria (Victoria, British Columbia, Canada)7 Hokkaido University (Hokkaido, Japan)8 University of Leeds (Leeds, United Kingdom)9 National Astronomical Observatory of Japan (Tokyo, Japan)10 Kogakuin University (Tokyo, Japan)11 The University of Tokyo (Tokyo, Japan)12 JAXA (Kanagawa, Japan)13 Institute of Space and Astronautical Science (Chungbuk, Korea

(the Republic of))

Pairs of azimuthal intensity decrements at near-symmetric locations have been seen in a number ofprotoplanetary disks. They are most commonly in-terpreted as the two shadows cast by a highly mis-aligned inner disk. Direct evidence of such an innerdisk, however, remains largely illusive, except in rarecases. In 2012, a pair of such shadows were discov-ered in scattered-light observations of the near face-on disk around 2MASS J16042165- 2130284, a tran-sitional object with a cavity 60 au in radius. Thestar itself is a “dipper,” with quasi-periodic dimmingevents on its light curve, commonly hypothesized ascaused by extinctions by transiting dusty structures

178

in the inner disk. Here, we report the detection of agas disk inside the cavity using Atacama Large Mil-limeter/submillimeter Array (ALMA) observationswith 0.2 arcsec angular resolution. A twisted but-terfly pattern is found in the moment 1 map of theCO (3–2) emission line toward the center, which isthe key signature of a high misalignment betweenthe inner and outer disks. In addition, the coun-terparts of the shadows are seen in both dust con-tinuum emission and gas emission maps, consistentwith these regions being cooler than their surround-ings. Our findings strongly support the hypothe-sized misaligned inner disk origin of the shadows inthe J1604-2130 disk. Finally, the inclination of the in-ner disk would be close to −45° in contrast with 45°;it is possible that its internal asymmetric structurescause the variations on the light curve of the host star.

501.05 — Polarizing Planetary Systems: New De-bris Disks Resolved on Solar System Scales byGPIES

Thomas M. Esposito1; Paul Kalas1,2; Michael P.Fitzgerald3; Maxwell A. Millar-Blanchaer4,7; Chris-tine Chen5; Marshall D. Perrin5; Schuyler Wolff8;Brenda Matthews9,10; Gaspard Duchene1,6; KatherineB. Follette11; Stanimir Metchev12,14; Pauline Arriaga3;Justin Hom13; Juan Sebastián Bruzzone12

1 Astronomy, UC Berkeley (Berkeley, California, United States)2 University of Victoria (Victoria, British Columbia, Canada)3 Physics and Astronomy Department, Amherst College (Amherst,

Massachusetts, United States)4 Department of Physics and Astronomy, Centre for Planetary

Science and Exploration, The University of Western Ontario (London,Ontario, Canada)

5 School of Earth and Space Exploration, Arizona State University(Tempe, Arizona, United States)

6 Department of Physics and Astronomy, Stony Brook University(Stony Brook, New York, United States)

7 SETI Institute (Mountain View, California, United States)8 Physics and Astronomy, University of California, Los Angeles

(Los Angeles, California, United States)9 NASA Jet Propulsion Laboratory (Pasadena, California, United

States)10 Space Telescope Science Institute (Baltimore, Maryland, United

States)11 Institut de Planetologie et d’Astrophysique de Grenoble, Univer-

site Grenoble Alpes / CNRS (Grenoble, France)12 NASA Hubble Fellow (Pasadena, California, United States)13 Leiden Observatory, Leiden University (Leiden, Netherlands)14 National Research Council of Canada Herzberg (Victoria, British

Columbia, Canada)

We will present new top-level results from our un-precedented debris disk survey of 100+ stars in near-

IR, polarized scattered light with the high-contrastGemini Planet Imager. This four-year survey is thefirst of its kind: a uniform probe of young plane-tary system environments for small dust on SolarSystem-like scales with polarimetry and high angu-lar resolution. Among the 26 detected disks we willpresent, seven are scattered-light discoveries, overa dozen are seen in polarized intensity for the firsttime, and all are resolved on spatial scales of 0.5–7.0au. On the population level, we constrain the pres-ence of micron-sized grains at stellar separations of1–20 au for our nearest observed stars and 20–200au for the farthest: a prime zone for planet forma-tion and migration. We now know the detailed mor-phologies of these disks, a handful of which eithermay be disturbed by interaction with low-mass com-panions or have had perturbing giant planets directlyimaged. In one case, an external stellar companionmay have truncated the primary star’s disk. We willalso link our data to ALMA observations, showingan empirical relationship between the radial loca-tions of small and large disk grains. Going forwardin a broader context, our data yield measurements ofthe disk-scattered light’s polarization fraction, whichwill be a key factor in determining grain composi-tions, sizes, and structures that are only weakly con-strained otherwise. Thus, GPIES disk data are push-ing the bounds of theory and models to explain ob-servations. In summary, we will use our ESS IV oralpresentation to share with the exoplanet communitythe results of the most information-rich survey ofscattered-light debris disks ever conducted, one thatwill inform the design and drive the science goals ofadvanced instrumentation and facilities that will seefirst light in the coming decade.

501.06 — The debris disk of HR 8799: do we needan extra planet?

Virginie Faramaz11 JPL-Caltech (Pasadena, California, United States)

HR 8799 is so far the only system where multipleplanets have been directly imaged. The planetarysystem is surrounded by an extremely faint debrisdisk, which detailed observations are now withinreach thanks to the sensitivity and resolution powerof ALMA. Prior observations of this disk at low (∼3)SNR in Band 6 (1.3 mm) led two different teams toderive two different values for the location of thedisk inner edge: one result suggests that an addi-tional planet should be present beyond the outer-most planet HR 8799 b to carve this inner edge. Theother one finds an inner edge location that is, on the

179

contrary, compatible with HR 8799 b carving this in-ner edge. We present here new high angular resolu-tion observations of this debris disk in Band 7 withALMA, the most sensitive that were obtained so far,thanks to which we derive an inner edge consistentwith the largest value derived from the modeling ofthe Band 6 emission, and thus confirm that the inneredge of the debris disk would require an additionalplanet beyond HR 8799 b to have been carved at thisdistance.

501.07 — The presence of gas in debris discs: whatdoes it imply?

Quentin Kral1; Alexis Brandeker21 LESIA, Paris Observatory (Meudon, France)2 Astronomy, Stockholm University (Stockholm, Sweden)

Gas is now discovered ubiquitously around main-sequence stars at a stage (>10Myr) where planetshave already formed. This gas is always discov-ered in systems with planetesimal belts (debris discstage) and is thought to being released from volatile-rich planetesimals when they collide with each otherand create the observed dust. I will present themost recent gas detections and new yet unpublishedALMA data showing new detections for the first timearound a G-type star (in addition to recent detec-tions around M and F stars), showing that gas re-lease around main sequence stars is not an A-starphenomenon as once thought. I will show what canbe learned from all these new gas detections in themain sequence phase for the planetary system as awhole. The main result so far is that, as we are prob-ing gas released from planetesimals, we have a directaccess to the volatile content of these exoplanetesi-mals, which is fundamental as they are the buildingblocks of planets and may also deliver volatiles whenimpacting onto Earth-like planets in these systems. Iwill show how gas evolution models (Kral et al. 2016,2017, 2019) can help to constrain the composition ofthese exoplanetesimals. From the gas models, wealso derive the viscosity of these gas discs and showthat it may be compatible with values given by themagnetorotational instability (MRI, see Kral & Lat-ter 2016), which may allow to test the MRI under newconditions, in low density environments where non-ideal effects (such as the ambipolar diffusion) maybe important. Last but not least, our new model isable to explain the most massive gas discs observed(with CO masses greater than 0.01 earth masses) asbeing of secondary origin as well (Kral et al. 2019),i.e. with gas released from planetesimals rather thanbeing primordial (i.e. a remnant of the protoplane-tary disc gas). This has important consequences con-

cerning planetary formation and the fate of proto-planetary discs that can be studied from these gasobservations.

502 — Habitability and Biosigna-tures502.01 — Are Exoplanets Orbiting M Dwarfs Ex-treme?

Philip Steven Muirhead1; Aurora Kesseli1; EunkyuHan1; Mark Veyette2

1 Astronomy, Boston University (Boston, Massachusetts, UnitedStates)

2 Lockheed Martin (Denver, Colorado, United States)

M dwarf stars have long spin-down timescales, longactivity lifetimes and persistent magnetic activity, allof which have implications for the potential habit-ability of orbiting planets. I will present results fromseveral research programs investigating M dwarf ro-tation, activity and evolution. I will discuss a newtechnique to measure chemical-kinematic ages ofmain-sequence M dwarf stars. We applied that tech-nique to a variety of nearby M dwarfs, both planethosts and non-planet hosts, and rapid (young) andslow (old) rotators. We find that relatively slow ro-tators (P∼100 days) do not appear to be α enriched,indicating that they are not over 10 Gyrs old. Sec-ond, for the rapid rotators, we see clear evidence ofZeeman enhancement of Y-band Ti I lines as a func-tion of Rossby number. While other activity indi-cators, such as H-α and X-ray emission, appear tosaturate with low Rossby number, Zeeman enhance-ment does not, indicating that the saturation mech-anism is confined to the chromosphere and corona.Finally, I will present new results on the M dwarfradius problem. Using spectral synthesis methods,we find that large magnetic star spot fractions areprimarily responsible for observed discrepancies be-tween model and measured stellar radii in fully con-vective M dwarf stars. As most M dwarfs appear dis-crepant, our results suggest the vast majority of Mdwarfs have large spot fractions and correspondinglyhigh localization of magnetic fields.

502.02 — Flare Statistics and High Resolution Spec-troscopy of a Volume Complete Sample of Mid-to-Late M dwarfs within 15 Parsecs

Amber Medina1; David Charbonneau1; JenniferWinters1; Jonathan Irwin1

1 Astronomy, Center for Astrophysics | Harvard and Smithsonian(Cambridge, Massachusetts, United States)

180

Main-sequence stars with masses less than 30% thatof the Sun are fully convective and are the most abun-dant stars in the galaxy. The question of how fullyconvective stars generate their magnetic field is of in-trinsic interest and also bears upon the habitability oftheir orbiting planets. These stars currently providethe best opportunities to study planets in the habit-able zone, so it is essential we characterize their mag-netic activity. We are currently undertaking a multi-epoch high-resolution spectroscopic survey in addi-tion to obtaining (through a TESS GI program) two-minute cadence data of a volume-complete sampleof stars with masses between 0.1-0.3 the solar valueand within 15 parsecs. The stars in the sample arewell-characterized with accurate masses and radii,and photometric rotation periods from the MEarthproject. We determined the statistics of flares onall mid-to-late M dwarfs within 15 parsecs observedby TESS to-date. We use our complementary high-resolution spectroscopic measurements of rotationalvelocities, H-α equivalent widths, along with ourgalactic space motions (calculated from our mea-sured radial velocities) to correlate the ages and ac-tivity levels of this population to the flare rates, lu-minosities, and durations.

This work is supported by grants from theJohn Templeton Foundation, the David and LucilePackard Foundation, and the US National ScienceFoundation.

502.03 — Volatile- and Water-rich Planetary Mate-rial Accreting onto a White Dwarf

Matthew Hoskin1; Odette Toloza1; Boris Gaensicke1;Roberto Raddi3; Detlev Koester2; Jay Farihi4

1 Physics, University of Warwick (Coventry, WEST MIDLANDS,United Kingdom)

2 Institut fur Theoretische Physik und Astrophysik, Christian-Albrechts-Universität zu Kiel (Kiel, Germany)

3 Dr. Karl Remeis-Sternwarte, Friedrich–Alexander UniversitätErlangen-Nürnberg (Bamberg, Germany)

4 Physics & Astronomy, University College London (London,United Kingdom)

We report the discovery of a white dwarf that has anunusually large amount of hydrogen, ∼5% by mass,within its helium atmosphere. Such a mixture cannotresult from the past evolution of this star alone (Rol-land+, 2018). The only plausible explanation is thatthis white dwarf has recently accreted ∼1022g of wa-ter — as much as 1% of the Earth’s oceans. Absorp-tion lines of the major mineral-forming elements (O,Mg, Si, Ca, Fe, Ni) and of volatile elements (C, S, P)detected in our VLT and Hubble Space Telescope spec-troscopy unambiguously demonstrate that the star

is currently accreting planetary debris. Our abun-dance analysis indicates a comet-like nature of thedisrupted planetesimal, carrying the material nec-essary for seeding terrestrial exo-planets with thebuilding-blocks of life. Small traces of hydrogen arecommon in helium atmosphere white dwarfs, andare often found alongside pollution by planetary de-bris, providing clear statistical evidence that water-rich rocky bodies prevale into the final stages of stel-lar and planetary evolution (Gentile Fusillo+, 2017).This connection is corroborated by this spectacularlypolluted white dwarf, which has accreted a sufficientamount of water to change its past and future spectraevolution.

502.04 — M-dwarf Activity Driven 3D Climate andPhotochemistry of Inner Habitable Zone Tidally-Locked Planets

Howard Chen1,2; Eric Wolf4; Zhuchang Zhan3; DanielHorton1,2

1 Department of Earth and Planetary Sciences, Northwestern Uni-versity (Evanston, Illinois, United States)

2 Center for Interdisciplinary Exploration and Research in Astro-physics (CIERA), Northwestern University (Evanston, Illinois, UnitedStates)

3 Department of Earth, Atmospheric and Planetary Sciences, Mas-sachusetts Institute of Technology (Cambridge, Massachusetts, UnitedStates)

4 Laboratory for Atmospheric and Space Physics, University ofColorado Boulder (Boulder, Colorado, United States)

Planets residing in circumstellar habitable zones(CHZs) offer our best opportunities to test hypothe-ses of life’s potential pervasiveness and complexity.Constraining the precise boundaries of habitabilityand its observational discriminants is thus critical tomaximizing our chances at remote life detection forfuture instruments. Conventionally, calculations ofthe inner edge of the habitable zone (IHZ) have beenperformed using both 1D climate models and 3Dgeneral circulation models. However, these modelslack interactive three-dimensional chemistry and donot resolve the observationally-critical mesosphereand lower thermosphere (MLT). Here we employ a3D chemistry-climate model (CCM) to simulate theatmospheres of synchronously- rotating planets or-biting at the inner edge of habitable zones of K- andM-dwarf stars (between Teff = 4000 K and 2600 K)with N2-O2-H2O-CO2 atmospheres. With the in-clusion of interactive chemistry, we find that sim-ulated runaway and moist greenhouse thresholdsare in good agreement with previous GCM studies.However, around quiescent stars, our prognostic hy-drogen mixing ratios are orders of magnitude lower

181

than previous diagnostic estimates, suggesting thatplanets in these systems are less vulnerable to desic-cation via water escape. Additionally, we find thataround active M-dwarfs, increases in upper atmo-spheric moisture and photodissociation rates allowhydrogen mixing ratios to approach that of watervapor, leading to elevated water loss efficiency viadiffusion-limited escape. Using our CCM results asinputs, translated transmission and emission spectrashow that both water vapor and ozone features couldbe detectable by future missions such as the JamesWebb Space Telescope.

502.05 — Dark water oceans on exoplanets orbitingcool stars

Lisa Kaltenegger11 Astronomy, Carl Sagan Institute Cornell University (Ithaca, New

York, United States)

Several thousand extrasolar planets orbiting otherstars provide a first glimpse into the diversity ofother worlds. We show that oceans on worlds or-biting different alien Suns will differ from Earth’soceans because the penetration depth of light canbe very different, altering ocean dynamics signifi-cantly. While Sunlight can penetrate up to 250m ina water ocean, light may penetrate as little as 2mfor oceans illuminated by cool red stars. Dynamicsand photosynthesis in water oceans on exoplanetsand exomoons orbiting other Suns can be very dif-ferent from Earth’s. We introduce a new paradigmfor the hydrosphere-atmosphere interaction in plan-etary models. The idea is fundamental – when youconvolute the absorption of water with wavelengthwith the irradiation exoplanets receive from differenthost stars it fundamentally changes how deep thatlight can penetrate water, especially for cool host starplanets like the detected, potentially habitable plan-ets around our neighboring stars Proxima-b and theplanets in the Trappist-1 system. Our paper showsthe huge impact the host star irradiation has on thethermal structure and resulting dynamics on wateroceans on extrasolar planets. Additionally, the depthbelow which there is generally insufficient light forphotoactive organisms in oceans on a red star will bemuch shallower than on Earth. Thus, photosyntheticocean life, if it exists, will be much closer to the oceansurface, and can be more readily detected on planetsand moons orbiting red stars.

502.06 — Prebiotic Planets: Evaluating PlanetaryConditions for Origins of Life

Dimitar Sasselov1

1 Astronomy, Harvard University (Cambridge, Massachusetts,United States)

We often discuss exoplanet habitability, but rarely fo-cus on the prebiotic planets – the ones with geochem-ical conditions conducive to the emergence of life.How could prebiotic exoplanets, if we could identifythem, help us solve life’s origins? In this talk I willfocus on the prebiotic synthesis of the nucleotides,amino acids and lipids needed for life as we know itand the planetary environmental context that makesthat synthesis possible. I will argue that, as we stillstruggle to understand life’s origins on Earth, thereare general predictions about the global planetaryconditions that are testable with upcoming spectro-scopic observations of the atmospheres of rocky ex-oplanets.

503 — Future Missions503.01 — Expanding our Telescope Toolkit: Exo-planet Science Opportunities with SmallSats

Evgenya L. Shkolnik11 School of Earth and Space Explorations, Arizona State University

(Tempe, Arizona, United States)

New technologies can disrupt the status quo by chal-lenging the current assumptions and opening upnew avenues of research. Small satellites, includ-ing CubeSats, have been growing in popularity inmany science and technology fields, yet are onlynow beginning to receive attention as tools for as-trophysics research. When deployed as space-basedtelescopes, SmallSats enable science experiments notpossible with existing or planned large space mis-sions. We trade some capabilities such as mirror sizefor lower cost and shorter build times for more fre-quent launch opportunities, with two additional andcrucial advantages over large, over-subscribed tele-scopes: SmallSats can monitor sources for weeks ormonths at time, and at wavelengths not accessiblefrom the ground such as the ultraviolet, far-infraredand low-frequency radio. Achieving high-impactastronomical research with SmallSats is becomingincreasingly feasible with advances in technologiessuch as precision pointing and compact sensitive de-tectors. SmallSats may also pair well with the largespace- and ground-based telescopes providing com-plementary data to better explain the physical pro-cesses observed. There are many possible exoplanet-focused science cases for SmallSats, several of whichare already in development and more ideas yet tobe proposed. I will share our experiences devel-

182

oping the NASA-funded SPARCS (Star-Planet Ac-tivity Research CubeSat) mission as a case study toexplore the challenges and opportunities of astro-physics SmallSats.

503.02 — Exoplanet Imaging with ELTs and theTMT’s Planetary Systems Imager

Andy Skemer11 Astronomy, UC Santa Cruz (Santa Cruz, California, United

States)

The combination of high angular resolution and sen-sitivity will allow ELTs to image hundreds of exo-planets ranging from gas-giants to super-earths andeven a handful of rocky planets. I will review thedifferent types of exoplanets that will be observablewith ELTs and also discuss how spectroscopy willenable measurements of molecular abundances, T-P profiles, cloud properties, spin rates, accretion,and weather. Finally I will present an overviewof the Planetary Systems Imager, the TMT’s multi-wavelength exoplanet imaging platform.

503.03 — The CHEOPS Mission: Launch Imminentfor ESA’s Next Exoplanet Mission

Christopher Broeg1; Willy Benz1; Andrea Fortier1;Thomas Beck1

1 Center for Space and Habitability, Space and planetary sciences,University of Bern (Bern, Switzerland)

The CHaracterising ExOPlanet Satellite (CHEOPS) isa mission jointly led by Switzerland and ESA whichwas selected in October 2012 as the first S-class mis-sion in the ESA Science Programme. CHEOPS willbe the first space observatory dedicated to search fortransits of exoplanets by means of ultrahigh preci-sion photometry on bright stars already known tohost planets. It will have access to more than 2/3of the sky and provide the unique capability of de-termining accurate radii for planets of known massfrom ground-based spectroscopic surveys. This willallow a first order characterisation of the planets’ in-ternal structure by determination of their mean den-sity, which provides direct insights into their com-position. CHEOPS will also provide precise radiifor new planets discovered by the next generation ofground- or space-based transits surveys.

To reach its goals, CHEOPS is designed to measurephotometric signals with a precision of 20 ppm in 6hours integration time on magnitude 9 stars, and 85ppm in 3 hour integrations on magnitude 12 stars.The CHEOPS payload is a single telescope of 30 cmclear aperture, which has a single CCD focal plane

detector. In Ritchey-Chrétien telescope optical con-figuration it provides a defocussed image of the tar-get star. The main design drivers are related to thecompactness of the optical system and to the capa-bility to reject the stray light. The nominal CHEOPSoperational orbit is a polar Sun-Synchronous Orbit(SSO) with an altitude of 700 km and a local time ofthe ascending node (LTAN) of 6 am; the orbit incli-nation is about 98° and the orbital period is 100 min.The nominal mission lifetime is 3.5 years, with a pos-sible extension to 5 years. CHEOPS will launch asauxillary passenger on a Soyuz from Kourou. Thelaunch is imminent with the launch window definedby Arianespace from 15 October to 14 November thisyear. This talk will review the CHEOPS mission,its scientific goals and mission design. We will dis-cuss the expected performances and also present lat-est analysis of the ground calibration campaign. Anoverview of the GTO programme will be presented.

503.04 — ARIEL: ESA’s Mission to Study the Natureof Exoplanets

Göran Pilbratt11 ESA/ESTEC (Noordwijk, Netherlands)

The ∼4000 exoplanets currently known display agreat diversity of physical parameters, and orbit starswith different properties and planetary system ar-chitectures. For most of them we know only eithermass or size, or both. However, planetary modellingbased on sizes and masses alone suffer from impor-tant degeneracies. To independently measure chem-ical composition is the next challenge. It would en-able improved modelling, which will enhance ourunderstanding of what planets are made of, howplanets and planetary systems form, and how plan-ets and their atmospheres evolve to what we observetoday.

The Atmospheric Remote-Sensing Infrared Exo-planet Large-survey (ARIEL) mission has been se-lected by ESA as M4 in the Cosmic Vision pro-gramme for a 2028 launch. ARIEL is dedicated toperforming measurements of the chemical compo-sition and dynamics of exoplanet atmospheres for alarge population (many hundreds) of known diversepreferentially warm and hot transiting planets, en-abling the understanding of the physics and chem-istry of these far away worlds.

The observations will probe atmospheric chem-istry and dynamics, by means of infrared spec-troscopy in three bands (covering 1.1-7.8 um) andvisible/NIR photometry in three bands (covering0.5-1.1 um). All six bands are observed simulta-neously with an off-axis Cassegrain telescope hav-

183

ing a ∼1.1 × 0.7 m aperture. Both transit andeclipse/occultation spectroscopy will be employedto obtain transmission and emission spectra. Thephotometry provides thermal and scattering proper-ties and monitors stellar activity.

ARIEL will conduct its observations from a largehalo orbit around the Sun-Earth L2 point. ARIELwants to embrace the general community, by offer-ing open involvement in target selection, and by pro-viding timely public releases of high quality dataproducts at various processing levels throughout themission.

In this presentation I will provide an overview ofall aspects of the mission, describe the current on-going activities in ESA, the ARIEL Consortium, andindustry, and the overall schedule.

503.05 — How to Determine Whether M-Dwarf Ter-restrial Planets Possess Atmospheres

Eliza Kempton1; Daniel D. B. Koll2; Megan Mansfield3;Matej Malik1; Edwin Kite3; Jacob L. Bean3; DorianAbbot3; Renyu Hu4

1 University of Maryland (College Park, Maryland, United States)2 MIT (Cambridge, Massachusetts, United States)3 University of Chicago (Chicago, Illinois, United States)4 JPL (Pasadena, California, United States)

In the era of TESS, we expect to detect legions ofplanets for which atmospheric characterization willbe possible with JWST. Perhaps the most excitingamong these planets are the rocky ones, which upuntil now have not been accessible to atmosphericstudies. Yet small rocky planets will still be challeng-ing targets for JWST, so the question arises of howbest to use JWST to make tangible progress towardunderstanding the atmospheres of terrestrial bodies.We posit that JWST is best suited to distinguish be-tween rocky planets that do and do not possess at-mospheres by photometrically observing their sec-ondary eclipses. The argument is as follows. Thedayside temperature of a tidally locked planet willbe reduced by the presence of an atmosphere, eitherbecause the atmosphere transports heat to the nightside of the planet or because atmospheric scattererssuch as clouds will increase the planet’s Bond albedo.There is therefore a maximal secondary eclipse depththat is representative of a hot dayside hemispherewith no atmosphere present. We focus on planets or-biting M stars because they are being discovered inlarge numbers by current facilities, they are withinthe observational grasp of JWST, and there is con-siderable skepticism as to whether these planets canretain atmospheres at all given the high-energy irra-diation from their host stars.

I will present the results from a multi-institutioncollaboration investigating the promise and the lim-its of secondary eclipse photometry as a test for can-didate atmospheres on rocky M-dwarf planets. Wehave developed a suite of general circulation mod-els and radiative-convective atmospheric structuremodels, and have developed our understanding ofrocky planet surface geochemistry, in order to ad-dress this topic. We have focused our efforts on threewarm transiting super-Earths that will be ideal tar-gets for secondary eclipse investigations with JWST.We find that JWST can distinguish between planetswith and without atmospheres in as little as a oneeclipse — a time investment that significantly out-performs phase curves and the more traditional tran-sit spectroscopy techniques.

503.06 — Cold exoplanets with WFIRST: demo-graphics with the microlensing survey, and char-acterization with the coronagraph instrument

Matthew Penny11 Physics & Astronomy, Louisiana State University (Baton Rouge,

Louisiana, United States)

As outlined by the 2010 Decadal Survey, NASA’snext flagship mission WFIRST will conduct wide-field infrared surveys and demonstrate space-basedcoronagraphy techniques necessary to directly im-age exoplanets in reflected visible light. We will givean update on the status of the WFIRST mission, fo-cusing on its two major exoplanet goals. Using itswide field instrument (WFI), WFIRST will carry outa large exoplanet microlensing survey toward theGalactic bulge. This survey is designed to statisti-cally explore exoplanet demographics over a widerange of orbital separations (from <∼1 AU to infinity[i.e., free-floating planets]), and five orders of mag-nitude in mass (super-Jupiters down to a few lu-nar masses). These broad statistics will provide vi-tal, and otherwise unobtainable, observational con-straints on the end products of the planet forma-tion process, and on the occurrence rates of low-mass planets in wider orbits than can be probed byradial velocity and transit techniques. Focusing onrecent and upcoming results from the WFIRST Mi-crolensing Science Investigation Team, we will de-scribe new estimates of WFIRST’s capabilities to de-tect bound and free-floating planets, the develop-ment of techniques required to measure microlens-ing planet masses, and the results of a data chal-lenge designed to test planet detection and mod-elling techniques on WFIRST-like simulated data.For the coronagraph instrument (CGI) we presenta few key highlights of CGI’s predicted capabilities

184

for characterizing nearby planetary systems via itstechnology demonstration of extreme contrast coro-nagraphic imaging and spectroscopy in visible light.CGI will demonstrate five main areas that aid fu-ture direct imaging missions such as LUVOIR andHabEX: exquisite wavefront control through a pairof deformable mirrors, suppression of an on-axisstar’s diffraction pattern through occulting masks orshaped pupils, the use of photon counting visible de-tectors, and post-processing techniques at high con-trast in space, and high contrast spectroscopy.

503.07 — Radial Velocity Science in the 2020s: TheFuture of Ground-based EPRV Surveys

Jennifer Burt11 Kavli Institute, MIT (Somerville, Massachusetts, United States)

The radial velocity community has delivered a va-riety of new and exciting instruments around theglobe over the past two years. While many of thesefacilities began operations during the end of the2010s, their true science impact will not be felt untilthe 2020s. Extreme precision radial velocity instru-ments such as ESPRESSO, EXPRES, and Neid willallow for detailed monitoring of our closest stellarneighbors on a scale that has never been seen before.They will obtain mass measurements for many of thesmallest transiting planets from missions like TESS,in addition to surveying nearby stars in search of theshort period, terrestrial planets that we expect basedon the Kepler planet occurrence rates. Meanwhile,near-infrared spectrographs like HPF, SPIRou, andIRD will facilitate searches for planets around thecoolest nearby stars, targeting a variety of stellar hostthat has not previously been surveyed by Doppler fa-cilities. I will discuss the upcoming advancementswithin these branches of radial velocity science andhow they are expected to expand the current bound-aries of exoplanet discovery space.

185