Application of Convolutional Neural Networks to Identify Stellar Feedback Bubbles in CO Emission

Abstract

We adopt the deep learning method CASI (Convolutional Approach to Shell Identification) and extend it to 3D (CASI-3D) to identify signatures of stellar feedback in molecular line spectra, such as 13CO. We adopt magneto-hydrodynamics simulations that study the impact of stellar winds in a turbulent molecular cloud as an input to generate synthetic observations. We apply the 3D radiation transfer code radmc-3d to model 13CO (J=1-0) line emission from the simulated clouds. We train two CASI-3d models: ME1 predicts only the position of feedback, while MF predicts the fraction of the mass coming from feedback in each voxel. We adopt 75% of the synthetic observations as the training set and assess the accuracy of the two models with the remaining data. We demonstrate that model ME1 identifies bubbles in simulated data with 95% accuracy, and model MF predicts the bubble mass within 4% of the true value. We use bubbles previously visually identified in Taurus in 13CO to validate the models and show both perform well on the highest confidence bubbles. We apply our two models on the full 98 deg2 FCRAO 13CO survey of the Taurus cloud. Models ME1 and MF predict feedback gas mass of 2894 M and 302 M, respectively. When including a correction factor for missing energy due to the limited velocity range of the 13CO data cube, model ME1 predicts feedback kinetic energies of 4.0*1e46 ergs and 1.5*1e47 ergs with/without subtracting the cloud velocity gradient. Model MF predicts feedback kinetic energy of 9.6*1e45 ergs and 2.8*1e46 ergs with/without subtracting the cloud velocity gradient. Model ME1 predicts bubble locations and properties consistent with previous visually identified bubbles. However, model MF demonstrates that feedback properties computed based on visual identifications are significantly overestimated due to line of sight confusion and contamination from background and foreground gas.

arXiv commentApJ Accepted

References in corpus (32)

• Mass Loss: Its Effect on the Evolution and Fate of High-Mass Stars
• Hi-GAL: the Herschel infrared Galactic Plane Survey
• The Energy Dissipation Rate of Supersonic, Magnetohydrodynamic Turbulence in Molecular Clouds
• The Taurus Spitzer Survey: New Candidate Taurus Members Selected Using Sensitive Mid-Infrared Photometry
• The COMPLETE Survey of Star-Forming Regions: Phase I Data
• The relation between gas and dust in the Taurus Molecular Cloud
• The Milky Way Project First Data Release: A Bubblier Galactic Disk
• The GALFA-HI Survey: Data Release 1
• Magnetically Regulated Star Formation in 3D: The Case of Taurus Molecular Cloud Complex
• The COMPLETE Survey of Outflows in Perseus
• The Five College Radio Astronomy Observatory CO Mapping Survey of the Taurus Molecular Cloud
• A Bubbling Nearby Molecular Cloud: COMPLETE Shells in Perseus
• Impact of Protostellar Outflows on Turbulence and Star Formation Efficiency in Magnetized Dense Cores
• The Greater Taurus-Auriga Ecosystem I: There Is A Distributed Older Population
• Turbulence driven by outflow-blown cavities in the molecular cloud of NGC 1333
• A Deep Learning Approach to Galaxy Cluster X-ray Masses
• New Young Star Candidates in the Taurus-Auriga Region as Selected From WISE
• Evidence For Cloud-Cloud Collision and Parsec-Scale Stellar Feedback Within the L1641-N Region
• A Stable, Accurate Methodology for High Mach Number, Strong Magnetic Field MHD Turbulence with Adaptive Mesh Refinement: Resolution and Refinement Studies
• The Milky Way Project: Leveraging Citizen Science and Machine Learning to Detect Interstellar Bubbles
• Outflows and Bubbles in Taurus: Star-formation Feedback Sufficient to Maintain Turbulence
• Impact of Winds from Intermediate-Mass Stars on Molecular Cloud Structure and Turbulence
• The Milky Way Project Second Data Release: Bubbles and Bow Shocks
• Molecular Outflows Identified in the FCRAO CO Survey of the Taurus Molecular Cloud
• TurbuStat: Turbulence Statistics in Python
• CASI: A Convolutional Neural Network Approach for Shell Identification
• Classifying structures in the ISM with Support Vector Machines: the G16.05-0.57 supernova remnant
• Identifying Tools for Comparing Simulations and Observations of Spectral-line Data Cubes
• A New Method for Constraining Molecular Cloud Thickness: A study of Taurus, Perseus and Ophiuchus
• Assessing the Performance of a Machine Learning Algorithm in Identifying Bubbles in Dust Emission
• An Exploration of the Statistical Signatures of Stellar Feedback
• The CARMA-NRO Orion Survey: Statistical Signatures of Feedback in the Orion A Molecular Cloud

Cited by in corpus (14)

• Turbulence, Coherence and Collapse: Three Phases for Core Evolution
• Quantifying the energy balance between the turbulent ionised gas and young stars
• Denoising Diffusion Probabilistic Models to Predict the Density of Molecular Clouds
• Application of Convolutional Neural Networks to Identify Protostellar Outflows in CO Emission
• Tracking down the origin of superbubbles and supergiant shells in the Magellanic Clouds with Minkowski tensor analysis
• PGNets: Planet mass prediction using convolutional neural networks for radio continuum observations of protoplanetary disks
• A Census of Outflow to Magnetic Field Orientations in Nearby Molecular Clouds
• A Census of Protostellar Outflows in Nearby Molecular Clouds
• Application of Convolutional Neural Networks to Predict Magnetic Fields Directions in Turbulent Clouds
• Effects of stellar feedback on cores in STARFORGE
• Probing the Physics of Star-Formation (ProPStar) III. No evidence for dissipation of turbulence down to 20 mpc (4 000 au) scale
• Modeling turbulent and self-gravitating fluids with Fourier neural operators
• Diagnosing Turbulence in the Neutral and Molecular Interstellar Medium of Galaxies
• Investigations of MWISP Bubbles: Identification and Analysis of Enclosed Molecular Bubbles by Weight Fields