Identifying new opportunities for exoplanet characterisation at high spectral resolution
Remco J. de Kok, Jayne Birkby, Matteo Brogi, Henriette Schwarz, Simon Albrecht, Ernst J.W. de Mooij, Ignas A.G. Snellen
aa r X i v : . [ a s t r o - ph . E P ] D ec Astronomy&Astrophysicsmanuscript no. crisens2 c (cid:13)
ESO 2018August 13, 2018
Identifying new opportunities for exoplanet characterisation athigh spectral resolution
R.J. de Kok1, J. Birkby2, M. Brogi2, H. Schwarz2, S. Albrecht3, E.J.W. de Mooij4, and I.A.G. Snellen2 SRON Netherlands Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht, The Netherlands e-mail:
[email protected] Leiden Observatory, Leiden University, Postbus 9513, 2300 RA Leiden, The Netherlands Department of Physics, and Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology,Cambridge, Massachusetts 02139, USA Department of Astronomy and Astrophysics, University of Toronto, 50 St. George Street, Toronto, Ontario M5S 3H4, Canada
ABSTRACT
Context.
Recently, there have been a series of detections of molecules in the atmospheres of extrasolar planets using high spectralresolution (R ∼ Aims.
We aim to identify new ways of increasing the planet signal in these kinds of high-resolution observations. First of all, wewish to determine what wavelength settings can best be used to target certain molecules. Furthermore, we want to simulate exoplanetspectra of the day-side and night-side to see whether night-side observations are feasible at high spectral resolution.
Methods.
We performed simulations of high-resolution CRIRES observations of a planet’s thermal emission and transit between 1-5 µ m and performed a cross-correlation analysis on these results to assess how well the planet signal can be extracted. These simulationstake into account telluric absorption, sky emission, realistic noise levels, and planet-to-star contrasts. We also simulated day-side andnight-side spectra at high spectral resolution for planets with and without a day-side temperature inversion, based on the cases of HD189733b and HD 209458b. Results.
Several small wavelength regions in the L-band promise to yield cross-correlation signals from the thermal emission of hotJupiters of H O, CH , CO , C H , and HCN that can exceed those of the current detections by up to a factor of 2-3 for the sameintegration time. For transit observations, the H-band is also attractive, with the H, K, and L-band giving cross-correlation signals ofsimilar strength. High-resolution night-side spectra of hot Jupiters can give cross-correlation signals as high as the day-side, or evenhigher. Conclusions.
We show that there are many new possibilities for high-resolution observations of exoplanet atmospheres that haveexpected planet signals at least as high as those already detected. Hence, high-resolution observations at well-chosen wavelengthsand at di ff erent phases can improve our knowledge about hot Jupiter atmospheres significantly, already with currently availableinstrumentation. Key words.
Planets and satellites: atmospheres – Infrared: planetary systems – Methods: data analysis – Techniques: spectroscopic
1. Introduction
The discovery of transiting extrasolar planets orbiting brightstars has opened up the possibility of characterising the atmo-spheres of these exoplanets through transmission and secondaryeclipse spectroscopy. Broadband observations of transits andsecondary eclipses have now become standard practice, bothfrom space and the ground, with the transit and eclipse depthsof many planets measured at a number of wavelengths (e.g.Seager & Deming 2010). On the other hand, real characterisa-tion of exoplanet atmospheres is still in its infancy, since thedata available for most planets contain only little informationregarding the planet’s gas abundances and temperature struc-ture. Medium-resolution spectra can increase the available in-formation on the planet significantly. However, some exciting re-sults on the most favourable targets have subsequently been dis-puted (Swain et al. 2010; Mandell et al. 2011; Tinetti et al. 2007;Gibson et al. 2011; Madhusudhan et al. 2011; Crossfield et al.2012), highlighting the demanding nature of these kinds of ob-servations (see also Bean et al. 2013). High-resolution spectroscopy, in which molecular bands areresolved into indivual absorption lines, can yield informationon exoplanet atmospheres that is complementary to broadbandtransit and secondary eclipse measurements. Despite several at-tempts in the past (e.g. Wiedemann et al. 2001; Barnes et al.2010), high-resolution spectra have only recently provided theirfirst detections of molecules on exoplanets. Snellen et al. (2010)detected absorption by carbon monoxide lines during a transitof HD 209458b using the CRyogenic high-resolution InfraRedEchelle Spectrograph (CRIRES, Kaeufl et al. 2004) on the VeryLarge Telescope, at a resolving power of R ∼ ± − with respect to the system velocity, possibly indicat-ing day-to-night winds. Subsequent modelling of the dynam-ics of HD 209458b indicates that such a Doppler shift in theplanetary transmission spectrum might indeed be expected un-der certain conditions (Miller-Ricci Kempton & Rauscher 2012;
1e Kok et al.: Exoplanets at high spectral resolution
Showman et al. 2013). Using CRIRES, absorption lines of COhave also been detected in the day-side thermal emission of thenon-transiting planets τ Boo b (Brogi et al. 2012; Rodler et al.2012) and possibly 51 Peg b (Brogi et al. 2013). In the lat-ter case, water vapour was also possibly detected. These de-tections allowed the determination of the orbital inclination ofthese planets, and hence their true mass. Furthermore, it wasfound that these planets’ temperature profiles do not show astrong inversion layer at the pressures probed by the CO lines(between ∼ − -1 bar), since the lines show up in absorp-tion. The same conclusion was drawn for the transiting planetHD 189733b, using from CRIRES (de Kok et al. 2013) andKeck / HIRES (Rodler et al. 2013) observations. Subsequently,Birkby et al. (2013) also found water vapour at 3.2 µ m on thisplanet. There are large degeneracies in determining the tempera-ture and molecular abundances from this type of high-resolutionobservations, mostly caused by the inability to determine the ab-solute continuum level, and by the general degeneracy betweentemperature profile and gas abundance. Nevertheless, constraintscan be placed on the depth of the absorption lines, which shouldreduce uncertainties in the retrieved temperature and composi-tion of transiting planets (de Kok et al. 2013). Detections of mul-tiple gases, and at multiple wavelengths, can also be used to de-termine relative abundances of gases, although the constraintsset by de Kok et al. (2013) and Birkby et al. (2013) are not strictenough to constrain the gas abundances of CO and H O bythemselves. During transit, the line depths can give more unam-biguous information regarding the composition, especially whenthe continuum opacity source is known and well-defined, whichconstrains the absolute pressure levels that are probed (e.g.Rayleigh scattering at visible wavelengths or H -H collision-induced absorption in the K-band. See also Benneke & Seager(2012)).We aim to assess how high-resolution observations can bestbe used to improve our knowledge of exoplanet atmospheres, byseeing how best to detect specific molecules and how the planetsignal can be measured at a higher signal-to-noise ratio in thenear-infrared. We present simulations of high-resolution planetspectra and observations, in order to explore these new possi-bilities for characterising exoplanet atmospheres. In Section 2,we investigate which wavelengths are best suited to search for arange of molecules, with a focus on the CRIRES instrument. InSection 3, we perform simulations to assess the detectability ofthe nightside atmosphere and day-night di ff erences. In Section4, we present conclusions.
2. Optimal wavelengths
Finding the best wavelengths to perform ground-based high-resolution observations of exoplanet atmospheres is far fromtrivial. On the one hand, a wavelength range should be chosenthat has a strong exoplanet signal, caused by strong absorptionlines. On the other hand, these strong absorption lines are of-ten present in the Earth’s atmosphere as well, greatly reducingthe transmission of the Earth’s atmosphere. Molecules that arenot abundant in the Earth’s atmosphere, but are abundant in theatmospheres of hot Jupiters, are therefore good candidates toobserve, as illustrated by the recent detections of CO. We per-formed a sensitivity analysis on simulated data to quantify howwell various molecular features can be observed, and from thisdetermine what the best wavelengths to perform these kinds ofobservations are. We focused on the CRIRES instrument, sinceit has proven to be the most e ffi cient for near-infrared high-resolution observations of exoplanets (e.g. de Kok et al. 2013), but lessons from this exercise can be applied to other instrumentsas well. Especially relevant will be the METIS instrument onthe European Extremely Large Telescope (Brandl et al. 2012).This instrument, which will greatly enhance the sensitivity ofthis method, will observe similarly fixed wavelength regions inthe near-infrared. The main focus in this section is on the planetHD 189733b, since the results of the sensitivity analysis can becompared to our detections of CO and H O. Results for otherplanets are briefly discussed.
The simulated data were created using ESO’s Exposure TimeCalculator (ETC), which simulates CRIRES’ perfomance in-cluding a high-resolution model of the Earth’s transmission andsky emission. The slit width was set to 0.2”, which results ina resolution of R ≈ ff ective temperature of the planet’shost star. This approach facilitated the insertion of a planet sig-nal later in the simulations (see below). The target magnitude(K = y t ), includ-ing telluric absorption, and the sky spectrum ( y s ), which takesinto account airglow lines and thermal background due to thetelescope and atmosphere, separately. Parts of the spectra thatare deemed unsuitable for scientific use, e.g. in between spec-tral orders, are given zero values by the ETC. We took the targetspectrum as the basis for our simulated observation. The noise( σ ) was calculated as is normally done by the ETC, but withoutthe small contributions from dark current and read-out noise: σ = . p y t + y s (1)For the thermal emission spectra, we created a hundredspectra, each being the target spectrum with random Gaussiannoise added, all with the above standard deviation, and in-serted planet signals at phases between 0.3-0.7. For transit ob-servations, we created fifty spectra between phases -0.05-0.05.The planet spectra are the result of line-by-line model calcula-tions, using HITEMP (Rothman et al. 2010) and HITRAN 2008(Rothman et al. 2009) absorption data. A discussion about theuse of line data is presented in Section 2.5. The calculated planetemission spectra were divided by the stellar blackbody spectrumused in the ETC, and scaled by (cid:16) R p R ∗ (cid:17) to obtain the correct planet-to-star contrast. This planet signal was then multiplied by theETC spectra and added to them to yield the simulated signalof the host star with an additional planet. The modelled transitspectra, which resulted from integrating over the entire plane-tary disc, are in units of fraction of starlight transmitted. Wemultiplied the ETC spectrum with these transit spectra to ob-tain the simulated transit observations. We did not simulate thee ff ect of changing airmass or throughput with time here to al-low fair comparisons between di ff erent phases. In the end, wesee very little variation in the correlation values for the di ff er-ent simulated spectra, and take the mean value in determining http: // / observing / etc / bin / gen / ......form?INS.NAME = CRIRES + INS.MODE = swspectr2e Kok et al.: Exoplanets at high spectral resolution the e ffi ciency of the planet detection. Since the only changingsignal is now the Doppler shift of the planet, we can simplifyour usual data analysis (Brogi et al. 2013; de Kok et al. 2013).‘Instead of e.g. fitting with airmass, we now simply subtractedthe mean spectrum for each simulated observation. This shouldbe equivalent to the real data with the stellar signals, telluric sig-nals, and any additional broadband features removed. We cross-correlated the model planet spectrum, which are identical to thespectra that were inserted into the simulated spectra, with thisreduced data at the known planet velocities. We did this both fordata that have the planet signal inserted, and for the same datafor which the planet signal was not inserted, to asses any po-tentially spurious cross-correlation signal created by the addednoise. The di ff erence in correlation between these two cases wasthen taken as a measure of the sensitivity of the measurement tothe inserted planet signal. In practise, the cross-correlation sig-nals from the added noise did not a ff ect the relative planet signalfor each wavelength setting significantly. Increasing the standarddeviation of the noise did of course reduce the cross-correlationsignal. The data was cross-correlated with the same model spec-trum that was inserted into the data, which implies we know theplanet spectrum exactly and only concern ourselves with the ef-ficiency of extracting this planet signal. This is of course not thecase in real measurements, but in practise, performing the cross-correlation analysis with a wide range of model spectra, and se-lecting the best one, will reduce the di ff erences between the real,disc-averaged, planet spectrum and the model spectrum. The cross-correlations for thermal emission models with a sin-gle trace gas in a hydrogen-dominated atmosphere for the pa-rameters of HD 189733b are shown in Fig. 1. The figure showsfour spectroscopically active gases that are thought to be rel-atively abundant for HD 189733b from theory or measure-ments: H O, CO, CO and CH . The temperature profile ischosen such that it is consistent with the retrieved profiles ofMadhusudhan & Seager (2009) and all molecules have an as-sumed volume mixing ratio of 10 − . Simulations with more re-alistic abundances are discussed below. The cross-correlationsare normalised by the cross-correlation of CO at 2.3 µ m, whichis where there is an already published detection (de Kok et al.2013).Fig. 1 shows that CO is best observed with CRIRES around2.3 µ m for HD 189733b, where it was indeed observed. There isalso an opportunity to observe CO in the M-band, where the COlines are stronger, the planet-to-star contrast is higher, but the skybackground becomes significant, and the flux from the planet islower. Specific wavelength settings in the L-band generally pro-vide the best opportunity to observe the thermal emission of theother gases, as plotted in Fig. 1. Especially around 3.5 µ m thetelluric transmission is very high and the balance between theflux from the planet and the noise from the star is optimal. Athigh temperatures, gases like CH , CO and H O have manyobservable lines even where they are very weak in the Earth’satmosphere, meaning that the advantage of a high telluric trans-mission weighs more heavily than the reduced strengths of theabsorption lines in the exoplanet’s atmosphere. The target spec-trum at 3.5 µ m from the ETC, showing the telluric lines, areshown in Fig. 2. This shows that there are still telluric lines here,but these are generally not strong and leave much space for apotential planet signal. We note that the detection of H O byBirkby et al. (2013) was achieved at 3.2 µ m, where we predictthe H O signal-to-noise to be rougly a factor of 2-3 smaller than at 3.5 µ m. In the H-band, the planet flux is reduced and the ex-pected signal is generally lower than in the K-band. e - / p i x e l / D I T Fig. 2.
Output target spectrum from the ETC for the wavelengthsetting that gives the best expected signal for H O and CO . DITis the integration time. All lines are telluric in nature.We performed the same analysis for the molecules C H and HCN. These gases are predicted to be abundant if pho-tochemistry is important, or if the planet has a high C / O ra-tio (Moses et al. 2013). We note that the line data for thesegases are very incomplete in HITRAN, with most notablymany weak lines missing, which can become more importantat higher temperatures. Self-consistent, more complete linelistsfor these gases will be available in the near-future (S. Yurchenko,pers. comm.), but we limit ourselves to the HITRAN list for now.Fig. 3 shows another promising wavelength region around 3.1 µ m where the carbon-bearing molecules CH , C H , and HCNall show a relatively good signal.In a real atmosphere there will be a combination of signa-tures from di ff erent molecules. The expected signal from spectraof HD 189733b with more realistic gas abundances is shown inFig. 4. The volume mixing ratios of the gases were guided byMoses et al. (2013) (green lines in their Fig. 2): [H O] = − ,[CO] = − , [CO ] = − , and [CH ] = × − . The best ob-tainable signal, assuming a perfectly accurate planet spectrum,again lies at 3.5 µ m, and is actually very similar to the results forpure water. This is not surprising, since the spectrum is domi-nated by water lines in this case. However, when performing thecross-correlation analysis with a spectrum from only one gas,which is then di ff erent from the spectrum that is inserted intothe data in this case, we see that the strong CH lines can have ashielding e ff ect at 3.5 µ m, reducing the signal by up to a factor oftwo for high CH volume mixing ratios. H O has a slightly lessstrong shielding e ff ect. Of course, including CH in the templatespectrum increases the correlation signal again. Decreasing theCH volume mixing ratio increases the ease with which the othermolecules are detected in that wavelength region. So, althoughH O and CH can have a shielding e ff ect, it does not prevent thedetection of other molecules as long as there are lines from theother molecule that are stronger than the H O or CH absorptionat that wavelength. For the assumed temperature profile, the ex-pected cross-correlation signal decreases by a factor of 2-3 forevery order of magnitude the volume mixing ratio of H O andCO is decreased from 10 − downwards to 10 − .
3e Kok et al.: Exoplanets at high spectral resolution CO R e l a t i v e c o rr e l a t i on H O R e l a t i v e c o rr e l a t i on CH R e l a t i v e c o rr e l a t i on CO R e l a t i v e c o rr e l a t i on Fig. 1.
Expected cross-correlation values, relative to the correlation of CO at 2.3 µ m, for the di ff erent wavelength settings ofCRIRES, for the thermal emission of a single gas in a hydrogen atmosphere. C H R e l a t i v e c o rr e l a t i on HCN R e l a t i v e c o rr e l a t i on Fig. 3.
As Fig. 1, but now for C H and HCN.In all the above analysis, the observing conditions, such asseeing and water vapour content, were not ideal. We repeatedsome of the calculations for more ideal situations, but this didnot qualitatively change the relative strengths of the signalsfor the di ff erent wavelength settings. Quantitatively, the cross-correlation signal increased by ∼
20% in the ideal case. This is not surprising, since the best wavelength settings are alreadythose least a ff ected by the Earth’s atmosphere.In our previous detections of the thermal emission ofhot Jupiters (Brogi et al. 2012, 2013; de Kok et al. 2013;Birkby et al. 2013), we showed the correlation values for a gridof planetary radial velocity amplitudes ( K p ) and systemic ve-locities ( V sys ), to show that the detected signal is not an outlier
4e Kok et al.: Exoplanets at high spectral resolution R e l a t i v e c o rr e l a t i on Fig. 4.
As Fig. 1, but now for a model spectrum with H O, CO,CH , and CO (see text).and lies at the expected position in this grid. The significance ofthe detection was also partly evaluated using this grid. For non-transiting planets, the correlation peak in this grid was used todetermine the inclination of the system. It is therefore interest-ing to see what the correlation values for such a grid are in oursimulations. Again, we calculated the di ff erence in correlationvalues for data with and without an inserted planet signal, butnow for a smaller phase range (0.38-0.48), which is closer to thereal measurements during a single night, for the grid of K p and V sys . The results are shown in Fig. 5, for CO at 2.3 µ m. Verysimilar patterns are seen here as are seen for the real data. Mostnotably, the maximum signal is elongated along slanted lines inthe grid. This arises because the planet radial velocities during asmall part of the orbit can be well reproduced by a range of K p and V sys combinations. When two nights are used that probe bothsides of phase 0.5, a cross-hatch pattern emerges (see Brogi et al.2013). Fig. 5 also shows negative wings around the peak correla-tion and positive peaks farther away in V sys . These are the resultof the correlation of the model spectrum with itself at a range ofDoppler shift (i.e. the autocorrelation function). In our previouswork, these structures are also included in our determination ofthe correlation noise, which determined the significance of thedetection, whereas in fact they can be seen as a reflection of theplanet signal. The analysis for transit signals of HD 189733b is shown inFigs. 6 and 7. Here, shorter wavelengths are more favourablecompared to the thermal emission spectra, since the planet sig-nal is to first order dependent on the atmospheric scale height,which does not depend on wavelength. Hence, it does not dropquickly towards shorter wavelengths, unlike the Planck functionat the exoplanet temperatures. The expected signals are there-fore more similar between the di ff erent atmospheric windows.The comparative signal of a single molecule at multiple wave-lengths can be used to search for haze and cloud signals, sincethe molecular lines will be less deep when haze is present.As can be seen in Fig. 7, a transit measurement around 1.5 µ m might be an excellent way to probe C H , which is predictedto be very abundant in the upper atmosphere for planets with ahigh C / O ratio. We note that this wavelength is also covered by -200 -100 0 100 200V sys (km s -1 )50100150200 K p ( k m s - ) Fig. 5.
Cross-correlation values as a function of K p and V sys for asimulated observation run at 2.3 µ m, targeting CO. Normalisedvalues range from -0.2 to 1. Dashed lines indicate the position ofthe inserted planet signal.HST / WFC3. Other molecules also give good signals in the H-band for these transit simulations. A decrease in the volume mix-ing ratio of a factor of 10 gives a decrease in cross-correlationsignal of a factor of ∼ In the above simulations we focused on HD 189733b. For otherhot Jupiters, the best wavelength setting to observe a certainmolecule will generally be the same, given a clear atmosphereand a lack of temperature inversion. There will be a di ff erencein the expected signal when two wavelengths are considered thatare far apart. For instance, the 3.5 µ m H O signal for τ Boo b and51 Peg b is not as strong relative to the 2.3 µ m CO signal com-pared to HD 189733b. However, the 3.5 µ m H O signal is stillhigher than the CO signal if the water abundance is as high asthe CO abundance. These di ff erences are present, because, gen-erally, the relative correlation signal as a function of wavelengthfrom one planet to the next roughly scales with the emission ortransmission signal from the planet, divided by the square rootof the emission of the star, since the star is the source of photonnoise. Di ff erences in the temperature structure, gas abundancesand haze opacity will give rise to further di ff erences in the ex-pected correlation signals. Besides the theoretical ease of detecting a certain moleculewith high-resolution spectroscopy, as discussed above, there areother aspects that can hinder detection. For instance, an accu-rate wavelength solution for the measured spectra is crucial inobtaining a good signal from the cross-correlation technique.In Snellen et al. (2010); Brogi et al. (2012, 2013); de Kok et al.(2013), we used telluric absorption lines to this end. Figs. 2 and8 show two spectra from the ETC in regions of high transmis-sion in the L-band and the H-band, which are very suitable forobservations of H O and CO respectively. The figures show thatin the H-band telluric absorption lines might be very sparse atsome wavelength regions, and one might need to rely on stellarlines, sky lines, or additional wavelength-calibration measure-ments for an accurate wavelength solution throughout the dura-
5e Kok et al.: Exoplanets at high spectral resolution CO R e l a t i v e c o rr e l a t i on H O R e l a t i v e c o rr e l a t i on CH R e l a t i v e c o rr e l a t i on CO R e l a t i v e c o rr e l a t i on Fig. 6.
Expected cross-correlation values, divided by the correlation of CO at 2.3 µ m, for the di ff erent wavelength settings ofCRIRES, for the transit spectrum of a single trace gas in a hydrogen-dominated atmosphere. C H R e l a t i v e c o rr e l a t i on HCN R e l a t i v e c o rr e l a t i on Fig. 7.
As Fig. 6, but now for C H and HCN.tion of the observation. The latter would result in a significant re-duction in e ffi ciency of the measurements, since the wavelengthcalibration will need to be performed several times during thecourse of a night. We note that the CO lines in the right part ofFig. 8 are actually very suitable for the wavelength-calibrationthere. The telluric spectrum from Fig. 2 shows a nice spread oflines across all wavelengths, which would be very suitable for the wavelength-calibrations in this wavelength setting. Changesin the telluric lines, due to, for instance, changes in the watervapour content or seeing, are generally well removed in the datareduction process before cross-correlating with the model spec-tra. The cross-correlation method for these high-resolution ob-servations is also very sensitive to our exact knowledge of the
6e Kok et al.: Exoplanets at high spectral resolution e - / p i x e l / D I T Fig. 8.
Output target spectrum from the ETC for the wavelengthsetting that gives a the best expected signal for CO in H-band.DIT is the integration time.line positions of the many lines in the exoplanet spectrum.If there is a random error in these line positions, the cross-correlation signal will be smeared out and sensitivity will belower. This is an issue when using ab initio calculations ofmolecular lines, which are often necessary to calculate spec-tra for high temperatures, in particular for lines that are weakat lower temperatures. Especially the line intensities extrapo-lated to high temperatures from Earth-focussed databases, likeHITRAN, can have large errors. On the other hand, ab initio cal-culations can have larger errors in the line positions (see e.g.Bailey & Kedziora-Chudczer 2012). For high-resolution obser-vations of certain less-studied molecules, the choice of absorp-tion data can thus be a choice between completeness and accu-rate line intensities on the one hand, versus accurate line posi-tions on the other. In the (near) future, initiatives like Exomol(Tennyson & Yurchenko 2012) and HITEMP (Rothman et al.2010) will improve on this situation. In this paper, we usedHITRAN for molecules that were not in HITEMP, with theaforementioned disadvantages noted.Another issue that has been ignored in the previous calcu-lations is the presence of stellar absorption lines. These showsmall (sub-pixel) Doppler shifts due to gravitational pull of theplanet. This will result in relatively large residuals in our usualdata reduction, which aims at removing signals that do not showany changes in barycentric velocity with time. When the targetedmolecule in the planet atmosphere is also present in the stellaratmosphere, these residuals in the data can give rise to relativelystrong cross-correlation signals, which hinder the detection ofthe planet signal near phases 0.0 and 0.5 (e.g. Brogi et al. 2013).Hence, during transit observations the stellar lines will be espe-cially bothersome and more steps in the data reduction will needto be taken to reduce the e ff ect of stellar lines, such as modellingand removing the stellar lines (Brogi et al., in prep.).
3. Night-side observations
One commonality among the present detections of the ther-mal emission of exoplanet atmospheres at high resolution isthat they probe the day-side, close to phase 0.5. Detecting thehigh-resolution planet signal throughout its orbit could revealdi ff erences in the composition and temperature structure be-tween the day-side and the night-side. Spitzer observations (e.g. Knutson et al. 2009, 2012) reveal that the night-side of a hotJupiter is generally colder and emits less broadband flux. Modelsof the dynamics of hot Jupiter atmospheres (e.g. Showman et al.2009; Dobbs-Dixon & Agol 2013; Rauscher & Menou 2013) in-dicate that at low latitudes the day-night di ff erences are largestin the upper atmosphere, at the millibar level, where radiativetime scales are short, and where the cores of absorption linesprobe in high-resolution observations. Deeper in the atmosphere,where the continuum of high-resolution spectra probe, the day-night di ff erences are generally smaller. This implies that in high-resolution spectra the depth of the lines at the night-side are notnecessarily less deep than at the day-side, since the day-nighttemperature di ff erences probed by the line cores are larger thanthose probed by the continuum.To explore the di ff erences in high-resolution spectra be-tween the day-side and the night-side we calculated spectra withtemperature structures based on dynamical model calculationsfrom the literature. Fig. 9 shows day-side and night-side tem-peratures of HD 189733b based on low latitude profiles fromShowman et al. (2009). The corresponding spectra, for an atmo-sphere with only H , and CO at volume mixing ratio of 10 − , areplotted in Fig. 10. The two spectra are very similar, with the maindi ff erence being an o ff set between the spectra. Despite the loweroverall flux level of the night-side spectra, the depths of the linesare actually almost identical in this case. The lines in the night-side in fact represent a larger di ff erence in brightness tempera-ture between the continuum and the line cores, since at lowertemperatures a larger brightness temperature shift is needed toobtain the same shift in flux.
600 800 1000 1200 1400 1600 1800 2000Temperature (K)10 -1 -2 -3 -4 -5 -6 P r e ss u r e ( ba r) Fig. 9.
Assumed temperature profiles for the day-side (solid line)and night-side (dashed line) of HD 189733b.Another example explored is HD209458b. This planet isthought to have a strong temperature inversion at the dayside,which presents itself as emission lines in the spectrum. To getan impression of how such a planet with a temperature inver-sion would manifest itself in correlation analyses across its or-bit, we calculated spectra at a range of emission angles to cre-ate a simple spatially inhomogeneous planet for which disc-integrated spectra were calculated along its orbit. The three-dimensional structure of the planet was strongly approximated,based on Parmentier et al. (2013), by assuming a day-side tem-perature profile along a 140 ◦ × ◦ area centred around the sub-solar point and a night-side temperature profile everywhere else.The temperature profiles for these two regions are plotted in
7e Kok et al.: Exoplanets at high spectral resolution
Day-side F p / F s Night-side F p / F s Fig. 10.
CO spectra, in units of planet-star contrast, for approxi-mated day-side and night-side temperatures of HD 189733b (seeFig. 9).Fig. 11. At low and high pressures, such an approach seems rea-sonable, given the dynamical model outputs for HD 209458b(e.g. Showman et al. 2009; Parmentier et al. 2013). At interme-diate pressures the real planet atmosphere is more complicatedthan assumed here, but it is su ffi cient to obtain a qualitative pic-ture of the possible high-resolution spectra along the planet’sorbit.
500 1000 1500 2000 2500Temperature (K)10 -1 -2 -3 -4 -5 -6 P r e ss u r e ( ba r) Fig. 11.
Assumed temperature profiles for the day-side (solidline) and night-side (dashed line) of HD 209458b.The disc-integrated spectra for phases between 0 and 0.5 areshown in Fig. 12. As can be seen, at intermediate phases thecontibutions from the inverted and non-inverted parts of the at-mosphere can cancel out for a large part, greatly reducing anypotential signal at high resolution. In all cases, the wings of COlines show up in absorption, since they probe the part of theatmosphere that is still getting cooler with increasing altitude.Depending on the exact location of the temperature minimumin the inverted temperature profile, these wings will be a strongor weak feature of the disc-integrated spectrum between phases ∼ P ha s e Fig. 12.
Simulated CO spectra for HD 209458b across half itsorbit for a day-side region with a temperature inversion.For an irradiation level that is in between HD 189733b andHD 209458b, the day-side atmosphere might be close to isother-mal at pressures probed by molecular absorption lines. The sec-ondary eclipse depths of several exoplanets also resemble thoseof blackbodies, which could indicate a day-side atmosphere thatis close to isothermal (e.g. WASP-18b (Nymeyer et al. 2011),WASP-12b (Crossfield et al. 2012), TRES-3b (Croll et al. 2010),HAT-P-1b (Todorov et al. 2010), WASP-24b (Smith et al. 2012),etc.). For such planets, the night-side might yield much deeperabsorption lines and hence a much larger high-resolution sig-nal, especially if the horizontal heat distribution at low pres-sures is known to be ine ffi cient (Cowan et al. 2012; Maxted et al.2013). We also point out that a single night of CRIRES measure-ments of τ Boo b yields a weaker CO signal (Brogi et al. 2012;Rodler et al. 2012) than HD 189733b (de Kok et al. 2013), de-spite the planet being hotter, and orbiting a nearer star. Hence, τ Boo b might also be a planet with a relatively small temperaturegradient at the day-side and night-side observations can possiblyyield a larger correlation signal.
4. Summary and conclusions
We simulated exoplanet spectra at high spectral resolution toassess how high-resolution observations can be used to fur-ther improve our knowledge of exoplanet atmospheres. We ap-plied a cross-correlation technique on simulated spectra fromthe CRyogenic high-resolution InfraRed Echelle Spectrograph(CRIRES) on the Very Large Telescope from which we identi-fied specific spectral regions that give the highest signal whenmeasuring hot Jupiter spectra at high spectral resolution. For de-tections of the thermal emission, specific regions in the L-bandgive the best results for H O, CO and CH , as well as C H and HCN. In this respect, the planned infrared instrument onthe European Extremely Large Telescope, METIS (Brandl et al.2012), which includes a high-resolution channel in the L-band,is expected to give excellent signals for hot Jupiters and will alsobe able to study colder objects in this way. For high-resolutiontransmission signatures shorter wavelengths also give good sig-nals, giving the opportunity to detect the same molecule at mul-tiple wavelengths. This would give constraints on haze opacity
8e Kok et al.: Exoplanets at high spectral resolution and possibly vertical variations in temperature and gas abun-dances.Other future instruments that show promise in detectingmolecules in exoplanet atmospheres at high resolution in thenear-infrared are CARMENES (Quirrenbach et al. 2012) andSPIRou (Thibault et al. 2012). These spectrographs have a lowerspectral resolution (R = = ∼ ffi ciently than CRIRES is notimmediately clear and will need further investigation. In thislight, the increase of wavelength coverage obtained by turningCRIRES into a cross-dispersed spectograph (Oliva et al. 2012)is very exciting and can improve the e ffi ciency of exoplanet at-mosphere detections with CRIRES by a factor of several.We also calculated example spectra of the day-side andnight-side of the planets HD 189733b and HD 209458b at highresolution. These indicate that, despite a lower broadband flux,the night-side does not necessarily give rise to shallower absorp-tion lines at high spectral resolution. For planets with a weakday-side temperature gradient, night-side observations may ac-tually be more suitable for high-resolution observations than theday-side. High-resolution observations covering the entire orbitof the planet can give great insight into the three-dimensionaltemperature structure and chemistry of hot Jupiters and greatlycomplements the information from Spitzer phase curves. Acknowledgements.
We thank the anonymous referee for his / her useful com-ments and suggestions. This work was funded by the Netherlands Organisationfor Scientific Research (NWO). References
Bailey, J. & Kedziora-Chudczer, L. 2012, MNRAS, 419, 1913Barnes, J. R., Barman, T. S., Jones, H. R. A., et al. 2010, MNRAS, 401, 445Bean, J. L., D´esert, J.-M., Seifahrt, A., et al. 2013, ApJ, 771, 108Benneke, B. & Seager, S. 2012, ApJ, 753, 100Birkby, J. L., de Kok, R. J., Brogi, M., et al. 2013, MNRAS, 436, L35Brandl, B. R., Lenzen, R., Pantin, E., et al. 2012, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 8446, Society ofPhoto-Optical Instrumentation Engineers (SPIE) Conference SeriesBrogi, M., Snellen, I. A. G., de Kok, R. J., et al. 2012, Nature, 486, 502Brogi, M., Snellen, I. A. G., de Kok, R. J., et al. 2013, ApJ, 767, 27Cowan, N. B., Machalek, P., Croll, B., et al. 2012, ApJ, 747, 82Croll, B., Jayawardhana, R., Fortney, J. J., Lafreni`ere, D., & Albert, L. 2010,ApJ, 718, 920Crossfield, I. J. M., Barman, T., Hansen, B. M. S., Tanaka, I., & Kodama, T.2012, ApJ, 760, 140de Kok, R. J., Brogi, M., Snellen, I. A. G., et al. 2013, A&A, 554, A82Dobbs-Dixon, I. & Agol, E. 2013, MNRAS, 435, 3159Gibson, N. P., Pont, F., & Aigrain, S. 2011, MNRAS, 411, 2199Kaeufl, H.-U., Ballester, P., Biereichel, P., et al. 2004, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 5492, Societyof Photo-Optical Instrumentation Engineers (SPIE) Conference Series, ed.A. F. M. Moorwood & M. Iye, 1218–1227Knutson, H. A., Charbonneau, D., Cowan, N. B., et al. 2009, ApJ, 690, 822Knutson, H. A., Lewis, N., Fortney, J. J., et al. 2012, ApJ, 754, 22Madhusudhan, N., Harrington, J., Stevenson, K. B., et al. 2011, Nature, 469, 64 Madhusudhan, N. & Seager, S. 2009, ApJ, 707, 24Mandell, A. M., Drake Deming, L., Blake, G. A., et al. 2011, ApJ, 728, 18Maxted, P. F. L., Anderson, D. R., Doyle, A. P., et al. 2013, MNRAS, 428, 2645Miller-Ricci Kempton, E. & Rauscher, E. 2012, ApJ, 751, 117Moses, J. I., Madhusudhan, N., Visscher, C., & Freedman, R. S. 2013, ApJ, 763,25Nymeyer, S., Harrington, J., Hardy, R. A., et al. 2011, ApJ, 742, 35Oliva, E., Hatzes, A., Piskunov, N., et al. 2012, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 8446, Society ofPhoto-Optical Instrumentation Engineers (SPIE) Conference SeriesParmentier, V., Showman, A. P., & Lian, Y. 2013, A&A, 558, A91Quirrenbach, A., Amado, P. J., Seifert, W., et al. 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 8446,Society of Photo-Optical Instrumentation Engineers (SPIE) ConferenceSeriesRauscher, E. & Menou, K. 2013, ApJ, 764, 103Rodler, F., K¨urster, M., & Barnes, J. R. 2013, MNRAS, 432, 1980Rodler, F., Lopez-Morales, M., & Ribas, I. 2012, ApJ, 753, L25Rothman, L. S., Gordon, I., & 42 colleagues. 2009, J. Quant. Spectro. Rad.Trans., 96, 139Rothman, L. S., Gordon, I. E., Barber, R. J., et al. 2010, J. Quant. Spectro. Rad.Trans., 111, 2139Seager, S. & Deming, D. 2010, ARA&A, 48, 631Showman, A. P., Fortney, J. J., Lewis, N. K., & Shabram, M. 2013, ApJ, 762, 24Showman, A. P., Fortney, J. J., Lian, Y., et al. 2009, ApJ, 699, 564Smith, A. M. S., Anderson, D. R., Madhusudhan, N., et al. 2012, A&A, 545,A93Snellen, I. A. G., de Kok, R. J., de Mooij, E. J. W., & Albrecht, S. 2010, Nature,465, 1049Swain, M. R., Deroo, P., Gri ffi th, C. A., et al. 2010, Nature, 463, 637Tennyson, J. & Yurchenko, S. N. 2012, MNRAS, 425, 21Thibault, S., Rabou, P., Donati, J.-F., et al. 2012, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 8446, Society ofPhoto-Optical Instrumentation Engineers (SPIE) Conference SeriesTinetti, G., Vidal-Madjar, A., Liang, M.-C., et al. 2007, Nature, 448, 169Todorov, K., Deming, D., Harrington, J., et al. 2010, ApJ, 708, 498Wiedemann, G., Deming, D., & Bjoraker, G. 2001, ApJ, 546, 1068th, C. A., et al. 2010, Nature, 463, 637Tennyson, J. & Yurchenko, S. N. 2012, MNRAS, 425, 21Thibault, S., Rabou, P., Donati, J.-F., et al. 2012, in Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, Vol. 8446, Society ofPhoto-Optical Instrumentation Engineers (SPIE) Conference SeriesTinetti, G., Vidal-Madjar, A., Liang, M.-C., et al. 2007, Nature, 448, 169Todorov, K., Deming, D., Harrington, J., et al. 2010, ApJ, 708, 498Wiedemann, G., Deming, D., & Bjoraker, G. 2001, ApJ, 546, 1068