Machine learning-accelerated chemistry modeling of protoplanetary disks

Abstract

Aims. With the large amount of molecular emission data from (sub)millimeter observatories and incoming James Webb Space Telescope infrared spectroscopy, access to fast forward models of the chemical composition of protoplanetary disks is of paramount importance. Methods. We used a thermo-chemical modeling code to generate a diverse population of protoplanetary disk models. We trained a K-nearest neighbors (KNN) regressor to instantly predict the chemistry of other disk models. Results. We show that it is possible to accurately reproduce chemistry using just a small subset of physical conditions, thanks to correlations between the local physical conditions in adopted protoplanetary disk models. We discuss the uncertainties and limitations of this method. Conclusions. The proposed method can be used for Bayesian fitting of the line emission data to retrieve disk properties from observations. We present a pipeline for reproducing the same approach on other disk chemical model sets.

arXiv comment11 pages, 5 figures

References in corpus (40)

• Machine learning and the physical sciences
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• ALMA Survey of Lupus Protoplanetary Disks I: Dust and Gas Masses
• The Spitzer ice legacy: Ice evolution from cores to protostars
• A three-phase chemical model of hot cores: the formation of glycine
• Radiation thermo-chemical models of protoplanetary disks I. Hydrostatic disk structure and inner rim
• Gas and grain chemical composition in cold cores as predicted by the Nautilus 3-phase model
• A Parametric Modeling Approach to Measuring the Gas Masses of Circumstellar Disks
• KROME - a package to embed chemistry in astrophysical simulations
• Warm Molecular Layers in Protoplanetary Disks
• The Ophiuchus DIsc Survey Employing ALMA (ODISEA) - I : project description and continuum images at 28 au resolution
• Chemistry in Disks. IV. Benchmarking gas-grain chemical models with surface reactions
• Chemical evolution of turbulent protoplanetary disks and the Solar nebula
• Cosmic-ray ionisation in circumstellar discs
• Setting the volatile composition of (exo)planet-building material. Does chemical evolution in disk midplanes matter?
• UCLCHEM: A Gas-Grain Chemical Code
• CO destruction in protoplanetary disk midplanes: inside versus outside the CO snow surface
• New extended deuterium fractionation model: assessment at dense ISM conditions and sensitivity analysis
• Chemo-dynamical deuterium fractionation in the early solar nebula: The origin of water on Earth and in asteroids and comets
• Protoplanetary Disk Structure With Grain Evolution: the ANDES Model
• A Unified Monte Carlo Treatment of Gas-Grain Chemistry for Large Reaction Networks. I. Testing Validity of Rate Equations in Molecular Clouds
• Chemical Modelling of Young Stellar Objects, I. Method and Benchmarks
• Molecules with ALMA at Planet-forming Scales (MAPS) VI: Distribution of the small organics HCN, C2H, and H2CO
• An ALMA Survey of H${}_2$CO in Protoplanetary Disks
• Molecules with ALMA at Planet-forming Scales (MAPS). IX. Distribution and Properties of the Large Organic Molecules HC${}_3$N, CH${}_3$CN, and $c$-C${}_3$H${}_2$
• Modeling protoplanetary disk SEDs with artificial neural networks: Revisiting the viscous disk model and updated disk masses
• Probing Planet Forming Zones with Rare CO Isotopologues
• Chemical Signatures of the FU Ori Outbursts
• Molecules with ALMA at Planet-forming Scales (MAPS) XIII: HCO${}^{+}$ and disk ionization structure
• Gas mass tracers in protoplanetary disks: CO is still the best
• ALMA chemical survey of disk-outflow sources in Taurus (ALMA-DOT): I. CO, CS, CN, and H2CO around DG Tau B
• Machine Learning of Interstellar Chemical Inventories
• Convolutional neural networks as an alternative to Bayesian retrievals
• Chemulator: Fast, accurate thermochemistry for dynamical models through emulation
• Predicting binding energies of astrochemically relevant molecules via machine learning
• Measuring elemental abundance ratios in protoplanetary disks at millimeter wavelengths
• An ALMA Survey of Chemistry in Disks around M4-M5 Stars
• Using HCO${}^{+}$ isotopologues as tracers of gas depletion in protoplanetary disk gaps
• History-independent tracers: Forgetful molecular probes of the physical conditions of the dense interstellar medium
• UCLCHEMCMC: A MCMC Inference tool for Physical Parameters of Molecular Clouds

Cited by in corpus (6)

• Neural network-based emulation of interstellar medium models
• Quantifying the informativity of emission lines to infer physical conditions in giant molecular clouds. I. Application to model predictions
• Mapping Synthetic Observations to Prestellar Core Models: An Interpretable Machine Learning Approach
• Using a neural network approach to accelerate disequilibrium chemistry calculations in exoplanet atmospheres
• Beetroots: spatially-regularized Bayesian inference of physical parameter maps -- Application to Orion
• Protoplanetary Disk Chemistry