Machine learning-accelerated chemistry modeling of protoplanetary disks
arXiv:2209.13336 · doi:10.1051/0004-6361/202244691
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.
11 pages, 5 figures
References in corpus (12)
- Rotation-invariant convolutional neural networks for galaxy morphology prediction
- Setting the volatile composition of (exo)planet-building material. Does chemical evolution in disk midplanes matter?
- UCLCHEM: A Gas-Grain Chemical Code
- Chemical Modelling of Young Stellar Objects, I. Method and Benchmarks
- An ALMA Survey of HCO in Protoplanetary Disks
- Gas mass tracers in protoplanetary disks: CO is still the best
- Machine Learning of Interstellar Chemical Inventories
- 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
- 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