Thinkshop 2022: High-resolution spectroscopy for exoplanet atmospheres and biomarkers

High resolution spectroscopy has proven a powerful and highly versatile tool in the characterization of exoplanet atmospheres. Over the last decade we have explored the rich detail that its sensitivity to the position and shape of exoplanet spectral lines provides, and with robustness thanks to the unique pattern of lines that each species exhibits at such high resolution. In this talk, I will review this playground for discovery in exoplanet atmospheres with an observer’s eye, focusing on the observational techniques it uses and their outputs. I will briefly explore its discoveries from abundances to the 3D dynamical nature of exoplanet atmospheres, from the ultra hot to the increasingly cooler and smaller planets. Despite a vast range of new discoveries, thanks to the deluge of new high resolution instruments from the EPRV community, we have barely only scratch the surface of what we can learn with high resolution spectroscopy. I will highlight current and future avenues for discovery, including with the ELTs. I will finish however with a primer for discussion during this workshop, on some of the challenges and the consensus that the field of high resolution spectroscopy of exoplanet atmospheres needs to face as we go towards the robust characterisation of potentially habitable worlds.

Ultra hot Jupiters around bright stars are the most favorable targets for high-resolution exoplanetary spectroscopy. We present our emission and transmission spectroscopy observations of the ultra hot Jupiter KELT-20 b/MASCARA-2 b from PEPSI on LBT, with R=130,000. With the LBT's large aperture we obtain some of the highest signal-to-noise ratio exoplanetary spectra to date (stellar spectra with total SNR~3400). We obtained a 16-sigma detection of Fe I emission from the planetary atmosphere via cross-correlation, and we consider the line profile and time variation of this signal in order to constrain the planetary atmospheric structure. We fail to detect any other species in emission despite the strong Fe I emission, including several species previously reported in the literature. We will discuss the implications of these non-detections, including placing limits on atmospheric inversion agents such as TiO and VO. These high signal-to-noise spectra give a preview of what we will be able to achieve for many more planets in the ELT era.

High-resolution (HR) ground-based spectrographs have drastically improved the investigation of exoplanet atmospheres.
We will present results obtained within the GAPS 2.0 long-term program for atmospheric characterization of hot giant planets using the near-infrared 0.95-2.45 \mu m) arm of GIARPS: the high-resolution (R /sim 50 000) spectrograph GIANO-B of the Telescopio Nazionale Galileo (TNG).
We will particularly focus on our ongoing efforts to probe exoplanetary atmospheres with the transmission spectroscopy technique. With the detection of six molecules in the atmosphere of the hot-Jupiter HD 209458b (Giacobbe+2021), we have demonstrated that exoplanetary atmospheres can show a chemical richness previously unknown. Here, we will present the atmospheric characterization of three warm planets, namely the super-Neptune WASP-107b (Giacobbe+ in prep.), the sub-Saturn WASP-69b (Guilluy+2022, accepted), and the sub-Jovian WASP-80b (Carleo+2022, accepted). The simultaneous detection of multiple molecules in their atmospheres reinforces our previous finding revealing a rich atmospheric composition also for warm giant planets. We will give an interpretation of our results in terms of the planet’s C/O ratio and metallicity, which allow us to derive important clues on the formation and migration histories. These analyses represent a new frontier in the characterization of exoplanetary atmospheres, and additional surprising discoveries are expected in the next future with both ground-based HR spectrographs, such as CRIRES+, SPIRou, and NIRPS, and the low-resolution (LR) spectrographs on board the JWST telescope.

We present High Resolution Cross Correlation Spectroscopy (HRCCS) observations and atmospheric retrieval results of the Ultra Hot Jupiter, MASCARA-1b. HRCCS is a novel technique used to characterize exoplanets with ground-based instruments by utilizing the planet's time-resolved Doppler shifted signal to separate it from the dominant stellar and telluric lines. The high spectral resolution along with wide wavelength coverage allows for robust constraints on molecular abundances and thermal structures using bayesian retrieval techniques.

We present our atmospheric retrieval results of MASCARA-1b using data taken from the IGRINS instrument (R~45000) at Gemini South. This high fidelity data provides robust constraints on molecular abundances in order to calculate C/O ratios and probe the thermal structure of the planet.

The holistic understanding of exoplanet atmospheres continues to elude us due to the difficulty in acquiring data for these far-away worlds. A wide array of techniques is necessary to probe their components, structure, and interactions. One of these techniques has proven to be particularly effective for large and highly irradiated exoplanets: Ground-based, narrow-band transmission spectroscopy in ultra high-resolution (uHRS, R > 80'000).

In uHRS, transmission spectroscopy allows us to detect the resolved lines of various atomic species, probing various orders of magnitude in pressure. Recently, the use of highly stabilized spectrographs has opened new avenues to extract information about the three dimensional nature of atmospheres: their composition in altitude, their dynamics, and, by proxy, even their magnetic fields. The refinement of uHRS observations and analysis techniques on large and highly irradiated planets as benchmark cases will ultimately help us push towards both smaller and cooler planets with future facilities.

State-of-the-art spectrographs on large ground-based telescopes have allowed significant discoveries in the structure, composition and dynamic processes of exoplanetary atmospheres through high-resolution transit spectroscopy. However, observations from ground-based facilities face two major challenges : The day-night cycle limits the length of observations, and incoming signals are distorted by Earth's atmosphere. The latter can be solved by adaptive optics and precise telluric correction, but the former is an issue for transit spectroscopy. This restricts the observable planets to ones with transit duration smaller than five hours.

Extremely inflated exoplanet KELT-11b is affected by this issue, as it transits its host star for more than seven hours. To bypass this limitation, we observed this system for series of three consecutive days: Before, during and after the transit, in order to collect a large number of in-transit and out-of-transit baseline spectra. This allowed the computation of a strong signal-to-noise high-resolution combined transmission spectrum. The individual spectra were corrected for telluric absorption using ESO atmospheric transmission code Molecfit.

This resulting transmission spectrum shows conclusive evidence of excess atomic sodium absorption through the 5'900 Å doublet of 0.29±0.05 % and 0.50±0.06 % in the Na D1 and D2 lines respectively. We model for wind patterns and find day-to-night side winds with surprisingly no vertical winds, considering KELT-11b's extreme scale height of 2'763 km. We also highlight the robustness of our three-day observation method which allows the study of exoplanets with long transits that haven't been studied with high-resolution ground-based spectrographs.

Other recent results from TESS, HST and CHEOPS combined with KELT-11b's hot atmosphere, bright host star, long transit, sub-Saturn mass and great bloatedness make it a unique and ideal candidate for atmospheric characterization with the next generation of telescopes, especially from space.

High-resolution transmission spectroscopy has opened new avenues in which to characterise the atmospheres of exoplanets. This technique has been used to unambiguously identify chemical species, map longitudinal variations in their abundance across the morning and evening limbs, as well as to infer atmospheric dynamics. We present multiple high-resolution transmission spectroscopy observations of the ultra-hot Jupiter WASP-121b, using the ESPRESSO spectrograph at the VLT, across which we consistently constrain the relative abundances of various neutral metals in its atmosphere. This is done via the cross-correlation technique, in which our observations are cross-correlated with Doppler shifted model transmission spectra. This cross-correlation value is then “mapped” to a likelihood value, which can then be folded into a Bayesian retrieval framework. From these retrievals, we can constrain atmospheric properties such as a species volume mixing ratio, the vertical T-P profile of the atmosphere, and atmospheric dynamic parameters which are then used to infer the planet’s orbital velocity, as well as offsets caused by winds and rotation. By constraining the relative abundance of species in a planet’s atmosphere, we can determine the planet’s bulk composition, and ultimately infer the planet’s formation and evolution mechanisms.

Planets on misaligned orbits around rapid rotators can experience “gravity-darkened seasons” as their orbits cross over the hot poles and cooler equators of rotationally-flattened stars. The periodically variable heating from these seasons presents a unique case for exploring how changes in stellar irradiation influence planetary atmospheric dynamics. We perform a homogeneous analysis of day-to-nightside winds on the ultra-hot Jupiters KELT-9 b and KELT-20 b, enabling the comparison of a planet experiencing gravity-darkened seasons (KELT-9 b) with a similar planet that does not (KELT-20 b). We conduct high-resolution transmission spectroscopy using two transits observed by the PEPSI spectrograph on the Large Binocular Telescope to empirically constrain supersonic ~10 km/s day-to-nightside winds traced by Fe II features in the atmosphere of KELT-9 b. Reconciling our findings with two archival HARPS-N datasets suggests multi-epoch variability ~5-8 km/s over timescales between weeks to years. In contrast, KELT-20 b’s day-to-nightside winds are less rapid (~2 km/s) and stable across four transits collectively observed by PEPSI and HARPS-N. The observed contrast of KELT-9 b’s wind variability and KELT-20 b’s stability is in accordance with our intuition on the effect of gravity-darkened seasons. A qualitative evaluation of our measured wind velocities and variability against current ultra-hot Jupiter GCMs reveals that KELT-9 b poses unique challenges for validating giant planet atmospheric models.

While first detections of atmospheric species were based on medium-to low resolution data from space, ground-based facilities have majorly contributed to our understanding of these far-away worlds since then.
To a large part this is owed to the use of high-resolution spectroscopy, which allows us to identify the resolved lines of metals and molecules in the optical and infrared wavelength regions using cross-correlation with synthetic models or studies of isolated lines. The success of these methods on ultra-hot to warm exoplanets have inspired the community to design instruments for the extremely large telescope facilities with the goal to extend these studies to the atmospheres of rocky worlds in the future. But not all our dreams for the bright future of exoplanet atmospheres are based on high-resolution instrumentation.
In parallel, studies at lower resolution have been continued both from the ground as well as from space. Especially with the recent launch of JWST and the planned launch of ARIEL in 2029 we can expect space based medium resolution studies to be a main driver of the advancement of our field in the next decades.
With the same planets being observed at different resolutions and thus analyzed with different methods the question emerges of how these different kinds of data relate to each other. In this talk I will focus on the individual challenges and differences presented by low and high-res data and the methods used to analyze them. I will address in which capacity the information derived from them is equivalent and/or complementary and what we can hope to gain when using both methods in unison.

Retrievals on high spectral resolution exoplanet observations have been recently made possible, allowing us now to place robust, quantitative constraints on their atmospheres in unprecedented detail. High resolution data can also be combined with low resolution, space-based data in retrievals, probing a wider range of altitudes and incorporating continuum information otherwise lost through detrending methods common in high resolution spectroscopy. We present retrieval results of combined observations of a benchmark hot Jupiter WASP-77A b, combining multiple nights taken with the IGRINS instrument on Gemini South (R~45,000), data from WFC3 on the Hubble Space Telescope (R~100), and IRAC/Spitzer photometry. We report a slightly superstellar C/O ratio and a substellar metallicity, indicative of diverse formation pathways for hot Jupiters.

The ultra-hot Jupiter WASP-33b shows evidence for a thermal inversion in the form of emission lines of atoms and molecules, including OH and Fe. Phase-curve measurements at optical wavelengths with TESS show a westward phase-offset while observations in the infrared with the Spitzer Space Telescope show an Eastward offset. We obtained phase-curve observations at high-spectral resolution at optical wavelengths with Subaru/HDS as well as CFHT/ESPaDOnS.

Using log-likelihood mapping and integrating a simple phase-curve model we demonstrate that it is possible to measure phase-curve variations, although unequal pre/post eclipse sampling can result in biases. For the Fe I emission, we show evidence that the peak is after eclipse at +22 +/- 12 degrees, consistent with the optical phase-curve. We also constrain the day-night contrast to be >0.9, indicating that, in the case of Fe I, the nightside contributes less than 10% of the day-side flux in our simple model.

We present Keck/KPIC high-resolution ($R\sim35,000$) $K$-band thermal emission spectroscopy of the ultra-hot Jupiter WASP-33b. The use of KPIC's single-mode fiber greatly improves both blaze and line-spread stabilities relative to slit spectrographs, enhancing the cross-correlation detection strength. We perform separate retrievals for the dayside and nightside emission spectra with a cross-correlation-to-logL nested sampling pipeline which fits for orbital parameters, the atmospheric pressure-temperature profile, and molecular abundances. While the dayside shows a thermal inversion which is absent on the nightside, we measure consistent CO and H$_2$O abundances for both hemispheres (nightside $\log\rm H_2O_{MMR} = -3.3^{+0.9}_{-0.6}$, $\log\rm CO_{MMR} = -1.5^{+0.9}_{-0.6}$ versus dayside $\log\rm H_2O_{MMR} = -4^{+1}_{-1}$, $\log\rm CO_{MMR} = -2^{+1}_{-1}$), suggesting limited dayside H$_2$O dissociation at the pressures probed. The retrieved abundances suggest a carbon- and possibly metal-enriched atmosphere, with a gas-phase C/O ratio of $1.0\pm0.1$, consistent with the accretion of high-metallicity gas near the CO$_2$ snow line and post-disk migration. The dayside hemisphere shows a systemic redshift compared to the nightside in excess of WASP-33b's rotation speed, suggesting day-to-night winds may be contributing to the offset. We also find tentative evidence for $\rm ^{13}CO/^{12}CO \sim 40$, consistent with values expected in protoplanetary disks. These observations demonstrate KPIC's ability to characterize close-in planets and the utility of KPIC's improved instrumental stability for cross-correlation techniques.

Characterisation of exoplanet atmospheres using high-resolution spectra provides a unique window into their composition, dynamics as well as their formation and evolution pathways. Our MEASURE data set contains a diverse set of 11 targets with a wide range of periods, equilibrium temperatures and masses. For one of the targets, the Ultra Hot Jupiter WASP-33 b, we report the first detection of CO emission lines at high spectral resolution from its thermal spectrum. These emission lines provide unambiguous evidence of a thermal inversion layer in its atmosphere. Moreover, by incorporating a Bayesian framework with 1D PHOENIX and 3D GCM atmospheric models, we show via Cross-Correlation-to-log-Likelihood mapping that the spectra indicate an eastward hotspot. Our resolution of R=15,000 pushes the High-Resolution Cross-Correlation Spectroscopy method to its lowest resolution limits. This bodes well for systems that may require the use of lower resolution spectra to improve photon collection, such as small planets orbiting in the close in habitable zones of small, faint M-dwarfs. Finally, we will show some preliminary results from the other targets in the MEASURE survey.

WASP-33b is an ultra hot Jupiter with a day-side temperature of 3100 K. In this talk, I will present our latest results in characterising its day-side atmosphere using the InfraRed Doppler instrument on the Subaru telescope. Using high-resolution cross-correlation spectroscopy, we confirmed our previous detection of OH emission but only after the secondary eclipse. Through an injection test, we show that the data quality before the eclipse is enough to detect the same signal if exists. More importantly, we were able to detect the emission of Fe and Si both before and after the secondary eclipse and Ti only after the secondary eclipse. This indicates that the spatial distribution of each chemical species is different and that we are probing the different parts of this 3D atmosphere. We also found evidence of the emission of Mg and Mn after combining all data sets, which, if confirmed, along with Ti add more chemical species that are detected on the day-side of an exoplanet.

Context. Transmission spectroscopy is a prime technique to study the chemical composition and structure of exoplanetary atmospheres. Strong excess absorption signals have been detected in the optical Na I D1, 2 Fraunhofer lines during transits of hot Jupiters, which are attributed to the planetary atmospheres and allow us to constrain its structure.
Aims. We study the atmosphere of WASP-7 b by means of high-resolution transit spectroscopy in the sodium lines.
Methods. We analyze a spectral transit time-series of 89 high-resolution spectra of the hot Jupiter WASP-7 b that was observed using the Ultraviolet and Visual Echelle Spectrograph (UVES). We use the telluric lines for an accurate alignment of the spectra and carry out a telluric correction with molecfit. Stellar magnetic activity is monitored by investigating chromospheric lines such as the Ca ii
H and K and hydrogen Hα lines. Finally, we obtain transmission spectra and light curves in the sodium lines region.
Results. The star shows no identifiable flares and, if any, marginal changes in activity during our observing run. The sodium transmission spectra and corresponding light curves clearly show signs of the Rossiter-McLaughlin effect (RM) and the stellar center-to-limb variation (CLV) that we model using synthetic spectra. A statistically significant, narrow absorption feature with a line contrast of 0.47 ± 0.06 % and a FWHM of 0.13 ± 0.03 Å is detected at the location of the Na I D2 line. For the Na I D1 line signal, we derive an upper limit of 0.12 %.

Measuring the chemical and thermal properties of planetary atmospheres is the main goal of current exoplanet science. It allows us to understand their true nature, origin, and eventually the ability to support life. Ground-based, high-resolution spectroscopy (HRS, R>25,000) has emerged as one of the leading methods to detect atomic and molecular species in the atmospheres of exoplanets. At infrared wavelengths, it allows us to resolve molecular bands into the individual spectral lines and extract their signal through cross correlation with model templates. As an added bonus, HRS can measure planet rotation, winds and orbital motion directly.
HRS has produced a rapidly growing inventory of detected species. Furthermore, recent developments of the technique have unlocked the ability to measure absolute abundances and temperatures, at a level comparable to expected JWST observations but using different and complementary spectral information. This ability sets the stage for a strong synergy between space and ground observations of hot and warm gaseous planets during the current decade.

High-resolution Doppler-resolved spectroscopy has opened up a new window into the atmospheres of both transiting and non-transiting exoplanets. While the 'classical' cross-correlation approach is efficient for finding atomic and molecular species, it is quite limited with its inability to recover quantitative information on the atmosphere such as abundances and temperature profiles - the very information we need to understand planetary atmospheres! Here, I will outline a new retrieval framework based on a simple Gaussian likelihood coupled to a fast model-filtering technique that is critical to account for the common pre-processing steps performed on time-series spectra. I will demonstrate its use including the first retrievals of high-resolution transmission spectra, showcasing multiple detections of neutral and ionised metals (Fe, V, Mg, Cr, FeII, etc) in ultra-hot Jupiters, which in turn place robust and precise constraints on abundance ratios and temperature-pressure profiles, as well as directly constrain the mean line-profile. However, to fully realise the power of high-resolution spectroscopy - which is sensitive to winds and atmospheric escape - it is clear that we need to treat highly irradiated planets as 3D objects. This is incompatible with the requirement for ultra-fast forward model atmospheres which make high-resolution retrievals viable in the first place. I will finish by discussing some open questions and current limitations of high-resolution atmospheric retrievals, with the goal of encouraging discussion of the best routes forward.

The PEPSI Exoplanet Transit Survey (PETS) is a spectroscopic survey of exoplanet transits, secondary eclipses, and host-star characterization. In this collaborative effort from AIP, MPIA, INAF, University of Arizona, Ohio State University and the LBT, a large number of targets have been observed with the goal to characterize planetary atmospheres in detail.

Towards the atmospheric characterization of smaller planets, the super-Earth exoplanet 55 Cnc e is one of the most promising terrestrial exoplanets studied to date. Here, we present a high-resolution spectroscopic transit observation of this planet within the PETS survey. Assuming the presence of Earth-like crust species on the surface of 55 Cnc e, from which a possible silicate-vapor atmosphere could have originated, we search in its transmission spectrum for absorption of various atomic and ionized species.

Not finding absorption for any of the investigated species, we are able to set absorption limits with a median value of 1.9 x Rp, ruling out the evidence of a widely extended silicate envelope on this super-Earth reaching several planetary radii.

The class of hot terrestrial worlds known as lava planets has come under significant investigation in the last decade. The brightest and most well-studied of these objects is 55 Cancri e, a nearby super-Earth with a remarkable 17-hour orbit. However, despite numerous studies, debate remains about the existence and composition of its atmosphere. We present upper limits on the iron content of 55 Cnc e derived from high-resolution time-series spectra taken with Gemini-N/MAROON-X and discuss implications for volcanic outgassing models for this planet.

Just looking through a telescope from your backyard you can see that the planets from our own Solar-System have widely different atmospheres. Such a large diversity is also expected for exoplanets, particularly since they cover a much wider parameter space. Now that we are able to measure exoplanet atmospheric spectrum our goal is to separate the effects of intrinsic differences, such as elemental chemical composition, from differences due to their specific circumstances, such as their rotation period or their equilibrium temperature.

I will review how radiative transfer, atmospheric circulation, chemistry and microphysics shape giant planet atmospheres and discuss how these processes can produce very different outcomes in planets that only posses small differences. I will show the expected and observed patterns in the population both at low and high spectral resolution and discuss how these differ from our expectation.

High-resolution transmission spectroscopy is one of the most fruitful techniques to infer the main opacity sources, thermospheric temperature and evaporation processes in the atmosphere of transiting exoplanets. Before retrieving the planetary features, however, observed spectra must be corrected for sky emission, residual atmospheric dispersion, presence of telluric features and interstellar lines, center to limb variation, and Rossiter-McLaughlin effect, with each of these steps possibly introducing systematic errors and altering the final transmission spectrum. The current lack of a public tool able to automatically extract a high-resolution transmission spectrum creates a problem of reproducibility of scientific results. As a consequence, it is very difficult to compare the results obtained by different research groups and to do a homogeneous characterization of the properties of exoplanetary atmospheres. In this talk I will present a standard, publicly available, user-friendly tool, named SLOPpy (Spectral Lines Of Planets with python, Sicilia et al. submitted), that automatically extracts the optical transmission spectrum of exoplanets and allows the perfect reproducibility of results. To validate the code and assess its performance, we performed a comparison with literature results of ideal targets for atmospheric characterization observed with HARPS and HARPS-N, finding a good agreement with published results. I will conclude the talk by highlighting a series of best practices that can greatly improve the reproducibility of results even when the analysis tools are not publicly released.

High resolution spectra of transit events enable the search for constituents in exoplanet atmospheres with a variety of methods.
Using cross correlation, the observations can be compared to models across a wide wavelength range, taking advantage of the large number of absorption lines of some species. For other species, the excess absorption in strong individual lines is measurable directly by comparing the spectra taken during the transit with those before and after the transit. We analyze transmission spectra of HAT-P-11 b obtained with CARMENES using these two methods to search for a multitude of species in the planetary atmosphere. We remove stellar and telluric lines with SYSREM and find a detection of CH$_4$ using cross correlation. In addition, we confirm the previously detected presence of He by examining the near-infrared helium triplet. Helium is known to be a tracer for extended atmospheres and to be related to stellar activity. The large eccentricity of HAT-P-11 b's orbit helps to disentangle the planetary from the stellar absorption lines. Furthermore, we search for other species that are expected to be present in the atmosphere of warm Neptunes.

The next generation of ground-based telescopes, which complement the space missions, are aimed at HR (R ~ 100,000) studies. Ground-based high resolution Doppler spectroscopy (HRS), pioneered by Snellen, has been used to characterise the atmospheres of a growing number of exoplanets in recent years. None of this is possible without the necessary laboratory data. While spectroscopic and other data needs for studies of Earth-clone exoplanets are well met by databases constructed for studying our own atmosphere, such as HITRAN, this is not the case for those exoplanets which will be the first exoplanets to be characterised. These planets are hot and most are subject to strong bombardment by starlight (and presumably stellar winds). Such studies require data on hot molecules which are not generally available and which are in many cases not easily amenable to laboratory experiments. In particular, current developments in exoplanetary observations are fueling the requirement for HR data. This creates new challenges for laboratory spectroscopy projects like ExoMol, which have been mainly concerned with completeness (important for low resolution-based retrievals) rather than HR studies. The field of the HRS of exoplanets is growing extremely fast and urgently demanding high precision molecular data. Failures to detect molecules in atmospheres of exoplanets can often be attributed to the lack of the underlying quality of the line positions.

The ExoMol database contains extensive line lists for approximately 80 molecules, in most cases for several isotopically substituted variants ("isotopologues"), with over 700 billion molecular transitions and two formal releases. The current ExoMol activity aims at providing high accuracy spectroscopic data for high resolution studies of exoplanetary atmospheres. Novel theoretical techniques are being used to provide similar high accuracy line lists for isotopically substituted species and to generate pressure broadening data.

It is only mildly controversial to insist that planets are three-dimensional. Whether or not 1D models can adequately be used to interpret exoplanet observations depends on some combination of how much spatial inhomogeneity exists on a planet and how sensitive the data are to its 3D structure. For atmospheric characterization using high resolution spectroscopy, our highest signal-to-noise targets are (ultra) hot Jupiters and these planets have huge spatial inhomogeneities. We are increasingly recognizing that these planets’ 3D properties influence, and are constrained by, high resolution spectroscopy in both transmission and emission. I will discuss hot Jupiter 3D atmospheric structure, including the pieces of complicated physics we are working to unravel, and briefly show how our expectations change as we move beyond the hot Jupiter population. I will then review the multiple ways in which 3D structures show up in high resolution spectroscopy, through spatial variations in physical properties and atmospheric winds. I will highlight the current edge of our capabilities and speculate about how this field may advance in the future.

Ultra-hot Jupiters are tidally locked gas giants with dayside temperatures high enough to dissociate hydrogen and other molecules. Featuring sharp chemical gradients and large temperature contrasts, their atmospheres are vastly non-uniform. In recent years, the wealth of data from high-resolution spectrographs such as HARPS-N, CRIRES, ESPRESSO and IGRINS has yielded spectacular insights into the chemical inventory, wind profiles and temperature structures of ultra-hot Jupiters. In particular, high-precision observations with ESPRESSO were able to sample the varying Doppler shift of the absorption lines in the spectrum of a transiting exoplanet.

In this talk, I will present a new framework for computing time-dependent, 3D transit spectra of exoplanets at high spectral resolution. We post-process the output of a state-of-the-art global circulation model (the SPART/MITgcm) through a GPU-optimised Monte-Carlo radiative transfer code, called gCMCRT. This allows us to correctly model the millions of spectral lines that can be observed with ground-based instruments, while accounting for the complex effects of thermal and chemical inhomogeneities, wind gradients and planet rotation in a unified framework.

I will demonstrate that different chemical species, such as water, CO, and iron, are distributed differently throughout the atmosphere of an ultra-hot Jupiter, resulting in unique transit signals for certain atoms and molecules. This is because thermal dissociation and condensation impact the chemical composition of the dayside and the nightside of the planet. I will show that our framework can qualitatively reproduce the time-dependent iron signal observed by Ehrenreich+ (2020) for WASP-76b. To this end, we either assume that iron condenses into clouds on the nightside, or that the morning limb is substantially cooler than expected.

In addition, I will illustrate how signals from other species, such as water (which is dissociated on the dayside) and CO (which is unaffected by temperature) can be used to bypass the inherent degeneracies associated with the observation of a single species. Finally, I will cover the results of our recent transit observations of WASP-121b with Gemini-S/IGRINS. These observations targeted water and CO absorption in the infrared, with the objective to obtain a similar signal-to-noise ratio as the WASP-76b iron signal observed with ESPRESSO. I will discuss how interpreting the absorption signals of these three chemical species (water, CO, iron) within the same framework allows us to place important constraints on the 3D structure of the limb of an ultra-hot Jupiter.

Exoplanet atmospheres are inherently three-dimensional systems in which thermal/chemical variation and winds can strongly influence spectra. Recently, the ultra-hot Jupiter WASP-76~b has shown evidence for condensation and asymmetric Fe absorption with time. However, it is unclear whether these asymmetries are driven by chemical or thermal differences between the two limbs, as precise constraints on variation in these have remained elusive due to the challenges of modelling these dynamics in a Bayesian framework as well as the procurement of such high quality observations. To address this we develop a new model, HyDRA-2D, capable of simultaneously retrieving morning and evening terminators with day-night winds, and use this on recent high-precision terminator observations of WASP-76~b with ESPRESSO/VLT. We explore variations in Fe, temperature profile, winds and opacity deck with limb and orbital phase. In this talk I will show that with HyDRA-2D we find Fe is more prominent on the evening for the last quarter of the transit. On the other hand the morning shows a lower abundance with a wider uncertainty, driven by degeneracy with the opacity deck and because the stronger evening signal dominates the overall spectrum. We also constrain a trend of higher temperatures for the more irradiated atmospheric regions, and a higher wind speed for the last quarter of the transit than the first. This new spatially- and phase-resolved treatment is statistically favoured by 4.9$\sigma$ over traditional 1D-retrievals, and thus demonstrates the power of such modelling for robust constraints with current and future facilities.

We are developing an open-source auto–differentiable spectral modeling code "ExoJAX" to perform atmospheric retrieval for high-resolution exoplanetary spectra by ab initio spectral fitting. This enables fully Bayesian inference of atmospheric parameters with the Hamiltonian Monte Carlo. Meanwhile, the parameter estimation of M dwarfs, which are important as host stars of various exoplanets, has mostly depended on pre-computed radiative-chemical equilibrium models. However, it has been reported in many literatures that the incompleteness of those models is a fundamental problem due to the spectral complexity caused by the cool atmosphere. We have retrieved the atmospheric parameters of the well-known M dwarf Barnard's star ($T_{\mathrm{eff}} \sim$ 3200 K) by applying the ExoJAX to its high-resolution near-infrared spectra ($R\sim$ 80000, $YJH$–band) independently of stellar models. We succeeded in reproducing well the molecular and atomic lines simultaneously at several wavelengths, and estimated the temperature-pressure profile and mixing ratios of each atomic and molecular species. Nevertheless, at other wavelengths, we found that the same way could not reproduce the absorption features at all, so we discuss the inaccuracy and missing absorption lines of the latest line list. The inspection of atmospheric retrieval of M dwarfs, for which is easier to obtain high-resolution spectra with high quality than in exoplanets, will highlight the path and caveats to accurate retrieval of exoplanets.

Ultrahot Jupiters are ideal candidates to explore with high-resolution emission spectra. Detailed theoretical studies are necessary to investigate the range of spectra we can expect to see from these objects throughout their orbit, because of the extreme temperature and chemical longitudinal gradients that exist across day and nightside regions. Using previously published 3D GCM models of WASP-76b with different treatments of magnetic drag, we post-process the 3D atmospheres to generate high-resolution emission spectra for two wavelength ranges and throughout the planet's orbit. We find that the high-resolution emission spectra vary strongly as a function of phase, at times showing emission features, absorption features, or both, which are a direct result of the 3D structure of the planet. At phases exhibiting both emission and absorption features, the Doppler shift differs in direction between the two spectral features, making them differentiable instead of canceling each other out. Through the use of cross-correlation, we find different patterns in net Doppler shift for models with different treatments of drag: the nightside spectra show opposite signs in their Doppler shift, while the dayside phases have a reversal in the trend of net shift with phase.

High resolution spectroscopy is quickly establishing itself as one of the most exciting ground-based observational techniques for exoplanets. Especially the doppler shifts of atomic and molecular lines are an excellent probe of dynamics and thus offer a unique window into the atmospheres of exoplanets. This information is highly complementary when compared to space-based telescopes, which observe at lower resolution. What is more, velocity differences observed for different absorbers are thought to probe their spatial distribution in the atmosphere, unlocking its 3-d nature. To interpret observations that probe the multi-d structure of atmospheres we are developing an open-source extension to our atmospheric retrieval package petitRADTRANS, called pRT-Orange. pRT-Orange divides the atmosphere of the planet into (orange) segments that share common properties such as temperature, abundance, or cloud structure. The location, extent and number of segments can be freely chosen, allowing for simple or complex atmospheric parameterizations. Arbitrary velocity fields can be specified for the Doppler shifts, ranging from planetary rotation to feeding in the dynamic solutions of general circulation models. At the moment pRT-Orange works for transmission only. In my talk I will present the current state of the code development, and grant early access to pRT-Orange at the meeting, if people are interested (with the obvious work-in-progress disclaimer).

The evolution of planetary atmospheres is very much dependent on the environment of their host stars (e.g., stellar radiation, stellar wind, stellar flares and Coronal Mass Ejections (CMEs)). For close-in planets, the stellar radiation evaporates the planetary atmosphere as a form of supersonic planetary outflow due to photoionization. This planetary outflow further interacts with the stellar wind, which shapes up the planetary atmosphere (sometimes producing a comet-like structure) and its mass loss rate. Moreover, flares and CMEs from the star will also have a great impact on planetary evaporation. In this talk, I will discuss the effect of stellar wind, and the impact of flares and CMEs on the atmosphere of hot Jupiters using a self-consistent 3D radiation hydrodynamic model. Among all the considered stellar environments, we find that CMEs are very effective in eroding planetary atmospheres. We also calculate synthetic Lyman-alpha transit signatures and find that the flare alone cannot explain the observed high blue shifted velocities seen in the Lyman-alpha observation. The CME, however, leads to an increase in the velocity of the escaping atmosphere, enhancing the transit depth at high blue shifted velocities. Finally, I will also discuss the effect of the different orientations of the CME magnetic field on the atmospheric escape and corresponding transit signatures.

Spectrographs for radial velocities evolved from the specifications required for Galactic dynamics studies to those set by exoplanet mass characterization. As time went by, the search for lower-mass planets on longer periods pushed for higher measurement precision and improved instrument stability. But the detection of a transiting planet in 2000 and an exo-atmosphere in 2010 reshuffled scientific priorities. Today, atmosphere characterization is done mostly on spectrographs designed for precise RVs, and yet is having a field day at it.

We will discuss the results from the Exoplanets with Gemini Spectroscopy (ExoGemS) survey, which has acquired high-resolution transit spectroscopy of over a dozen exoplanets. Because our survey covers a large range of radii, masses, and Teq, we can compare how these properties affect the resulting transmission spectrum. The high-resolution of GRACES allows us to analyze the line profile and study the atmosphere at higher altitudes. In particular, we will highlight the detections from WASP-76 b, HAT-P-32 b, WASP-31 b, and WASP-85 Ab. We will also discuss the effect of stellar activity on these results.

We aim to characterize transiting exoplanets spanning a wide range of sizes, masses and effective temperatures using a variety of high-resolution spectrographs, including through our on-going Gemini Large Program “Exploring the Diversity of Exoplanet Atmospheres at High Spectral Resolution” (ExoGemS, for short) with Gemini-N/GRACES. Here I will share a few early results from ExoGems, and present our new, complementary study of GJ 486b, a 1.3-Earth-radius planet around a nearby M dwarf. Given its high transmission spectroscopy metrics, GJ 486b is a compelling target for constraining an atmosphere in the terrestrial size regime. We observed three planetary transits with Subaru/IRD, Gemini-S/IGRINS and CFHT/SPIRou, and searched for absorption by a multitude potential atmospheric species. We are able to rule out a H/He-dominated atmosphere with solar abundances to a confidence of >5σ and a 100% water atmosphere to a confidence of 3σ. We also investigate the implications of our results for the upcoming JWST transit observations of GJ 486b. Our findings suggest that terrestrial planets orbiting M-dwarf stars may experience significant atmospheric loss.

Characterization of cooler atmospheres of super-Earths and Neptune sized objects at low-resolution is often thwarted by the presence of clouds, hazes and aerosols which effectively flatten the transmission spectra. High-Resolution Spectroscopy (HRS) presents an opportunity to overcome this limitation by having the ability to detect molecular species whose spectral line cores extend above the level of clouds in these atmospheres. We analyse High-Resolution observations of the warm Neptune GJ 3470b taken over one transit using CARMENES (R ~ 80400) and two transits using GIANO (R ~ 50000) and look at the possibility of signatures for first H$_2$O in isolation and then H$_2$O and CH$_4$ together. We find a tentative detection of H$_2$O using either the peak of the cross correlation signal or a cross-correlation-to-likelihood metric while comparing it to just the best fit model, when all three nights are combined. The detection becomes even weaker when both H$_2$O and CH$_4$ are used for abundance and cloud deck layer values close to the best fit model. This decrease is in line with results from the Hubble Space Telescope, at much lower resolution. However, accounting for the effects of data analysis on the compared model produces a strong detection close to the expected exoplanet position using just one night of CARMENES. Following this, we are planning to use a Bayesian retrieval tool to put simultaneous constraints on the abundance of molecular species and the pressure of the cloud top-deck. Such a tool will also allow us to directly compare/combine our results with published HST observations.

Directly imaged planets orbiting at very wide separation present a challenge to current planet formation theories. We present the analysis of new observations of beta Pictoris b with the refurbished and enhanced CRIRES+, with new constraints on its spin-rotation and C/O ratio. Furthermore, we will introduce a python package for the reduction of CRIRES+ data specifically aimed at directly imaged planets, and our future plans for observing directly imaged planets with CRIRES+.

I will discuss the smallsat SEEJ (Structure & Evolution of ExoJupiter Atmospheres). It will measure both the fluxes of high-energy photons emanating from a flux limited sample of planet-hosting stars and the absorption depth of X-rays in the atmospheres of hot Jupiter, Saturn, and Neptune analogs. SEEJ will measure how often high energy flares of a given size occur, and establish, for the first time, the statistics of these crucial events on exoplanet hosts. Specifically, the SEEJ investigation will determine the degree to which stellar high-energy photons inflate nearby exoplanet atmospheres and the physical characteristics of driven planetary winds. The investigation will determine the bulk composition of the inflated atmosphere and will assess the presence of dense evaporation tails resulting from this interaction. The experimental objective is to measure the high energy fluence through stellar monitoring and understand the impact of the high energy fluence by measuring the atmospheres using the X-ray transit technique.

To accomplish this goal, we need to be able to demonstrate to the astrophysical community at-large a clear connection from the observations to reality. We need to explain what the transit results from a limited set of planets imply for the atmospheric evolution of planets more broadly. This path starts with the detection of a transit followed by a fitted light curve and then inferences about the obstructing material. Finally, we need to demonstrate the implications that the obscuring material has for the underlying atmosphere. In this discussion, I highlight some of our communal shortcomings. For example, while we have recently improved our simulations and the inference they provide, the single existing positive detection of an X-ray transit (HD 189733b) makes it difficult to evaluate or quantify astrophysical and instrumental systematic issues that are associated with general transit simulations. The utility of solar templates for the diverse and different spot coverage of young stars and M dwarfs is undemonstrated.

The search for signs of life on other worlds is an exciting and now key component of exoplanet science. In the exoplanet context, biosignatures are potentially-detectable impacts of life on a global planetary environment. Biosignatures may include identified and sought after molecules in a planetary atmosphere, reflectivity signals on its surface, or the characteristic variation of these or other characteristics as a function of time. Biosignatures may also be “agnostic”, meaning that they are not associated with a known or postulated metabolism, thereby allowing us to search for unknown forms of life. Agnostic biosignatures consist of anomalous patterns or molecules that are unlikely to be generated by abiotic planetary processess. All biosignatures, and particularly agnostic biosignatures, must be interpreted in the context of their planetary environment. This context can provide additional information to help rule out biosignature “false positives” where instead of being generated by life, a sought after environmental characteristic is instead generated by abiotic processes like volcanism and photochemistry. In the near-term, the search for life on exoplanets will focus exclusively on M dwarf exoplanets, which, due to the coevolution with their star, may undergo a very different evolutionary path than our own Earth. These searches will be undertaken using low-medium resolution transmission spectroscopy with JWST, and high-resolution transmission and reflected light spectroscopy with ground-based telescopes. In the longer term, a large-aperture space-based telescope will use direct imaging techniques to obtain reflected light from planets orbiting a broader swath of host stars, including the more “Sun-like” FGK dwarfs. This will expand our search for life in the universe to planetary systems that are more analogous to our own. In this talk I will introduce the field of biosignatures, and describe the potential capabilities for biosignature searches using high-resolution spectroscopy with ground-based telescopes, placing these opportunities in the context of what might be possible with space-based telescopes over the next two decades.

The upcoming class of extremely large telescopes (ELTs) will provide an unprecedented opportunity to use high-resolution spectroscopy to characterize terrestrial exoplanets for habitability and life. In particular, these telescopes are likely the best near-term tools for detecting molecular oxygen in nearby exoplanet atmospheres. However, determining whether oxygen is more likely to have a biological origin requires contextual information from the planetary environment to support the identification and rule out false positives. Studies which investigate the ELTs’ capacity to detect other gases can enhance the science return from these telescopes and expand our abilities to search for signs of life. We have developed a novel pipeline to simulate telescope observations and estimate the detectability of a suite of gases—CH$_4$, CO$_2$, CO, O$_3$, and H$_2$O—that can help give context to ELT O$_2$ detections in terrestrial exoplanet atmospheres. As input, we used a suite of photochemically self-consistent simulations of M dwarf planets with modern/Archean Earth-like atmospheres, and worlds with abiotic O$_2$ buildup due to photochemical generation and ocean loss processes. We find that CO$_2$ and CH$_4$ are two of the most detectable molecules in M dwarf planetary atmospheres, and may be detectable on TRAPPIST-1 e in less than 35 transits. This may be the only known biosignature pair accessible with ELT high-resolution spectroscopy for this target. However, for closer targets, the ELTs alone may be capable of discriminating an inhabited world from one without life with tens of hours of observation time under ideal conditions. Additionally, we find the detectability of all gases is strongly dependent on host star type—planets orbiting late-type M dwarfs may require less overall observation time to achieve a significant detection. Finally, we develop an observing protocol that prioritizes the most detectable gas absorption bands to maximize the science output of ELT observations, and inform instrument development beyond the first light capabilities.

Among more than 5000 exoplanets discovered up to now, around 60 are believed to be potentially habitable. The Sun-Earth system provides a unique example based on which detailed insights into the properties, formation, evolution, and thus habitability of exoplanets could be gained. However, observing the Sun as a star and the Earth as an exoplanet has been rare. In this talk, I will briefly introduce the SOlar Terrestrial Habitability Explorer (SOTHE) to be deployed to the Sun-Earth L1 point around 2025. SOTHE will carry 5 payloads to obtain the spectra of the Sun and the Earth at the same time, together with images of the Earth at 7 unique passbands and the local plasma and magnetic field parameters at the L1 point. The core scientific goal of SOTHE is to conduct the first-ever simultaneous spectral observations of the Sun and Earth to explore key characteristics related to the habitability of the Sun-Earth system and provide a unique baseline for habitable exoplanets exploration.

The most common type of exoplanet discovered to date has a size that falls between Earth-sized and Neptune-sized, and orbits its star with a period less than 100 days. Atmospheric characterization of these planets often finds featureless transmission spectra, which is taken as evidence for the presence in the of clouds or hazes in the upper atmosphere that obscure the presence of deeper lying atmospheric constituents. We will present the results of a new model that suggests hazy atmospheres are innate byproducts of planet formation, where volatile ingredients are outgassed from the mantle and transformed in the upper atmosphere after exposure to stellar photons. Our model relates the initial mantle composition of the planet to its formation zone around its star, factoring in the relative contributions of refractories (metals and silicates) and volatile components (solid state organics, water vapor/ice, and hydrogen-dominated nebular gas). We predict that a population of super-Earths will form in particular locations in their protoplanetary disks such that they receive significant inventories of organics, but very low amounts of water. These hydrocarbon-rich planets are not carbon worlds, but rather carbon-rich planets whose subsequent evolution would differ from typically assumed for terrestrial planets. We will show models that encompass the geochemical equilibrium of the mantle and atmosphere, compute the resulting atmospheric chemical equilibrium, haze production, and predicted spectra.

With advances in exoplanet detection, thousands exoplanets have been discovered including almost 200 confirmed rocky exoplanets. This leaves us with a large number of possible targets in the search for biomarkers. A quick look at our own solar system shows that not every planet can hold on to a significant or habitable atmosphere for an extended amount of time. Whether a planet can hold on to its atmosphere depends strongly on the XUV radiation of the star and the chemical composition of the atmosphere. To this end we use the Kompot code, an upper atmosphere thermo-chemical code, developed at the University of Vienna. By modelling the interaction between these two factors we can exclude scenarios where the atmosphere would not survive long enough for possible life to form. This work will also present how grids of such model atmospheres can be used to constrain the chemical composition and identify which biomarkers remain abundant under the stellar radiation.

The upcoming Extremely Large Telescopes (ELTs) will have the collecting area required to detect potential biosignature gases such as molecular oxygen, O2, in the atmosphere of terrestrial planets around nearby stars. To maximize our capability to detect O2 using this method, extreme high spectral resolution R=300,000-500,000 is required to fully resolve the absorption lines in an exoplanet atmosphere and disentangle telluric lines from our own atmosphere. Current high-resolution spectrographs typically achieve spectral resolution of R=100,000. We demonstrate a new approach with an ultra-high spectral resolution booster to be coupled in front of a high-resolution spectrograph. The booster is a chained Fabry Perot array which imposes a hyperfine chained spectral profile. With a prototype we developed, our on-sky observations of the solar spectrum around the O2 A-band demonstrate a resolving power of R=450,000. The capabilities of this instrument for exoplanet characterization are substantiated by detection of multiple atomic species, Ni I, Fe I, Mg I, K, and Si, hidden among the molecular oxygen feature.