Table Of Content

Characterizing Transiting Planet Atmospheres through 2025 N.B. Cowan1, T. Greene2, D. Angerhausen3, N.E. Batalha4, M. Clampin3, K. Colo´n5, I.J.M. Crossfield6, J.J. Fortney7, B.S. Gaudi8, J. Harrington9, N. Iro10, C.F. Lillie11, 5 1 J.L. Linsky12, M. Lopez-Morales13, A.M. Mandell2, and K.B. Stevenson14, on behalf of 0 ExoPAG SAG-10 2 1Amherst College ([email protected]) n 2NASA Ames Research Center a J 3NASA Goddard Space Flight Center 0 4Pennsylvania State University 3 5Lehigh University 6University of Arizona ] P 7University of California, Santa Cruz E 8The Ohio State University . 9University of Central Florida h p 10Universita¨t Hamburg - 11Lillie Consulting o r 12University of Colorado st 13Harvard-Smithsonian Center for Astrophysics a 14University of Chicago [ Executive Summary to maximize the scientific returns of TESS is to 1 adopt a triage approach. A space mission con- v The discovery of planets around other stars is sisting of a 1 m telescope with an optical–NIR 4 revolutionizing our notions of planet formation ∼ 0 spectrographcould measure molecular absorption and is poised to do the same for planetary cli- 0 for non-terrestrial planets discovered by TESS, 0 mate. Studying transiting planets is complemen- as well as eclipses and phase variations for the 0 tarytoeventualstudiesofdirectly-imagedplanets: hottest jovians. Such a mission could observe 2. 1) we can readily measure the mass and radius of up to 103 transits per year, thus enabling it to 0 transiting planets, linking atmospheric properties survey a large fraction of the bright (J < 11) 5 tobulkcompositionandformation,2)manytran- hot-Jupiters and warm sub-Neptunes TESS is ex- 1 siting planets are strongly irradiated and exhibit pected to find. The James Webb Space Telescope : v novel atmospheric physics, and 3) the most com- (JWST) could be used to perform detailed at- i mon temperate terrestrial planets orbit close to X mospheric characterization of the most interest- red dwarf stars and are difficult to image directly. ing transiting targets(transit, eclipse,and—when r a We have only been able to comprehensively char- possible—phase-resolved spectroscopy). TESS is acterize the atmospheres of a handful of transit- also expected to discover a few temperate terres- ing planets, because most orbit faint stars. The trialplanetstransitingnearbyM-Dwarfs. Charac- TransitingExoplanetSurveySatellite(TESS)will terizingtheseworldswillbetime-intensive: JWST discover transiting planets orbiting the brightest willneedmonthstoprovidetantalizingconstraints stars, enabling, in principle, an atmospheric sur- onthe presence ofanatmosphere,planetary rota- vey of 102–103 brighthot Jupiters and warmsub- tionalstate,clouds, andgreenhousegases. Future Neptunes. Uniform observations of such a statis- flagship missions should be designed to provide tically significant sample would provide leverage better constraints on the habitability of M-Dwarf to understand—and learn from—the diversity of temperate terrestrial planets. short-period planets, and would identify the mi- nority of truly special planets worthy of more in- tensive follow-up. We argue that the best way 1 1. Context plate tectonics, etc. However, the characteriza- tion of exoplanets presents a challenge familiar to The study of exoplanet atmospheres has ex- astronomers: our targets are so distant that we ploded in the past decade. In 2013, the Ex- only see them as unresolved dots. oplanet Exploration Analysis Group (ExoPAG) Manyaspectsofplanetarysciencearecurrently created—with approval from NASA’s Astro- accessible for exoplanets, or soon will be. Since physics Subcommittee—a tenth Study Analysis weobserveexoplanetarysystemsfromtheoutside, Group (SAG-X) to consider what NASA could the easiest aspect to constrain is the architecture do in the next decade to better understand the of planetary systems, and indeed our theories of atmospheres of transiting planets. SAG-X had planet formation are currently being revolution- openmembershipandinvolvedthreepresentations ized by our growing knowledge of planetary de- to the exoplanet community. The first presenta- mographics and architecture. Transiting planets tion, outlining the challenges and opportunities are crucial to our understanding of planet forma- ofstudyingtransitingexoplanetatmospheres,was tion, because they are the only planets for which made at the ExoPAG 8 meeting preceding the wecanhopetoknowtheorbitalarchitecture,bulk Oct. 2013 Division of Planetary Sciences meet- density, and atmospheric composition. ing in Denver, CO. We held a mini-workshop on The study of individual planets is likewise pro- the capabilities of the James Webb Space Tele- gressing from the top–down: first the exospheres, scope for characterizing transiting exoplanets at then the atmospheres, and the surfaces last. It ExoPAG 9 preceding the Jan. 2014 American As- is therefore difficult to make definitive statements tronomical Society meeting in Washington, DC. about the surface conditions of exoplanets, and Interestedmembersofthecommunityhelpeddraft their interiors will only be known to us through the current document over the course of the 2014 their bulk density, and surface character. The calendar year and we presented it at ExoPAG 11, studyofplanetaryatmospheres,however,ispoised preceding the Jan. 2015 AAS meeting in Seat- to be revolutionized by observations of “exo- tle, WA. Multiple drafts of this report have been climes.” The first suchmeasurements constrained circulated to the ExoPAG membership and the planetary-scale temperature structure and com- exoplanet community at large and we have done position, while the second generation of measure- our best to implement feedback. This document ments leveraged the orbital motion of the planet is therefore a consensus view of what can and to infer the horizontal temperature structure of shouldbe done in the field of transiting exoplanet the planet. Planned instruments will enable 3D atmospheres in the next decade. measurements of atmospheric composition, tem- perature structure, and winds. 2. Planetary Science from the Top–Down Our knowledge of Earth and the So- 3. Planetary Climate lar System planets will always exceed our A planet’s albedo determines how much knowledge of any individual exoplanet, but radiation it absorbs, atmospheric composi- the diversity of exoplanets enables the sta- tion dictates how energy trickles up via ra- tistical study of planets to crack difficult diation and convection, while atmospheric problems in planetary science. (andoceanic)dynamicsdeterminehowheat What started as a trickle in the mid 1990’s is is transported from regions receiving more now a torrent, with over one thousand extrasolar sunlight to those receiving less. planets currently known, and thousands of can- Theprimaryaspectofplanetaryclimateistem- didates awaiting confirmation. The study of ex- perature, averaged over time, and often space. A oplanets has already revolutionized our view of detailed calculation of climate involves radiative planet formation, and will soon do the same to transfer, fluid dynamics, chemistry and, in the our understanding of planetary atmospheres and case of an inhabited planet, biology. Stripped to interiors. The diversity of exoplanets gives us the itsessentials,however,climatedescribestheinter- leverage to crack hard problems in planetary sci- action of star light with a planetary atmosphere ence: cloud formation, atmospheric circulation, 2 (Figure 1). ate terrestrial planets that rotate slowly allow us to empirically verify the effects of Coriolis forces on atmospheric circulation, super-Jupiters let us test the effects of surface gravity on Jovian atmo- spheres, etc. While the most detailed atmospheric studies will always focus on Earth, and in situ measure- mentswillbelimitedtotheSolarSystem,thevast majority of planets are extrasolar. This means thatthemostextremeworlds,andthosemostlike theEarth,areexoplanets. Onlybystudyingthese planetscanwehopetodevelopcomprehensivethe- Fig. 1.— Planetaryclimate is determined by stel- ories of climate. To paraphrase Kipling (1891): lar radiation interacting with a planetary atmo- What should they know of Earth who only Earth sphere. Incoming radiation (solid yellow lines) is know? either reflected (dashed yellow lines) or absorbed. The warm air emits at longer wavelengths than 4. Transiting Planets the incoming light. Longwave radiation is read- Transiting planets are representative of ily absorbed by greenhouse molecules in the at- planets on short-period orbits. Exoplan- mosphere, resulting in radiative diffusion (small ets that transit Sun-like stars tend to be red arrows). The inefficient upward heat trans- hot, while those transiting low-mass stars portmeansthatthelowerpartsoftheatmosphere are merely warm, and represent the major- tendtobehotterthantheoverlyingregions,andin ity of temperate terrestrial planets. Tran- practicemostplanetaryatmospheresconvectover sit spectroscopy probes atmospheric opac- some pressure range (blue arrows). The cooler, ity, while measurements of thermal emis- thinnerupperregionsoftheatmosphereemitther- sion constrain vertical and horizontal tem- malradiation(dashedredlines)thatbalancesthe perature structure. High cadence, high- absorbed shortwaveradiation. precisionemissionspectroscopy throughout a planet’s orbit enables 3D mapping of its Asimpleenergybalancemodel,shownschemat- atmosphere. ically in the left panel of Figure 2, can therefore A transiting planet is a planet that passes in be used to predict planetary climate. Unfortu- betweenitshoststarandtheobserver. Anyplanet nately, despite over a century of research into that orbits a star may be a transiting planet, but Earth’s climate, there are currently no compre- transitingplanetsareessentiallysynonymouswith hensive, predictive theories for cloud formation short-period planets, since the probability that a (planetary albedo), volatile cycling (greenhouse planet transits is the ratio of the stellar radius to gasabundances),orwindspeeds(heattransport). theplanet–starseparation. Thismakesthetransit The empirical approachto this challenge is to ac- methodanimprobablewaytostudy SolarSystem quire observations for many different planets in analogs. the hopes of uncovering the principles of climate Direct imaging is a promising approach to (right panel of Figure 2); the first step in that di- studyingtheatmospheresofSolarSystemanalogs, rectionhas beenthe study ofSolarSystemworlds but close-in planets are difficult to study in this over the last half-century. way, because any means of blocking/nulling the The logical next step, already underway, is to starlight is liable to block the planet, too. We characterize the atmospheres of extrasolar plan- instead study the combined light of the plane- ets. Exoplanets are more diverse than the planets tary system, so the dominant source of photons orbiting our Sun, and hence provide more lever- (and photon noise) is the star rather than the agefor testing theories. Onthe other hand, many planet. Thesignal-to-noiseratioformeasurements exoplanets are similar to those in our Solar Sys- of planetary light is therefore proportional to the tem, providing crucial Rosetta stones: temper- 3 !"#$%&’ /0*12,-:9+",)2$’’ !"#$%&’ ;$<$+-*1+$’’ 3)$+-0&2+&).’ 3)$+-0&2+&).’ 3,4)"$’ 3,4)"$’ 3708*+$’ 3708*+$’ ()*+,-.’ 5",4*-$’ ()*+,-.’ 5",4*-$’ /$4)$0*-70$’ /$4)$0*-70$’ 6&%$"’ 6&%$"’ =>*2$:9+",)2$’ 94,22,&1’’ /0*12)&0-’ 6*)),1?’ /0*12)&0-’ 3)$+-0&2+&).’ Fig. 2.—Left: Itispossibletopredictaplanet’sclimategivenitsalbedo(reflectiveness),atmosphericopacity (greenhousegas abundances),and heattransport(wind speeds and oceancirculation). Unfortunately, these critical inputs cannot be predicted in general. Right: Fortunately, it is possible to measure planetary albedo,atmosphericcomposition,andheattransport,evenforexoplanets(inmanycases,there aremultiple independent means of constraining atmospheric properties; we list only a few in the interest of clarity). This makes it possible to empirically determine a planet’s climate, even in the absence of a fully predictive theoryofplanetaryclimate. In the longrun, suchobservationsmayrevealsomeofthe underlyingprinciples governing cloud formation, volatile cycling, and large-scale circulation. planetary flux, rather than its square-root; the ofhotJupitersoftenresultsinmassloss,andsome usualbiasesinfavorofbiggerandbrightertargets arethoughtto be undergoingRochelobe overflow are especially true for transiting planets. onto their host star (Li et al. 2010). Characterizing transiting planets would be of The most common currently detectable tran- merely theoretical interest, except that many ex- siting planets in the Galaxy are sub-Neptunes in oplanets orbit much closer to their stars than in tight (10–100 days) orbits around their host stars ourSolarSystem. Someoftheseunfamiliarworlds (Howard et al. 2012; Fressin et al. 2013). These are actually rare, like hot Jupiters (Wright et al. warm sub-Neptunes have masses dominated by 2012), while others represent a common outcome rock and ice, but covered in thick H+He atmo- of planet formation, as with packed systems of spheres, and are sufficiently cool (500–1000 K) warm sub-Neptunes (Lissauer et al. 2011). that photochemistry often trumps thermochemi- A hot Jupiter is a jovian planet that orbits calequilibrium(Moses2014). Theapriori transit its host star with a period of .1 week.1 Such probabilityofsuchplanets is low,butthey arein- planets only exist around 1% of Sun-like stars trinsicallycommon,occurringaroundroughlyhalf (Wright et al. 2012), but h∼ave a transit probabil- ofstars. Assuch,theyformthebulkoftheKepler ity of 10%, so there is a transiting hot Jupiter crop. The densities of these planets suggest they for every few thousand FGK stars. These ex- are largely made of ice, and/or have substantial treme worlds experience strong radiative forcing H+He atmospheres, hence their name. Our expe- resulting in day–night temperature contrasts of rience from the Solar System suggests that these hundreds to thousands of K (Showman & Guillot planets likely have high atmospheric metallicities 2002). The high atmospheric temperatures ther- (Fortney et al. 2013). In short, these planets are mally ionize alkali metals, effectively coupling at- expected to be less extreme than hot Jupiters in mospheric dynamics to planetary magnetic fields terms of radiative forcing, but probably more in- (Perna et al.2010a,b). Theradiativeenvironment teresting in terms of chemistry. Temperate terrestrial planets transiting M- 1Hot Jupiters should not be confused with young Jupiters, dwarfstarsareoftentoutedasthepoor-astronomer’s thetypicaltargetsofcurrentdirect-imagingsurveys,which Earth analog, since they are easier to detect arealsohot. 4 and characterize than a true Earth twin. Based on what we currently know, however, M-Dwarf planets are the most common habitable worlds. !"#$%&’( That is because: 1. rocky planets are much more !7*’&&( 6,7*’&&( common in the temperate zones of M-Dwarfs 1$2234&( (Dressing & Charbonneau 2013; Morton & Swift 2014) than in the temperate zones of Sun-like ./+&’&( stars(Petigura et al.2013;Foreman-Mackey et al. 54+*-’*( 2014; Farr et al. 2014), 2. small stars are more common than big stars (e.g., Bochanski et al. 0*’&"’,-( 2010), 3. the tidally-locked nature of these plan- )*+,&$-( ets is not a challenge to climate (Joshi et al. 1997;Merlis & Schneider2010;Edson et al.2011) and may double the width of the habitable zone (Yang et al. 2013), 4. the red stellar ra- diation results in a weaker ice-albedo feedback Fig. 3.— A planet whose orbit is nearly edge-on and hence stabler climate (Joshi & Haberle 2012; will transit in front of its star. This is more likely Shields et al. 2013, 2014), and 5. the slow main to occur for planets that orbit close to their star, sequence evolution of M-Dwarfs means that a ge- so transiting planets are often synonymous with ological thermostat is not strictly necessary to short-periodplanets. Theamountoflightblocked maintain habitable conditions for billions of years during transit tells us the planet’s size, while the (Kasting et al. 1993). Studying temperate terres- transmission spectrum during transit is sensitive trial planets around red dwarfs is our best shot at to scatteringandabsorptioninthe planet’s upper understanding habitability writ large. atmosphere. During transit we see the planet’s nightside, but the planet’s orbital motion even- 4.1. Transit tually brings the dayside into view. The dayside reflectssunlightandisoftenhotterthanthenight- When a planet passes in front of its host star, side, causing variations in brightness known as it blocks a fraction of the star’s light equal to the phases. Atthetopoftheorbit,thestarwilleclipse planet/star area ratio. Inferring the stellar radius the planet, allowing us to measure the brightness from its color and surface gravity,it is possible to ofthestarwithoutcontaminationfromtheplanet; convertthisrelativesizeintoaphysicaldimension. a difference measurement yields the planet’s day- Ontopoftheopaqueplanetarydiskthereisan side brightness. Reflected light eclipse measure- annulus of partially transparent atmosphere that ments are sensitive to the planetary albedo and filters starlight (Figure 3). The spectrum of a reflected light spectroscopy constrains the nature planet in transit therefore contains an imprint of of scattering. Thermal eclipse measurements can scatteringandabsorptionthatoccursintheupper be used to estimate the planet’s dayside temper- atmosphere near the planet’s day-night termina- ature, while thermal spectroscopy is sensitive to tor. Evenif the bulk composition can be taken as atmosphericcompositionandverticaltemperature agiven(e.g.,H+Heforjovianworlds),itisstilldif- profile. (from Cowan 2014) ficult to nail down the abundances of trace gases, andhigh-altitudehazescanwashoutspectralfea- tures(Burrows2014b): eveniftheskylookedclear 4.2. Eclipse to an inhabitant, it may very well be opaque to the grazing rays relevant for transit spectroscopy. It is possible to use eclipses of the planet by Nonetheless, transmission spectroscopy is a pow- its star to isolate planetary light. A planet that erfulcharacterizationtoolthatcanonlybeapplied passes directly in front of its host star usually to transiting planets. passes directly behind it half an orbit later. The brightness of the planetary system immediately before and after occultation is compared to the 5 brightness during eclipse, and the difference is a 4.3. Phases measure of the planet’s dayside brightness. Horizontal and temporal differences in plane- We may then convert the eclipse measurement tary temperature produce time variationsin ther- into an estimate of the planet’s geometric albedo mal emission. For example, the dayside of a or dayside brightness temperature, depending on slowly-rotating planet might appear warmer and whether the instrument is sensitive to visible or hence brighter than its nightside. Most known thermal radiation. Spectrally resolved eclipse exoplanets have curiously eccentric orbits and measurements can constrain atmospheric scatter- thereforeexperience significantseasonsdue to the ing, composition, and vertical temperature profile changing star-planet separation. Rocky planets (Burrows 2014b). are expected to form with randomly orientedspin High-resolutionemissionspectraareableto re- axes, which leads to obliquity seasons. Disentan- solve molecular lines (as opposed to bands), pro- gling the diurnal cycle, eccentricity seasons and viding two novelcapabilities: performing Doppler obliquity seasons based on thermal phase varia- measurements on the planet itself, and probing tions is a work in progress (Cowan et al. 2012b). the compositions of cloudy worlds. For tran- Tides damp obliquity, slow planetary rotation, siting planets the system inclination and plane- and damp orbital eccentricity (e.g., Heller et al. tary mass are known, but Doppler measurements 2011), so most short period planets do not ex- might be used to infer high-altitude wind veloc- perience seasons, but probably have permanent ities (Snellen et al. 2010a) and planetary rota- day and night hemispheres. The day–night tem- tion (Snellen et al. 2014), but these are arguably perature contrast is therefore an indirect mea- accessible via thermal phase curve and eclipse sure of atmospheric heat transport (Cowan et al. mapping measurements (Rauscher & Kempton 2007,2012a). High-precisionthermalphasecurves 2014), or transit morphology (Carter & Winn can be inverted to construct coarse longitudi- 2010). Clouds, on the other hand, have been nal temperature maps of short-period exoplan- most problematic in grazing transit geometry ets (Knutson et al. 2007; Cowan & Agol 2008), (Kreidberg et al.2014b;Knutson et al.2014). Al- and even contain indirect information about the though many emission spectra have so far been planet’slatitudinalfluxdistribution(Cowan et al. featureless, this is likely due to coarse spectral 2013). Full-orbit observations of emission spectra resolution, large measurement uncertainties, and enablespatially-resolvedinferencesoftemperature roughly isothermal atmospheres (Hansen et al. structureandcomposition(Stevenson et al.2014). 2014, and references therein). With full-phase observations at high-cadence, Theeclipsedepthissensitivetothehemisphere- high signal-to-noise, and high spectral resolu- averagedpropertiesofaplanet,whiletheverybe- tion, it should be possible to constrain the 3- ginning and end of an eclipse offer a means of re- dimensional composition and temperature of an solving the planet’s dayside: as the planet disap- exoplanet’satmosphere. Thiswouldallow,forthe pears behind its star and reappears (ingress and first time, realistic initialization and/or testing of egress, respectively), the star’s edge scans across a general-circulation model with active chemistry the planet. It is possible to invert these raster (as opposed to fixed chemistry). This is the holy scans to construct a coarse two-dimensional map grail of atmospheric characterization, and a top of the planet’s dayside. The first-order effect is priority for future exoplanet observations. the phase offset of the eclipse due to a zonally- advected hot-spot (Williams et al. 2006) and was 5. Lessons Learned first detected by Agol et al. (2010). The eclipse timing offset is largelydegenerate with orbital ec- The principle lessons learned from the centricity,butmaybeteasedoutifitischromatic. first dozen years of exoplanet atmospheric The detailed morphology of ingress and egress observations are: (1) short-period plan- provides a smaller but more robust signal about ets are not all alike, nor are they a one- the 2D flux distribution of the planet’s dayside parameter family, (2) observations are usu- (Rauscher et al. 2007), observed by Majeau et al. ally systematics-limited, making out-of- (2012) and de Wit et al. (2012). occultation data critical to modeling detec- 6 tor behavior, and dictating that repeatabil- reduction, and analysis of eclipse measurements ity is the only reliable test of accuracy. have improved, the broadband emission measure- ments of hot Jupiters have trended toward fea- 5.1. Short-Period Planets tureless Planck spectra (Hansen et al. 2014). Dayside broadband emission measurements Observations of exoplanet atmospheres have of hot Jupiters suggest that the hottest plan- so far been weakly constraining, but they have ets cannot effectively transport heat to their provided a few robust surprises (for more com- nightside (Cowan & Agol 2011), and this trend plete reviews, see Burrows 2014a,b; Bailey 2014; has been corroborated by thermal phase mea- Heng & Showman 2014). surements of two planets (Cowan et al. 2012a; Alkali metals have been detected in the atmo- Maxted et al. 2013). Thermal phase and eclipse spheres of some hot Jupiters (Charbonneau et al. maps are consistent with an equatorial hotspot 2002), but are obscured by Rayleigh-scattering (Majeau et al. 2012; de Wit et al. 2012) and sug- hazesinothers(Pont et al.2013);aRayleighscat- gest eastward equatorial winds at a variety of tering slope has also been seen in the transmis- depths (Knutson et al. 2007, 2009, 2012) with a sion spectrum of a warm ice giant (Biddle et al. characteristic advective time comparable to the 2014). Hazes are present on a warm sub- radiative timescale (Agol et al. 2010). Thermal Neptune (Kreidberg et al. 2014b) and a hot Nep- phase measurements of hot Jupiters on eccen- tune (Knutson et al. 2014), but absent on an- tric orbits suggest radiative times less than a day other (Fraine et al. 2014). Some hot Jupiters and equatorial super-rotation (Lewis et al. 2013; areslowlyevaporating(Vidal-Madjar et al.2004), Wong et al. 2014). while there is questionable evidence that one The measured geometric albedos of transit- planet is overflowing its Roche lobe and accret- ing planets are generally small (Rowe et al. 2008; ing onto its star (Fossati et al. 2010; Cowan et al. Kipping & Spiegel 2011) and blue (Evans et al. 2012a). 2013) but with notable exceptions (Demory et al. Water vapor absorption has now been se- 2011). Moreover, the one hot Jupiter with high curely detected in hot Jupiters with the WFC3 albedo appears to have spatially inhomogeneous instrument on HST using both transit spec- clouds (Demory et al. 2013). troscopy(Deming et al.2013;Huitson et al.2013; Wakeford et al. 2013; Mandell et al. 2013) and 5.2. Spitzer eclipse spectroscopy(Kreidberg et al.2014a), and one hot Neptune shows evidence of water as By operating in the mid-infrared without the well (Fraine et al. 2014). However, claims of confounding telluric IR background, the Spitzer molecular absorption in transit and eclipse mea- Space Telescope overcame a key barrier and mea- surements with other instruments remain con- suredthe firstexoplanetaryphotons in the Fall of troversial. Early detections in NICMOS spec- 2004. Teams led by Charbonneau et al. (2005) troscopy data (Swain et al. 2008, 2009a,b) have and Deming et al. (2005) observed secondary been called into question (Gibson et al. 2011, eclipses of TrES-1 and HD 209458b, respectively, 2012;Crouzet et al.2012),whileanalysesofmulti- submitting their independent reports simultane- bandeclipsephotometrywithSpitzer showingev- ously and participating in a joint NASA press idence of temperature inversions (Knutson et al. releasein April 2005. Spitzer remainedthe obser- 2008), disequilibrium chemistry (Stevenson et al. vational tool of choice for most of the subsequent 2010; Swain et al. 2010) and super-Solar C/O decade, and it is useful to summarize what we (Madhusudhan et al. 2011a) have been disputed bysubsequentstudies(Beaulieu et al.2011;Mandell et al.Madhusudhanetal. (2011b) forces plausible chemistry, 2011; Crossfield et al. 2012; Zellem et al. 2014; Line&Yung (2013) and Lineetal. (2014) do not. Since Diamond-Lowe et al. 2014).2 As the acquisition, eclipsespectroscopyrepresentsdisk-averagedemissionfrom regionswithdifferenttemperaturestructures andpossibly different chemistries, it is not obvious which strategy is 2Even when the data reduction scheme is taken at face moresound. A pragmatic solution might beto report the value, disagreements may occur based on the assump- best-fit “reasonable chemistry” solution, but with uncer- tions that go into spectral retrieval. For example, while taintyestimatesthatare“chemistryagnostic.” 7 have learned during this time. ever, would have greatly benefited from instru- In its cryogenic mission (2003 – 2009), Spitzer ments designed to stare at bright point sources, had six photometric channels useful for exo- and the same observations would have had much planet characterization, centered at 3.6, 4.5, 5.8, more scientific value had they been obtained in a 8, 16, and 24 µm. Channels ranged from 1– uniform way and reduced with a uniform, open- 3 µm wide. Spitzer’s InfraRed Spectrograph had sourcepipeline. The Spitzer experience ultimately a low-resolution, 5.3–14 µm mode useful for ex- justifies the expense of a purpose-built mission to oplanets. Unfortunately, the rate of transiting- perform a survey of exoplanet atmospheres. planet detection was low until around 2010,when Spitzer’s after-launch data analysis efforts at- the cryogen was already gone. In Warm Spitzer tempted to characterize the varying sensitivity only the 3.6 and 4.5 µm photometric channels across the faces of individual pixels and to fit its remain operational; spectroscopy is unavailable. temporal response curves. However, pre-launch Thus, only HD 209458b and HD 189733b were laboratory calibration measurements at better everobservedspectroscopically;only they andGJ thanthe0.01%levelcouldhavedoneabetterjob, 436b were observed at 24 µm; and only GJ 436b, as could calibrating long stares at point sources. TrES-1, TrES-4, HD 189733b, and HD 149026b For example, the prominent, time-dependent sen- were observed at 16 µm. However, numerous sitivity“ramp”at5.8,8,and16µmwasunknown planets have measurements in the four shortest prior to launch because lab calibrations on bright bands, and even more have published 3.6 and 4.5 sources lasted only a short time, rather than the µm eclipse depths (Hansen et al. 2014). hours-long timescale of the ramp. Learning from Withanapertureofjust85cm,Spitzer wasde- thisexperience,instrumentsthatreduceintrapixel signed for 10% absolute and 1% relative photom- sensitivity variations, that have long-term stabil- etry, an order of magnitude poorer precision than ity, and that point consistently are now being requiredforexoplanetstudies. Theracetoachieve designed and proposed for dedicated missions. results thus became a contest of systematic cor- Even JWST, which was designed before some rectionmethods. Bayesiansampling(e.g.,Markov key lessons were learned from Spitzer, is being ChainMonteCarlo)replacedmodelfittingbysim- calibrated with enhanced emphasis on observing ple χ2 minimization, Bayesian methods for com- modes suitable for exoplanet studies. paring models with different numbers of free pa- rameters became the norm, and a variety of nu- 6. Exoplanet Observatories Through 2025 mericalapproachesmodeledSpitzer’ssystematics. Manyground-and spaced-basedobserva- Although known for a decade or more, improved toriescan beusedtostudytheatmospheres image centering and photometric extraction tech- of hot Jupiters and warm sub-Neptunes in niques also finally entered common use. Ulti- the next decade. For terrestrial planets, es- mately, contrast uncertainties better than 0.01% peciallythe temperatevariety, the choice of for single eclipses became possible. While an im- instruments is much more limited: only the pressive improvement over Spitzer’s design, the Extremely Large Telescopes and the James 0.01% uncertainty still left the broadband mea- WebbSpace Telescopewillbecapable ofat- surements at S/N of only 10 for most observed ∼ mospheric observations. planets. The 1–3µmwidth andsmallnumberof photometric channelsfurther confoundedspectro- 6.1. 4m telescopes scopic retrieval. Most Spitzer exoplanet observations consist of 4-meter class telescopes are an underused re- unrepeatedeclipseobservationsobtainedwithdif- sourcein the atmospheric characterizationofexo- ferent observing schemes and analyzed by dis- planets. Ground-based, optical multi-object spec- parate groups using a variety of evolving reduc- trographs typically capture light from 0.4–1.0 mi- tion pipelines. The poor signal-to-noise is mostly cronsandaresensitiveto thepresenceofH2O,al- a testament to the relative faintness of currently kali metals such as Na and K, and metal hydrides known transiting planets; NASA’s TESS mission in cloud-free atmospheres. Additionally, in atmo- will addressthis problem. The observations,how- spheres with clouds or hazes, we can use spectral 8 informationtodeterminethesizedistributionand Jupiter-size planets. Indeed, the first ground- altitude of the cloud/haze particles. based detection of sodium in an exoplanet atmo- In order to achieve the equivalent precision sphere came from the Hobby-Eberly Telescope and be competitive with larger telescopes, these (Redfield et al.2008). Current10m-classfacilities smaller telescopes must acquire multiple transits include the Keck telescopes, the Gran Telescopio of a single target. For comparison, four transits Canarias(GTC),Hobby-EberlyTelescope,South- of a single target with a 4-meter telescope, such ern African Large Telescope, Subaru, Very Large as Mayall or Blanco, have the equivalent photon- Telescope, Large Binocular Telescope, and Gem- limited precision of two transits with an 8-meter ini. These facilities offer a number of instruments telescope,suchasGemini. However,acquiringnu- suitable for exoplanet atmospheric characteriza- merous transit observations with the smaller tele- tion. For example, the primaryinstrumenton the scope will mitigate residual atmospheric effects GTC suitable for exoplanet observations is the that limit our precision with ground-based obser- Optical System for Imaging and low Resolution vations, plus these observations will likely have Integrated Spectroscopy (Cepa et al. 2000, 2003, a higher duty cycle, thereby outperforming the OSIRIS), which offers a moderately sized field of largertelescope. Withasufficientnumberoftran- view of 7.8 7.8 arcmin. OSIRIS has the capa- × sits, 4-meter class telescopes can contribute ex- bility for standard long-slit spectroscopy, and it citing, cutting-edge research like with larger tele- also offers a unique tunable filter imaging mode, scopes, but at a fraction of the cost. allowing the user to specify custom optical band- passes with FWHM = 1.2–2.0 nm. In this mode, High precision photometry in the near-infrared observationscanbeconductedinmultipletunings of exoplanet occultations provides a direct mea- nearly-simultaneously, thus allowing for narrow- surement of thermal emission from hot planets, with single-eclipse precisions of 2 10 4 now all band spectrophotometry. Bean et al. (2010) used − × multi-object spectroscopy on the VLT to obtain but routine (Croll et al. 2014). Wide-field NIR transit spectra of GJ 1214b, with an epoch-to- photometry with 4m class telescopes is a viable epoch white-light accuracy of 2 10 4. Other means of achieving high precision photometry for ∼ × − up-and-coming facilities include GMOS on Gem- gas and ice giants orbiting relatively bright stars. ini andMOSFIRE onKeck. Thanks to their large Thewidefieldofviewallowsforasignificantnum- aperture 10 m telescopes are capable of the high ber of reference stars for differential photometry, precision spectroscopy or spectrophotometry re- which is critical for minimizing effects of Earth’s quired tomeasurethe small signals from exoplanet atmosphere (typically the limiting factor for high atmospheres, e.g., alkali metals in hot Jupiter at- precisionphotometryinthenear-infraredfromthe mospheres (Sing et al. 2011) or methane in warm ground). sub-Neptunes (Wilson et al. 2014). Despite the advantages of these types of facili- ties, only a handful exist: WIRCam on the 3.6m 6.3. Extremely Large Telescopes CFHT, WFCAM on the 3.8m UKIRT, NEW- FIRM on the KPNO 4m Mayall Telescope, and By 2025,we expect 1–3 of the extremely large, the Spartan IR Camera on the 4.1m SOAR. Of ground-based telescopes (ELTs) currently being these, only WIRCam on CFHT and WFCAM planned (GMT, TMT, E-ELT) to be operational. on UKIRT have been demonstrated for exoplanet These facilities may provide several possible ben- studies (e.g., Croll et al. 2011, Col´on & Gaidos efits: their larger apertures will enable studies of 2013). largernumbersoftransitingplanetsaroundfainter stars, with techniques used today on 8–10 m tele- 6.2. 10m telescopes scopes; if systematic effects can be controlled, the largeraperturesmayalsoallowplanetsinbrighter As demonstrated by Col´on & Ford (2009), systems to be studied at higher precision; finally, large ground-based telescopes are capable of con- byrelievingsomepressureonexisting8–10mtele- tributing significantly to photometric follow-up scopes it may be possible to use these older facili- efforts for both small, long-period planets dis- ties for large-scale survey science precluded today covered by missions like Kepler and for larger, due to the limited availability of observing time. 9 The ELTs will be optimized to observe faint source of systematic noise for ground-based ob- targets,oftenusingadaptiveoptics–implyingrel- servations at shorter NIR wavelengths, as these atively narrow fields of view. Thus fewer nearby telluric molecules are also the species of interest systems will be accessible to the differential pho- in the exoplanetary atmospheres. The SOFIA tometric and spectroscopic techniques so popular telescope, operating at much lower temperatures today, which use multiple comparison stars to re- (240 K) than ground-based telescopes, reduces move telluric and instrumental noise (Bean et al. thermal background contributions that are the 2010; Gibson et al. 2013; Crossfield et al. 2013). dominant noise source for transit observations at Today, these efforts perform within factors of 2–3 wavelengths longer than 3 microns. SOFIA can of the spectroscopic photon noise limit, but there observe time-critical events, such as planetary oc- is no consensus as to whether the extra noise cultations,underoptimizedconditions,asdemon- comes from detectors, instruments, telescopes, or strated by an observation of a Pluto occultation the Earth’s atmosphere. For ELT multi-object in June 2011 (Person et al. 2013). SOFIA’s ini- transitspectroscopy tosucceed, these noisesources tial forays into transit observations have yielded must be identified and mitigation strategies incor- precisions within a factor of 2 ofthe Poissonlimit porated into the new instruments currently being (Angerhausen et al. submitted). In contrast to designed. space telescopes it is possible to update SOFIA High-dispersion spectroscopy may be the niche with state-of-the-art instrumentation, e.g. with a inwhichELTs canbest makesignificantprogress. dedicated 2nd generation exoplanet instrument Observations at high dispersion (λ/∆λ&20,000) (such as NIMBUS, McElwain et al. 2012). canmeasureuniquemolecularsignatures,thermal 6.5. Hubble Space Telescope profiles, global atmospheric circulation, and or- bitalmotionandhasalreadybeenusedtocharac- Planetary emission peaks in the near-to-mid- terize the atmospheres of several transiting plan- infrared, while stellar emission falls dramati- ets (Snellen et al. 2010b; Crossfield et al. 2011; cally longward of its peak in the visible to Rodler et al. 2012; Birkby et al. 2013). Such ob- near-infrared. In addition, prominent molecular servations also come within 20% of the photon rotational-vibrational bands for numerous abun- noiselimit(Brogi et al.2014),whichindicatesthe dant molecules, such as water, methane, car- technique is less susceptible to the systematic ef- bon monoxide, and carbon dioxide, occur in the fects that limit multi-object observations. Thus, near- and mid-infrared. The IR has thus be- ELT high-dispersion spectroscopy may be espe- come the most productive spectral regionfor exo- cially well-suited to high-precision atmospheric planet characterization,despite the disadvantages studies of transiting planets. of poorer technology, orders-of-magnitude worse High dispersion spectroscopy with theELTs will thermalbackground,andfewerphotonscompared be also key to detecting atmospheric signatures to the visible range. of transiting Earth-like planets hosted by nearby HST/WFC3 can be used for NIR transit spec- stars, to be discovered by TESS and PLATO troscopy of any planet with large scale-height (see Snellen et al. 2013; Rodler & Lo´pez-Morales atmosphere, and emission spectroscopy for the 2014). hottest planets. It has proven capable of 30 ppm spectrophotometry(infifteen7-pixelspectralbins; 6.4. SOFIA Kreidberg et al.2014b),sufficienttodetectmolec- SOFIA combines a number of advantages for ularfeatures( 5.1). While the limitedwavelength § extremely precise time-domain optical and near- range prohibits detailed atmospheric retrieval us- infrared spectrophotometric observations using ing HST measurements alone, it is sufficient to its HIPO (Dunham et al. 2004) and FLITE- determine which exoplanets are hazy. It is also CAM (McLean et al. 2006) instruments, in par- possible to stitch together multiple HST orbits to ticular when used in simultaneous FLIPO mode obtain continuous phase measurements of an ex- (Angerhausen et al.2010;Smee et al.2012). SOFIA oplanet, hence constraining its global albedo and is able to avoid most of the perturbing varia- heat transport (Stevenson et al. 2014). tionsofatmospherictracegasesthatarethe main HST/COS can be used to perform reflection 10

ebook img

Characterizing Transiting Planet Atmospheres through 2025 PDF

0.57 MB·English
by  N. B. Cowan
#journals #arxiv
Save to my drive
Quick download
Download
Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Characterizing Transiting Planet Atmospheres through 2025

Characterizing Transiting Planet Atmospheres through 2025 N.B. Cowan1, T. Greene2, D. Angerhausen3, N.E. Batalha4, M. Clampin3, K. Colo´n5, I.J.M. Crossfield6, J.J. Fortney7, B.S. Gaudi8, J. Harrington9, N. Iro10, C.F. Lillie11, 5 1 J.L. Linsky12, M. Lopez-Morales13, A.M. Mandell2, and K.B. Stevenson14, on behalf of 0 ExoPAG SAG-10 2 1Amherst College ([email protected]) n 2NASA Ames Research Center a J 3NASA Goddard Space Flight Center 0 4Pennsylvania State University 3 5Lehigh University 6University of Arizona ] P 7University of California, Santa Cruz E 8The Ohio State University . 9University of Central Florida h p 10Universita¨t Hamburg - 11Lillie Consulting o r 12University of Colorado st 13Harvard-Smithsonian Center for Astrophysics a 14University of Chicago [ Executive Summary to maximize the scientific returns of TESS is to 1 adopt a triage approach. A space mission con- v The discovery of planets around other stars is sisting of a 1 m telescope with an optical–NIR 4 revolutionizing our notions of planet formation ∼ 0 spectrographcould measure molecular absorption and is poised to do the same for planetary cli- 0 for non-terrestrial planets discovered by TESS, 0 mate. Studying transiting planets is complemen- as well as eclipses and phase variations for the 0 tarytoeventualstudiesofdirectly-imagedplanets: hottest jovians. Such a mission could observe 2. 1) we can readily measure the mass and radius of up to 103 transits per year, thus enabling it to 0 transiting planets, linking atmospheric properties survey a large fraction of the bright (J < 11) 5 tobulkcompositionandformation,2)manytran- hot-Jupiters and warm sub-Neptunes TESS is ex- 1 siting planets are strongly irradiated and exhibit pected to find. The James Webb Space Telescope : v novel atmospheric physics, and 3) the most com- (JWST) could be used to perform detailed at- i mon temperate terrestrial planets orbit close to X mospheric characterization of the most interest- red dwarf stars and are difficult to image directly. ing transiting targets(transit, eclipse,and—when r a We have only been able to comprehensively char- possible—phase-resolved spectroscopy). TESS is acterize the atmospheres of a handful of transit- also expected to discover a few temperate terres- ing planets, because most orbit faint stars. The trialplanetstransitingnearbyM-Dwarfs. Charac- TransitingExoplanetSurveySatellite(TESS)will terizingtheseworldswillbetime-intensive: JWST discover transiting planets orbiting the brightest willneedmonthstoprovidetantalizingconstraints stars, enabling, in principle, an atmospheric sur- onthe presence ofanatmosphere,planetary rota- vey of 102–103 brighthot Jupiters and warmsub- tionalstate,clouds, andgreenhousegases. Future Neptunes. Uniform observations of such a statis- flagship missions should be designed to provide tically significant sample would provide leverage better constraints on the habitability of M-Dwarf to understand—and learn from—the diversity of temperate terrestrial planets. short-period planets, and would identify the mi- nority of truly special planets worthy of more in- tensive follow-up. We argue that the best way 1 1. Context plate tectonics, etc. However, the characteriza- tion of exoplanets presents a challenge familiar to The study of exoplanet atmospheres has ex- astronomers: our targets are so distant that we ploded in the past decade. In 2013, the Ex- only see them as unresolved dots. oplanet Exploration Analysis Group (ExoPAG) Manyaspectsofplanetarysciencearecurrently created—with approval from NASA’s Astro- accessible for exoplanets, or soon will be. Since physics Subcommittee—a tenth Study Analysis weobserveexoplanetarysystemsfromtheoutside, Group (SAG-X) to consider what NASA could the easiest aspect to constrain is the architecture do in the next decade to better understand the of planetary systems, and indeed our theories of atmospheres of transiting planets. SAG-X had planet formation are currently being revolution- openmembershipandinvolvedthreepresentations ized by our growing knowledge of planetary de- to the exoplanet community. The first presenta- mographics and architecture. Transiting planets tion, outlining the challenges and opportunities are crucial to our understanding of planet forma- ofstudyingtransitingexoplanetatmospheres,was tion, because they are the only planets for which made at the ExoPAG 8 meeting preceding the wecanhopetoknowtheorbitalarchitecture,bulk Oct. 2013 Division of Planetary Sciences meet- density, and atmospheric composition. ing in Denver, CO. We held a mini-workshop on The study of individual planets is likewise pro- the capabilities of the James Webb Space Tele- gressing from the top–down: first the exospheres, scope for characterizing transiting exoplanets at then the atmospheres, and the surfaces last. It ExoPAG 9 preceding the Jan. 2014 American As- is therefore difficult to make definitive statements tronomical Society meeting in Washington, DC. about the surface conditions of exoplanets, and Interestedmembersofthecommunityhelpeddraft their interiors will only be known to us through the current document over the course of the 2014 their bulk density, and surface character. The calendar year and we presented it at ExoPAG 11, studyofplanetaryatmospheres,however,ispoised preceding the Jan. 2015 AAS meeting in Seat- to be revolutionized by observations of “exo- tle, WA. Multiple drafts of this report have been climes.” The first suchmeasurements constrained circulated to the ExoPAG membership and the planetary-scale temperature structure and com- exoplanet community at large and we have done position, while the second generation of measure- our best to implement feedback. This document ments leveraged the orbital motion of the planet is therefore a consensus view of what can and to infer the horizontal temperature structure of shouldbe done in the field of transiting exoplanet the planet. Planned instruments will enable 3D atmospheres in the next decade. measurements of atmospheric composition, tem- perature structure, and winds. 2. Planetary Science from the Top–Down Our knowledge of Earth and the So- 3. Planetary Climate lar System planets will always exceed our A planet’s albedo determines how much knowledge of any individual exoplanet, but radiation it absorbs, atmospheric composi- the diversity of exoplanets enables the sta- tion dictates how energy trickles up via ra- tistical study of planets to crack difficult diation and convection, while atmospheric problems in planetary science. (andoceanic)dynamicsdeterminehowheat What started as a trickle in the mid 1990’s is is transported from regions receiving more now a torrent, with over one thousand extrasolar sunlight to those receiving less. planets currently known, and thousands of can- Theprimaryaspectofplanetaryclimateistem- didates awaiting confirmation. The study of ex- perature, averaged over time, and often space. A oplanets has already revolutionized our view of detailed calculation of climate involves radiative planet formation, and will soon do the same to transfer, fluid dynamics, chemistry and, in the our understanding of planetary atmospheres and case of an inhabited planet, biology. Stripped to interiors. The diversity of exoplanets gives us the itsessentials,however,climatedescribestheinter- leverage to crack hard problems in planetary sci- action of star light with a planetary atmosphere ence: cloud formation, atmospheric circulation, 2 (Figure 1). ate terrestrial planets that rotate slowly allow us to empirically verify the effects of Coriolis forces on atmospheric circulation, super-Jupiters let us test the effects of surface gravity on Jovian atmo- spheres, etc. While the most detailed atmospheric studies will always focus on Earth, and in situ measure- mentswillbelimitedtotheSolarSystem,thevast majority of planets are extrasolar. This means thatthemostextremeworlds,andthosemostlike theEarth,areexoplanets. Onlybystudyingthese planetscanwehopetodevelopcomprehensivethe- Fig. 1.— Planetaryclimate is determined by stel- ories of climate. To paraphrase Kipling (1891): lar radiation interacting with a planetary atmo- What should they know of Earth who only Earth sphere. Incoming radiation (solid yellow lines) is know? either reflected (dashed yellow lines) or absorbed. The warm air emits at longer wavelengths than 4. Transiting Planets the incoming light. Longwave radiation is read- Transiting planets are representative of ily absorbed by greenhouse molecules in the at- planets on short-period orbits. Exoplan- mosphere, resulting in radiative diffusion (small ets that transit Sun-like stars tend to be red arrows). The inefficient upward heat trans- hot, while those transiting low-mass stars portmeansthatthelowerpartsoftheatmosphere are merely warm, and represent the major- tendtobehotterthantheoverlyingregions,andin ity of temperate terrestrial planets. Tran- practicemostplanetaryatmospheresconvectover sit spectroscopy probes atmospheric opac- some pressure range (blue arrows). The cooler, ity, while measurements of thermal emis- thinnerupperregionsoftheatmosphereemitther- sion constrain vertical and horizontal tem- malradiation(dashedredlines)thatbalancesthe perature structure. High cadence, high- absorbed shortwaveradiation. precisionemissionspectroscopy throughout a planet’s orbit enables 3D mapping of its Asimpleenergybalancemodel,shownschemat- atmosphere. ically in the left panel of Figure 2, can therefore A transiting planet is a planet that passes in be used to predict planetary climate. Unfortu- betweenitshoststarandtheobserver. Anyplanet nately, despite over a century of research into that orbits a star may be a transiting planet, but Earth’s climate, there are currently no compre- transitingplanetsareessentiallysynonymouswith hensive, predictive theories for cloud formation short-period planets, since the probability that a (planetary albedo), volatile cycling (greenhouse planet transits is the ratio of the stellar radius to gasabundances),orwindspeeds(heattransport). theplanet–starseparation. Thismakesthetransit The empirical approachto this challenge is to ac- methodanimprobablewaytostudy SolarSystem quire observations for many different planets in analogs. the hopes of uncovering the principles of climate Direct imaging is a promising approach to (right panel of Figure 2); the first step in that di- studyingtheatmospheresofSolarSystemanalogs, rectionhas beenthe study ofSolarSystemworlds but close-in planets are difficult to study in this over the last half-century. way, because any means of blocking/nulling the The logical next step, already underway, is to starlight is liable to block the planet, too. We characterize the atmospheres of extrasolar plan- instead study the combined light of the plane- ets. Exoplanets are more diverse than the planets tary system, so the dominant source of photons orbiting our Sun, and hence provide more lever- (and photon noise) is the star rather than the agefor testing theories. Onthe other hand, many planet. Thesignal-to-noiseratioformeasurements exoplanets are similar to those in our Solar Sys- of planetary light is therefore proportional to the tem, providing crucial Rosetta stones: temper- 3 !"#$%&’ /0*12,-:9+",)2$’’ !"#$%&’ ;$<$+-*1+$’’ 3)$+-0&2+&).’ 3)$+-0&2+&).’ 3,4)"$’ 3,4)"$’ 3708*+$’ 3708*+$’ ()*+,-.’ 5",4*-$’ ()*+,-.’ 5",4*-$’ /$4)$0*-70$’ /$4)$0*-70$’ 6&%$"’ 6&%$"’ =>*2$:9+",)2$’ 94,22,&1’’ /0*12)&0-’ 6*)),1?’ /0*12)&0-’ 3)$+-0&2+&).’ Fig. 2.—Left: Itispossibletopredictaplanet’sclimategivenitsalbedo(reflectiveness),atmosphericopacity (greenhousegas abundances),and heattransport(wind speeds and oceancirculation). Unfortunately, these critical inputs cannot be predicted in general. Right: Fortunately, it is possible to measure planetary albedo,atmosphericcomposition,andheattransport,evenforexoplanets(inmanycases,there aremultiple independent means of constraining atmospheric properties; we list only a few in the interest of clarity). This makes it possible to empirically determine a planet’s climate, even in the absence of a fully predictive theoryofplanetaryclimate. In the longrun, suchobservationsmayrevealsomeofthe underlyingprinciples governing cloud formation, volatile cycling, and large-scale circulation. planetary flux, rather than its square-root; the ofhotJupitersoftenresultsinmassloss,andsome usualbiasesinfavorofbiggerandbrightertargets arethoughtto be undergoingRochelobe overflow are especially true for transiting planets. onto their host star (Li et al. 2010). Characterizing transiting planets would be of The most common currently detectable tran- merely theoretical interest, except that many ex- siting planets in the Galaxy are sub-Neptunes in oplanets orbit much closer to their stars than in tight (10–100 days) orbits around their host stars ourSolarSystem. Someoftheseunfamiliarworlds (Howard et al. 2012; Fressin et al. 2013). These are actually rare, like hot Jupiters (Wright et al. warm sub-Neptunes have masses dominated by 2012), while others represent a common outcome rock and ice, but covered in thick H+He atmo- of planet formation, as with packed systems of spheres, and are sufficiently cool (500–1000 K) warm sub-Neptunes (Lissauer et al. 2011). that photochemistry often trumps thermochemi- A hot Jupiter is a jovian planet that orbits calequilibrium(Moses2014). Theapriori transit its host star with a period of .1 week.1 Such probabilityofsuchplanets is low,butthey arein- planets only exist around 1% of Sun-like stars trinsicallycommon,occurringaroundroughlyhalf (Wright et al. 2012), but h∼ave a transit probabil- ofstars. Assuch,theyformthebulkoftheKepler ity of 10%, so there is a transiting hot Jupiter crop. The densities of these planets suggest they for every few thousand FGK stars. These ex- are largely made of ice, and/or have substantial treme worlds experience strong radiative forcing H+He atmospheres, hence their name. Our expe- resulting in day–night temperature contrasts of rience from the Solar System suggests that these hundreds to thousands of K (Showman & Guillot planets likely have high atmospheric metallicities 2002). The high atmospheric temperatures ther- (Fortney et al. 2013). In short, these planets are mally ionize alkali metals, effectively coupling at- expected to be less extreme than hot Jupiters in mospheric dynamics to planetary magnetic fields terms of radiative forcing, but probably more in- (Perna et al.2010a,b). Theradiativeenvironment teresting in terms of chemistry. Temperate terrestrial planets transiting M- 1Hot Jupiters should not be confused with young Jupiters, dwarfstarsareoftentoutedasthepoor-astronomer’s thetypicaltargetsofcurrentdirect-imagingsurveys,which Earth analog, since they are easier to detect arealsohot. 4 and characterize than a true Earth twin. Based on what we currently know, however, M-Dwarf planets are the most common habitable worlds. !"#$%&’( That is because: 1. rocky planets are much more !7*’&&( 6,7*’&&( common in the temperate zones of M-Dwarfs 1$2234&( (Dressing & Charbonneau 2013; Morton & Swift 2014) than in the temperate zones of Sun-like ./+&’&( stars(Petigura et al.2013;Foreman-Mackey et al. 54+*-’*( 2014; Farr et al. 2014), 2. small stars are more common than big stars (e.g., Bochanski et al. 0*’&"’,-( 2010), 3. the tidally-locked nature of these plan- )*+,&$-( ets is not a challenge to climate (Joshi et al. 1997;Merlis & Schneider2010;Edson et al.2011) and may double the width of the habitable zone (Yang et al. 2013), 4. the red stellar ra- diation results in a weaker ice-albedo feedback Fig. 3.— A planet whose orbit is nearly edge-on and hence stabler climate (Joshi & Haberle 2012; will transit in front of its star. This is more likely Shields et al. 2013, 2014), and 5. the slow main to occur for planets that orbit close to their star, sequence evolution of M-Dwarfs means that a ge- so transiting planets are often synonymous with ological thermostat is not strictly necessary to short-periodplanets. Theamountoflightblocked maintain habitable conditions for billions of years during transit tells us the planet’s size, while the (Kasting et al. 1993). Studying temperate terres- transmission spectrum during transit is sensitive trial planets around red dwarfs is our best shot at to scatteringandabsorptioninthe planet’s upper understanding habitability writ large. atmosphere. During transit we see the planet’s nightside, but the planet’s orbital motion even- 4.1. Transit tually brings the dayside into view. The dayside reflectssunlightandisoftenhotterthanthenight- When a planet passes in front of its host star, side, causing variations in brightness known as it blocks a fraction of the star’s light equal to the phases. Atthetopoftheorbit,thestarwilleclipse planet/star area ratio. Inferring the stellar radius the planet, allowing us to measure the brightness from its color and surface gravity,it is possible to ofthestarwithoutcontaminationfromtheplanet; convertthisrelativesizeintoaphysicaldimension. a difference measurement yields the planet’s day- Ontopoftheopaqueplanetarydiskthereisan side brightness. Reflected light eclipse measure- annulus of partially transparent atmosphere that ments are sensitive to the planetary albedo and filters starlight (Figure 3). The spectrum of a reflected light spectroscopy constrains the nature planet in transit therefore contains an imprint of of scattering. Thermal eclipse measurements can scatteringandabsorptionthatoccursintheupper be used to estimate the planet’s dayside temper- atmosphere near the planet’s day-night termina- ature, while thermal spectroscopy is sensitive to tor. Evenif the bulk composition can be taken as atmosphericcompositionandverticaltemperature agiven(e.g.,H+Heforjovianworlds),itisstilldif- profile. (from Cowan 2014) ficult to nail down the abundances of trace gases, andhigh-altitudehazescanwashoutspectralfea- tures(Burrows2014b): eveniftheskylookedclear 4.2. Eclipse to an inhabitant, it may very well be opaque to the grazing rays relevant for transit spectroscopy. It is possible to use eclipses of the planet by Nonetheless, transmission spectroscopy is a pow- its star to isolate planetary light. A planet that erfulcharacterizationtoolthatcanonlybeapplied passes directly in front of its host star usually to transiting planets. passes directly behind it half an orbit later. The brightness of the planetary system immediately before and after occultation is compared to the 5 brightness during eclipse, and the difference is a 4.3. Phases measure of the planet’s dayside brightness. Horizontal and temporal differences in plane- We may then convert the eclipse measurement tary temperature produce time variationsin ther- into an estimate of the planet’s geometric albedo mal emission. For example, the dayside of a or dayside brightness temperature, depending on slowly-rotating planet might appear warmer and whether the instrument is sensitive to visible or hence brighter than its nightside. Most known thermal radiation. Spectrally resolved eclipse exoplanets have curiously eccentric orbits and measurements can constrain atmospheric scatter- thereforeexperience significantseasonsdue to the ing, composition, and vertical temperature profile changing star-planet separation. Rocky planets (Burrows 2014b). are expected to form with randomly orientedspin High-resolutionemissionspectraareableto re- axes, which leads to obliquity seasons. Disentan- solve molecular lines (as opposed to bands), pro- gling the diurnal cycle, eccentricity seasons and viding two novelcapabilities: performing Doppler obliquity seasons based on thermal phase varia- measurements on the planet itself, and probing tions is a work in progress (Cowan et al. 2012b). the compositions of cloudy worlds. For tran- Tides damp obliquity, slow planetary rotation, siting planets the system inclination and plane- and damp orbital eccentricity (e.g., Heller et al. tary mass are known, but Doppler measurements 2011), so most short period planets do not ex- might be used to infer high-altitude wind veloc- perience seasons, but probably have permanent ities (Snellen et al. 2010a) and planetary rota- day and night hemispheres. The day–night tem- tion (Snellen et al. 2014), but these are arguably perature contrast is therefore an indirect mea- accessible via thermal phase curve and eclipse sure of atmospheric heat transport (Cowan et al. mapping measurements (Rauscher & Kempton 2007,2012a). High-precisionthermalphasecurves 2014), or transit morphology (Carter & Winn can be inverted to construct coarse longitudi- 2010). Clouds, on the other hand, have been nal temperature maps of short-period exoplan- most problematic in grazing transit geometry ets (Knutson et al. 2007; Cowan & Agol 2008), (Kreidberg et al.2014b;Knutson et al.2014). Al- and even contain indirect information about the though many emission spectra have so far been planet’slatitudinalfluxdistribution(Cowan et al. featureless, this is likely due to coarse spectral 2013). Full-orbit observations of emission spectra resolution, large measurement uncertainties, and enablespatially-resolvedinferencesoftemperature roughly isothermal atmospheres (Hansen et al. structureandcomposition(Stevenson et al.2014). 2014, and references therein). With full-phase observations at high-cadence, Theeclipsedepthissensitivetothehemisphere- high signal-to-noise, and high spectral resolu- averagedpropertiesofaplanet,whiletheverybe- tion, it should be possible to constrain the 3- ginning and end of an eclipse offer a means of re- dimensional composition and temperature of an solving the planet’s dayside: as the planet disap- exoplanet’satmosphere. Thiswouldallow,forthe pears behind its star and reappears (ingress and first time, realistic initialization and/or testing of egress, respectively), the star’s edge scans across a general-circulation model with active chemistry the planet. It is possible to invert these raster (as opposed to fixed chemistry). This is the holy scans to construct a coarse two-dimensional map grail of atmospheric characterization, and a top of the planet’s dayside. The first-order effect is priority for future exoplanet observations. the phase offset of the eclipse due to a zonally- advected hot-spot (Williams et al. 2006) and was 5. Lessons Learned first detected by Agol et al. (2010). The eclipse timing offset is largelydegenerate with orbital ec- The principle lessons learned from the centricity,butmaybeteasedoutifitischromatic. first dozen years of exoplanet atmospheric The detailed morphology of ingress and egress observations are: (1) short-period plan- provides a smaller but more robust signal about ets are not all alike, nor are they a one- the 2D flux distribution of the planet’s dayside parameter family, (2) observations are usu- (Rauscher et al. 2007), observed by Majeau et al. ally systematics-limited, making out-of- (2012) and de Wit et al. (2012). occultation data critical to modeling detec- 6 tor behavior, and dictating that repeatabil- reduction, and analysis of eclipse measurements ity is the only reliable test of accuracy. have improved, the broadband emission measure- ments of hot Jupiters have trended toward fea- 5.1. Short-Period Planets tureless Planck spectra (Hansen et al. 2014). Dayside broadband emission measurements Observations of exoplanet atmospheres have of hot Jupiters suggest that the hottest plan- so far been weakly constraining, but they have ets cannot effectively transport heat to their provided a few robust surprises (for more com- nightside (Cowan & Agol 2011), and this trend plete reviews, see Burrows 2014a,b; Bailey 2014; has been corroborated by thermal phase mea- Heng & Showman 2014). surements of two planets (Cowan et al. 2012a; Alkali metals have been detected in the atmo- Maxted et al. 2013). Thermal phase and eclipse spheres of some hot Jupiters (Charbonneau et al. maps are consistent with an equatorial hotspot 2002), but are obscured by Rayleigh-scattering (Majeau et al. 2012; de Wit et al. 2012) and sug- hazesinothers(Pont et al.2013);aRayleighscat- gest eastward equatorial winds at a variety of tering slope has also been seen in the transmis- depths (Knutson et al. 2007, 2009, 2012) with a sion spectrum of a warm ice giant (Biddle et al. characteristic advective time comparable to the 2014). Hazes are present on a warm sub- radiative timescale (Agol et al. 2010). Thermal Neptune (Kreidberg et al. 2014b) and a hot Nep- phase measurements of hot Jupiters on eccen- tune (Knutson et al. 2014), but absent on an- tric orbits suggest radiative times less than a day other (Fraine et al. 2014). Some hot Jupiters and equatorial super-rotation (Lewis et al. 2013; areslowlyevaporating(Vidal-Madjar et al.2004), Wong et al. 2014). while there is questionable evidence that one The measured geometric albedos of transit- planet is overflowing its Roche lobe and accret- ing planets are generally small (Rowe et al. 2008; ing onto its star (Fossati et al. 2010; Cowan et al. Kipping & Spiegel 2011) and blue (Evans et al. 2012a). 2013) but with notable exceptions (Demory et al. Water vapor absorption has now been se- 2011). Moreover, the one hot Jupiter with high curely detected in hot Jupiters with the WFC3 albedo appears to have spatially inhomogeneous instrument on HST using both transit spec- clouds (Demory et al. 2013). troscopy(Deming et al.2013;Huitson et al.2013; Wakeford et al. 2013; Mandell et al. 2013) and 5.2. Spitzer eclipse spectroscopy(Kreidberg et al.2014a), and one hot Neptune shows evidence of water as By operating in the mid-infrared without the well (Fraine et al. 2014). However, claims of confounding telluric IR background, the Spitzer molecular absorption in transit and eclipse mea- Space Telescope overcame a key barrier and mea- surements with other instruments remain con- suredthe firstexoplanetaryphotons in the Fall of troversial. Early detections in NICMOS spec- 2004. Teams led by Charbonneau et al. (2005) troscopy data (Swain et al. 2008, 2009a,b) have and Deming et al. (2005) observed secondary been called into question (Gibson et al. 2011, eclipses of TrES-1 and HD 209458b, respectively, 2012;Crouzet et al.2012),whileanalysesofmulti- submitting their independent reports simultane- bandeclipsephotometrywithSpitzer showingev- ously and participating in a joint NASA press idence of temperature inversions (Knutson et al. releasein April 2005. Spitzer remainedthe obser- 2008), disequilibrium chemistry (Stevenson et al. vational tool of choice for most of the subsequent 2010; Swain et al. 2010) and super-Solar C/O decade, and it is useful to summarize what we (Madhusudhan et al. 2011a) have been disputed bysubsequentstudies(Beaulieu et al.2011;Mandell et al.Madhusudhanetal. (2011b) forces plausible chemistry, 2011; Crossfield et al. 2012; Zellem et al. 2014; Line&Yung (2013) and Lineetal. (2014) do not. Since Diamond-Lowe et al. 2014).2 As the acquisition, eclipsespectroscopyrepresentsdisk-averagedemissionfrom regionswithdifferenttemperaturestructures andpossibly different chemistries, it is not obvious which strategy is 2Even when the data reduction scheme is taken at face moresound. A pragmatic solution might beto report the value, disagreements may occur based on the assump- best-fit “reasonable chemistry” solution, but with uncer- tions that go into spectral retrieval. For example, while taintyestimatesthatare“chemistryagnostic.” 7 have learned during this time. ever, would have greatly benefited from instru- In its cryogenic mission (2003 – 2009), Spitzer ments designed to stare at bright point sources, had six photometric channels useful for exo- and the same observations would have had much planet characterization, centered at 3.6, 4.5, 5.8, more scientific value had they been obtained in a 8, 16, and 24 µm. Channels ranged from 1– uniform way and reduced with a uniform, open- 3 µm wide. Spitzer’s InfraRed Spectrograph had sourcepipeline. The Spitzer experience ultimately a low-resolution, 5.3–14 µm mode useful for ex- justifies the expense of a purpose-built mission to oplanets. Unfortunately, the rate of transiting- perform a survey of exoplanet atmospheres. planet detection was low until around 2010,when Spitzer’s after-launch data analysis efforts at- the cryogen was already gone. In Warm Spitzer tempted to characterize the varying sensitivity only the 3.6 and 4.5 µm photometric channels across the faces of individual pixels and to fit its remain operational; spectroscopy is unavailable. temporal response curves. However, pre-launch Thus, only HD 209458b and HD 189733b were laboratory calibration measurements at better everobservedspectroscopically;only they andGJ thanthe0.01%levelcouldhavedoneabetterjob, 436b were observed at 24 µm; and only GJ 436b, as could calibrating long stares at point sources. TrES-1, TrES-4, HD 189733b, and HD 149026b For example, the prominent, time-dependent sen- were observed at 16 µm. However, numerous sitivity“ramp”at5.8,8,and16µmwasunknown planets have measurements in the four shortest prior to launch because lab calibrations on bright bands, and even more have published 3.6 and 4.5 sources lasted only a short time, rather than the µm eclipse depths (Hansen et al. 2014). hours-long timescale of the ramp. Learning from Withanapertureofjust85cm,Spitzer wasde- thisexperience,instrumentsthatreduceintrapixel signed for 10% absolute and 1% relative photom- sensitivity variations, that have long-term stabil- etry, an order of magnitude poorer precision than ity, and that point consistently are now being requiredforexoplanetstudies. Theracetoachieve designed and proposed for dedicated missions. results thus became a contest of systematic cor- Even JWST, which was designed before some rectionmethods. Bayesiansampling(e.g.,Markov key lessons were learned from Spitzer, is being ChainMonteCarlo)replacedmodelfittingbysim- calibrated with enhanced emphasis on observing ple χ2 minimization, Bayesian methods for com- modes suitable for exoplanet studies. paring models with different numbers of free pa- rameters became the norm, and a variety of nu- 6. Exoplanet Observatories Through 2025 mericalapproachesmodeledSpitzer’ssystematics. Manyground-and spaced-basedobserva- Although known for a decade or more, improved toriescan beusedtostudytheatmospheres image centering and photometric extraction tech- of hot Jupiters and warm sub-Neptunes in niques also finally entered common use. Ulti- the next decade. For terrestrial planets, es- mately, contrast uncertainties better than 0.01% peciallythe temperatevariety, the choice of for single eclipses became possible. While an im- instruments is much more limited: only the pressive improvement over Spitzer’s design, the Extremely Large Telescopes and the James 0.01% uncertainty still left the broadband mea- WebbSpace Telescopewillbecapable ofat- surements at S/N of only 10 for most observed ∼ mospheric observations. planets. The 1–3µmwidth andsmallnumberof photometric channelsfurther confoundedspectro- 6.1. 4m telescopes scopic retrieval. Most Spitzer exoplanet observations consist of 4-meter class telescopes are an underused re- unrepeatedeclipseobservationsobtainedwithdif- sourcein the atmospheric characterizationofexo- ferent observing schemes and analyzed by dis- planets. Ground-based, optical multi-object spec- parate groups using a variety of evolving reduc- trographs typically capture light from 0.4–1.0 mi- tion pipelines. The poor signal-to-noise is mostly cronsandaresensitiveto thepresenceofH2O,al- a testament to the relative faintness of currently kali metals such as Na and K, and metal hydrides known transiting planets; NASA’s TESS mission in cloud-free atmospheres. Additionally, in atmo- will addressthis problem. The observations,how- spheres with clouds or hazes, we can use spectral 8 informationtodeterminethesizedistributionand Jupiter-size planets. Indeed, the first ground- altitude of the cloud/haze particles. based detection of sodium in an exoplanet atmo- In order to achieve the equivalent precision sphere came from the Hobby-Eberly Telescope and be competitive with larger telescopes, these (Redfield et al.2008). Current10m-classfacilities smaller telescopes must acquire multiple transits include the Keck telescopes, the Gran Telescopio of a single target. For comparison, four transits Canarias(GTC),Hobby-EberlyTelescope,South- of a single target with a 4-meter telescope, such ern African Large Telescope, Subaru, Very Large as Mayall or Blanco, have the equivalent photon- Telescope, Large Binocular Telescope, and Gem- limited precision of two transits with an 8-meter ini. These facilities offer a number of instruments telescope,suchasGemini. However,acquiringnu- suitable for exoplanet atmospheric characteriza- merous transit observations with the smaller tele- tion. For example, the primaryinstrumenton the scope will mitigate residual atmospheric effects GTC suitable for exoplanet observations is the that limit our precision with ground-based obser- Optical System for Imaging and low Resolution vations, plus these observations will likely have Integrated Spectroscopy (Cepa et al. 2000, 2003, a higher duty cycle, thereby outperforming the OSIRIS), which offers a moderately sized field of largertelescope. Withasufficientnumberoftran- view of 7.8 7.8 arcmin. OSIRIS has the capa- × sits, 4-meter class telescopes can contribute ex- bility for standard long-slit spectroscopy, and it citing, cutting-edge research like with larger tele- also offers a unique tunable filter imaging mode, scopes, but at a fraction of the cost. allowing the user to specify custom optical band- passes with FWHM = 1.2–2.0 nm. In this mode, High precision photometry in the near-infrared observationscanbeconductedinmultipletunings of exoplanet occultations provides a direct mea- nearly-simultaneously, thus allowing for narrow- surement of thermal emission from hot planets, with single-eclipse precisions of 2 10 4 now all band spectrophotometry. Bean et al. (2010) used − × multi-object spectroscopy on the VLT to obtain but routine (Croll et al. 2014). Wide-field NIR transit spectra of GJ 1214b, with an epoch-to- photometry with 4m class telescopes is a viable epoch white-light accuracy of 2 10 4. Other means of achieving high precision photometry for ∼ × − up-and-coming facilities include GMOS on Gem- gas and ice giants orbiting relatively bright stars. ini andMOSFIRE onKeck. Thanks to their large Thewidefieldofviewallowsforasignificantnum- aperture 10 m telescopes are capable of the high ber of reference stars for differential photometry, precision spectroscopy or spectrophotometry re- which is critical for minimizing effects of Earth’s quired tomeasurethe small signals from exoplanet atmosphere (typically the limiting factor for high atmospheres, e.g., alkali metals in hot Jupiter at- precisionphotometryinthenear-infraredfromthe mospheres (Sing et al. 2011) or methane in warm ground). sub-Neptunes (Wilson et al. 2014). Despite the advantages of these types of facili- ties, only a handful exist: WIRCam on the 3.6m 6.3. Extremely Large Telescopes CFHT, WFCAM on the 3.8m UKIRT, NEW- FIRM on the KPNO 4m Mayall Telescope, and By 2025,we expect 1–3 of the extremely large, the Spartan IR Camera on the 4.1m SOAR. Of ground-based telescopes (ELTs) currently being these, only WIRCam on CFHT and WFCAM planned (GMT, TMT, E-ELT) to be operational. on UKIRT have been demonstrated for exoplanet These facilities may provide several possible ben- studies (e.g., Croll et al. 2011, Col´on & Gaidos efits: their larger apertures will enable studies of 2013). largernumbersoftransitingplanetsaroundfainter stars, with techniques used today on 8–10 m tele- 6.2. 10m telescopes scopes; if systematic effects can be controlled, the largeraperturesmayalsoallowplanetsinbrighter As demonstrated by Col´on & Ford (2009), systems to be studied at higher precision; finally, large ground-based telescopes are capable of con- byrelievingsomepressureonexisting8–10mtele- tributing significantly to photometric follow-up scopes it may be possible to use these older facili- efforts for both small, long-period planets dis- ties for large-scale survey science precluded today covered by missions like Kepler and for larger, due to the limited availability of observing time. 9 The ELTs will be optimized to observe faint source of systematic noise for ground-based ob- targets,oftenusingadaptiveoptics–implyingrel- servations at shorter NIR wavelengths, as these atively narrow fields of view. Thus fewer nearby telluric molecules are also the species of interest systems will be accessible to the differential pho- in the exoplanetary atmospheres. The SOFIA tometric and spectroscopic techniques so popular telescope, operating at much lower temperatures today, which use multiple comparison stars to re- (240 K) than ground-based telescopes, reduces move telluric and instrumental noise (Bean et al. thermal background contributions that are the 2010; Gibson et al. 2013; Crossfield et al. 2013). dominant noise source for transit observations at Today, these efforts perform within factors of 2–3 wavelengths longer than 3 microns. SOFIA can of the spectroscopic photon noise limit, but there observe time-critical events, such as planetary oc- is no consensus as to whether the extra noise cultations,underoptimizedconditions,asdemon- comes from detectors, instruments, telescopes, or strated by an observation of a Pluto occultation the Earth’s atmosphere. For ELT multi-object in June 2011 (Person et al. 2013). SOFIA’s ini- transitspectroscopy tosucceed, these noisesources tial forays into transit observations have yielded must be identified and mitigation strategies incor- precisions within a factor of 2 ofthe Poissonlimit porated into the new instruments currently being (Angerhausen et al. submitted). In contrast to designed. space telescopes it is possible to update SOFIA High-dispersion spectroscopy may be the niche with state-of-the-art instrumentation, e.g. with a inwhichELTs canbest makesignificantprogress. dedicated 2nd generation exoplanet instrument Observations at high dispersion (λ/∆λ&20,000) (such as NIMBUS, McElwain et al. 2012). canmeasureuniquemolecularsignatures,thermal 6.5. Hubble Space Telescope profiles, global atmospheric circulation, and or- bitalmotionandhasalreadybeenusedtocharac- Planetary emission peaks in the near-to-mid- terize the atmospheres of several transiting plan- infrared, while stellar emission falls dramati- ets (Snellen et al. 2010b; Crossfield et al. 2011; cally longward of its peak in the visible to Rodler et al. 2012; Birkby et al. 2013). Such ob- near-infrared. In addition, prominent molecular servations also come within 20% of the photon rotational-vibrational bands for numerous abun- noiselimit(Brogi et al.2014),whichindicatesthe dant molecules, such as water, methane, car- technique is less susceptible to the systematic ef- bon monoxide, and carbon dioxide, occur in the fects that limit multi-object observations. Thus, near- and mid-infrared. The IR has thus be- ELT high-dispersion spectroscopy may be espe- come the most productive spectral regionfor exo- cially well-suited to high-precision atmospheric planet characterization,despite the disadvantages studies of transiting planets. of poorer technology, orders-of-magnitude worse High dispersion spectroscopy with theELTs will thermalbackground,andfewerphotonscompared be also key to detecting atmospheric signatures to the visible range. of transiting Earth-like planets hosted by nearby HST/WFC3 can be used for NIR transit spec- stars, to be discovered by TESS and PLATO troscopy of any planet with large scale-height (see Snellen et al. 2013; Rodler & Lo´pez-Morales atmosphere, and emission spectroscopy for the 2014). hottest planets. It has proven capable of 30 ppm spectrophotometry(infifteen7-pixelspectralbins; 6.4. SOFIA Kreidberg et al.2014b),sufficienttodetectmolec- SOFIA combines a number of advantages for ularfeatures( 5.1). While the limitedwavelength § extremely precise time-domain optical and near- range prohibits detailed atmospheric retrieval us- infrared spectrophotometric observations using ing HST measurements alone, it is sufficient to its HIPO (Dunham et al. 2004) and FLITE- determine which exoplanets are hazy. It is also CAM (McLean et al. 2006) instruments, in par- possible to stitch together multiple HST orbits to ticular when used in simultaneous FLIPO mode obtain continuous phase measurements of an ex- (Angerhausen et al.2010;Smee et al.2012). SOFIA oplanet, hence constraining its global albedo and is able to avoid most of the perturbing varia- heat transport (Stevenson et al. 2014). tionsofatmospherictracegasesthatarethe main HST/COS can be used to perform reflection 10

See more

The list of books you might like

Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.
DMCA & Copyright: Dear all, most of the website is community built, users are uploading hundred of books everyday, which makes really hard for us to identify copyrighted material, please contact us if you want any material removed.
Copyright @ PDFdrive.to, 2022 | We love our users