Machine learning inference of the interior structure of low-mass exoplanets

Abstract

We explore the application of machine learning based on mixture density neural networks (MDNs) to the interior characterization of low-mass exoplanets up to 25 Earth masses constrained by mass, radius, and fluid Love number $k_2$. We create a dataset of 900000 synthetic planets, consisting of an iron-rich core, a silicate mantle, a high-pressure ice shell, and a gaseous H/He envelope, to train a MDN using planetary mass and radius as inputs to the network. For this layered structure, we show that the MDN is able to infer the distribution of possible thicknesses of each planetary layer from mass and radius of the planet. This approach obviates the time-consuming task of calculating such distributions with a dedicated set of forward models for each individual planet. While gas-rich planets may be characterized by compositional gradients rather than distinct layers, the method presented here can be easily extended to any interior structure model. The fluid Love number $k_2$ bears constraints on the mass distribution in the planets' interior and will be measured for an increasing number of exoplanets in the future. Adding $k_2$as an input to the MDN significantly decreases the degeneracy of the possible interior structures.

arXiv comment14 pages, 7 figures, accepted for publication in ApJ

References in corpus (45)

• The Transiting Exoplanet Survey Satellite
• The PLATO 2.0 Mission
• The WASP Project and the SuperWASP Cameras
• Planetary Radii across Five Orders of Magnitude in Mass and Stellar Insolation: Application to Transits
• Mass-Radius Relationships for Solid Exoplanets
• A super-Earth transiting a nearby low-mass star
• The evaporation valley in the Kepler planets
• Understanding the Mass-Radius Relation for Sub-Neptunes: Radius as a Proxy for Composition
• Growth Model Interpretation of Planet Size Distribution
• Internal Structure of Massive Terrestrial Planets
• KEPLER's First Rocky Planet: Kepler-10b
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• A New Family of Planets ? "Ocean Planets"
• Can we constrain interior structure of rocky exoplanets from mass and radius measurements?
• A Framework for Quantifying the Degeneracies of Exoplanet Interior Compositions
• Detailed Models of super-Earths: How well can we infer bulk properties?
• The Next Generation Transit Survey (NGTS)
• Three Possible Origins for the Gas Layer on GJ 1214b
• A Detailed Model Grid for Solid Planets from 0.1 through 100 Earth Masses
• Ocean Planet or Thick Atmosphere: On the Mass-Radius Relationship for Solid Exoplanets with Massive Atmospheres
• HATSouth: a global network of fully automated identical wide-field telescopes
• New indication for a dichotomy in the interior structure of Uranus and Neptune from the application of modified shape and rotation data
• A generalized bayesian inference method for constraining the interiors of super Earths and sub-Neptunes
• Exoplanetary Atmospheres - Chemistry, Formation Conditions, and Habitability
• Long-term tidal evolution of short-period planets with companions
• Thermal evolution and structure models of the transiting super-Earth GJ 1214b
• A grid of upper atmosphere models for 1--40 MEARTH planets: application to CoRoT-7 b and HD219134 b,c
• Searching for Exoplanets Using Artificial Intelligence
• HAT Variability Survey in the High Stellar Density "Kepler Field" with Millimagnitude Image Subtraction Photometry
• Determination of the Interior Structure of Transiting Planets in Multiple-Planet Systems
• Constraints on Super-Earths Interiors from Stellar Abundances
• ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks
• The Transiting System GJ1214: High-Precision Defocused Transit Observations and a Search for Evidence of Transit Timing Variation
• Interior Structures and Tidal Heating in the TRAPPIST-1 Planets
• The Pressure and Temperature Limits of Likely Rocky Exoplanets
• Revised Masses and Densities of the Planets around Kepler-10
• On the degeneracy of the tidal Love number k2 in multi-layer planetary models: application to Saturn and GJ436b
• An estimate of the $k_2$ Love number of WASP-18Ab from its radial velocity measurements
• In hot water: effects of temperature-dependent interiors on the radii of water-rich super-Earths
• Classifying Exoplanet Candidates with Convolutional Neural Networks: Application to the Next Generation Transit Survey
• Detectability of shape deformation in short-period exoplanets
• Retrieval of the fluid Love number $k_2$ in exoplanetary transit curves
• Using Deep Neural Networks to compute the mass of forming planets
• Matrix-propagator approach to compute fluid Love numbers and applicability to extrasolar planets
• Secondary atmospheres on HD 219134 b and c

Cited by in corpus (24)

• GJ 367b: A dense ultra-short period sub-Earth planet transiting a nearby red dwarf star
• Detection of the tidal deformation of WASP-103b at $3\sigma$ with CHEOPS
• ExoMDN: Rapid characterization of exoplanet interior structures with Mixture Density Networks
• Transmission spectroscopy for the warm sub-Neptune HD3167c: evidence for molecular absorption and a possible high metallicity atmosphere
• Convolutional neural networks as an alternative to Bayesian retrievals
• Exoplanet Characterization using Conditional Invertible Neural Networks
• MAGRATHEA: an open-source spherical symmetric planet interior structure code
• Possible Atmospheric Diversity of Low Mass Exoplanets, some Central Aspects
• Unveiling the internal structure and formation history of the three planets transiting HIP 29442 (TOI-469) with CHEOPS
• NGTS-14Ab: a Neptune-sized transiting planet in the desert
• Redox state and interior structure control on the long-term habitability of stagnant-lid planets
• HST/STIS capability for Love number measurement of WASP-121b
• Why do more massive stars host larger planets?
• Evidence of apsidal motion and a possible co-moving companion star detected in the WASP-19 system
• H${}_2$-H${}_2$O demixing in Uranus and Neptune: Adiabatic structure models
• Characterizing WASP-43b's interior structure: unveiling tidal decay and apsidal motion
• NeuralCMS: A deep learning approach to study Jupiter's interior
• Searching for Hot Water World Candidates with CHEOPS: Refining the radii and analysing the internal structures and atmospheric lifetimes of TOI-238 b and TOI-1685 b
• Exo-Geoscience Perspectives Beyond Habitability
• Tidally Heated Sub-Neptunes, Refined Planetary Compositions, and Confirmation of a Third Planet in the TOI-1266 System
• TOI-2458 b: A mini-Neptune consistent with in situ hot Jupiter formation
• Exploring the habitability and interior composition of exoplanets lying within the extended habitable zone
• Discovery of a transiting hot water-world candidate orbiting Ross 176 with TESS and CARMENES
• Improved characterization of the TOI-2141 system: a dense sub-Neptune with non-transiting inner and outer companions