329.01 — Towards a more complex description of chemical profiles in exoplanets retrievals: A 2-layer parameterisation
Quentin Changeat1; Billy Edwards1; Ingo Waldmann1;
Giovanna Tinetti1
1 Physics and Astronomy, University College London (London, United Kingdom)
State of the art spectral retrieval models of exoplanet atmospheres assume chemical profiles which are constant with altitude/pressure. This assumption is justified by the information content of currently available datasets which do not allow, in most cases, for the molecular/atomic abundances as a function of atmospheric pressure/altitude to be constrained. In the context of the next generation of space tele- scopes, a more accurate description of chemical pro- files with additional levels of flexibility may become crucial to interpret observations and gain new in- sights into atmospheric physics. We explore here the possibility of retrieving pressure-dependent chemi- cal profiles from transit spectra as recorded by fu- ture space observatories, without injecting any pri- ors from theoretical chemical models in the retrieval algorithms. The “2-layer” retrieval parameterisation presented here allows for the independent extraction of molecular/atomic abundances above and below a certain atmospheric pressure. By simulating var- ious cases, we demonstrate that this evolution from the assumption of constant chemical abundances is justified by the information content of transit spec- tra provided by future space instruments. Compar- isons with traditional retrieval models show that as- sumptions made on chemical profiles may signifi-
cantly impact retrieved parameters, such as the at- mospheric temperature, and justify the attention we give here to this issue. We find that the 2-layer re- trieval is able to accurately capture discontinuities in the vertical chemical profiles, which could be caused by disequilibrium processes — such as ver- tical mixing or photo-chemistry — or the presence of clouds/hazes. The 2-layer retrieval could also help to constrain the composition of clouds and hazes by exploring the correlation between the chemical changes in the gaseous phase and the pressure at which the condensed/solid phase occurs. The 2- layer retrieval presented here therefore represents an important step forward in our ability to constrain theoretical chemical models and cloud/haze compo- sition from the analysis of future observations.
329.02 — The Formation of Zonal Flow and a Hot- spot Shift on Tidally Locked Planets
Mark Hammond1; Raymond Pierrehumbert1 1 Physics, University of Oxford (Oxford, United Kingdom) The global atmospheric circulation and temperature distribution of tidally locked exoplanets is key to in- terpreting observations such as phase curves, eclipse maps, and atmospheric retrievals. The dominant dynamical process in their atmospheres is a super- rotating zonal equatorial jet. The meridional cir- culation on tidally locked planets is rarely investi- gated, as it is assumed to be secondary to the zonal day-night circulation. This poster will show that the meridional circulation is in fact vital to the formation of the zonal flow, via the same ”Gierasch-Rossow- Williams” mechanism that operates on Venus. It will show how the meridional circulation adds eastward angular momentum to the atmosphere and trans- ports this angular momentum upwards. This mech- anism will then be used to explain the scaling be- haviour of the zonal jet position and speed in a suite of GCM simulations, by predicting how the various terms in the zonal momentum equation scale with stellar flux and rotation rate.
Finally, I will link this to our recently published study using a rotating shallow-water model lin- earized about this eastward zonal flow. I will show that the response to day-night forcing is modified by this jet, producing the distinctive eastward hot-spot shift that appears in observations. The key result is that the shift is not produced by advection of heat, but rather is a result of interactions between forced stationary waves and the mean flow. This theoretical prediction of the equilibrium zonal flow and the re- sulting temperature distribution could be useful for interpreting observations of tidally locked planets.
329.03 — Atmospheric Mass Loss due to Giant Im- pacts
Almog Yalinewich1
1 CITA (Toronto, Ontario, Canada)
Exoplanet systems (especially Kepler 36b/c and Ke- pler 107b/c) exhibit large density variations between neighbouring planets. This difference is attributed to atmospheric content. One mechanism that is usu- ally implicated in atmospheric mass loss is photoe- vaporation, but this mechanism cannot explain the Kepler 107b/c dichotomy, for which the planet with the atmosphere is closer to the host star and less mas- sive. Another explanation is giant impacts that occur shortly after the dispersal of the protoplanetary disc. In this talk I will present new results from a state of the art, moving mesh, hydrodynamic simulations of such collisions. These simulations capture the prop- agation of the shock wave through the interior of the planet. Using these results it is possible to calculate the amount of atmospheric mass loss from a wide range of parameters. In contrast to other atmospheric loss processes, giant impacts can remove both the primordial and secondary atmospheres with a larger molecular weight. In this talk I will present new, yet unpublished results that include the effects of an oblique collision and take into account the finite speed of sound in the interior of the target planet. With these more realistic simulations I will show that under the right circumstances, giant collisions can account for the differences in atmospheric content in Kepler 36b/c and Kepler 107b/c.
329.04 — Do Magnetic Fields Prevent Atmospheric Escape?
Hilary Egan1,5; Riku jarvinen2,3; Yingjuan Ma4; David Brain5,1
1 Astrophysical and Planetary Science, University of Colorado Boulder (Boulder, Colorado, United States)
2 Department of Electronics and Nanoengineering, Aalto Univer- sity (Espoo, Finland)
3 Finnish Meteorological Institute (Helsinki, Finland)
4 Department of Earth Planetary and Space Science, University of California Los Angeles (Los Angeles, California, United States)
5 Laboratory for Atmospheric and Space Physics, University of Colorado (Boulder, Colorado, United States)
Atmospheric escape is capable of shaping a planet’s atmospheric composition and total mass, and thus the planet’s long-term habitability. Loss to space of atmospheric particles has played a key role in the atmospheric evolution of both Mars and Venus. Intrinsic planetary magnetic fields like the Earth’s
have long been thought to shield planets from at- mospheric erosion via stellar winds; however, re- cent arguments have suggested that a magnetic field will increase the interaction area with the solar wind, collecting correspondingly more energy that can be used to drive increased escape.
Using a set of global three-dimensional hybrid plasma simulations validated via observations at Mars and Venus, we find that neither of these paradigms are complete descriptions. Rather than solely inhibiting or driving ion escape, there is a value of magnetic field strength associated with maximum ion outflow. For weaker magnetic fields, ion escape is enhanced due to shielding of the southern hemisphere from “misaligned” ion pickup forces. For stronger magnetic fields ion escape de- creases due to trapping associated with closed mag- netic field lines. The peak escape rate occurs where the intrinsic magnetosphere (caused by the planetary magnetic field) reaches the induced magnetosphere (caused by ionospheric conductivity). As the size of the intrinsic magnetosphere is determined by pres- sure balance between the incoming solar wind and the planetary magnetic field, the magnetic field as- sociated with peak escape is critically dependent on the solar wind pressure.
Where possible we have fit power laws for the vari- ation of fundamental parameters (escape rate, es- cape power, polar cap opening angle and effective interaction area) with magnetic field, and assessed upper and lower limits for the relationships. Such power laws can be used in generalized studies of at- mospheric escape and potential habitability to better characterize a wide variety of systems.
329.05 — THOR version 2: a GPU-enabled, non- hydrostatic general circulation model for extra- solar planets
Russell Deitrick1; Joao M. Mendonca2; Urs Schroffenegger1; Shang-Min Tsai3; Simon Grimm1; Kevin Heng1
1 Center for Space and Habitability, University of Bern (Bern, Bern, Switzerland)
2 National Space Institute, Technical University of Denmark (Kon- gens Lyngby, Denmark)
3 University of Oxford (Oxford, United Kingdom)
We present the first major update to THOR, a non- hydrostatic, GPU enabled 3D general circulation model, which is the culmination of 8 years of work in the Heng group in Bern. THOR is the first GCM that has been built from the ground up for the study of exoplanets. Thus, it is entirely free of tunings to- ward solar system planets and contains as few as-
sumptions as possible. It is also publicly available and we actively encourage the community to become involved in further development. With this model, we have the capability to model atmospheres with or without the hydrostatic approximation, indepen- dent of the additional approximations that lead to the primitive equations of meteorology. We use the model to study whether the climate structures of ex- oplanets are robust to the assumption of hydrostatic equilibrium. We demonstrate that the hydrostatic approximation alone is sufficient to significantly al- ter the zonal and vertical winds of hot jupiters. This implies that aerosol sizes derived from spectra may be miscalculated, if the wind velocities are based upon hydrostatic GCMs. The divergence between hydrostatic and non-hydrostatic simulations appears to be a function of temperature. We further dis- cuss improvements and additions to the model that have been implemented since the release of version 1.0, including grey radiative transfer, chemical trac- ers, and an insolation scheme that allows for arbi- trary orbits and rotation parameters. We have be- gun adapting the model for terrestrial planets with the goal of studying atmospheric collapse on tidally- locked worlds. Additionally, we reproduce a num- ber of benchmark tests for dynamical cores.
329.06 — Effect of disequilibrium chemistry on the spectra of exoplanet atmospheres
Yui Kawashima1; Michiel Min1
1 SRON Netherlands Institute for Space Research (Utrecht, Nether- lands)
Recently, transmission and/or emission spectra of exoplanet atmospheres have been observed by both space- and ground-based telescopes. Forthcoming space missions such as JWST and ARIEL are ex- pected to enable high-precision observation of these spectra. Most of the current retrieval models used to derive the atmospheric properties from observed spectra assume the abundance profiles of chemical species in the atmospheres to be thermochemical- equilibrium or constant ones throughout the atmo- sphere for simplicity. However, in reality, the abun- dance profiles can depart from the thermochemi- cal equilibrium ones by the disequilibrium processes such as quenching effect. In this study, we exam- ine how the quenching process affects the spectra of exoplanet atmospheres for some atmospheric prop- erties such as temperature and eddy diffusion co- efficient. For this purpose, we have developed 1-D model to simulate the abundance profiles consider- ing both thermochemical reactions and eddy diffu- sion transport with the use of the chemical timescale
for each species derived by Tsai et al. (2018). We dis- cuss the conditions in which we can assume equilib- rium chemistry or disequilibrium chemistry.
329.07 — A coherent disentanglement of the finger- prints of physics, chemistry and dynamics of exo- planets on their atmospheric spectra
Karan Molaverdikhani1; Thomas Henning1; Paul Mollière2
1 Max Planck Institute for Astronomy (Heidelberg, Germany) 2 Sterrewacht Leiden, Huygens Laboratory (Leiden, Netherlands) Characterization of planetary atmospheres has been always a challenge. While the next generation of fa- cilities, such as E-ELT, JWST, and ARIEL, will help to improve the status, the number of well-characterized exoplanet atmospheres will still be limited. Large- scale simulations could assist us by predicting the diversity of the planetary atmospheres, and point- ing toward the regions on the parameter space where we have a higher chance of finding interesting targets with desired properties.
We present the results of an extensive investiga- tion, with a three-step strategy, to understand the fingerprints of physics, chemistry and dynamics of exoplanets on their atmospheric spectra. In the first step, we study the synthetic spectra of 28,224 self-consistent cloud-free models; assuming effective temperature, surface gravity, metallicity, C/O ratio of the planet, and host star’s stellar type as the free parameters. We propose a new classification scheme and find a region (Methane Valley) between 800 and 1500 K, where a greater chance of CH4 detection is
expected. The first robust CH4 detection on an irra- diated planet places HD102195b within this region; supporting our prediction.
We then investigate the fingerprints of disequilib- rium chemistry on the atmospheric spectra by per- forming 84,672 full chemical network kinetic simu- lations with ChemKM. We find that the quenching pressure decreases with the effective temperature of planets, but it also varies with other atmospheric parameters. We show that the atmospheric mixing does not change the shape of the two main color- populations in the Spitzer color-maps and thus any deviation of observational points from these popula- tions are likely due to the presence of clouds and not disequilibrium processes. However, we find some colder planets (Teff<900 K) with very low C/O ra- tios (<0.25) that show significant deviations; making these planets interesting cases for further investiga- tions.
We further present the results of 38,500 self- consistent cloudy models to demonstrate how this
picture changes when the radiative feedback of clouds is included in the models.
329.08 — Retrieving Exoplanet Spectra using deep learning
Ingo Waldmann1
1 Physics & Astronomy, University College London (London, United Kingdom)
The field of exoplanetary spectroscopy is as fast mov- ing as it is new. Analysing currently available ob- servations of exoplanetary atmospheres often invoke large and correlated parameter spaces that can be dif- ficult to map or constrain. This is particularly true for the theoretical modelling of their atmospheres and the atmospheric parameter retrieval from observed data. Issues of low signal-to-noise data and large, non-linear parameter spaces are nothing new and commonly found in many fields of engineering and the physical sciences. Recent years have seen vast im- provements in statistical data analysis and machine learning that have revolutionised fields as diverse as telecommunication, pattern recognition and medical physics. In this talk, I will discuss the use of ma- chine and deep learning in inverse retrievals of ex- oplanetary atmospheres. I will present the ExoGAN retrieval framework, which learns the intrinsic like- lihood surface of a radiative transfer retrieval using generative adversarial networks (GANs) and com- pare this approach to other classical and machine learning solutions in the recent literature. As we firmly move into the era of ‘big data’ and increas- ing model complexities in the era of JWST, ELTs and ARIEL, intelligent algorithms will play an important part in facilitating the analysis of these rich data sets in the future.
329.09 — Nucleation of TiO2molecular clusters in the context of cloud formation on hot Jupiters
Jan Philip Sindel1; David Gobrecht1
1 Instituut voor Sterrenkunde, KU Leuven (Leuven, Belgium) Clouds form when a super-saturated gas condenses on small dust grains, leading to an optically thick gas-liquid mixture. On rocky planets like earth the processes are well-understood, as small dust grains are easily transferred from the surface to the up- per atmosphere by winds. On hot Jupiters however, there is no solid surface that supplies dust grains to act as condensation cores. Yet, clouds have been ob- served in the atmospheres of hot Jupiters. The cur- rent understanding of the formation process is that the required condensation cores are formed through
nucleation of molecular clusters of highly refractory molecules, similar to the nucleation of dust in stellar outflows.
In this work we investigate the cluster-formation of Titanium dioxide (TiO2) as a potential candidate
for the seed-nucleus. In order to establish a selfcon- sistent nucleation pathway, it is crucial to obtain the bond-energies for all isomers and geometries of the clusters with high accuracy. We achieve this by using quantum chemical density functional theory (DFT) calculations. We found that the B3LYP/cc-PVTZ level of theory comes closest to the experimental val- ues from the Janaf-Nist tables. Since the DFT cal- culations quickly get too computationally expensive with cluster size, we employ a force-field approach to approximate the energies of larger clusters as ac- curate as possible. Using an interatomic bucking- ham pair potential we find new cluster-geometries and their respective energies using a simulated an- nealing technique. We use the new clusters and their energies as input for a kintetic nucleation model that serves as the basis of a cloud formation code. Finally, we investigate the impact of the cluster geometries and energies on the nucleation process and thereby on the cloud formation on hot Jupiters.
329.10 — The effect of internal gravity waves on cloud evolution in sub-stellar atmospheres
Amy Parent1; Ruth Falconer1; Karen Meyer1; Craig R.
Stark1
1 Division of Computing and Mathematics, University of Abertay Dundee (Dundee, United Kingdom)
Substellar objects exhibit photometric variability which is believed to be caused by a number of pro- cesses such as magnetically-driven spots or inhomo- geneous cloud coverage. Recent substellar models have shown that turbulent flows and waves, includ- ing internal gravity waves, may play an important role in dust cloud evolution. The aim of this paper is to investigate the effect of internal gravity waves on dust cloud nucleation and dust growth, and whether observations of the resulting cloud structures could be used to recover atmospheric density information. For a simplified atmosphere in two dimensions, we numerically solve the governing fluid equations to simulate the effect on dust nucleation and mantle growth as a result of the passage of an internal grav- ity wave. Furthermore, we derive an expression that relates the properties of the wave-induced cloud structures to observable parameters in order to de- duce the atmosphere density. Numerical simula- tions show that the density, pressure and tempera- ture variations caused by gravity waves lead to an
up to 600-fold increase of the dust nucleation rate and an up to 80% increase of the dust growth rate in the linear regime. These variations lead to banded areas in which dust formation is much more pro- nounced. We show that internal gravity waves in substellar atmospheres lead to banded cloud struc- tures similar to those observed on Earth. Using the proposed method, potential observations of banded clouds could be used to estimate the atmospheric density of substellar objects.
329.11 — Atmospheric escape: new windows, longer baselines and demographic influences
John McCann2,1; Ruth Murray-Clay1; Mark Krumholz4; Kaitlin M. Kratter3
1 Astronomy and Astrophysics, UC Santa Cruz (Santa Cruz, Cali- fornia, United States)
2 Physics, UC Santa Barbara (Santa Barbara, California, United States)
3 Astronomy and Steward Observatory, Univ. of Arizona (Tucson, Arizona, United States)
4 Astronomy and Astrophysics, Australian National University (Canberra, Australian Capital Territory, Australia)
We present new quasi-global 3-D radiative-