Mapping Synthetic Observations to Prestellar Core Models: An Interpretable Machine Learning Approach
arXiv:2502.07874 · doi:10.1051/0004-6361/202453266
Abstract
Observations of molecular lines are a key tool to determine the main physical properties of prestellar cores. However, not all the information is retained in the observational process or easily interpretable, especially when a larger number of physical properties and spectral features are involved. We present a methodology to link the information in the synthetic spectra with the actual information in the simulated models (i.e., their physical properties), in particular, to determine where the information resides in the spectra. We employ a 1D gravitational collapse model with advanced thermochemistry, from which we generate synthetic spectra. We then use neural network emulations and the SHapley Additive exPlanations (SHAP), a machine learning technique, to connect the models' properties to the specific spectral features. Thanks to interpretable machine learning, we find several correlations between synthetic lines and some of the key model parameters, such as the cosmic-ray ionization radial profile, the central density, or the abundance of various species, suggesting that most of the information is retained in the observational process. Our procedure can be generalized to similar scenarios to quantify the amount of information lost in the real observations. We also point out the limitations for future applicability.
Accepted A&A
References in corpus (76)
- The photodissociation and chemistry of CO isotopologues: applications to interstellar clouds and circumstellar disks
- Formation of Primordial Stars in a LCDM Universe
- Rotational quenching of CO due to H collisions
- Modelling and interpreting spectral energy distributions of galaxies with BEAGLE
- The 2014 KIDA network for interstellar chemistry
- Investigating the Cosmic-Ray Ionization Rate in the Galactic Diffuse Interstellar Medium through Observations of H3+
- Observing the gas temperature drop in the high-density nucleus of L 1544
- KROME - a package to embed chemistry in astrophysical simulations
- The anatomy of the Orion B Giant Molecular Cloud: A local template for studies of nearby galaxies
- Cosmic-ray ionisation in circumstellar discs
- Bayesian analysis of cosmic-ray propagation: evidence against homogeneous diffusion
- VLBA determination of the distance to nearby star-forming regions II. Hubble 4 and HDE 283572 in Taurus
- Insights into the content and spatial distribution of dust from the integrated spectral properties of galaxies
- Metal and molecule cooling in simulations of structure formation
- Metal and molecule cooling in simulations of structure formation
- Structure and kinematics of the Taurus star-forming region from Gaia-DR2 and VLBI astrometry
- The Different Structures of the Two Classes of Starless Cores
- A New Reference Chemical Composition for TMC-1
- A Nearby Galaxy Perspective on Dust Evolution. Scaling relations and constraints on the dust build-up in galaxies with the DustPedia and DGS samples
- 2MASS wide field extinction maps: II. The Ophiuchus and the Lupus cloud complexe
- Nitrogen chemistry and depletion in starless cores
- The dynamics of collapsing cores and star formation
- A Simple and Accurate Network for Hydrogen and Carbon Chemistry in the ISM
- Low-Metallicity Star Formation : Prestellar Collapse and Protostellar Accretion in the Spherical Symmetry
- Structure of photodissociation fronts in star-forming regions revealed by observations of high-J CO emission lines with Herschel
- Cosmic-ray and X-ray Heating of Interstellar Clouds and Protoplanetary Disks
- pyspeckit: A spectroscopic analysis and plotting package
- The Distribution of Ortho-H_2D+(1_1,0 - 1_1,1) in L1544: Tracing the Deuteration Factory in Prestellar Cores
- On the Three-Dimensional Structure of Local Molecular Clouds
- Non-equilibrium chemistry and cooling in the diffuse interstellar medium - II. Shielded gas
- The observed chemical structure of L1544
- Simulations of protostellar collapse using multigroup radiation hydrodynamics. II. The second collapse
- The Deuterium Fractionation Timescale in Dense Cloud Cores: A Parameter Space Exploration
- High sensitivity maps of molecular ions in L1544: I. Deuteration of N2H+ and HCO+ and first evidence of N2D+ depletion
- Rate of three-body recombination of hydrogen molecules during primordial star formation
- ROHSA: Regularized Optimization for Hyper-Spectral Analysis - Application to phase separation of 21 cm data
- Electron-impact rotational and hyperfine excitation of HCN, HNC, DCN and DNC
- Rate Coefficient for H^+ + H_2(X ^1Sigma_g^+, nu=0, J=0) --> H(1s) + H_2^+ Charge Transfer and Some Cosmological Implications
- Simulating the formation of massive seed black holes in the early Universe. II: Impact of rate coefficient uncertainties
- Deuteration of ammonia in the starless core Ophiuchus/H-MM1
- A Dust spectral energy distribution model with hierarchical Bayesian inference. I. Formalism & benchmarking
- Simulations of protostellar collapse using multigroup radiation hydrodynamics. I. The first collapse
- PDR Model Mapping of Physical Conditions via Spitzer/IRS Spectroscopy of H2: Theoretical Success toward NGC 2023-South
- Oxygen fractionation in dense molecular clouds
- Towards an astronomical foundation model for stars with a Transformer-based model
- Tracing Interstellar Heating: An ALCHEMI Measurement of the HCN Isomers in NGC 253
- A grid of 1D low-mass star formation collapse models
- Constraining physical conditions for the PDR of Trumpler 14 in the Carina Nebula
- A detailed framework to incorporate dust in hydrodynamical simulations
- Hydrodynamics with gas-grain chemistry and radiative transfer: comparing dynamical and static models
- The core population and kinematics of a massive clump at early stages: an ALMA view
- Kinematics of dense gas in the L1495 filament
- Reducing the complexity of chemical networks via interpretable autoencoders
- An Interferometric View of H-MM1. I. Direct Observation of NH3 Depletion
- Inferring the HII region escape fraction of ionizing photons from infrared emission lines in metal-poor star-forming dwarf galaxies
- A new proxy to estimate the cosmic-ray ionisation rate in dense cores
- Reaction Studies of Neutral Atomic with using a Merged-Beams Apparatus
- Photoevaporation of protoplanetary discs with PLUTO+PRIZMO I. Lower X-ray-driven mass-loss rates due to enhanced cooling
- Chemulator: Fast, accurate thermochemistry for dynamical models through emulation
- LOC program for line radiative transfer
- The 3D structure of CO depletion in high-mass prestellar regions
- Multi-line observations of CHOH, c-CH and HNCO towards L1544: Dissecting the core structure with chemical differentiation
- ROBO: a Model and a Code for the Study of the Interstellar Medium
- Quantitative inference of the column densities from 3 mm molecular emission: A case study towards Orion B
- A 3D physico-chemical model of a pre-stellar core. I. Environmental and structural impact on the distribution of CHOH and -CH
- Bayesian Inference of Reaction Rates in Icy Mantles
- Chemistry and dynamics of the prestellar core L1544
- Chemistry as a diagnostic of prestellar core geometry
- Chemical post-processing of magneto-hydrodynamical simulations of star-forming regions: robustness and pitfalls
- Understanding the Formation and Evolution of Interstellar Ices: A Bayesian Approach
- Machine learning-accelerated chemistry modeling of protoplanetary disks
- Mapping the HD and NH emission towards prestellar cores. Testing dynamical models of the collapse using gas tracers
- The evolution of HCO in molecular clouds using a novel chemical post-processing algorithm
- Neural network-based emulation of interstellar medium models
- Do simulated molecular clouds look like real ones?
- Quantifying the informativity of emission lines to infer physical conditions in giant molecular clouds. I. Application to model predictions